Financing and Delivery of Health Services NCMCH

Background Papers of the National Commission on Macroeconomics and Health
Financing and Delivery of Health Care Services in India Background Papers of the National Commission on Macroeconomics and Health
Background Papers
Financing and Delivery of Health Care Services in India
National Commission on Macroeconomics and Health
MINISTRY OF HEALTH AND FAMILY WELFARE GOVERNMENT OF INDIA, 2005
EQUITABLE DEVELOPMENT
HEALTHY FUTURE
NCMH Background Papers
National Commission on Macroeconomics and Health Ministry of Health & Family Welfare Government of India, New Delhi August 2005
Ministry of Health & Family Welfare, Government of India

September 2005 ISBN 81-7525-632-8
This Report does not address tertiary care and related areas such as super speciality hospital development in the public or private sector, telemedicine, medical tourism, environmental pollution or food safety etc. though they are all equally important. The Commission Report is based on background papers which can be accessed from the NCMH website www.mohfw.nic.in. They have also been published in two companion volumes. This report was written during the period April 1, 2004 - March 31, 2005.
Printed at: Cirrus Graphics Private Limited B 261, Phase I, Naraina Industrial Area, New Delhi 110 028 Tel: + 91 11 51411507/1508 Fax: +91 11 51417575 email: cirrusgraphics@touchtelindia.net Editors: Pranay G. Lal and Byword Editorial Consultants
Cover design: Quote Design Studio
ii
Preface
IN PURSUANCE OF THE RECOMMENDATIONS MADE BY THE COMMISSION ON MACROECONOMICS AND Health, WHO, India established the National Commission on Macroeconomics and Health (NCMH) in March, 2004. The main objective of the NCMH was to establish the centrality of health to development and make an evidence-based argument to increase investment in health. The Terms of Reference of the NMCH were mainly centered on identifying a package of essential health interventions that ought to be made available to all citizens and also list systemic constraints that need to be addressed for ensuring universal access to this package of services. The NCMH was also to indicate the resources required and targets that ought to be achieved by 2015. The Terms of Reference of the NCMH were very widespread and spanned across a wide range of issues.For addressing each of the major concerns a broad outline of the approach to be adopted was prepared and shared with a large number of researchers, policy makers, experts from donor agencies and health activists. Based on the suggestions received, topics to be addressed were identified and studies / papers commissioned. Every paper was also peer reviewed by experts in that field. In all over 35 papers were commissioned. Due to limitations on time and resources, original field surveys were limited to a hundred percent facility survey in eight districts of Khammam(AP), Ujjain(MP), Varanasi(UP),Udaipur(Rajasthan), Kozhikode(Kerala), Jalna(Maharashtra), Nadia ( West Bengal) and Vaishali (Bihar). For arriving at the estimates of public spending, we obtained information from other government departments, PSU's, FII's etc. and analyzed the data under the National Health Accounts Framework. Analysis of consumer surveys, the 57th. Round Survey National Sample Survey Organization on establishments, and other data bases related to drug manufacture and sales, import and export of medical devices etc. were also analyzed. Principal focus was on critically evaluating the current status of the health system - its organizational structure, financing mechanisms, regulatory frameworks etc. The three key drivers of health costs - namely human resources, drugs and technology were specially studied in detail as the main concern for the future is going to be the rapid escalation of costs. Such analysis highlighted and reiterated several shortcomings in the country's health system which are well known and have been recognized for long. Clearly, a well conceived and sequenced system of reform emerged to be the priority area for policy attention so as to develop the capacity to absorb the promised funding of 2-3% of GDP in the next five years committed in the Common Minimum Program. What also emerged were that solutions for many of the issues have been known for long, but routinely ignored and not acted upon. It was impossible not to conclude that if only timely attention to the large number of recommendations already available had been accorded, the health system need not have been so inefficient, insensitive, dysfunctional and in such a crises as we find it today.
iii
The background papers formed the basis for the main report of the Commission and its recommendations. We have attempted to bring into the public domain all the data and analysis that has been carried out by the NCMH, both in printed form ( 2 volumes) as well as in the website of the NCMH - www.ncmh.org. The main purpose has been to stimulate greater debate and research that would be useful for policy formulation. If this has been achieved even in a small measure, we would be content that our efforts have been worthwhile. I wish to thank my colleagues at the Sub-Commission - Dr. Ajay Mahal, Dr. Avtar Dua, Dr. Sakthivel, Dr. Somil Nagpal, Ms. Madhurima Nundy and Shri Sunil Nandraj and Dr. Rama Baru for their help and assistance. I also thank all the contributors and reviewers for taking time off to write the paper or review it and helping us in every possible way, very often at short notice. And finally a special thanks to Dr. Ranjit RoyChaudhury , member of the NCMH and chair of the sub-commission for his constant support, encouragement and advise. I am grateful to each and every one of them.
Sujatha Rao Secretary, NCMH
iv
List of Contributors and Reviewers

Authors
AJAY MAHAL Asstt. Prof. Harvard School of Public Health Boston, U.S.A. ANIL VARSHNEY Consultant 90/2, Malaviya Nagar, Opp. Govt. Senior Secondary School, New Delhi ANUP K. KARAN Fellow Institute for Human Development IAMR Building, 3rd Floor, I.P. Estate, New Delhi ASHOK D.B. VAIDYA Medical and Research Director Bhartiya Vidya Bhavan's SPARC, 13th N.S. Road, J.V.P.D. Scheme, Juhu, Mumbai 400049 AVTAR SINGH DUA Asstt. Prof., Deptt. of PSM SMS Medical College, Jaipur MS. CONSUELO ESPINOSA MARTY Senior Health Economist and Advisor, Health Care Reforms Ministry of Finance, Chile DHIRENDRA KUMAR Associate Professor Indian Institute of Health Management & Research 1, Prabhu Dayal Marg, Sanganer Airport, Jaipur 302011 K. SUJATHA RAO Principal Secretary, Government of Andhra Pradesh, Hyderabad Andhra Pradesh LALIT MOHAN NATH Former Dean (AIIMS) E-21, Defence Colony New Delhi 110003 M. GOVINDA RAO Director National Institute of Public Finance & Policy, 18/2, Satsang Vihar Marg, Special Institutional Area, Near JNU, New Delhi 110067 MADHURIMA NUNDY Research Scholar, Centre for Social Medicine in Community Health, School of Social Sciences, JNU New Delhi MARCELO TOKMAN Director, Economic Policy Ministry of Finance Chile MITA CHOUDHARY Economist National Institute of Public Finance & Policy, 18/2, Satsang Vihar Marg, Special Institutional Area, Near JNU, New Delhi 110067 MUKESH ANAND Senior Economist National Institute of Public Finance & Policy, 18/2, Satsang Vihar Marg, Special Institutional Area, Near JNU, New Delhi 110067 N. RAVICHANDRAN Assistant Professor Indian Institute of Health Management & Research 1, Prabhu Dayal Marg, Sanganer Airport, Jaipur 302011 N. VEERABHRAIAH Andhra Pradesh Vaidya Vidhan Parishad Department of Health Govt. of Andhra Pradesh, Hyderabad P. DURAISAMY Professor Department of Econometrics University of Madras Chepauk, Chennai - 600005
SHIV CHANDRA MATHUR Director State Institute of Health & Family Welfare Jhalana Institutional Area, South of DD Kendra Jaipur 302004 S.D. GUPTA Director Indian Institute of Health Management & Research 1, Prabhu Dayal Marg, Sanganer Airport, Jaipur 302011 S. SAKTHIVEL Research Associate, Institute of Economic Growth Delhi University Enclave, Delhi 110007
SOMIL NAGPAL WHO Consultant, TB Division New Delhi S. SELVARAJU Consultant, BD-3 G, DDA Flats, Munirka, New Delhi T. DILEEP KUMAR Advisor (Nursing), Dte.GHS and President, Indian Nursing Council Nirman Bhavan, New Delhi
vi
Reviewers
ALAKA SINGH World Health Organisation, Geneva, Switzerland ANURAG BHARGAVA Consultant Jan Swasthya Sahyog, Village & Post: Ganiyar, District Bilaspur 495112 Madhya Pradesh BARUN KANJILAL Dean Indian Institute of Health Management & Research, 1 Prabhu Dayal Marg, Sanganer Airport, Jaipur C.H.S. SASTRY Director(Retd.), National Institute of Ayurveda, Jaipur 3-599/4, Congress Office Road, Near Ayappa Temple, Undavalli, Tidapalli (Mandal) Distt. Guntur, Andhra Pradesh CHARU GARG World Health Organisation Geneva, Switzerland D. NARAYANA Professor, Department of Economics, Centre for Development Studies, Thiruvananthapuram DARSHAN SHANKAR Director Foundation for Revitalisation of local Health Traditions 74/2, Jarakabanda Kaval P.O. Attur, Via Velahanka Bangalore- 560064 DINESH AGARWAL Technical Advisor, UNFPA, 53, Jor Bagh New Delhi D.K. SRINIVAS Rajiv Gandhi University of Health Sciences, 4-T Block, Jayanagar, Bangalore (Karnataka) GANGA MURTHY Additional Economic Advisor Ministry of Health & Family Welfare Nirman Bhavan, New Delhi GIRISH CHATURVEDI Joint Secretary (Insurance) Ministry of Finance Jeevandeep Building, Parliament Street, New Delhi GIRISH N. RAO Managing Director TTK Health Care Services Pvt. Ltd., #7, Jeevan Bhima Nagar, Main Road HAL III Stage, Bangalore-560075 G.P. DUBEY Professor Department of Biofeedback, Institute of Medical Sciences, Banaras Hindu University, Varanasi, Uttar Pradesh INDRANI GUPTA Institute of Economic Growth University Enclave, Delhi - 110 007 J.V. MEENAKSHI IFPRI, Washington JAYAPRAKASH MULIYIL Principal Christian Medical College, Vellore K.S. RAGHAVAN Consultant 102, Jyothi Manor, Plot No.41, Srinagar Colony, Hyderabad 500073 M.S. VALIATHAN Honorary Advisor, Manipal Academy of Higher Education, Manipal 576104 MIRA SHIVA Senior Consultant Voluntary Health Association of India B-40, Qutub Institutional Area New Delhi- 110 016 N.K. SETHI Director National Institute of Health & Family Welfare, Munirka, New Delhi
VII
NARENDRA BHATT Vice President Indian Association for the Study of Traditional Asian Medicine 15 - Bachubhai Bldg. J. Bhatnagar Marg, Parel Mumbai- 400 012 PRAKIM SUCHAXAYA Faculty of Nursing Chiang Mai University, Chiang Mai, Thailand RAVI NARAYAN Global Secretariat C/o Community Health Cell No.359 (Old No.367) Srinivas Nilaya, Jakkssadlu, First Main, 1st Block, Kormangala, Bangalore 560002 RAMESHWAR SHARMA Constultant B-32, Vijay Path Tilak Nagar, Jaipur-302004 R.D. BANSAL Consultant Kothi No.3059 Sector 19 D Chandigarh-19 RAVI DUGGAL Coordinator, CEHAT Aram Society Road, Vakola, Santacruz(E) Mumbai -400055 R.L. MISHRA Former Secretary Health No.4403, Qutub Enclave, DLF Phase IV, Gurgaon 122002 RAMESH BHAT Prof. of Finance Indian Institute of Management Vastrapur, Ahmedabad- 380 015
RAMA BARU Centre for Social Medicine & Community Health School of Social Sciences, Jawaharlal Nehru University, New Delhi 1100067 SEETA PRABHU United Nations Development Programme Lodhi Estate New Delhi SRINIVASAN R Former Secretary (Health) D-402, Kaveri Apartments, Alaknanda, New Delhi S. SRINIVASAN LOCOST, 1st Floor, Premanand Sahitya Sabha Hall, Opp. Lakadi Pool, Dandiya Bazar, Baroda 390001 SUNIL NANDRAJ National Professional Officer, Health Systems Developments, WHO, Nirman Bhavan, New Delhi V.N. PANDIT Sri Sathyasai Institute for Higher Learning, Prasantinilayam, Distt. Ananthapur, Andhra Pradesh 515134 VAIDYANATHAN A. Madras Institute of Development Studies 79, Second Main Road, Gandhinagar, Adyar, Chennai 600020 WILAWAM SEMARATAMA Assistant Professor Chiang Mai University Chiang Mai, Thaniland
VIII
Contents
Preface List of Contributors and Reviewers SECTION I: Health, Poverty and Economic Growth in India Health, Poverty and Economic Growth in India Health, nutrition and poverty: Linking nutrition to consumer expenditures SECTION II: Delivery of Health Care Services in India Primary Health Care in India: Review of Policy, Plan and Committee Reports Delivery of health services in the public sector Training for effective delivery of health services Effective Integration of Indian Systems of Medicine in Health Care Delivery: People's Participation, Access and Choice in a Pluralistic Democracy Delivery of health services in the private sector The not-for-profit sector in medical care Peoples Partnership for Health Towards a Healthy Public in India SECTION III: Drivers of Health Care Costs Human Resources for Health Nursing for the delivery of essential health interventions Access to Essential Drugs and Medicine Appropriate Policies for Medical Device Technology: The Case of India Annexure 1: Medical equipment use pattern in the public and private sectors in India: Policy implications SECTION IV: Financing of Health Care in India Financing of Health in India Annexure 1: National Health Accounts for India User charges in Indias health sector: An assessment Health insurance in India Resource Devolution from the Centre to States: Enhancing the Revenue Capacity of States for Implementation of Essential Health Interventions
iii v 1 3 19 37 39 43 65 77 89 125 135 151 153 175 185 213 226
237 239 256 265 275 297
vii
SECTION I Health, Poverty and Economic Growth in India
SECTION I
Health, Poverty and Economic Growth in India
P. DURAISAMY
DEPARTMENT OF ECONOMETRICS UNIVERSITY OF MADRAS CHENNAI 600005, INDIA E-MAIL: pduraisamy@hotmail.com
AJAY MAHAL
HARVARD SCHOOL OF PUBLIC HEALTH DEPARTMENT OF POPULATION AND INTERNATIONAL HEALTH BOSTON MA 02115, USA E-MAIL: amahal@hsph.harvard.edu
HE IMPORTANCE OF ECONOMIC GROWTH, MEASURED BY INCREASES IN GROSS domestic product (GDP) and GDP per capita, for policy purposes can hardly be overemphasized. Economic growth is commonly used as an indicator of a nations economic performance, and the level of GDP per capita is a key component of the Human Development Index of the United Nations Development Programme, a popular indicator of national well-being. The benefits of economic growth are so pervasive that it has been a central agenda everywhere and countries have accorded top priority to achieving high rates of growth. Some experts and policy-makers have also argued that it is difficult to achieve declines in poverty rates by relying on redistribution strategies alone, without a concomitant improvement in size of the national economic cake, as reflected in the magnitude of real GDP and real GDP per capita. It is difficult to imagine a sustained decline in poverty unaccompanied by a simultaneous improvement in aggregate economic performance. There is now a large body of theoretical and empirical research on the determinants of economic growth. Much of the early work highlighted growth in labour and the stock of physical capital as the key determinants of economic growth. However, early empirical work was unable to explain a significant portion of the growth in GDP and GDP per capita, by the growth in labour force and capital alone, and so attention turned to other factors-most notably technological change embodied in capital goods, and on the quality and quantity of labour, referred to as human capital, in promoting economic growth. Two key elements of human capital are the extent to which the labour force is educated, and the level of its health. Recent empirical work has sought to assess the association between human capital and aggregate economic performance and found that, given labour and capital, improvement in health status and education of the population lead to a higher output (Barro and Sala-iMartin 2004). The role of health in influencing economic outcomes has been well understood at the micro level. Healthier workers are likely to be able to work longer, be generally more productive than their relatively less healthy counterparts, and consequently able to secure higher earnings than the latter, all else being the same; illness and disease shorten the working lives of people, thereby reducing their lifetime earnings. Better health also has a positive effect on the learning abilities of children, and leads to better educational outcomes (school completion rates, higher mean years of schooling, achievements) and increases the efficiency of human capital formation by individuals and households (Strauss and Thomas 1998; Schultz 1999). However, more recent research has also established a strong causal association running from health to aggregate economic performance. Thus Bloom, Canning and Sevilla (2004) report evidence from more than a dozen cross-country studies and all these studies, with a single exception, show that health has a positive and statistically significant effect on the rate of growth of GDP per capita. The causal relationship does not run in only one direction-from health to aggregate economic performance-and there is strong case for considering a reverse link, running from wealth to health. Higher incomes potentially permit individuals (and societies) to afford better nutrition, better health care and, presumably, achieve better health. There is some cross-country evidence that such a relationship holds at the national level (Pritchett and Summers 1996; Bhargava et al. 2001). Several experts believe, however, that the causal direction from health to economic performance is stronger. The previous empirical findings have implications for the role of health improveFinancing and Delivery of Health Care Services in India
SECTION I
ments among workers in influencing another key policy objective-poverty reduction. First, to the extent that improvements in health result in improvements in national income, poverty could decline on account of both the standard trickle-down effects and an increased financial capacity of nations to set up safety nets. There is a good deal of evidence suggesting that countries that experience a steep rise in growth rates of real GDP per capita also experience impressive declines in poverty (Barro and Sala-i-Martin 2004). Second, improvements in health, when directed at the poor, can contribute more directly to poverty reduction and serve as an element of a pro-poor growth strategy. The poor bear a disproportionately higher burden of illness, injury and disease than the rich. The poor suffer ill health due to a variety of causes, poor nutrition for instance, which reduces the ability to work and weakens their resistance to disease. With their body often being their main income-earning asset, sickness and disability have significant adverse implications in terms of loss of work and incomes, compounded by their inability to obtain adequate health care. Frequently, treatment expenditure and loss of earnings force poor families to exhaust their savings and assets, and take recourse to borrowing, leading to more poverty and poor health status. This paper contributes to existing analyses of the healthpoverty-income nexus by examining these relationships at the State level in India, using the most recent empirical methods available in the literature (Bloom et al. 2004). Our analysis is carried out using a cross-State panel dataset for 14 major Indian States for the years 1970/71, 1980/81, 1990/91 and 2000/01, spanning a thirty-year period. The analysis of this paper is important for several reasons. First, there is no denying the policy significance of understanding the determinants of economic growth and its relationship with poverty and improvements in health. If health turns out to have significantly influenced Indias economic performance, this may call for investing more public funds in health, given that health budgets have been severely resource-constrained in recent years. One way this could happen is by greater emphasis on the commitments India has made to meet the targets set by the Millennium Declaration. These targets include significant improvement in health through reduction in infant and child mortality by two-thirds by 2015 (World Bank 2004). Conversely, this also calls for understanding better the impact of economic growth on health, so that one can assess the improvements in economic performance necessary to achieve the desired goals. Second, unlike the existing literature which relies on crossnational data, our paper examines the interlinkages between health and economic performance within a single country. Intracountry analysis has the advantage of being much better equipped to handle data-comparability issues relating to health, education and economic performance. At the same time, the significant variation in inter-State performance in health and economic achievement means that our estimation procedures yield estimates that are reasonably robust. Third, significant inter-State differences in Indias economic performance call for enhanced efforts in understanding
these differentials. Unfortunately, there have been only a few recent attempts at examining the relative growth performances of the States in India (Ahluwalia 2001; Sachs et al. 2002), and most of the major studies do not emphasize the role of health in influencing economic performance. The only study that sought to do this in the Indian context was one by Gupta and Mitra (2003), which examined the link between growth, health and poverty in India. While useful, the chief drawback of this paper is that its empirical specification was essentially ad hoc, and not influenced by developments in the economic growth literature. As a consequence, there are legitimate concerns with their model specifications, including the criteria used for the inclusion (or exclusion) of explanatory variables. There are now newer and more powerful methods to assess the links between health, poverty and economic growth. For these reasons, we believe that the estimates reported in their paper are unlikely to be robust.
Economic growth and health: A review of cross-country and regional studies

Modern growth literature includes, in addition to the standard labour and capital variables, indicators of human capital-the stock of education and health-among the determinants. Particularly, the influential works in this area are the cross-country studies by Barro (1991, 1997) and Barro and Sala-i-Martin (2004) and the theoretical framework developed by Mankiw, Romer and Weil (1992). A comprehensive review of empirical evidence on the new macroeconomics of growth is contained in Temple (1999). Barro (1991) used a cross-sectional framework and the human capital variable was restricted to school enrolment rates at the primary and secondary levels. He showed, using cross-section data for 98 countries, that the growth rate of real GDP per capita over the period 1960-85 was positively related to the initial (1960) enrolment rate, and inversely related to the starting (1960) level of real per capita GDP. In subsequent analyses, Barro (1997) and Barro and Sala-i-Martin (2004) used a panel dataset of countries, and included health as a determinant (life expectancy at birth [LEB]) besides years of educational attainment and other factors that could potentially influence the growth of real income per capita. Their results indicate that the log of LEB has a positive and statistically significant effect on growth rate with a coefficient of 0.042, which implies an annual rate of increase of per capita real GDP of 4.2%. Fogel (1994) showed that about one-third of the increase in income in Britain during the nineteenth and twentieth centuries could be attributed to improvements in health and nutrition. Mayer (2001) concluded that improvements in adult survival were causally linked to improvements in growth performance in Brazil and Mexico; and Weil (2001) found that health (indicated by average height and LEB) explained about 17% of the variation in income per capita across countries. Gyimah-Brempong and Wilson (2004) find that 22% and 30% of the growth rate of per capita income in sub-Saharan Africa and OECD countries, respectively, can be attributed to health. Bloom, Canning and Sevilla (2004) review several studies
SECTION I
that include health as an explanatory variable in growth equations, in addition to presenting new results, based on a crossnational panel dataset for countries. They use a production function model of economic growth with a measure for human capital which takes account of the indicators of health, education and labour market experience. There are two noteworthy findings from their analysis. First, their analysis reconciles microeconomic analyses of the rate of return to schooling with macroeconomic analyses of returns to education. Second, they report a positive and statistically significant effect of health on economic growth. Their empirical findings reveal that an increase of one year in LEB raises the growth rate of GDP by 4%. Bhargava et al. (2001) found that the adult survival rate (ASR) has a positive effect on growth rate of per capita GDP and that a 1% increase in ASR increases the growth rate by 0.05% for the poorest countries. While there is compelling evidence that health contributes significantly to economic growth, there is also voluminous literature that focuses on causality in the reverse direction-from income to health. Much of this work is based on micro-level data that focus on the impact of income on the health status of households and their members (Behrman and Deolalikar 1988; Strauss and Thomas 1998). There has also been some recent work at the macro-level, using cross-national panel datasets; and much of the current work using cross-country time series data has tended to account for reverse causality and inter-dependence between health, income and economic growth. Thus, Pritchett and Summers (1996) estimate the effect of income on health, measured by infant and child mortality as well as life expectancy. Some authors have also inquired into the distributional aspects of the income-health relationship. For instance, Preston (1975) used cross-country evidence to suggest that the effect of income improvements on health was greater for the poorest countries than for the richest countries. Deaton (2001) argued that income inequality is not a major determinant of health of the population. How about the relationship between health, income and poverty? In a purely accounting sense, increases in real GDP per capita will be accompanied by simultaneous declines in the number of people living in poverty, provided the distribution of income remains more or less constant. Growth may be essential to reducing poverty and one might presume that policies promoting distributional improvements will prove difficult to sustain in the absence of long-term increases in real GDP per capita-that is, economic growth. Empirically, Barro and Sala-i-Martin (2004) demonstrated that regions of the world that experienced higher growth rates also witnessed steeper declines in poverty. Bourgoignon (2004) cites studies that provide evidence on the poverty-reducing impact of growth given that income distribution remains the same, and of increases in poverty with a worsening of income distribution. In India, poverty levels have declined the fastest over periods that experienced the highest growth rates, during the1990s (Ahluwalia 2001). According to Srinivasan (2003), there was no perceptible decline in poverty in India until growth accelerated in 1980s and hence a necessary condition for eradicating mass poverty is to accelerate average annual rate of
aggregate GDP growth to at least 8%-10% and sustain it at that level for a sufficiently long period. Bourguignon (2004) examines theoretically the interrelationship between growth, inequality and poverty, and shows that both growth and changes in inequality contribute to changes in poverty. However, the relative effects of these phenomena may be country-specific and depend on initial income level and inequality. It was noted earlier that health improvements contribute to income improvements or growth. With much evidence also pointing to the growth-poverty reduction nexus, better health can be seen as a factor that contributes to poverty reduction via some form of trickle-down mechanism. When health improvements are concentrated among people living close to, or below the poverty line, both a trickle-down mechanism and a redistributive one work to reduce poverty. Rough computations by the World Bank, using National Sample Survey (NSS) data, suggest that ill-health and associated economic losses cause as much as 22 lakh Indians, most living marginally above a poverty line standard of living, to temporarily fall below the poverty line each year, owing to a combination of income losses on account of being unable to work and declines in non-medical care consumption. The NSS for India for 1995-96 also reveal that when the poor fall sick, they are often unable to afford treatment, and even when they do decide to get treated, tend to sell off productive assets and rely on borrowing, all of which have the potential of decreasing their long-run earning capacity-and the capacity to take advantage of any trickle-down labour market advantages offered by a growing economy. There are several studies in India on health status and healthseeking behaviour. In an early attempt Kannan et al. (1991) analysed the linkages between health, development and socioeconomic factors in Kerala. Vaidyanathan (1995) examined the measurement issues related to nutritional and health status and the adequacy of currently available data for assessing nutrition-health status. A number of studies examined levels and changes in morbidity and health expenditure using the National Sample Survey 1986-87 health survey data (Visaria and Gumber 1994, Krishnan 1995, Duraisamy 1995, and Gumber 1997). Sundar (1995) studied the levels and changes in health status and health expenditure based on NCAER survey. These studies are mainly descriptive and refer to earlier periods. The relationship between income, health and productivity has been analysed at the household level based on microeconometric framework. Duraisamy (1998, 2001) found evidence of a strong negative effect of income or total consumption expenditure on morbidity and household assets emerged as an important determinant of child survival and preventive health care (Duraisamy and Duraisamy 1995). Deolalikar (1988) demonstrated that health was a significant determinant of labour productivity using farm level data. A study on health, wages and labour supply by Duraisamy and Sathiyavan (1998) revealed that a 10% increase in the body mass index of males and females increased their wage rate by 7% and 2% respectively and labour supply by 20% and 11% respectively.
SECTION I
At the macro level, very little is known on the association between income/economic growth and health (Gupta and Mitra 2003, World Bank 2004). Gupta and Mitra (2003) examined the relationship between health, poverty and economic growth in India for the years 1973/74, 1977/78, 1983, 1987/88, 1993/94, 1999/2000 based on data for 15 Indian States. Their econometric analysis showed that per capita public health expenditure positively influences health status, that poverty declines with better health, and that growth and health have a positive two-way relationship. Despite reporting what appear to be significant findings, this study suffers from certain methodological drawbacks as indicated earlier. Identification restrictions in the model specification appear to be arbitrary rather than based on economic theory, or empirical literature. Their empirical specification with growth of net State domestic product (NSDP) as the dependent variable uses NSDP (not per capita NSDP) in the base year as an explanatory variable, a procedure not used previously in the literature, and for which no justification is provided. The same specification omitted population as an explanatory variable, an assumption which appears not to be standard (Bloom and Freeman 1986; Bloom and Williamson 1998). Many of the estimated coefficients in their analysis turned out not to be statistically significant. For example, in the growth equation (growth of NSDP), poverty, infant mortality rate, initial NSDP and literacy are statistically not significant even at the 10% level of significance. In the same equation, the infrastructure (INF) variable has a significant negative effect on growth rate. In a World Bank (2004) study, the effects of per capita GDP, per capita health expenditure and female literacy on infant mortality were examined using State-level data for the period 1980-99 based on econometric framework. The results show that both per capita public spending on health and per capita GDP are inversely related to IMR, but they are not very robust to alternative specifications of the model. However this study does not examine the effect of per capita income on LEB, an alternative and perhaps better measure of health status of the population. The lack of consistent findings in the literature, and possibly specification problems in the early works, lend further justification to the empirical analysis that we pursue in this paper.
Database
To empirically examine the linkages between health, poverty and economic growth at the sub-national (State) level in India, we constructed a panel dataset of 14 States, including observations every ten years-1970/71, 1980/81, 1990/91 and 2000/01. This study is confined to the major Indian States for which consistent time series data are available. The States excluded from the study are: Jammu and Kashmir, Goa and Himachal Pradesh, eight north-eastern States, and seven Union Territories. In the year 2000, three of the States included in our sample, Bihar, Madhya Pradesh and Uttar Pradesh, were bifurcated. We have merged the data on the new States (Chattisgarh, Jharkhand and Uttaranchal) with their respective parent States and constructed a comparable series of all the
variables for the study period. The States included for the study account for 90% of Indias population and 83% of the countrys total land area at present. State-level income and per capita income are represented by the respective States NSDP and the per capita NSDP (PCNSDP). Data on the NSDP and PCNSDP are produced on a regular basis by the Central Statistical Organisation (CSO) of the Government of India. We obtained these data from publications of the EPW Research Foundation (2002a, 2003) and CSO (2004). The value of NSDP and PCNSDP in these is reported in current prices and this has been converted into constant price series using a GDP deflator. The poverty variable is the head count measure, i.e. the proportion of the population living below the poverty line. In India, the poverty line is defined as the minimum expenditure required for achieving a basic calorie requirement, plus comparable non-food consumption expenditures. The source of poverty data for this paper is the Planning Commission, which computed poverty levels from the National Sample Survey Organization (NSSO) consumer expenditure surveys using the expert group methodology. Poverty data are available for the years 1972/73, 1983, 1993/94 and 1999/2000, respectively, and for the purposes of our statistical analysis, are taken to correspond to the years 1971, 1981, 1991 and 2001. The health status of the population is captured through two indicators-LEB and the infant mortality rate (IMR). Data on these two health indicators were obtained from the Registrar General of India (1999) and updated for recent years using the Sample Registration System (SRS) Bulletin published by the Registrar General of India. Data for 1971 for Bihar and West Bengal were extrapolated using the time series data of the concerned States. LEB estimates for 1961 were taken from the estimates published by the Registrar General of India, which is based on the population census of that year. Apart from health, human capital is measured along two additional dimensions-average years of schooling and work experience. First, we computed years of schooling using census data on completed levels of education by age and sex of the population. The completed years of education for various levels are assumed to be as follows: literate below primary-4 years; primary-5 years; middle-8 years; secondary10 years; higher secondary/pre-university-12 years; technical and non-technical diploma-13 years; graduate and above16 years. The variable for average years of schooling is constructed from the census tables on completed levels of education by age and sex of the population given in the Social and Cultural Tables, Census of India, published by the Registrar General of India for various census years, weighted by its appropriate population share. Second, following Bloom, Canning and Sevilla (2004), the years of labour market experience is constructed using the age and gender distribution of workers provided in the General Economic Tables, Census of India, Registrar General of India for various years. The years of experience is defined as age minus years of schooling minus six, the age of entry into schools as used in the micro-studies in labour economics. The average work
SECTION I
experience is the weighted average of the age- and sex-group specific potential experience with the respective groups share of the total. The data on the number of workers for various census years were obtained from the General Economic Tables of the population census (Registrar General of India [various years]). Total workers include both main and marginal workers. The total population in the working age groups of 1559 years were also collected from the decennial population census for the respective years. As the age distribution of the population for 2001 was not available when this work was completed, projected instead of actual population by age groups was used. Physical capital is another key explanatory variable in analyses of economic growth. Unfortunately, data on gross capital formation or the level of investment at the State level comparable with the national-level data on physical capital from national accounts statistics are not available. Data on gross fixed capital formation (GFCF) is available only for a few States from 1993-94 onwards. However, data on the value of fixed capital for the industrial sector are available from the Annual Survey of Industries (ASI) published by the CSO and compiled and published by the EPW Research Foundation (2002b). These values were expressed in current prices and have been converted into a constant price series using the GDP deflator. Public expenditure on health is an important determinant of the health status of the population. State-level
government expenditures on health, water supply and sanitation, and family welfare were compiled from the RBI Bulletin for various years. We also constructed two variables to represent political power: (i) the percentage of votes gained by the ruling party at the Centre in the Assembly elections; and (ii) the percentage of votes secured by socialist and communist parties in the respective State Assembly elections. The data for these variables were gathered from the Election Commission. Using the above data we first present a descriptive analysis to understand the association between some of the variables used in the study. This is followed by the specification of the econometric model and discussion of the results.
Health, poverty and economic growth: Inter-State descriptive analysis

The basic socioeconomic characteristics of the 14 States and for all of India are given in Table 1. Clearly, there is large inter-State variation in the level of PCNSDP for the most recent year (2000/01). The richest State is Punjab, with a per capita income of Rs 15,390; with Bihar being the State with the lowest income per capita of Rs 4123. The estimated growth rates of real per capita income over the thirty-year period 1970-2000, also shown in Table 1, reveal similar trends. The range of variation in growth rates is from a low of about 0.9% and 1% respectively in Madhya Pradesh
Table 1 Basic characteristics of the States included in the study

State PCNSDP 2000/01 (Rs)Life Annual average rate of real PCNSDP growth 1970-2000 (%) Life expectancy at birth 1995-99 (years) IMR 2000 (per 1000 live-births) Poverty 1999-2000 (% below poverty line) Population 2001 (in thousands) Annual average rate of population growth 1971-2001
Andhra Pradesh Bihar Gujarat Haryana Karnataka Kerala Madhya Pradesh Maharashtra Orissa Punjab Rajasthan Tamil Nadu Uttar Pradesh West Bengal India
9,982 4,123 12,975 14,331 11,910 10,627 7,620 15,172 5,187 15,390 7,937 12,779 5,770 9,778 10,376
2.6 1.0 3.6 2.8 3.5 1.9 0.9 3.8 1.7 3.1 2.5 3.5 1.2 2.8 2.4
63.1 60.2 62.8 61.5 64.0 73.5 56.4 65.8 57.7 68.1 60.5 64.6 58.4 63.4 61.7
55 62 62 67 57 14 87 48 95 52 79 51 83 51 68
15.77 42.6 14.07 8.74 20.04 12.72 37.43 25.02 47.15 6.16 15.28 21.12 31.13 27.02 26.1
75,728 82,879 50,597 21,083 52,734 31,839 60,385 96,752 36,707 24,289 56,473 62,111 166,053 80,221 1,027,015
1.8 1.3 2.1 2.5 2.0 1.3 1.2 2.2 1.7 1.9 2.6 1.4 2.1 2.0 2.1
Note: 1. Data for Bihar, Madhya Pradesh and Uttar Pradesh include the three newly formed States of Jharkand, Chhattisgarh and Uttaranchal, respectively. 2. The data for India includes all States and Union Territories. 3. Per capita income (PCY) refers to real per capita NSDP . Sources: PCNSDP from EPW Research Foundation (2003), growth rate in PCNSDP is based on the authors computation, LEB and IMR are from the Sample Registration System Bulletin (2004) published by the Registrar General of India, poverty estimates are from Planning Commission (from www.indiastat.com), population for 2001 is from the Registrar General of India, GOI and the growth rate in population is computed by the authors.
SECTION I
and Bihar to a high of 3.8% in Maharashtra. The relationship between initial real per capita income (1970/71) and annual average rates of growth of real income per capita is indicated in Fig. 1. In general, States with low initial incomes also witnessed low growth rates except Andhra Pradesh, West Bengal and Karnataka. Conversely, States with higher starting incomes experienced higher growth rates, with the notable exceptions of Kerala and Madhya Pradesh. Next, we examine the relationship between economic growth and initial per capita income pooling the data for the three periods, 1970-80, 1980-90 and 1990-2000. The computed growth rate is the decadal rate for the periods and the initial income corresponds to the beginning year of the respective decade. The scatter plot with a trend line is exhibited in Fig. 2. It is amply evident that there is a positive association between
the graph, and the declining slope of the curve indicates that the effect of LEB increases faster at lower than at higher income levels. The relationship is similar to the cross-country evidence
Fig 2 Relationship between initial income and growth rate

NSDP: net State domestic product
Fig 1 Per capita income and growth rate by States, 1970-2000

Fig 3 Trends in life expectancy at birth (LEB), 1970-99, India
initial income and growth rate. At first glance, this is at variance with the cross-country results and the regional evidence reported in Barro and Sala-i-Martin (2004). However, the simple association of Fig. 2 does not control for confounding factors such as human capital stock, and additional analyses are called for to reach firmer conclusions. This issue will be explored further later in the paper. Figure 3 presents all-India trends in life expectancy at birth (LEB) during the period 1970-2000. It is immediately apparent that India experienced a remarkable improvement in LEB over this period, from 49.7 years during 1970/75 to 61.7 years during 1995/99. The inter-State disparity in LEB in 1995/99 is laid out in Table 1. LEB is highest in Kerala (73.5 years) and lowest in Madhya Pradesh (56.4 years), implying a difference of 18.1 years. Bihar, which is one of the States with the lowest per capita income, seems to have fared better than Madhya Pradesh, Orissa and Uttar Pradesh in this health status indicator. It is instructive to compare the simple association between LEB and per capita income, pooling the three-period data, as shown in the scatter plot (Fig. 4). The positive association between income and life expectancy is vividly brought out in
on the association between LEB and per capita income (in 1985 purchasing power parity (PPP) in dollars) shown by Pritchett and Summers (1996) as well as in Preston (1975). The positive association between LEB and per capita income could be due to (i) increased income causing better health; or (ii) healthier workers being more productive and hence having higher incomes; or (iii) a common factor that leads to both better health and higher incomes. Thus, the simple association between LEB and per capita income cannot tell us exactly what the nature of the relationship is. This issue is further examined below using multivariate techniques. The inter-State variation in the second health status indicator-IMR-is seen in Table 1. Kerala again stands out with
SECTION I
the lowest IMR of 14 per 1000 live-births, compared to 95 in Orissa, which has the highest IMR. Interestingly, the second lowest IMR is 48 in Maharashtra, nearly three times higher than Keralas IMR. This clearly shows that even States with better health status than all of India have a long way to go
Fig 4 Life expectancy at birth (LEB) and per capita NSDP, 1970-2000
Our discussion on the correlations that exist between indicators of economic growth-income and health-cannot be interpreted as a cause-effect relationship, including also the possibility of two-way causality among the above-mentioned variables. We now develop an econometric framework to examine the causal or simultaneity relationships among these three variables.
Model specification and estimation issues

Following the cross-country empirical studies on the determinants of economic growth (Barro 1991; Barro and Sala-iMartin 2004; Bloom and Canning 2004, the growth rate of real per capita income function can be specified as (1) Git = 0 + 1lnYit + 2lnHit + 3 Sit + 4 lnWit + a5GWit + uit, i=1,2,..,N States, t=1,2,..T periods Here Git =1/m[lnYit+1 - lnYit] is the growth in per capita real income over the period t and t+1, lnY is the natural logarithm of initial real per capita income, lnH is the logarithm of health indicator, namely LEB, S is another dimension of human capital, namely average years of schooling of the adult population, W is the ratio of working age to total population, GW is the rate of growth in W, m is the length of t (t+1), i are parameters to be estimated and uit is the random disturbance term distributed with zero mean and constant variance (see Bloom and Canning 2004 for the theoretical derivation of the model). Several other variables to capture the economic geography and quality of governance such as openness, institutional quality, ethnolinguistic fractionalization, landlocked, tropical area, average government savings rates, access to ports, government consumption ratio, rule of law, etc. were included in the cross-country analysis (Barro 1997); Bloom and Williamson 1998). However, some of these variables are not relevant for a study such as this (e.g. openness) and data on many of the variables such as governance, investment or savings ratio, rule of law, etc. were not available at the State level. Religious and caste composition (percentage of the population belonging to various religions, and schedule caste and schedule tribes), urbanization and population density were considered. Due to high correlation between these and other variables, particularly LEB and schooling, these were not included in the final analysis. In addition to average years of schooling, we also tried including years of labour market experience but due to high collinearity between schooling and labour market experience, the experience variable turned out to be statistically insignificant and hence was dropped in the final analysis. The coefficient of the initial income variable Yit is an indicator of whether there is a conditional convergence in income per capita or not among countries or regions (States) within a country. The conditional convergence hypothesizes a negative sign of the initial income coefficient. A positive sign would imply increased income dispersion among rich and poor countries (States). A problem with the initial income per capita is that this
to catch up with Kerala. The association between initial per capita income and IMR, pooling the three-period data is shown in Fig. 5. Per capita income and IMR are negatively related. The decline in IMR as income increases is not uniform across all income levels. The decline is higher at the low-income levels and lower at high-income levels.
Econometric model and empirical analysis
Fig 5 Infant mortality rate (IMR) and NSDP real per capita
SECTION I
variable is potentially endogenous and also measured with error. The procedure adopted in the growth literature is to predict the per capita income using lagged values and the predicted values are used to compute the growth rate as well as for initial income (Barro 1997). We have also adopted this procedure. Pritchett and Summers (1996), Bhargava et al. (2001) and others argue that health cannot be treated as an exogenous determinant of growth. That is, increased income leads to more investment in health and thus there is strong case for reverse causality. The current level of health status depends upon the initial income per capita and mean years of schooling of the population as specified in Pritchett and Summers (1996). The determinants of health function can be specified as (2) lnHit = 1 + 2lnYit + 3lnHExpit + 4Sit + 5Pit + eit , i=1,2,..,N States, t=1,2,T periods where H, Y and S are as defined above, HExp is the per capita State government expenditure on health, water supply and sanitation, and family welfare, hereafter referred to as health expenditure in this study, P is a measure of political power, i are the parameters to be estimated and e is the random error term assumed to be distributed with zero mean and constant variance. Increases in per capita income of the people and public expenditure are expected to improve the health status of the population. The political power factor should influence public spending in a welfare state. Two variables are considered. One is the percentage of votes gained by socialist and communist parties in the elections in the decade. The larger the share of votes gained by the socialist and communist parties, the greater their influence on public policy decisions such as government spending on welfare measures like health. Hence it is expected that political power will exert a positive effect on health status. Another political variable considered is the percentage of Assembly seats gained by the ruling party at the Centre. The higher the number of Assembly seats won by the ruling party at the Centre in the State, the more likely the State to get a higher share in central fund allocation. This variable is thus expected to have a positive effect on health. Given that elections are held once every five years under normal conditions, there were at least two elections in a decade. Hence we assigned a weight equal to the number of years a particular government stayed in power in a decade. The initial income per capita is likely to be endogenous and researchers have instrumented initial income using lagged values of the per capita income variable (Barro 1997). However, Bhargava et al. (2001) argue that lagged variables should be treated as endogenous. Pritchett and Summers (1996) experimented with alternative instruments-terms of trade shocks, investment/GDP ratio, black market premium and price level distortions in their cross-country study. These variables are probably less relevant within a country, and information on these is not available at the State level in India. In the above formulation, initial income is used to control for the transitional dynamics induced by factor accumulation. If, on the other hand, data on factor inputs are available, it is possible to formulate a model in which the change
in output is regressed on changes in inputs. Let the aggregate production function be of Cobb-Douglas form: (3) Yit = AitK it L itH itS itE it where Y is aggregate output, A is a technology parameter, K is physical capital stock, L is labour force, H is health (life expectancy), S is mean years of schooling, E is an experience vector (experience and experience squared) and , , , , and are the parameters. Taking logs of the Cobb-Douglas aggregate production function (3), we can obtain the following model (4) lnYit = it + lnKit + lnLit + lnHit + Sit + Eit where it is lnAit. The inputs K, L and H are endogenous and also measured with errors. To overcome these problems, the practice adopted in literature is to instrument the inputs using their lagged values and this approach has been used in this study for capital stock, total workers and LEB. Some limitations in the data should be noted. For instance, the definition of workers has changed between 1961 and 1981. The SRS provides data on LEB only from 1970-71 and hence the LEB for the year 1960-61 is based on the estimates of the population census of 1961.The estimation methodology depends upon the assumption we make about the technology parameter it. If it is assumed to be the same for all States over a period of time, then the production function (4) can be estimated by the OLS or by instrumental variables (IV) methods. On the other hand, if all the States are at an identical technology level but that technology itself changes over time as shown below (5) it = t + wit where wit is the random disturbance, t is the time-specific constant. Under this assumption, the equation can be estimated by the fixed effects (time) method or by including a set of dummy variables for time. However, if the technology remains the same over a period of time but varies across States, then the assumption about the technology parameter can be stated as (6) it = t + wit Specification (6) of the model can be estimated by introducing a set of dummy variables for States or by the fixed effects (states) method. The fixed effects method enables us to control for unobserved time-specific or State-specific fixed factors such as genetic factors, climatic conditions, regionspecific health problems, etc. An alternative approach is the random effects model that can be estimated by the feasible GLS method (Bloom et al. 2004). We have tested for the fixed versus random effects specification of the model using Hausmans (chi-sqaure) specification test.

Empirical results: Estimates of economic growth and health equations

Table 2 contains the OLS and two-stage least squares (2SLS) estimates of the Barro-type growth equation (1). The depend-
10
SECTION I
ent variables are the growth rates of real per capita income over the three periods 1970-80, 1980-90 and 1990-2000, and the explanatory variables are initial levels of income, LEB, years of schooling, ratio of working age population to total population and the growth rate in the ratio of working age over total population. The first column provides the OLS estimates of the initial levels of log per capita income and log LEB. The effect of initial income on growth is positive but not statistically significant even at the 10% level. The positive sign of the coefficient of initial income implies that Indian States do not converge to a steady-state growth of per capita income. This is largely at variance with the cross-country evidence that supports the conditional convergence hypothesis. However, similar findings emerge from a study by Sachs et al. (2002) using panel data on Indian States for the period 1980-98. In the case of China, there are marked differences in findings, with convergence and divergence in various subperiods, due to major shifts in economic policy (Sachs et al. 2002). The effect of log LEB on economic growth is positive and the coefficient is statistically significant at the 5% level. In the second specification, we include years of schooling and we observe that the effect of both log LEB and schooling turns out to be statistically insignificant. The high correlation between these two variables (r=0.87), indicates that there is a problem of multicollinearity. Education is expected to influence health and the relationship is apparently quite strong at the macro level. Hence, the schooling variable was excluded in the remaining specifications of the model.
The 2SLS estimates are reported in column 3. The effect of log LEB is positive but the coefficient is significant only at the 10% level. The last specification includes two demographic variables-ratio of working age over total population and its growth rate over the decade. The effect of working age over total population is positive and statistically significant at the 5% level. An increase in the share of working age population increases the potential labour force which in turn increases the growth rate. However, the growth in the share of working age over total population is negative and not statistically significant. The IV estimates of the effect of income on health are reported in Table 3. The estimates, given in column 1, show that both log per capita income and per capita health expenditure have a positive and statistically significant effect on LEB, as expected. A 10% increase in per capita income would increase the LEB by about 2% while a thousand rupee increase in per capita health expenditure would lead to 1.3% increase in LEB. Next, the average number of years of schooling is added in the specification and the results reported in column 2 reveal that a substantial effect of per capita income and health expenditure is taken away by the schooling variable. Its effect is positive and highly significant (at the 1% level or better). The next specification (column 3) includes the percentage of votes gained by the socialist and communist parties in the Assembly elections. The effect of the political factor variable is positive but not significant. The other measure of the political variable, namely the per cent of Assembly seats won by the
Table 2 The effect of health on economic growth in India, 1970-2000 Dependent variable: Growth rate in per capita NSDP over the decade
Explanatory variable OLS 1 OLS 2 2SLS 3 2SLS 4
Table 3 The instrumental variable (IV) estimates of the effect of per capita income and health expenditure on LEB, India, 1970-2000 Dependent variable: Log of life expectancy at birth (LEB)
Explanatory variable 1 2 3
Log per capita NSDP* Log initial per capita NSDP Log initial LEB Initial average years of schooling Log initial working age over total population Growth of working age over total population Constant Adjusted R2 Number of States Number of observations 1.534 (1.58) 5.567 (2.23) 1.519 (1.53) 5.172 (1.29) 0.463 (0.13) 1.419 2.875 (1.27) (2.63) 5.990 -1.801 (1.86) (0.43) Per capita health expenditure (in thousands) Average years of schooling
0.174 (3.01) 1.265 (2.35)
0.078 (1.85) 0.156 (0.39) 0.069 (6.74)
-33.223 -31.659 -33.690 0.346 0.329 0.377 14 14 14 42 42 42
14.162 (2.47) -9.837 (1.37) -5.880 0.304 14 42
Average percentage of votes secured by socialist and communist parties in the Assembly election Constant 2.480 Adjusted R2 0.521 Number of observations 42
0.820 (1.77) 0.179 (0.42) 0.0671 (5.17) 0.000170 (0.21) 3.092 0.770 42
3.123 0.776 42
NSDP: net State domestic product Note: t values are given in parentheses * Instrumented using lagged values of per capital NSDP Source: Authors' calculation
NSDP: net State domestic product; LEB: life expectancy at birth; OLS: Ordinary least squares; 2SLS: two-stage least squares. Note: t values are given in parentheses Source: Authors' calculation
11
SECTION I
ruling party at the Centre, also had a positive effect on LEB but the effect was not statistically significant.
Table 4 Estimates of the aggregate production function, India, 1970-2000 Dependent variable: Log (NSDP)
Inputs 1 OLS (levels) 2 IV (levels) 1 2
Panel data estimates of the aggregate Cobb-Douglas production function

The effect of health on output at the State level is examined by estimating a production function as specified in equations (4-6). Output is measured by the real NSDP. To overcome the problem of measurement errors and year to year fluctuations in NSDP, predicted rather than actual values of NSDP are used. The NSDP for a particular year is predicted using its lagged values. Two conventional inputs-capital and labour-are used in the production functions. At the State level, there is no information on capital stock or investment, even though data on these two variables are available at the national level over a period of time. In this study, we use the value of fixed capital net of depreciation for the manufacturing sector to capture the capital input. The capital stock should include public sector investment as well as private investment in other sectors also. In the absence of such comprehensive data, the capital measure used in this study captures only the partial and not the full effect of capital on aggregate output. The labour input refers to the total number of workers including main and marginal workers. The measure of health-LEB-and average years of schooling are as defined in the previous section. Potential experience and its squared term were also computed and included but due to the small variation in these variables and high collinearity between the two variables, the parameter estimates turned out to be imprecise and hence were dropped from the final analysis. The input variables-log capital, log labour and log LEB-are instrumented using their lagged variables as in Bloom, Canning and Sevilla (2004). As there is high correlation between the two human capital variables of LEB and schooling, all the models are estimated with and without the schooling variable. The OLS and IV estimates of the aggregate production function (4), based on the assumption that the technology is constant over time and across States, are reported in Table 4. The OLS and IV results reported in columns 1 and 3 indicate that the conventional inputs-labour and capital-and health (LEB) exert a positive and statistically significant effect (1% level) on output. The magnitude of the coefficient of LEB is high, which is puzzling. The average number of years of schooling is included in specification 2. The effect of the schooling variable is positive and statistically significant at the 1% level. However, once the schooling variable is included, the effect of LEB on output became statistically insignificant, which is due to the high correlation between the two variables as discussed above. The coefficient estimates from the OLS and IV methods are similar in sign but the standard errors of the coefficients are somewhat higher in the case of IV estimates. The estimates of the fixed effects model under the assumptions made in equations (5) and (6) are reported in Table 5. The Hausman specification test statistic suggests that the error terms are correlated with the inputs and thus the null hypothesis that the random effects model is appropriate
Log Labour Log Capital Log LEB Years of schooling
0.461 (7.15) 0.414 (7.58) 1.134 (4.51)
Constant -7.639 R2 0.934 F statistics 244.00 Number of States 14 Number of observations 56
0.475 (7.74) 0.384 (7.23) 0.163 (0.37) 0.105 (2.60) -3.905 0.941 204.89 14 56
0.678 (10.86) 0.355 (5.66) 1.960 (5.74)
-13.793 0.909 180.48 14 56
0.686 (11.11) 0.341 (5.46) 1.250 (2.27) 0.0718 (1.64) -11.064 0.915 140.40 14 56
NSDP: net State domestic product; OLS: Ordinary least square; IV: instrumental variable; LEB: Life expectancy at birth Note: t values are given in parentheses. The input variables (log labour, log capital and log LEB) in IV (levels) columns are instrumented using lagged values of their values. Source: Authors' calculation
stands rejected. We begin the discussion with specification 1. The effect of the changes in the two conventional inputs-labour and capital-and LEB on the change in output is positive and statistically significant (1% level) on output. The results suggest that a 1% improvement in LEB would result in a 1%-2% increase in output. The effect of health on output is much higher than the effect of the two conventional inputs. Specification 2 includes the average years of schooling along with health and other conventional inputs. Schooling exerts a positive and statistically significant effect at the 10% level. Both the magnitude and significance of the health effect on output are reduced due to inclusion of the education variable.
Economic growth, poverty and health: Theory and empirical evidence

Poverty is a measure of income that indicates inadequate command over material resources. The level of poverty in a country or region depends upon the level of income as well as its distribution. Any policies or programmes which alter the distribution of income would affect poverty. In a country or State with a large income inequality there would be a relatively large number of poor people or people with a low income (below a fixed poverty line), even if the country/State has a high per capita income. A higher rate of economic growth would reduce poverty if growth affects the distribution of income in ways that pulls up the bottom tail of the distribution. Countries that pursue a growth-oriented strategy firmly believe that
12
SECTION I
Table 5 Estimates of the aggregate production function, India, 1970-2000 Dependent variable: Log (NSDP )
Inputs Fixed effects (States) 1 2 Fixed effects (time) 1 2
Log labour Log capital Log LEB Years of schooling
0.910 (6.82) 0.318 (4.91) 1.510 (2.71)
Constant -15.317 R2 0.883 F statistics 197.91 Chi-square (p value) 3.05 (fixed vs. random effects) (0.38) Number of States 14 Number of observations 56
0.787 (5.41) 0.202 (2.27) 1.026 (1.70) 0.154 (1.84) -10.992 0.905 158.52 3.80 (0.43) 14 56
0.607 (8.24) 0.451 (4.78) 1.817 (3.73)
-13.344 0.910 85.37 3.23 (0.36) 14 56
0.612 (8.46) 0.405 (4.21) 0.870 (1.19) 0.079 (1.72) -9.248 0.916 67.35 3.71 (0.45) 14 56
NSDP: net State domestic product; LEB: life expectancy at birth Notes: 't' values are given in parentheses. The inputs variables (log labour, log capital and log LEB) are instrumented using lagged values of their values. Source: Authors' calculation
growth will have its trickle-down effects that will help reduce poverty. Bourguignon (2004) argues that while rapid elimination of poverty (absolute poverty) is a meaningful development goal, attainment of the goal also requires that the growth strategy be combined with distribution measures that are countryspecific. Poverty reduction at a given point in time in a country is fully determined by the rate of growth of mean income and changes in income distribution in the population. A change in income distribution can be decomposed into a growth effect (the effect of a proportional change in all incomes with the distribution of relative income remaining unchanged) and a distributional effect (change in relative incomes). He also points out that there is a case for strong interdependence between growth and distribution. What do empirical verifications suggest? The studies reviewed in Bourguignon (2004) point to ambiguous and contradictory results. Cross-sectional studies have come out with the finding that countries with more inequality in income distribution have experienced sluggish growth. But when country specific (regions) effects were controlled for, the inequality effect turned insignificant. Decadal country data, on the other hand, found a positive relationship between growth and inequality. On the poverty-health link, some argue that poverty can cause poor health while others maintain that low income and poor health are caused by some common factor such as genetic endowments or education. Poverty can have an adverse impact on health because of malnutrition and also due to poor sanitation, unsafe drinking water supply, etc. Much of
the disease burden in developing countries is due to the intake of an inadequate diet. Since expenditure on food forms a major portion of the budget of the poor, eradicating poverty could be instrumental in reducing malnutrition and the resulting ill health (Wagstaff 2001). The association among growth rate of PCNSDP, level of per capita income and poverty is first examined using data pertaining to 14 major States over a period of 30 years (1970/711999/2000). The relationship between long-run growth of per capita income and level of poverty is shown in Fig. 6. The poverty level is above the all-India average in the BIMARU States (with the exception of Rajasthan), Orissa and West Bengal. It is interesting to note that States which experienced higher levels of growth over the thirty-year period witnessed a lower level of poverty except Kerala. Similarly, in States where the long-run growth rate is lower, the current level of poverty is higher. A notable exception here is West Bengal. It is worth mentioning that the two exception States have many similarities, particularly in respect of political ideology and policy decisions. The growth-poverty link seems to suggest that rapid growth of per capita income may be required for States to achieve poverty reduction. Such growth would be able to generate productive employment and thus increase per capita incomes. The simple association between per capita income and poverty across the States over a period of time is displayed in Fig. 7. As one may expect, increase in per capita income and the percentage of population living below the poverty line are negatively related and the decline in poverty is sharp, espe-
Fig 6 Growth rate and poverty by States, India
cially at lower levels of per capita income. Possibly the growth in income in the past three decades has had the desirable trickle-down effect. Next, we turn to the association between poverty for the periods 1972/73, 1983, 1993/94 and 1999/2000, and fiveyear average life expectancy corresponding to the above years at the national level. As shown in Fig. 8, the percentage of
13
SECTION I
Fig 7 Per capita income-poverty association, panel data, 1970-71 to 1999-2000
population living below the poverty line has considerably declined over the period and there is a negative relationship between poverty and LEB, which shows a small increase during the period. The scatter plot in Fig. 9, constructed using State-level information on poverty and LEB clearly points to the inverse relationship between the two variables. An important question in this context is whether poverty is the cause or consequence of poor health status. Both may be interdependent.
Conclusion and policy suggestions

This study examines the determinants of economic growth and health using a panel data of 14 major Indian States for the period 1970/71-2000/01. The association between initial per capita income, growth rate, and health across the
Indian States has been explored using scatter plots and charts. The interesting findings are as follows: A strong positive association is observed between initial per capita income and long-run economic growth in per capita income across the States. That is, States with a higher initial income have grown faster than States with a lower initial income. This has the effect of widening the gap between the rich and poor States. There is also a strong association between per capita income and health status (LEB and IMR) of the population. The nexus between growth, poverty and health based on cross-sectional data of Indian States over a period of time point to the following: States that have experienced higher (lower) levels of growth over the thirty-year period witnessed a lower (higher) level of poverty, except Kerala and West Bengal. Per capita income and the percentage of the population living below the poverty line are negatively related; possibly the growth in income in the past three decades has had the desirable trickle-down effect. There is an inverse relationship between poverty and LEB. The descriptive analysis indicates only associations between the variables and it is not possible to infer any cause-effect or simultaneous relationships among them. We have formulated an econometric framework based on the recent developments in growth theory and this is applied to inter-State panel data for the years 1970-71, 1980-81, 1990-91 and 2000-01. The following important findings emerge from our econometric analysis: There is a two-way causation between economic growth and health status. The effect of health measured by life expectancy is positive and significant on economic growth even after controlling for initial income levels. There is evidence of a significant effect of per capita income and per capita public expenditure on health on LEB. Average number of years of schooling emerges as the most sig-
Fig 8 Trends in poverty and life expectancy at birth, India 1972/73-1999/2000

LEB: life expectancy at birth
Fig 9 Health-poverty nexus, India, panel data, 1970-71 to 1999-2000

LEB: life expectancy at birth
14
SECTION I
nificant determinant of LEB. Our analysis shows that a thousand rupee increase in per capita health expenditure would lead to a 1.3% increase in LEB, while a 10% increase in per capita income is required to increase the LEB by about 2%. The production function estimates indicate that the effect of health (LEB) on NSDP is very high, in fact, much higher than the effect of the conventional inputs of capital and labour. The following policy suggestions are made based on the empirical findings of our study:
Increasing investment in health is a required policy intervention for accelerating the economys growth rate. Growth-oriented policies would result in bringing about improvements in the health status of the population. Policies promoting growth would also have the desirable effect of reducing poverty. Overall, there is a compelling reason for stepping up both public and private investment in health which would pay off in the long run.
15
SECTION I
References
Ahluwalia MS. State level performance under economic reforms in India. Working Paper No. 96. Stanford University, USA: Center for Research on Economic Development and Policy Reform; 2001. Barro RJ. Economic growth in a cross-section of countries. Quarterly Journal of Economics 1991. 106:407-43. Barro RJ. Determinants of economic growth: A crosscountry empirical study. Cambridge, Massachusetts: MIT Press; 1997. Barro R, Sala-i-Martin X. Economic growth. 2nd Ed. New Delhi: Prentice-Hall of India; 2004. Behrman JR, Deolalikar AB. Health and nutrition. In: Chenery H, Srinivasan TN (eds). Handbook of development economics, Vol. I, Amsterdam: North-Holland Press; 1988. Bhargava ADT, Jamison Lau LJ, Murray CJL. Modelling the effects of health on economic growth. Journal of Health Economics 2001;20:423-40. Bloom D, Canning D. Global demographic change: Dimensions and economic significance. Working Paper No. 10817. Cambridge, Massachusetts: National Bureau of Economic Research; 2004. Bloom D, Freeman RB. The effects of rapid population growth on labour supply and employment in developing countries. Population and Development Review 1986;12 (Supplement):381-414. Bloom D, Williamson J. Demographic transitions and economic miracles in emerging Asia. World Bank Economic Review 1998;12:419-55. Bloom D, Canning D, Sevilla J. The effect of health on economic growth: A production function approach. World Development 2004;32:1-13. Bourguignon F. The poverty-growth-inequality triangle. Working Paper No. 125. New Delhi: ICRIER; 2004. Central Statistical Organisation (CSO). National Accounts Statistics, 2002. New Delhi: Ministry of Statistics & Programme Implementation, Government of India; 2004. Deaton A. Health, inequality, and economic development. CMH Working Paper No. WGI: 3, Commission on Macroeconomics and Health, 2001. Deolalikar A. Do health and nutrition influence labor productivity in agriculture? Econometric estimation for rural south India. Review of Economics and Statistics 1988;70:406-13. Dreze J, Sen A. India: Economic development and social opportunity. Delhi: Oxford University Press; 1995. Duraisamy P. Health status and curative health care in rural India. Working Paper Series No. 78. New Delhi: National Council of Applied Economic Research; 2001. Duraisamy P. Morbidity in Tamil Nadu: Levels, differentials and determinants. Economic and Political Weekly 1998;33:982-90. Duraisamy P, Duraisamy M. Determinants of investment in health of boys and girls: Evidence from rural households of Tamil Nadu, India. Indian Economic Review 1995; XXX:51-68. Duraisamy P, Sathiyavan D. Impact of health status on wages and labour supply of men and women. Indian Journal of Labour Economics 1998,41:67-84. EPW Research Foundation. National Accounts Statistics of India, 1950-51 to 2000-01. Mumbai: EPW Research Foundation; 2002a. EPW Research Foundation. Annual Survey of Industries, 1973-74 to 1997-89: A database on the industrial sector in India. Mumbai: EPW Research Foundation; 2002b. EPW Research Foundation. Domestic products of the States of India, 1960-61 to 2000-01. Mumbai: EPW Research Foundation; 2003. Fogel RW. Economic growth, population theory, and philosophy: The bearing of long-term processes on the making of the economic policy. American Economic Review 1994;84:369-95. Gumber A. Burden of disease and cost of ill-health in India: Setting priorities for health intervention during the ninth plan. Margin 1997;29:33-72. Gupta I, Mitra A. Economic growth, health, and poverty: An exploratory study on India. In: Misra R, Chatterjee R, Rao S (eds). India Health Report. New Delhi: Oxford University Press; 2003.
16
SECTION I
Gyimah-Brempong K, Wilson M. Health human capital and economic growth in sub-Saharan Africa and OECD countries. Quarterly Review of Economics and Finance 2004;44:296-320. Kannan KP, Thankappan KR, Raman Kutty V, Aravindan KP. Health and development in rural Kerala: A study of the linkages between socioeconomic status and health status. Trivandrum: Kerala Sastra Sahitya Parishad; 1991. Krishnan TN. Access to health and burden of treatment in India: An inter-state comparison. Working paper No. 2, UNDP Research Project. Thiruvananthapuram: Centre for Development Studies; 1995. Mankiw NG. Romer D, Weil DN. A contribution to the empirics of economic growth. Quarterly Journal of Economics 1992;107:407-37. Mayer D. The long-term impact of health on economic growth in Mexico, 1950-1995. Journal of International Development 2001;13:123-6. Preston SH. The changing relation between mortality and level of economic development. Population Studies 1975; 29:231-48. Pritchett L, Summers LH. Wealthier is healthier. Journal of Human Resources 1996;31:841-68. Registrar General of India (various years). Social and cultural tables. Population Census. Government of India. Registrar General of India (1999). Compendium of Indias fertility and mortality indicators, 1971-1997 based on the sample registration system (SRS). New Delhi: Registrar General of India. Sachs JD, Bajapi N, Ramiah A. Understanding regional economic growth in India. CID Working Paper No. 88. USA: Harvard University; 2002.
Schultz TP. Health and schooling investments in Africa. Journal of Economic Perspectives. 1999;13:67-88. Srinivasan TN. Indian economic reform: A stocktaking. Working Paper No. 190. USA: Stanford Center for International Development, Stanford University; 2003. Strauss J, Thomas D. Health, nutrition and economic development. Journal of Economic Literature 1998;36:766-817. Sundar R. Household survey of health care utilisation and expenditure. Working Paper No. 53, New Delhi: National Council of Applied Economic Research; 1995. Temple J. The new growth evidence. Journal of Economic Literature 1999;37:112-56. Vaidyanathan A. An assessment of nutritional and health status. Discussion Paper No. 3, UNDP Research Project, Thiruvananthapuram: Centre for Development Studies; 1995. Visaria P, Gumber A. Utilisation of and expenditure on health care in India, 1986-7. (Unpublished report) Ahmedabad: Gujarat Institute of Development Research; 1994. Wagstaff A. Poverty and health. Commission on Macroeconomics and Health Working Paper Series WGI: 5, Geneva: WHO; 2001. Weil D. Accounting for the effects of health on economic growth (mimeo). Economic Department, Brown University; 2001. World Bank. Attaining the Millennium Development Goals in India: Role of public policy and service delivery, Human Development Unit, South Asia Region The World Bank; 2004.
17
SECTION I
Health, nutrition and poverty: Linking nutrition to consumer expenditures
ANUP K. KARAN
INSTITUTE FOR HUMAN DEVELOPMENT NIDM BUILDING, I.P. ESTATE, MAHATMA GANDHI MARG NEW DELHI 110002 E-MAIL: akkaran@yahoo.com
AJAY MAHAL
Assistant Professor DEPARTMENT OF POPULATION AND INTERNATIONAL HEALTH BOSTON MA 02115, USA E-MAIL: amahal@hsph.harvard.edu
HERE IS NOW SUBSTANTIAL RECENT LITERATURE ON THE IMPACT OF IMPROVEMENTS in the health status of a country's population on its aggregate economic performance (Bloom and Canning 2000; Bhargava et al. 2001). The main conclusion of this set of literature, with a few exceptions, is that improvements in health provide a substantial boost to the economies of countries where they occur. There is also evidence that the aggregate economic performance of a country can influence the health status of its population (Pritchett and Summers 1996). While there is some debate about the actual magnitude of this effect (see Subramanian 2004 and Ruger et al. 2001 for a review), its overall direction is not subject to much debate. In fact, more can be said about the association between increased income and health empirically. Increases in the average income are also associated with declines in the poverty ratio, especially when the overall distribution of income does not simultaneously worsen too much. To the extent that commonly used measures of absolute poverty incorporate expenditure required to achieve the consumption of a minimal basket of food items,' or the purchase of food items required to achieve a minimal level of energy defined in calories', it is reasonable to argue that increases in the average income, taking account of disparities, will tend to be associated with improvements in the nutrition of the poorest. In this regard, Bhargava (1999) suggests that when people can afford to do so, they do consume healthier diets. Bhargava (1991) presents evidence from International Crops Research Institute for the Semi-Arid Tropics (ICRISAT) data that improvements in permanent' income are positively associated with improvements in the consumption of food items and some nutrients. (This debate was triggered by the pioneering study of Behrman and Deolalikar (1987) which showed, in (six) ICRISAT villages of south India, the income elasticity of calorie intake was quite low, and not significantly different from zero in statistical terms. The authors note that even among the very poor, as incomes rise, households mostly purchase additional taste'.) Nutrition has been positively associated with anthropometrical indicators such as height and weight (e.g. Jamison et al. 2003). Nutritional deficiency (e.g. of iron, calcium, vitamins A, B and C) in the human body has been associated with a variety of adverse health conditions (Willett 1998). Since the 1980s, there has been a consistent decline in India's poverty rate, together with a significant growth in the real income per capita. During 1980-2000, India's real per capita income has nearly doubled, having grown at an annual average rate of 3.3%. In 1983, as per the head-count measure, India's poverty ratio was 45.7% in rural areas and 40.8% in urban areas, which declined to 27.1% and 23.6%, respectively, by the year 2000 (Planning Commission 2001). This has been used to bolster the claim that India's rapid economic growth can be effective in substantially reducing poverty (Lal et al. 2001; Datt and Ravallion 2002). This ought, by implication of the discussion in the preceding paragraph, to enhance the ability of the Indian poor to reduce the level of their malnutrition. This paper adds two wrinkles to the above set of issues relating to poverty in the Indian context. First, we broaden the definition of minimal consumption' that, for purposes of measuring poverty, focuses only on energy intake (in calories). We do this by incorporating in the notion of minimal consumption the requirements of a balanced diet across a vector of nutrients, such as proteins, vitamins, fats, carbohydrates, etc. Second, we move away from the focus on the minimal basket of pre-identified food items' to allow for variation in the relative proportions of different items consumed, as well as the inclusion of newer items in the food consumption basket.
19
SECTION I
Further, we argue that this departure does not loosen the comparability of the poverty line over time. This is because the poverty line is constructed by the method of estimating the smallest consumption expenditure required to achieve a fixed (minimum) nutritional requirement. We provide some justification for these modifications to the way in which poverty is assessed in India. In particular, there is one advantage of these modifications over the traditional approach to measuring poverty. This is the benefit that results from being better able to account for the nutritional impact of relative price changes in food commodities, and also of the evidence from the nutritional literature of the need for a balanced diet' that includes a variety of nutrients. Moreover, we argue that many of the arguments against using these modifications are not as debilitating as has been suggested in the literature. Using this perspective, we construct revised estimates of the head count measure of poverty in India for two periods1993-94 and 1999-2000-and for individual provinces, further classified into rural and urban populations. We found first the poverty line (PL) based on a balanced diet' measure to be higher than that calculated by the Planning Commission methodology. Second, the poverty ratio in our framework declines at a rate that is markedly slower than the poverty measure used by the Planning Commission at the all-India level. In addition, there are major differences in trends in the two sets of poverty estimates at both the provincial levels, as well as for rural and urban populations. Our main conclusions are as follows. If the focus of the foodpoverty line is a minimal level of nutritionally balanced diet, the official method of estimating poverty that typically fails to fully account for the impact of changes in relative prices and a diet that includes the consumption of micro-nutrients, is inadequate for assessing the impact of rising incomes on the health of the poor. Moreover, why relative prices change and the role that governments sometimes play in bringing about such change is a subject of crucial policy importance.
Poverty Line Measurement with Reference to a Nutritionally Balanced Diet

This section examines the way poverty is usually measured in India, and lays out the case for the approach taken in this paper.
Measuring Absolute Poverty

The standard approach to measuring poverty is to define a poverty line' level of income (or expenditure), and then to estimate the proportion of total population that lives below the PL, to arrive at the so-called head count' measure of poverty. The PL defines an absolute minimum level of consumption of food and non-food items that is necessary for sustenance and acceptable to a society. The standard approach is to first define a food-poverty line (FPL)' which estimates the expenditures needed to fund a minimum level of nutrition. There
are two main (but related) ways of doing this. First, one can estimate the minimum amount of expenditure needed to achieve a certain minimum level of nutrition typically expressed in energy units (kilocalories). This approach allows for variation in the proportion of food items consumed (as well as the types of food consumed), and the combination chosen is the one that minimizes the cost of achieving a pre-specified calorie intake. The second method also focuses on the expenditures used to achieve a minimum energy level but, unlike the first method, it requires predetermining the combination of food items and their proportions at some base-year level, and then calculating the expenditure needed on this combination to achieve the requisite energy level. It is entirely possible and, in fact, very likely that the first approach will yield a lower PL than the second in the base year. Having defined the FPL by either of the above techniques, the challenge is then add some non-food expenditure component to the FPL that reflects a minimal level of consumption of non-food items. In India, the share of nonfood expenditure in total spending for the household on the margin of poverty is based on an exogenously determined normative standard. As per the approach adopted by the Planning Commission, on an average, this share amounted to about 26.50 per cent of total spending in urban areas and 19.58 per cent of total expenditure in rural areas (Malhotra 1997). Another approach has been to estimate this additional amount as being the non-food expenditures of the individual whose food expenditure equals the food poverty-line level of spending. Adding the food- and the non-food components of the PL yields the PL level of expenditures. There are several conceptual and empirical issues that arise with these methods of arriving at the PL. These methods need to be understood clearly since they have implications for the methods used in this paper. First, it has been argued that the method of estimating the FPL using a predetermined food basket is more desirable than the approach that estimates a minimum-cost food combination. This is because the latter method can sometimes give rise to cost-minimizing combinations that may not be culturally acceptable, or that are not tasty' (Stigler 1945). Moreover, some have noted that the cost-minimizing approach poses difficulties in solving linear programming problems and so, may be expensive in terms of computing time (Lanjouw 1997). Finally, it has been suggested that the cost-minimizing approach is less well suited for making inter-temporal comparisons since it is not comparing like with like' combination of goods. It is not apparent to us, however, that the above, in fact constitute reasons enough to discard the least costminimizing approach. First, extremely powerful and relatively cheap mathematical computing software are now available (e.g. MATHEMATICA and MATLAB), which can solve, what appear at the first sight to be complex linear programming problems. Second, it is not apparent to us that a base-year combination of food items can be taken as more-or-less descriptive of population tastes. Patterns of food consumption (including proportions of items consumed) depend on the price levels,
20
SECTION I
relative prices and income, and they may reflect the nutritional needs (e.g. Bhargava 1991). If food consumption is allowed to change along these lines, it is no longer obvious why a base-year food-consumption basket possesses any more validity than a food-consumption basket for any other year. Food habits may be slow to change, but change they will; and that will pose challenges for extended inter-temporal comparisons under the currently preferred method of a predetermined food basket. We do not deny that certain habits such as vegetarianism (Bhargava 1991) may be difficult to give up, but that is a much less restrictive imposition than the share of different items in the food basket. One can think of various methods to incorporate culturally acceptable food consumption patterns, although none of them are perfect. One way is to calculate the PLs for each State, and by rural and urban populations in each State, to account for the differences across space. Another way is to pre-specify the tastes, and directly incorporate them as additional constraints (or a constraint) in the linear programming exercise. There will be some degree of arbitrariness in describing these constraints but this problem can be addressed by taking account of the patterns of food consumption by populations over fairly long periods of time. Finally, we do not see why comparing like with like' of food items is so crucial for inter-temporal comparisons of poverty ratios. Suppose the issue is one of achieving a minimal level of energy intake for people living at the very margin of survival. What is relevant is comparing like with like in energy units', which is taken account of by the cost-minimizing combination method. As one obvious, albeit extreme, example of survival needs determining food intake, and not merely culturally determined tastes, one has to consider only the food habits of the survivors of the recent tsunami in Indonesia and the Andaman Islands, many of whom lived off coconuts and the bark of coconut trees (Gray 2005). Moreover, the predetermined food basket method for estimating PLs and comparing poverty ratios over time is not equipped to handle non-trivial changes in relative prices, especially if individuals are likely to change their food consumption patterns in response. The approach taken in this paper has another attractive feature, which has to do with the concern about substitution elasticity between different types of food, mainly on account of the high price elasticity of the demand for non-cereal food items. Nutritionists see high substitution elasticities as a cause for concern, at least among the poor, since the nutritional status is thereby threatened by price increase (Deaton and Muellbauer 1980). This does cause a difficulty with the standard food-basket formulation of the FPL, which may remain unchanged even when the relative prices change, and therefore are unable to capture nutritional deficiencies among the poor that might result from a change in the relative prices. Notice that a definition based on estimating the smallest expenditure required to achieve some minimum nutritional level will reflect this, by means of an upward shift in the FPL. By emphasizing the role of relative prices in influencing the level of nutrition, this method can help draw policy attention to a variety of government policies that affect prices
of different food items-price support systems, public distribution systems, and the like. A second set of conceptual issues arises with respect to the calculation of the non-food component of the PL level of expenditures. As mentioned above, the most popular approach has been to use as the non-food poverty line-the average nonfood expenditures of individuals whose food expenditures equal the FPL. Caution must be exercised, however, because if this estimate is constructed separately for each year for which data are available, it may lead to unsatisfactory results. For instance, rising food prices may lead individuals to consume more non-food items even if non-food item prices are unchanged (because their prices relative to food items have fallen), and thus there will be a simultaneous increase of the food-and non-food poverty lines. That, in turn, may lead to sometimes spurious findings of increasing poverty, or higher urban poverty than rural poverty (Ravallion and Bidani 1994). One approach to fix this would be to use either some base year level of the non-food poverty line' (scaled up to reflect the inflation over time); alternatively, one could use a baseyear ratio of non-food to food expenditures. The Planning Commission estimates in India effectively use the base year numbers updated to calculate price increases over time.
A Nutritionally Balanced Diet

We have argued that a cost-minimizing approach to estimate the FPL is not only readily feasible, but may also be conceptually more satisfying than an approach that works with a fixed basket of food items, apart from being flexible enough to capture most of the good points of the latter method. We now make the case for estimating an FPL that requires the estimation of the lowest expenditures needed to achieve a minimum combination' of nutrients. The minimum combination' is defined not just as the energy requirement (in calories) but also as the requirement for a vector of nutrients, including both macro- and micronutrients, such as carbohydrates, proteins, fats, Vitamins A, B, and C, carotene, iron, riboflavin and calcium. Why does one need to go beyond the criterion of energy intake, irrespective of the source (carbohydrates, fats or proteins)? For one, even if energy intake were the sole objective, it matters how such energy is obtained. Thus, Jamison et al. (2003) report that the proportion of energy consumed in the form of proteins matters much more for anthropometric indicators such as height and weight, than the overall energy intake. Bhargava (1991) notes that the consumption of proteins without carbohydrates in the diet has an adverse effect on the ability of the dietary proteins to replace body proteins. Some of these theories have been disputed. For instance, Jamison et al.s paper has been criticized by a number of influential commentators for its weak methodology and, moreover, the impact of protein on the human body has been questioned previously by Sukhatme (Sukhatme 1974; Martorell 2003). However, other research has tended to agree with Jamison et al.'s conclusions (Bhargava and Guthrie 2002). Second, nutrients matter in ways that go beyond a narrow
21
SECTION I
focus on energy intake. The lack of calcium, vitamins A, B1, B2, and C, and iron has been associated with a higher frequency of certain types of cancers, cardiovascular conditions, and other serious health disorders. In fact, it has been suggested that the results of Jamison et al. were confounded by the presence of other micronutrients that could also influence height and weight. Again, there is one key contentious issue. Researchers have noted that the energy intake levels tend to be highly correlated with the intake of other nutrients as well, because most diet and foods that constitute major energy sources also contain at least some quantity of key micronutrients (Willett 1998). That might suggest a preferential focus on energy intake only. Two considerations militate against this viewpoint. First, focusing only on energyintensive diets may be an economically inefficacious way of obtaining the requisite level of micronutrients. Moreover, nutrition research suggests that the quantity of various micronutrients in the diet continues to impact for disease risk, even after being scaled by the level of energy intake (Willett 1998). This calls for attention to individual components of the minimum nutrition vector discussed above. What precisely should the minimum be even if one agrees in principle with the notion that a certain minimum combination of nutrients is necessary? For instance, energy consumption (and therefore needs) typically varies among individuals by their weight, level of physical exertion/activity (including in occupation), and the metabolic rate (the efficacy with which the body absorbs energy-providing foods). For people living on the margins of poverty, one can reasonably construct some estimate of their daily energy needs based on the nature of their jobs and some intelligent guesses about weight for given age and sex. In India, the average minimum energy requirement has been stated to be 2100 calories for an average urban resident and 2400 calories for someone living in a rural area. In our framework, for each category of minimum energy intake, we must also define a corresponding quantity of micronutrients to be consumed. Given the current state of scientific knowledge, this is possible only roughly, by defining a recommended dietary allowance' (RDA), based on research that shows the efficacy of different types of nutrient consumption per calorie consumed in influencing specific types of disease risk. As defined in the literature, RDA is not some minimum requirement, however, and we are still only learning about the possible consequences of having too much of a specific nutrient. Moreover, if we were to set out the ideal RDA as one that achieves some minimum desirable health status (e.g. impact on overall mortality risk), then the task of coming up with an RDA is well-nigh impossible since we still know very little about interactions between various nutrients and how they translate as a combination into mortality risk, for instance. The difficulty outlined in the previous paragraph possibly explains the dominant focus in the poverty literature on energy intake. However, that ought not to divert us from emphasizing the role of other micronutrients. To address the lack of precision about the amount of other nutrients required, we
can assess alternative PLs and poverty ratios for different levels of the RDA vector. A second rationale for focusing solely on energy intake is the following argument: provided that the minimum energy needs are met, the poor may be able to participate effectively in the labour market and the additional incomes earned from their labour can then be used to support the purchase of food items containing other nutrients (Bhargava 1991). This sort of hierarchy of human needs', even if observed in practice, appears to us to be unsatisfactory as a rationale for defining the FPL. The FPL, however defined, is a static concept, which describes the expenditure that is just enough to meet the minimum nutrition requirements. If household income/expenditure is just enough to purchase minimum energy requirements, and falls short of what is needed to purchase other desirable nutrients, then that is all the information we have. It is difficult to conclude from this fact alone the future prospects of the individual, or the household, since future earnings depend on a host of other exogenous variables that can affect labour market conditions. More significantly, behaviour by desperate households living on the margins of below survival levels of income cannot be taken as an indicator of a normative standard to which the FPL is closer in spirit. In the Indian context, there is a long history of debate between experts who have sought to incorporate the notion of a balanced diet, or adequate nutrition in the definition of poverty (e.g. Rao 1997; Sukhatme 1977, 1978), and those who have underplayed it (e.g. Dandekar 1996). The latter perspective has tended to dominate the calculations of PL in India. However, as biomedical research increasingly highlights the importance of micronutrients for health as against a pure energy intake, it is difficult to bypass this perspective on nutrition. Dandekar (1996) has argued that poverty and undernutrition are different and that, want of adequate income, howsoever defined, is poverty; deficiency of energy appropriately defined is under-nourishment. These two are related in the sense that statistically they go together. But the two are not identical; in fact they are two different phenomena.' If we take poverty as more than just income deprivation, and include it to mean deprivation in other areas, such as health and nutrition, as per Sen's capability approach (Sen 1985), efforts to divorce under-nutrition from the notion of poverty appear less justified.
Poverty Line Measurement: A How to' for this paper

Estimating the food- and the non-food poverty line The standard approach to measuring poverty is to define a PL level of income (or expenditure), and then to estimate the proportion of total population that lives below the PL, to arrive at the so-called head count' measure of poverty. We estimated the balanced diet-based PLs for all the major States, and associated rural and urban areas of India, for two years, i.e. 1993-94 and 1999-2000. The estimation of the PL involved two steps. First, we calculated the minimum expenditure required for meeting the predefined nutritional basket
22
SECTION I
of the Indian Council of Medical Research (ICMR). This provided us with the FPL. Second, we constructed an estimate for minimal non-food expenditures. These two components added together, yielded the required PLs. The methods of estimating these two components of the PL are described below.
Finally, the PL level of expenditure (food and non-food) was estimated by using the formula: FPL x 1/(1-) where FPL = food expenditure cut-off (or the FPL) = non-food expenditure ratio to the total household expenditure for the base year 1993-94. This specification implies that to estimate the PL, we scale up the minimum expenditure required for nutritionally balanced food expenditure, or the FPL, by an allowance for some minimum needed non-food expenditure.
Food expenditure cut-off (FPL)

We will estimate the least possible expenditures required to achieve a predefined minimum nutritional basket, taken for our purposes, the RDA. Let this basket be denoted by the n x 1 vector N, where N = [N1 N2 Nn]. Suppose there are m types of possible food items, whose quantities consumed are described by the m x 1 vector f, where f = [f1 f2 fm]. Let the corresponding per unit prices of each food item be described by m x 1 vector P, where P = [P1 P2 Pm]. Finally, let each food item Fi (i = 1 m) have a corresponding n x 1 nutrient content vector fi = [Ni1, Ni2 Nin]. Then the problem of solving for the FPL becomes one of solving the following linear programming problem: (1) Minimize f'P Subject to Ff N and N 0
Sources of data
Our definition of the RDA was obtained from the ICMR. This information was available at the individual level, classified by age and sex. A simple average across age and sex of the RDA for 10 nutrients was calculated to arrive at the per capita RDA. The per capita RDA has been used as the minimum threshold of nutritional requirement at the household level. On the basis of the ICMR recommendations of different nutritional requirements for different age and sex,
Table 1 Average per capita RDA per day of various nutrients

Nutrients Per capita RDA per day
Here F is the nutrient content matrix, of dimension n x m. The first constraint Ff N ensures that the nutrient intake equals or exceeds the RDA. The second constraint ensures that only nonnegative amounts are consumed. It is also possible to introduce other constraints to satisfy requirements of tastiness' or other cultural characteristics'. We will discuss some of these extra constraints later. Solving (1) yields the cost-minimizing expenditure E* that is a scalar product of two vectors-a given price vector P, and the optimal combination of food items f*. We undertook this exercise for all States, rural and urban areas of India, and for years 1993-94 and 1999-2000. The resulting estimates yield the FPL, which also indicates minimum expenditure required for the minimum balanced nutrition on a per capita basis.
Calorie Protein Fat Iron Calcium B carotene Riboflavin Thiamin Niacin Vitamin C
RDA: recommended dietary allowance Source: Based on ICMR, 2002
2300 60 40 28 400 2400 1.4 1.2 16 40
Non-Food Expenditure Allowance

For estimation of the non-food component of the PL for the corresponding years, we estimated the State, and rural- and urban-specific ratios of food and non-food expenditure for the marginally poor (defined here as individuals belonging to households with per capita expenditures that lie in a band of 10% (5% above and 5% below) of the PL as defined by the Planning Commission methodology1 for the year 1993-94. The non-food expenditure ratio so worked out was used to calculate the minimum necessary allowance for non-food consumption items.
Poverty Line
the calculated average per capita RDA for the 10 nutrients are given in Table 1. In addition, we needed information on the price vector P, the nutrient content matrix F, and the various types of food items available. Various types of food available and consumed were listed based on the data collected by the National Sample Survey Organization (NSSO) in four nationally representative consumer expenditure surveys, 1983-84, 1987-88, 1993-94 and 1999-2000. These surveys provided information on 125 different food items. The information on prices was obtained from the previously
1 We worked out three parallel estimates of the share of non-food expenditure. These are non-food expenditure share of (i) all poor below the Planning Commission poverty line; (ii) the marginally poor, who are around (5% below and 5% above the Planning Commission poverty line; and (iii) bottom 30% of the population as per the monthly per capita expenditure (MPCE). Poverty line estimates produced under these different procedures are not very different from each other.
23
SECTION I
mentioned consumer expenditure surveys as well. Unfortunately, there are several complications in defining these prices. First, different households face different food prices, and they may be purchasing different quality products for which no information is available in the survey. To an extent, some of these differences can be addressed by constructing PLs for different States, and across rural and urban areas within a State. That may still not fully address the problem. To this end, we defined implicit' prices, as reflected by household-level information on expenditures for specific items, and the quantities purchased of each. The ratio of the total expenditures to the total quantity purchased is taken as the implicit price. The nutrient content of food items is described in Appendix I. This information was obtained from publications of the National Institute of Nutrition (Gopalan et al. 1989). The number of food items covered by Gopalan et al. (1989) is approximately 450 and there is a perfect match for more than 100 items between those in their study and the consumption baskets of households in the NSSO consumer expenditure surveys. However, the data for about 15 food items do not match, either because these items were cooked, or were less commonly consumed. For these items, estimates of the nutrient content were based on their closest substitute in the list of Gopalan et al.. The nutrient values per 100 g of food items were calculated. Although the data are extremely comprehensive (and valuable), several cautionary remarks are in order. First, the nutrient content of a food item can vary considerably across geographical areas, depending on a variety of conditions including the quality of soil (Willett 1998). Second, the theoretical nutrient content of food and the actual nutrient consumption of a person may vary markedly depending on the method of food processing, and the combinations in which they are consumed. For instance, cut fruits rapidly lose some of their nutrients if not consumed soon; chemical properties of foods change in the process of cooking; the properties of yoghurt, milk and skimmed milk are vastly different and surveys may not always distinguish these products (Willett 1998). Third, nutrient consumption is not synonymous with nutrient intake. For example, as noted earlier, for dietary proteins to effectively replace body proteins, the simultaneous consumption of carbohydrates appears to be essential (Bhargava 1991).
in India has gone down from approximately 9% in the early 1980s to less than 6% in the late 1990s, with the value of production falling from 1.64% to 0.71% during the same period. Similarly, the area under cultivation and the value of
Table 2 Area under cultivation and production of jowar, bajra, ragi and small millets during the 1980s and 1990s
Crop 1981-82 Years (Triennium average) 1991-92 1998-99
Cultivated Area as a Share of Total Cultivated Area (%) Jowar 9.01 7.55 Bajra 6.35 5.71 Ragi 1.39 1.21 Small millets 2.08 1.32 Share in Total Agricultural Production (%) Jowar 1.64 1.16 Bajra 0.75 0.66 Ragi 0.36 0.21 Small millets 0.19 0.08
Source: Government of India, 2000.
5.69 5.08 0.92 0.83 0.71 0.49 0.15 0.06
Findings
Based on the methodology and data sources mentioned above, we solved the linear programming problem described above under three different scenarios. Under scenario I, we imposed only the constraint that the consumption of any food item must be non-negative, i.e. (fi 0). Under scenario II, in addition to the constraint in scenario I, we imposed the constraint that the consumption of coconut, jowar, bajra, ragi and millets equals zero. The main justification for this assumption is that coarse grains such as jowar, bajra, ragi and millets are increasingly vanishing from the average Indian diet, as is their production. For example, the area under jowar cultivation
production of bajra, ragi and small millets have declined considerably over the years (Table 2). There are two main reasons for imposing the requirement that the consumption of coconut is zero. First, in many parts of India, coconut is consumed in limited amounts. Second, including the possibility of coconut consumption in the linear programming exercise leads to unrealistic solutions where large amounts of coconut are consumed, and items commonly observed in the consumption basket are excluded. Under scenario III, we impose the further requirement that at least 200 g rice is consumed. This requirement was imposed because rice did not figure in the optimum consumption basket in any of the State except three States viz. Andhra Pradesh, Assam and Tamil Nadu. However, a long-term consumption pattern of different food items in India shows that, on an average, 200 g per capita per day of rice is consumed in almost all the States. (For consumption trends of different cereals see Mahendradev et al. 2004.) We estimated the FPL by solving for the least cost combination of foods separately for rural and urban areas in all states. For illustrative purposes, however, we present in Tables 3 and 4, the solution that would obtain if one was interested in an FPL at the all India level, separately for rural and urban areas. In Table 3, scenario I describes the solution to the linear programming problem without imposing any constraint other than the RDA requirement and non-negativity of the consumption of food items. Here the optimum consumption basket is 214 grams of Jowar, 361 grams of Bajra, 48 grams of Ragi, 32 grams of Spinach, 40 grams of coconut and 14 grams of guava per person per day in rural areas. Similarly, in urban areas the optimum quantity of consumption is 213 grams of Jowar 361 grams of Bajra, 48 grams of Ragi, 32 grams of
24
SECTION I
Table 3 Three scenarios of optimum consumption basket and minimum expenditure required for recommended nutritional intakes in rural and urban areas in India, 1993-94
Region and food items Price in Rs per 100g Amount (x 100 g) Scenario I Total expenditure (Rs) Amount ( x 100 g) Scenario II Total expenditure (Rs) Amount (x 100 g) Scenario III Total expenditure (Rs)
RURAL Rice 0.51 Khoi, lawa Wheat/Atta 0.47 Jowar and products 0.32 Bajra and products 0.34 Ragi and products 0.39 Milk Other milk products 1.70 Mustard oil 3.24 Spinach/other leafy vegetables 0.32 Coconut 0.42 Guava 0.44 Per capita per day total expenditure required Per capita per month total expenditure required URBAN Rice 0.54 Khoi, lawa Wheat/Atta 0.55 Jowar and products 0.40 Bajra and products 0.42 Ragi and products 0.46 Milk Other milk products 0.80 Mustard oil 3.26 Spinach/other leafy vegetables 0.43 Coconut 0.45 Guava 0.56 Per capita per day total expenditure required Per capita per month total expenditure required
0.00 0.00 2.14 3.61 0.48 0.00 0.00 0.32 0.40 0.14
0.00 0.00 0.68 1.23 0.19 0.00 0.00 0.10 0.17 0.06 2.43 72.90
0.00 7.44
0.00 3.50
2.00 0.30 5.00
1.03 0.27 2.35
0.02 0.24 0.38 0.14
0.04 0.78 0.12 0.06 4.50 135.00
0.49 0.16 0.18 1.38 0.13
0.33 0.27 0.60 0.44 0.06 5.35 160.50
0.00 0.00 2.13 3.61 0.48 0.00 0.00 0.32 0.40 0.14
0.00 0.00 0.86 1.51 0.22 0.00 0.00 0.14 0.18 0.08 2.99 89.70
0.00 7.44
0.00 4.09
2.00 0.34 4.66
1.07 0.34 2.56
0.06 0.24 0.38 0.14
0.05 0.79 0.16 0.08 5.17 155.10
0.49 0.16 0.19 1.39 0.13
0.40 0.50 0.63 0.60 0.07 6.17 185.10
Note: Scenario I: Food items consumed must be greater than or equal to zero. Scenario II requires the constraint that not only are food items consumed greater than zero, but also that no ragi, jowar, bajra, millets, and coconut are consumed. Scenario III incorporates, in addition to the constraints under scenarios I and II, that at least 200 g of rice per capita per person is consumed. Source: Authors Estimates.
Spinach, 40 grams of coconut, and 14 grams of guava. Taking their respective unit prices in rural and urban areas separately, the total expenditure comes to be Rs. 2.44 per person per day, i.e. Rs. 73.14 per person per month in rural areas and Rs. 2.99 per person per day, i.e. Rs. 89.66 per person per month in urban areas. These figures indicate the minimum required expenditure by households per capita per month in order to secure the minimum level of nutrition. However, it is likely that the optimum consumption basket is so uninteresting that no one can be expected to accept that as a balanced diet. In particular, the coconut consumption seems quite a bit excessive and the absence of rice and wheat from the diet appears unrealistic, given that the consumption of coarse grains such as jowar and bajra is low in India at the present time. Hence, scenario II requires that the
consumption of coarse grains such as jowar, bajra, and ragi and of coconut to be equal to zero in the linear programming problem. In the revised solution, the optimal consumption basket includes wheat, milk and oil in different proportions both in rural as well as urban areas. This solution gives a more diversified consumption basket relative to that under scenario I, but with the imposition of additional constraints the cost is a bit higher. Now the minimum food expenditure required by an individual is Rs 135 and Rs 155.10 in rural and urban areas per person per month, respectively. Figure 1 summarizes the estimates of the food poverty line under the three scenarios at the all India level for 1993-94 and 1999-2000, and for rural and urban areas separately. Scenario III introduced the additional constraint of a minimum consumption of 200 grams of rice (or related prodFinancing and Delivery of Health Care Services in India
25
SECTION I
Table 4 Three different scenarios of optimum consumption basket and minimum expenditure required for recommended nutritional intakes in rural and urban areas in 1999-2000.
Region and food items Price in Rs per 100g Amount (x 100 g) Scenario I Total expenditure (Rs) Amount ( x 100 g) Scenario II Total expenditure (Rs) Amount (x 100 g) Scenario III Total expenditure (Rs)
RURAL Rice 1.09 Khoi, lawa Wheat/atta 0.82 Bajra and products 0.68 Ragi and products 0.64 Milk: liquid 1.06 Other milk products 3.19 Vanaspati, margarine 4.10 Mustard oil 4.26 Groundnut oil 4.14 Spinach/other leafy vegetables 0.59 Coconut 0.86 Guava 0.73 Groundnut 2.99 Per capita per day total expenditure required Per capita per month total expenditure required URBAN Rice 1.33 Khoi, lawa Wheat/atta 0.94 Bajra and products 1.69 Ragi and products 0.78 Milk: liquid 1.24 Other milk products 6.09 Vanaspati, margarine 4.18 Mustard oil 4.24 Groundnut oil 4.27 Spinach/other leafy vegetables 0.81 Coconut 0.91 Guava 0.89 Per capita per day total expenditure required Per capita per month total expenditure required
0.00 0.00 0.00 5.99 0.34 0.00 0.00 0.00 0.00 0.00 0.29 0.15 0.15 0.08
0.00 0.00 0.00 4.06 0.22 0.00 0.00 0.00 0.00 0.00 0.17 0.13 0.11 0.23 4.92 147.6
0.00 0.00 7.44
0.00 0.00 6.08
2 0.00 4.90
2.19 0.00 4.00
0.00 0.02 0.06 0.00 0.18 0.38 0.14 0.00
0.00 0.07 0.24 0.00 0.75 0.22 0.10 0.00 7.46 223.8
0.63 0.13 0.23 0.81 0.08
0.67 0.42 0.99 0.48 0.06 8.81 264.3
0.00 0.00 2.47 0.00 4.09 0.00 0.00 0.00 0.00 0.00 0.49 0.73 0.12
0.00 0.00 2.33 0.00 3.19 0.00 0.00 0.00 0.00 0.00 0.39 0.66 0.11 6.68 200.40
0.00 0.00 7.44
0.00 0.00 7.01
2 0.02 4.87
2.66 0.05 4.59
0.07 0.00 0.13 0.11 0.00 0.37 0.14
0.09 0.00 0.56 0.46 0.00 0.30 0.12 8.54 256.20
1.20 0.00 0.00 0.24 0.00 0.83 0.00 0.08
1.49 0.00 0.00 1.03 0.00 0.67 0.00 0.07 10.56 316.80
Scenario I: Food items consumed must be greater than or equal to zero. Scenario II requires the constraint that not only are food items consumed greater than zero, but also that no ragi, jowar, bajra, millets, and coconut are consumed. Scenario III incorporates, in addition to the constraints under scenarios I and II, that at least 200 g of rice per capita per person is consumed. Source: Authors Estimates.
ucts) per person per day. The resulting solution for the consumption basket included rice, wheat, oil, milk, and guava (fruits) and is indicated in the last two columns of Tables 3 and 4. The total expenditure required to command this food basket came to be Rs 160 per person per month in rural areas and Rs 185 per person per month in urban areas for 1993-94. The solution under scenario III was taken to be the FPL for 1993-94 in Table 2. Similarly, the solution under scenario III for the FPL during 1999-2000 was Rs 264 per person per month in rural areas and Rs 317 per person per month in urban areas. It is interesting to note that in the solution of the linear
programming exercise under scenario I (i.e. without imposing any additional constraint other than the RDA itself) in 1999-2000 gives exactly the same composition of the consumption basket as in 1993-94, with the sole difference being that jowar in 1993-94 is replaced by groundnut in rural and by wheat in urban areas in 1999-2000. Under scenarios II and III, milk and milk products figure in the 1999-2000 consumption basket prominently indicating that the relative prices of milk and milk products may have declined between 1993-94 and 1999-2000. In addition to milk, the optimal 1999-2000 basket contains vanaspati, gram and other edible oils.
26
SECTION I
In general, a comparison of the least cost consumption basket of 1993-94 with that of 1999-2000 (scenario III in both cases) indicates that the latter has a higher proportion of expenditures allocated to cereals. This occurs presumably because of the comparatively slower price increase of cereals between 1993-94 and 1999-2000, relative to non-cereal food item prices during the same period. However, since the consumption of cereals does not suffice for a balanced diet, the least cost expenditure required for achieving nutritional requirements in 1999-2000 consumption basket may be much more (in real terms) than the basket of 1993-94.
Fig 1 Estimates of food poverty line (in Rs) under three scenarios at the all India level in rural and urban areas for 1993-94 and 1999-2000
Source: Authors Estimates.
State-specific poverty lines and poverty ratios

To derive State-, rural- and urban-specific PLs, FPLs, estimated on the basis of methods discussed earlier are scaled up by using non-food expenditure ratios for the year 1993-94. Poverty line estimates for the three scenarios outlined above at the all-India level are given in Table 5, for 1993-94 and 1999-2000. Each of the FPL and PL in Table 5 refers to a specific cultural' constraint (reflecting the three scenarios).
The three FPLs (FPL1, FPL2 and FPL3) indicate three different levels of cut-off of the minimum expenditure required for a nutritionally balanced food basket for the three different sets of cultural constraints. The three PL (PL1, PL2 and PL3) on the other hand, indicate the corresponding minimum expenditure required to cover both the nutritional minimum, as well as the minimum non-food allowance. We prefer to use PL3 for subsequent analyses, although PL1 or PL2 may also be used. Using the standardized RDA and the added constraints for scenario III, we estimated State-, rural- and urban-specific FPLs. Then, using State-, ruraland urban-specific non-food expenditure ratios of the marginally poor, we arrived at the corresponding PLs. The PLs and the head-count poverty ratios for 17 major States in rural and urban areas for two years (1993-94 and 1999-2000) are given in Tables 6 and 7. These estimates were also used to derive the all India level poverty ratios for rural and urban populations and the overall poverty ratio. Corresponding official estimates for poverty ratios produced by the Planning Commission are given in Table 8. The national level poverty ratios under both approaches produced by aggregating the population living in poverty at the State level (nutrition and the Planning Commission) are further brought together in Table 9. On comparing the two methods (i.e. nutritionally balanced' and that of the Planning Commission) we find that the PLs as well as the poverty ratios are much higher under the nutritionally balanced' approach than the Planning Commission approach. During 1993-94 and 1999-2000, the all-India poverty ratio declined from 37.3% to 27.1% in the rural population; and from 32.4% to 23.6% in the urban population.2 The head-count ratio of nutrition-based poverty also shows a decline in 1999-2000, compared to 1993-94. However, the head-count ratio of nutritional poverty was 38.8% in rural populations and 27.5% in urban populations in 1999-2000 as against the official poverty ratios 27.1% and 23.6% among rural and urban populations, respectively in the same period. The ratios of nutritional poverty are higher than those of official poverty both in 1993-94 and 1999-2000. Apart from the fact that the nutritional poverty is much
Table 5 Food poverty lines, non-food expenditure ratios and poverty lines under different scenarios for rural and urban India, 1993-94 and 1999-2000
FPL1 FPL2 FPL3 NFE ratio PL1 PL2 PL3
1993-94
Rural Urban 73.14 89.66 134.85 155.26 160.20 185.17 0.27 0.33 100.19 133.82 184.73 231.73 219.45 276.37
1999-2000
Rural Urban 147.47 200.44 223.94 256.30 264.09 316.76 0.27 0.33 202.01 299.16 306.77 382.54 361.77 472.78
NOTE: 1, 2 and 3 indicate scenario I, II and III, respectively. Scenario I: Food items consumed must be greater than or equal to zero. Scenario II requires the constraint that not only are food items consumed greater than zero, but also that no ragi, jowar, bajra, millets, and coconut are consumed. Scenario III incorporates, in addition to the constraints under scenarios I and II, that at least 200 g of rice per capita per person is consumed. FPL: food poverty line; NFE: ratio: non-food expenditure; PL: poverty line. Source: Authors Estimates.
27
SECTION I
Table 6 Food (nutrition-based) poverty line, non-food expenditure ratios and poverty line for 17 major States, and rural and urban areas in India for 1993-94 and 1999-2000
State 1993-1994 Rural Urban Food poverty line 1999-2000 Rural Urban NF expenditure ratio 1993-94 Rural Urban 1993-1994 Rural Urban Poverty line 1999-2000 Rural Urban
Andhra Pradesh Assam Bihar Gujarat Haryana Himachal Pradesh Jammu and Kashmir Karnataka Kerala Madhya Pradesh Maharashtra Orissa Punjab Rajasthan Tamil Nadu Uttar Pradesh West Bengal All India
Source: Authors Estimates.
156.23 179.81 164.84 159.25 164.56 172.08 173.23 141.25 178.68 151.34 154.98 154.10 168.25 158.22 152.14 160.85 175.14 160.20
175.57 193.88 172.09 195.31 171.85 178.19 202.05 213.25 212.21 205.43 213.69 209.62 179.92 183.40 203.25 178.60 178.90 185.17
260.69 269.85 264.40 232.46 283.37 293.06 294.76 254.70 303.08 239.42 246.11 264.07 294.12 268.08 275.68 266.99 283.35 264.09
302.77 288.29 271.58 258.94 309.01 332.63 325.53 342.84 358.96 326.83 349.88 318.74 296.31 328.16 368.20 302.41 301.30 316.76
0.28 0.25 0.25 0.26 0.31 0.29 0.26 0.31 0.29 0.29 0.31 0.24 0.30 0.32 0.26 0.29 0.24 0.27
0.34 0.22 0.28 0.30 0.36 0.35 0.23 0.34 0.28 0.35 0.35 0.30 0.31 0.34 0.32 0.32 0.29 0.33
216.99 239.75 219.78 215.20 238.49 242.36 234.10 204.71 251.66 213.15 224.61 202.76 240.36 232.67 205.59 226.55 230.45 219.45
266.01 248.57 239.01 279.02 268.51 274.14 262.40 323.10 294.74 316.05 328.75 299.45 260.76 277.88 298.90 262.65 251.97 276.37
362.00 360.37 354.38 369.24 409.65 413.48 398.53 369.19 428.13 336.68 356.76 348.55 423.87 395.02 372.69 369.29 372.40 359.45
458.59 365.21 376.39 461.76 482.82 512.21 422.81 520.62 499.85 505.23 537.91 461.93 425.33 497.11 532.82 444.98 425.21 466.28
higher in comparison to the official poverty in both rural and urban areas, the decline in the nutritional poverty has been
much lower than the official poverty between 1993-94 and 1999-2000. As against a decline of approximately 10 percentage points in official poverty estimates over the period 1993-94 and Table 7 1999-2000, the decline in nutritional Poverty ratios using the nutrition poverty line in India, poverty has been approximately 6% dur1993-94 and 1999-2000 ing the same period (Fig. 2). As in the case of the all-India average, 1993-94 1999-2000 the decline in the nutrition poverty ratios State Rural Urban Rural Urban
Andhra Pradesh Assam Bihar Gujarat Haryana Himachal Pradesh Jammu and Kashmir Karnataka Kerala Madhya Pradesh Maharashtra Orissa Punjab Rajasthan Tamil Nadu Uttar Pradesh West Bengal All India
Source: Authors Estimates. Source: Authors Estimates.
42.24 49.23 60.51 27.61 29.63 33.01 18.33 38.59 27.89 49.82 49.89 55.19 13.32 33.34 36.98 47.53 45.51 45.28
35.08 15.26 35.08 23.87 18.99 11.49 6.72 44.80 27.50 47.81 35.13 41.25 12.78 30.20 40.39 36.58 23.81 33.83
38.36 38.31 51.52 22.49 14.04 15.48 12.15 32.22 16.58 44.13 32.44 54.97 12.95 23.51 37.29 40.23 37.00 38.27
27.61 9.58 34.59 12.71 15.75 7.71 3.37 25.75 22.62 44.34 26.49 40.48 8.25 24.38 29.19 35.11 16.84 27.54
Fig 2 Relative decline in the nutritional and official poverty ratios
28
SECTION I
Table 8 Poverty line and head-count ratio of poverty by the Planning Commission estimates
State 1993-94 Rural Urban Poverty Line (in Rs) 1999-2000 Rural Urban 1993-94 Rural Urban Poverty ratio (%) 1999-2000 Rural Urban
Andhra Pradesh Assam Bihar Gujarat Haryana Jammu and Kashmir Karnataka Kerala Madhya Pradesh Maharashtra Orissa Punjab Rajasthan Tamil Nadu Uttar Pradesh West Bengal All India
SOURCE: Planning Commission
163.02 232.05 212.16 202.11 233.79 233.79 186.63 243.84 193.1 194.94 194.03 233.79 215.89 196.53 213.01 220.74 205.84
278.14 212.42 238.49 297.22 258.23 253.61 302.89 280.54 317.16 328.56 298.22 253.61 280.85 296.63 258.65 247.53 281.35
262.94 365.43 333.07 318.94 362.81 367.45 309.59 374.79 311.34 318.63 323.92 362.68 344.03 307.64 336.88 350.17 327.56
457.4 343.99 379.78 474.41 420.20 420.20 511.44 477.06 481.65 539.71 473.12 388.15 465.92 475.6 416.29 409.22 454.11
15.92 45.01 58.21 22.18 28.02 30.34 29.88 25.76 40.64 37.93 49.72 11.95 26.46 32.48 42.28 40.8 37.27
38.33 7.73 34.5 27.89 16.38 9.18 40.14 24.55 48.38 35.15 41.64 11.35 30.49 39.77 35.39 22.41 32.36
11.05 40.04 44.3 13.17 8.27 3.97 17.38 9.38 37.06 23.72 48.01 6.35 13.74 20.55 31.22 31.85 27.09
26.63 7.47 32.9 15.59 9.99 1.98 25.25 20.27 38.44 26.81 42.83 5.75 19.85 22.11 30.89 14.86 23.62
across most of the major States has been much slower in comparison with the official poverty ratios (Table 10). A comparison of changes in the two poverty ratios across States shows that in 10 of the 16 major States (viz. Andhra Pradesh, Bihar, Gujarat, Haryana, Jammu and Kashmir, Punjab, Rajasthan,
Table 9 Head-count ratios of nutritional and official poverty among rural and urban populations in India, 1993-94 and 1999-00
Region Nutritional poverty ratio(%) 1993-94 1999-2000 Official poverty ratio(%) 1993-94 1999-2000
Rural Urban Combined
45.28 33.83 42.27
38.77 27.51 35.65
37.27 32.36 35.97
27.09 23.62 26.10
Source: Authors Estimates and Planning Commission.
Tamil Nadu, Uttar Pradesh and West Bengal), the nutrition poverty ratios have declined at a slower rate in both rural and urban areas. Further, in 3 states (viz. Karnataka, Kerala and Orissa) rural nutritional poverty has declined at a rate slower than officially estimated rural poverty ratios. In Madhya Pradesh, the nutritional poverty ratio has declined at a faster rate in
rural areas but at a slower rate in urban areas. Assam is the only State that shows a faster decline in the nutritional poverty ratio in comparison with the official poverty ratio, both in rural as well as urban areas. A comparison of decline in nutritional and official poverty ratios in rural areas of 16 major States between 1993-94 and 1999-2000 is presented in Fig. 3. The comparison of the two PLs across States also indicates that the difference between the two estimates has been highest in Andhra Pradesh followed by Maharashtra, Madhya Pradesh and Rajasthan, particularly in rural areas. The case of Andhra Pradesh needs special mention. Official estimates of the rural poverty ratio in Andhra Pradesh have been a subject of controversy because the Planning Commission estimates much lower poverty in rural than in urban Andhra Pradesh. However, the nutritional poverty ratio in Andhra Pradesh is not only much higher than the official poverty ratio but also rural Andhra shows higher poverty ratio than the urban Andhra Pradesh. Similarly, in rural Jammu and Kashmir, the official poverty ratio shows a drastic decline between 1993-94 and 1999-2000 from more than 30% in 1993-94 to less than 4% in 1999-2000. The nutritional poverty ratio, on the other hand shows a systematic decline in rural poverty in Jammu and Kashmir from 18% in 1993-94 to 12% in 1999-2000.
2 The official estimates of poverty ratios in 1999-2000 have been a subject of intense discussion on account of their different methodology of recall period. Most of these discussions have centred on the magnitude of and intensity of decline in poverty as reported by the Planning Commission (2001) based on the 55th round of the National Sample Survey (NSS). For a review of this literature see Visaria 2000; Deaton 2001; Deaton and Dreze 2002; Kozel and Parker 2002, Sen and Himanshu 2004; and Sundaram 2003
29
SECTION I
Fig 3 Decline in nutritional and official poverty ratios (rural) between 1993-94 and 1999-2000 in various States in India
AP: Andhra Pradesh, AS: Assam, BI: Bihar, GU: Gujarat, HA: Haryana, HP: Himachal Pradesh, J&K: Jammu and Kashmir, KA: Karnataka, KE: Kerala, MA: Maharashtra, MP: Madhya Pradesh, OR: Orissa, PU: Punjab, RA: Rajasthan, TN: Tamil Nadu, UP: Uttar Pradesh, WB: West Bengal Source: Authors Estimates and Planning Commission.
Table 10 Decline in the poverty ratios between 1993-94 and 1999-2000 in rural and urban areas in the States of India
State Nutritional Rural Urban Official Rural Urban
are able to purchase these commodities. Theoretically, this ought to lead to a setting where a much smaller cultivable area is devoted to food commodities that are relatively richer (relative to cereals) in other nutrients that are components of the RDA, with obvious implications for affordability.
Conclusions
The main aim of this paper is to present the case for and to develop an indicator of poverty for India that highlights the need to achieve a balanced diet-in terms of a minimum set of required nutrients. Such an indicator is valuable both to take account of our increasing knowledge about the role of micro-nutrient consumption in influencing health outcomes; as well as to begin the process of examining the question of how the policies of the government on agricultural prices may have affected the health of Indians in general and of the poor, in particular. Our main findings are the following. First, estimates of PLs that focus on the expenditures needed to achieve a nutritionally balanced diet are readily constructed and typically are higher than the official PL. Thus, poverty ratios based on the nutrition-adjusted PL exceed official estimates of head-count poverty. Second, trends over the period 1993-94 to 1999-2000 suggest that poverty ratios based on the nutrition-adjusted PL declined more slowly than poverty ratios based on the official PL given by the Planning Commission. Third, there were considerable inter-State and regional (urban versus rural) differences in the poverty ratios in the 1990s. The incidence of nutritionally poor population is highest in Orissa and Bihar, followed by Madhya Pradesh, Uttar Pradesh and Andhra Pradesh. The official estimates show a higher poverty ratio (and also the absolute number of poor) in urban Andhra Pradesh compared to rural Andhra Pradesh. Similarly, the poverty ratio in Jammu and Kashmir is as low as
Andhra Pradesh Assam Bihar Gujarat Haryana Jammu and Kashmir Karnataka Kerala Madhya Pradesh Maharashtra Orissa Punjab Rajasthan Tamil Nadu Uttar Pradesh West Bengal
3.68 10.91 8.79 5.12 15.59 6.18 6.37 11.31 5.69 17.45 0.22 0.37 9.83 -0.31 7.3 8.51
7.26 5.68 0.48 11.16 3.24 3.35 19.05 4.88 3.47 8.64 0.77 4.53 5.82 11.2 1.47 6.97
4.87 11.7 4.97 0.26 13.91 1.6 9.01 12.3 19.75 6.39 26.37 7.2 12.5 14.89 16.38 4.28 3.58 9.94 14.21 8.34 1.71 -1.19 5.6 5.6 12.72 10.64 11.93 17.66 11.06 4.5 8.95 7.55
Source: Authors Estimates and Planning Commission.
A slower decline in the nutritional poverty ratio in comparison to the official poverty ratio suggests at least prima facie, the need for examining carefully the potential role that relative prices might play in influencing the nutritional well-being of people. This is particularly so in India, where the State has long subsidized, on the one hand, cereal production (wheat and rice) through various price support schemes while simultaneously subsidizing prices at which consumers
30
SECTION I
approximately 3% in the official estimates registering a decline of more than 25 percentage points during 1993-94 to 1999-2000. Our estimates of the least cost balanced diet-based poverty ratio show more consistent results in these two States. Kerala shows much lower incidence of nutritional poverty compared to that of Karnataka and Tamil Nadu. These trends and estimates raise obvious policy issues that need further examination and lie well beyond the scope of this exploratory paper. In particular, the paper strengthens arguments of those who state that increases in incomes alone are not enough to eliminate poverty and malnutrition quickly; that relative prices of essential nutrients may also need policy attention. This, in turn, may require additional attention to government policies with respect to the prices of cereals such as rice and wheat.
Although our paper focuses on using a PL approach to estimate the lack of affordability of nutrients, other authors have taken an alternative route that equates inadequate nutrition with poverty. We have shown that, in the context of a balanced diet of macro- and micronutrients, the two approaches give rise to markedly different results, even if their general direction is the same. We conclude that a lack of purchasing power (and not simply choice) offers much in terms of developing an understanding of the roots of inadequate nutrient intake in India.
Acknowledgements:
Alok Bhargava, K. Sujatha Rao, S. Sakthivel, Himanshu, Sandip Sarkar, J.V. Meenakshi
31
SECTION I
References
Behrman JR, Deolalikar AB. Will developing country nutrition improve with income? A case study of rural south India. Journal of Political Economy 1987:95:108-38. Bhargava A. Estimating short- and long-run income elasticities of foods and nutrients for rural south India. Journal of the Royal Statistical Society Series A, 1991;154:157-74. Bhargava A. Modelling the effects of nutritional and socioeconomic factors on the growth and morbidity of Kenyan school children. Am J Hum Biol 1999;11:317-26. Bhargava A, Guthrie J. Unhealthy eating habits, physical exercise and macronutrient intakes are predictors of anthropometric indicators in the Womens Health Trial: Feasibility study in minority populations. British Journal of Nutrition 2002;88:719-28. Bhargava A, Jamison D, Lau L, Murray C. Modeling the effects of health on economic growth. Journal of Health Economics 2001;20:423-40. Bloom D, Canning D. The health and wealth of nations. Science 2000;287:1207-9. Dandekar VM. Population, poverty and employment. New Delhi: Sage Publications; 1996. Datt G, Ravallion M. Is India's economic growth leaving the poor behind? Journal of Economic Perspectives 2002;16:89-108. Deaton A. Adjusted Indian poverty estimates for 19992000 (draft). Paper presented at the Planning Commission/ World Bank. Workshop on Poverty Measurement, Monitoring, and Evaluation., 2001, New Delhi. Deaton A, Dreze J. Poverty and inequality in India: A re-examination. Economic and Political Weekly 2002. September : 3729-3748 Deaton A, Muellbauer J. Economics and consumer behaviour. Cambridge, United Kingdom: Cambridge University Press; 1980. Gopalan. C,Rama Shashtri BV and Balasubramanian SC. Nutritive value of Indian foods. Hyderabad: National Institute of Nutrition; 1989. Government of India, Statistical Abstract, 2000-2001, Ministry of Statistics and Programme Implementation, New Delhi, India; 2000. Government of India. Economic Survey 2003-4. New Delhi, India: Ministry of Finance; 2004. Gray D. Frail survivors rescued. The Tampa Tribune, 4 January 2005. Available from URL: http://www.tampatrib.com/News/MGBZ1EZ7K3E.html. Indian Council of Medical Research (ICMR). Nutrient Requirements and Recomended Dietary Allowances for Indias. Indian Council of Medical Research. 2002. Jamison D, Leslie J, Musgrove P. Malnutrition and dietary protein: Evidence from China and from international comparisons. Food and Nutrition Bulletin 2003;24:145-54. Kozel V, Parker B. A profile and diagnostic of poverty in Uttar Pradesh. Paper presented in a seminar, at the National Council for Applied Economic Research (NCAER), 2002, New Delhi, (mimeo). Lal D, Mohan R, Natarajan I. Economic reforms and poverty alleviation: A tale of two surveys. Economic and Political Weekly 2001;36:1017-28. Lanjouw J. Demystifying poverty lines (draft); 1997 Available from URL: http://www.undp.org/ poverty/publications/pov_red/Demystifying_Poverty_Lines.pdf#s earch='jean%20%20Lanjouw%20demystifying%20poverty%20lines'. Mahendradev S, Ravi C, Viswanathan B, Gulati A, Ramachander S. Economic liberalisation, targeted programmes and household security: A case study of India. Washington, D.C: International Food Policy Research Institute; 2004. Malhotra R. Incidence of poverty in India: Towards a consensus on estimating the poor. The Indian Journal of Labour Economics 1997;40:67-102. Martorell R. Commentary 3. Food and Nutrition Bulletin 2003;24:158-9. Planning Commission. Poverty in India (Press release). New Delhi: Planning Commission, Government of India; 2001. Pritchett L, Summers L. Wealthier is healthier. Journal of Human Resources 1996;31:841-68. Rao VKRV. Nutritional norms by calorie intake and measurement of poverty. Bulletin of the International Statistical Institute Proceedings of the 41st Session, December 1997;XLVII.
32
SECTION I
Ravallion M, Bidani B. How robust is a poverty profile. World Bank Economic Review 1994;8:75-102. Rugger JP, Jarrison DT, Bloom DE. Health and the Economy. In Merson MH, Black RE, Mills AJ (eds.) International Public Health. Gaithersburg, Aspen. 2001. Subramanian, A. Are INcome Calories Elasticitys really high in developing countries?: Some implications for nutrition and income. National Council for Applied Economic Research, New Delhi. 2001. Sen A. Commodities and capabilities, North-Holland, Amsterdam; 1985. Sen A, Himanshu. Poverty and inequality in India: Getting closer to the truth. Economic and Political Weekly 2004. September: 4247-4263 Sundaram K. Poverty has declined in the 1990s: A resolution of comparability problems in NSS consumer expenditure data. Economic and Political Weekly 2003. January: 327-337.
Stigler G. The cost of subsistence. Journal of Farm Economics 1945;27:303-14. Sukhatme P. The protein problem, its size and nature. Journal of the Royal Statistical Society Series A 1974;137:166-99. Sukhatme P. Malnutrition and poverty. The 9th Lal Bahadur Shastri Memorial Lecture, New Delhi: Indian Agricultural Research Institute; 1977. Sukhatme P. Assessment of adequacy of diets at different income levels. Economic and Political Weekly 1978; Special Number, August. Visaria P. Poverty in India during 1994-98: Alternative Estimates, processed, Delhi: Institute of Economic Growth; 2000. Willett W. Nutritional epidemiology. 2nd Edition. New York: Oxford University Press; 1998.
33
SECTION I
Appendix I Nutrient composition of food items (Nutrient content per 100 grams of different food items)
Food item Calories Protein Fat Iron Calcium Carotene Riboflavin Thiamin Niacin Vitamin C
Rice Chira Muri Other rice products Wheat/atta Maida Suji, rawa Sewai, noodles Bread (bakery) Other wheat products Arhar (tur) Gram (whole) Gram (split) Moong Masur Urad Peas Soya bean Kesari Gram products Besan Other pulse products Milk: liquid (litre) Baby food Milk: condensed/ powder Curd Ghee Butter Other milk products Vanaspati, margarine Mustard oil Groundnut oil Coconut oil Fish, prawn Goat meat/mutton Beef/ buffalo meat Pork Chicken Others (birds, crab, oyster, tortoise, etc.) Potato Onion Radish Carrot Turnip Beet Sweet potato Arum Pumpkin Gourd Bitter gourd
345 346 325 346 290 348 348 352 244 245 335 372 360 348 323 347 315 432 345 369 372 336 117 67 357 60 850 729 421 900 900 900 900 219 156 86 114 109 130.7 97 50 32 48 29 43 120 120 25 12 25
7 7 14 6 12 11 10 9 9 8 22 21 17 25 24 24 20 43 28 23 21 23 4 3 38 3 0 0 15 0 0 0 0 43.5 20 19 19 26 17 1.6 1 1 1 1 2 1 1 1 0 2
1 1 16 1 2 1 1 0 1 1 2 6 5 1 1 1 1 20 1 5 6 4 7 4 0 4 100 81 31 100 100 100 100 5 8.5 1 4 1 5.7 0.1 0 0 0 0 0 0 0 0 0 0
1 20 35 1 5 3 2 2 2 1 3 5 5 4 9 4 7 10 6 10 5 6 0 0 1 0 0 0 6 0 0 0 0 2.5 1.5 0.8 2 0 0 0.5 1 0 1 0 1 0 0 0 0 1
10 20 67 9 48 23 16 22 18 11 73 56 202 75 77 154 75 240 90 58 56 101 210 120 1370 149 0 0 650 0 0 0 0 500 81 3 30 25 542 10 47 50 80 30 18 46 46 10 20 20
9 0 0 2 29 25 0.15 0.12 0 0 132 129 189 49 12 38 39 426 120 113 129 12 48 420 0 31 600 960 500 750 162 37 0 0 9 18 0 0 425 24 15 3 6460 0 0 1810 6 1160 0 126
0.12 0.05 0.01 0.05 0.17 0.07 0.03 0.05 0.17 0.17 0.19 0.18 0.15 0.21 0.2 0.2 0.19 0.39 0.17 0 0.18 0.2 0.1 1.36 1.64 0.16 0 0 0.41 0 0.26 0.13 0.01 0.1 0.14 0.04 0.09 0.14 0.4 0.01 0.01 0.02 0.02 0.04 0.09 0.04 0.04 0.04 0.01 0.09
0.06 0.21 0.21 0.21 0.49 0.12 0.12 0.19 0.49 0.49 0.45 0.48 0.3 0.47 0.51 0.42 0.47 0.73 0.39 0.2 0.48 0.51 0.04 0.31 0.45 0.05 0 0 0.23 0 0.65 0.9 0.08 0 0.18 0.15 0.54 0 0.1 0.1 0.08 0.06 0.04 0.04 0.04 0.08 0.08 0.06 0.03 0.07
2.5 4 4.1 3.8 4.3 2.4 1.6 1.8 4.3 4.3 2.9 2.4 2.9 2.4 1.3 2 3.4 3.2 0.17 1.3 2.4 1.3 0.1 0.8 1 0.1 0 0 0.4 0 0.4 0 3 2.1 0 5.8 2.8 0 0.1 1.2 0.4 0.4 0.6 0.5 0.4 0.7 0.7 0.5 0.2 0.5
0 0 0 0 0 0 0 0 0 0 0 1 3 0 0 0 0 0 0 0 1 0 1 2 5 1 0 0 6 0 0 0 0 15 0 0 2 0 0 17 11 17 3 43 10 24 24 2 0 88
34
SECTION I
Appendix I
Food item
Calories
Protein
Fat
Iron
Calcium
Carotene
Riboflavin
Thiamin
Niacin
Vitamin C
Cucumber Parwal/patal Jhinga/torai Snake gourd Cauliflower Cabbage Brinjal Lady's finger Spinach/other leafy vegetables French beans and barbati Tomato Peas Chillies (green) Capsicum Plantain (green) Jackfruit (green) Other vegetables Jackfruit Watermelon Pineapple Guava Orange, Mausambi Mango Watermelon Pear (Naspati) Berries Apple Grapes Coconut (Kopra) Groundnut Dates Cashewnut Walnut Raisin (kishmish, monacca etc.) Other dry fruits Sugar Gur Candy (misri) Honey Turmeric (gm) Black pepper (gm) Dry chillies (gm) Garlic (gm) Tamarind (gm) Ginger (gm) Curry powder (gm) Other spices (gm)
SOURCE: Adapted from Gopalan et al. 1998
13 20 17 18 30 27 24 35 26 26 21 93 29 24 64 51 25 88 16 46 51 48 74 17 52 49.7 59 51.5 662 567 317 596 687 308 687 398 383 398 319 349 304 246 145 283 67 108 250
0 2 1 1 3 2 1 2 2 2 1.3 7 3 1 1 3 2 2 0 0 1 1 1 0 1 1 0 1 7 25 3 21 16 2 16 0 0 0 0 6 12 16 6 3 2 6 8
0 0 0 0 0 0 0 0 1 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0.3 1 0 62 40 0 47 65 0 65 0 0 0 0 5 7 6 0 0 1 1 7
1 2 0 2 1 1 0 0 1 1 1.3 2 4 1 6 2 1 1 8 2 0 0 1 1 1 2 1 0.5 8 3 7 6 3 8 3 0 3 0 1 68 12 2 1 17 4 1 25
10 30 18 26 33 39 18 66 73 50 38.7 20 30 10 10 30 34 20 11 20 10 26 14 32 8 46.7 10 22.5 400 90 120 50 100 87 100 12 80 12 5 150 460 160 30 170 20 830 120
0 153 33 96 30 120 74 52 5580 132 351 83 1007 427 30 175 325.3 130 0 18 0 1104 1990 169 28 1248 0 3 0 37 26 60 6 2.4 0 0 0 0 0 30 1080 345 0 60 40 7560 304
0 0.06 0.01 0.06 0.1 0.09 0.11 0.1 0.26 0.06 0.06 0.01 0.39 0.05 0.02 0.13 0.06 0.13 0.04 0.12 0.03 0 0.09 0.08 0.03 0.13 0 0.03 0.01 0.13 0.02 0.19 0.4 0.19 0 0 0 0 0 0 0.14 0.43 0.23 0.07 0.03 0.21 0.19
0.03 0.05 0 0.04 0.04 0.06 0.04 0.07 0.03 0.08 0.12 0.25 0.19 0.55 0.05 0.03 0.08 0.03 0.02 0.2 0.03 0 0.08 0.11 0.06 0.04 0 0.12 0.08 0.9 0.01 0.63 0.45 0.07 0 0 0 0 0 0.03 0.09 0.93 0.06 0 0.06 0.08 0.22
0.2 0.5 0.2 0.3 1 0.4 0.9 0.6 0.5 0.3 0.4 0.8 0.9 0.1 0.3 0.4 0.46 0.4 0.1 0.1 0.4 0 0.9 0.3 0.2 0.5 0 0.2 3 19.9 0.9 1.2 1 0.7 0 0 0 0 0 2.3 1.4 9.5 0.4 0.7 0.6 2.3 1.2
7 29 5 0 56 124 12 13 28 24 27 9 111 137 24 7 30.55 7 1 39 212 30 16 26 0 30 1 31 7 0 3 0 0 1 0 0 0 0 0 0 0 50 13 3 6 4 50
35
SECTION II Delivery of Health Care Services in India
SECTION II
Primary Health Care in India: Review of Policy, Plan and Committee Reports
MADHURIMA NUNDY
CENTRE OF SOCIAL MEDICINE AND COMMUNITY HEALTH, JAWAHARLAL NEHRU UNIVERSITY, NEW DELHI, INDIA E-MAIL: madhurima.nundy@gmail.com
RIMARY HEALTH CARE' IS A TERM THAT IS USED EXTENSIVELY WORLDWIDE BY policy-makers. What does this term imply? Is it merely a term or does it hold within it a much wider and deeper significance to the concept of health? The Alma Ata Declaration in 1978 gave an insight into the understanding of primary health care. It viewed health as an integral part of the socioeconomic development of a country. It provided the most holistic understanding to health and the framework that States needed to pursue to achieve the goals of development. The Declaration recommended that primary health care should include at least: education concerning prevailing health problems and methods of identifying, preventing and controlling them; promotion of food supply and proper nutrition, and adequate supply of safe water and basic sanitation; maternal and child health care, including family planning; immunization against major infectious diseases; prevention and control of locally endemic diseases; appropriate treatment of common diseases and injuries; promotion of mental health and provision of essential drugs. It emphasized the need for strong first-level care with strong secondary- and tertiary-level care linked to it. It called for an integration of preventive, promotive, curative and rehabilitative health services that had to be made accessible and available to the people, and this was to be guided by the principles of universality, comprehensiveness and equity. In one sense, primary health care reasserted the role and responsibilities of the State, and recognized that health is influenced by a multitude of factors and not just the health services. It also recognized the need for a multisectoral approach to health and clearly stated that primary health care had to be linked to other sectors. At the same time, the Declaration emphasized on complete and organized community participation, and ultimate self-reliance with individuals, families and communities assuming more responsibility for their own health, facilitated by support from groups such as the local government, agencies, local leaders, voluntary groups, youth and women's groups, consumer groups, other non-governmental organizations, etc. The Declaration affirmed the need for a balanced distribution of available resources (WHO 1978). Keeping this definition in mind, we now discuss whether this holistic concept has been utilized as a framework to guide policy-makers to develop various health policy documents, health committee reports and the five-year plans since Independence so as to impact on the health system. After Independence, India adopted the welfare state approach, which was dominant worldwide at that time. As with most post-colonial nations, India too attempted to restructure its patterns of investment. During that time, India's leaders envisaged a national health system in which the State would play a leading role in determining priorities and financing, and provide services to the population. If it were possible to evaluate the loss, which this country annually suffers through the avoidable waste of valuable human material and the lowering of human efficiency through malnutrition and preventable morbidity, we feel that the result would be so startling that the whole country would be aroused and would not rest until a radical change had been brought about' (Bhore Committee Report 1946). The emphasis of the first health report, i.e. the Health Planning and Development Committee's Report, 1946 (popularly known as the Committee Report) on the role of the State was explicit. It was a plan equivalent to Britain's National Health Service. The Report was based on a countrywide survey in British India. It is the first organized set of health care data for India. The poor health status was attributed to the prevalence
39
SECTION II
of insanitary conditions; malnutrition and undernutrition leading to high infant and maternal mortality rates; inadequacy of the existing medical and preventive health organizations; lack of general and health education; unemployment and poverty that produced adverse effects on health and resulted in inadequate nutrition; improper housing and lack of medical care. Intersectoral linkages were well discussed with nutrition, housing and employment as essential precursors for healthy living. It considered that the health programme in India should be developed on a foundation of preventive health work and proceed in the closest association with the administration of medical relief. The Committee strongly recommended a health services system based on the needs of the people, the majority of whom were deprived and poor. It felt the need for developing a strong basic health services structure at the primary level with referral linkages. It also recommended the need to invest in the pharmaceutical sector to develop indigenous capabilities and reduce excessive reliance on multinational companies. India was therefore one of the few developing countries which adopted a health policy that integrated the principles of universality and equity. Community participation and cooperative efforts to promote preventive and curative health work was important to achieve a vibrant health system. The Committee felt that large sections of the people were living below the normal subsistence level and they could not afford to pay for or contribute to the health services. It was decided that medical benefits would have to be supplied free to all at the point of delivery and those who could afford to pay should channel contributions through the mechanism of taxation. Though the report stated that it will be for the governments of the future to decide ultimately whether medical service should remain free to all classes of the people or whether an insurance scheme would be more in accordance with the economic, social and political requirements of the country at the time' (Bhore Committee Report 1946), one point was apparent-that no individual should fail to secure adequate medical care, curative and preventive because of the inability to pay for it. They recommended that State Governments should spend a minimum of 15% of their revenues on health activities. The National Planning Committee (NPC) set up by the Indian National Congress in 1938 under the chairmanship of Colonel S. Sokhey stated that the maintenance of the health of the people was the responsibility of the State, and the integration of preventive and curative functions in a single state agency was emphasized. The Sokhey Committee Report was not as detailed as the Bhore Committee Report but endorsed the recommendations of the Bhore Committee Report and commented that it was of the utmost significance' (Banerji 1985). The objectives of the First (1951-56) and Second Five-Year (1956-61) Plans were to develop the basic infrastructure and manpower visualized by the Bhore Committee. Though health was seen as fundamental to national progress, less than 5% of the total revenue was invested in health. The following priorities formed the basis of the First Five-Year Plan: provision of water supply and sanitation; control of malaria; preventive health care of the rural population through health units and
mobile units; health services for mothers and children; education, training and health education; self-sufficiency in drugs and equipment; family planning and population control. Starting from the first plan, vertical programmes started, which became the centre of focus. The Malaria Control Programme, which was made one of the principal programmes, apart from other programmes for the control of TB, filariasis, leprosy and venereal diseases, was launched. Health personnel were to take part in vertical programmes. However, the first plan itself failed to create an integrated system by introducing verticality. The concern of the Health Survey and Planning Committee (Mudaliar Committee 1962) was limited to the development of the health services infrastructure and the health cadre at the primary level. It felt the growth of infrastructure needed radical transformation and further investment. Another major shift came in the Third Plan (1961-66) when family planning received priority for the first time. Increase in the population became a major worry and was seen as a hurdle to the development process. Although the broad objective was to bring about progressive improvement in the health of the people by ensuring a certain minimum level of physical wellbeing and to create conditions favourable for greater efficiency, there was a shift in focus from preventive health services to family planning. During the Fourth Plan (1969-74), efforts were made to provide an effective base for health services in rural areas by strengthening the PHCs. The vertical campaigns against communicable diseases were further intensified. During the Fifth Plan (1974-79), policy-makers suddenly realized that health had to be addressed alongside other development programmes. The Minimum Needs Programme (MNP) promised to address all this but became an instrument through which only health infrastructure in the rural areas was to be expanded and further strengthened. It called for integration of peripheral staff of vertical programmes but the population control programme got further impetus during the Emergency (1975-77) and most of the basic health workers got sucked into the family planning programme. Meanwhile the Chaddha Committee Report (1963), the Kartar Singh Committee Report on Multipurpose Workers (1974) and the Srivastava Committee Report on Medical Education and Support Manpower (1975) remained focused on giving recommendations on how the health cadres at the primary level should be distributed. With the widespread disillusionment with vertical programmes worldwide and the need to provide universal health services came the Primary Health Care Declaration at Alma Ata in 1978, which India was a signatory to. The Sixth Plan (1980-84) was influenced by two policy documents: the Alma Ata Declaration and the ICMR/ICSSR report on Health for All by 2000'. The ICMR/ICSSR Report (1980) was in fact a move towards articulating a national health policy that was thought of as an important step to realize the Alma Ata Declaration. It was realized that one had to redefine and rearticulate and get back into track an integrated and comprehensive health system that policy-makers had wavered from. It reiterated the need to integrate the development of the health system with the overall plans of socioeconomic and political change.
40
SECTION II
It recommended that the Government formulate a comprehensive national health policy dealing with all dimensions-environmental, nutritional, educational, socioeconomic, preventive and curative. The National Health Policy, 1983 attempted to incorporate all these. Provision of universal, comprehensive primary health services was its goal. A large number of private and voluntary organizations who were active across the country in the health field were to support the Government in its efforts to integrate health services. Evolving a decentralized system of health care and nationwide chain of epidemiological stations were some of the main recommendations. Once again, a selective approach to health care became the focus when a strong lobby questioning the financial repercussions of the primary health care approach came up. Verticality was reintroduced as an interim' arrangement and interventions of immunization, oral rehydration, breastfeeding and antimalarial drugs were suggested (Warren 1988). This was seen as a technical solution even before comprehensive primary health care could be realized. UNICEF too came out with its report on The state of the world's children and suggested immunization as the spearhead in the selective GOBI-FF (growth monitoring, oral rehydration, breastfeeding, immunization, food supplements for pregnant women and children, and family planning) approach (Rifkin and Gill 1986). Programme-driven health policies were once again the central focus. The plan documents henceforth, emphasized on restructuring and developing the health infrastructure, especially at the primary level. The Seventh Plan (1985-90) restated that the rural health programme and the three-tier health services system need to be strengthened and that the government had to make up for the deficiencies in personnel, equipment and facilities. The Eighth Plan (1992-97) distinctly encouraged private initiatives, private hospitals, clinics and suitable returns from tax incentives. With the beginning of structural adjustment programmes and cuts in social sectors, excessive importance was given to vertical programmes such as those for the control of AIDS, tuberculosis, polio and malaria funded by multilateral agencies with specified objectives and conditions attached. Both the Ninth (1997-2002) and the Tenth Five-Year Plans (2002-2007) start with a dismal picture of the health services infrastructure and go on to say that it is important to invest more on building good primary-level care and referral services.
Both the plans highlight the importance of the role of decentralization but do not state how this will be achieved. The National Health Policy (2002) includes all that is wanted from a progressive document and yet it glosses over the objective of NHP 1983 to protect and provide primary health care to all. The Policy document talks of integration of vertical programmes, strengthening of the infrastructure, providing universal health services, decentralization of the health care delivery system through panchayati raj institutions (PRIs) and other autonomous institutions, and regulation of private health care but fails to indicate how it achieves the goals. It encourages the private sector in the first referral and tertiary health services.
Conclusion
The overview of the plans and policy reports not only throws light on the gap between the rhetoric and reality but also the framework within which the policies have been formulated. There has been an excessive preoccupation with singlepurpose driven programmes. Above all, the spirit of primary health care has been reduced to just primary level care. The health reports and plans mostly concentrated on building the health services infrastructure and even this lacked a sense of integration. Most of the policy reports miss out on the importance of a strong referral system. Instead, there has been more emphasis on building the primary level care and even that has lacked proper implementation. The Bhore committee report and later, the Primary Health Care Declaration discussed the operational aspects of integrating the other sectors of development related to health. The multisectoral approach that is much needed and the intersectoral linkages that are essential for a vibrant health system have not been well thought out, and there has been no plan drawn out for it later. The outline of plan documents and their implementation have been incremental rather than being holistic. It is important to question whether it is only the low investment in health that is the main reason for the present status of the health system or is it also to do with the framework, design and approach within which the policies have been planned.
Acknowledgement
I thank Dr Rama Baru for the valuable insights.
41
SECTION II
References
Banerji D. Health and family planning services in India: An epidemiological, socio-cultural and political analysis and a perspective. New Delhi: Lok Paksh; 1985. Government of India. Report of the Health Survey and Development Committee, Vol. II (Chairman: Bhore). Delhi: Manager of Publications; 1946. Government of India. First Five-Year Plan (1951-56) to Tenth Five-Year Plan (2002-2007). New Delhi: Planning Commission of India; 1961. Government of India. Report of the Health Survey and Planning Committee, (Chairman: Mudaliar). New Delhi: Ministry of Health; 1961. Government of India. Report of the Committee of Multipurpose Workers under Health and Family Planning Programme (Chairman: Kartar Singh). New Delhi: Ministry of Health and Family Planning; 1973. Government of India. Health Series and Medical Education: A Programme for Immediate Action: A Report of the Group on Medical Education and Support Manpower (Chairman: Srivastava). New Delhi: Ministry of Health and Family Planning; 1975. Government of India. Statement on National Health Policy. New Delhi: Ministry of Health and Family Welfare; 1983. Government of India. National Health Policy 2002. New Delhi: Ministry of Health and Family Welfare; 2002. Indian Council of Social Science Research (ICSSR) and Indian Council of Medical Research (ICMR). Health for All: An alternative strategy. Report of a Study Group. New Delhi: ICSSR; 1980. Rifkin SB, Gill W. Why health improves: Defining the issues concerning comprehensive primary health care and selective primary health care. Social Science and Medicine 1986;23:559-66 Warren KS. The evolution of selective primary health care. Social Science and Medicine 1988;26:891-8. World Health Organization. Primary Health Care: Report of the International Conference on Primary Health Care. Geneva: WHO; 1978.
42
SECTION II
Delivery of health services in the public sector
K. SUJATHA RAO
SECRETARY NATIONAL COMMISSION ON MACROECONOMICS AND HEALTH, GOVERNMENT OF INDIA NEW DELHI E-MAIL: ksujatharao@hotmail.com
RE HEALTH SYSTEMS AN END IN THEMSELVES OR A MEANS TO ACHIEVING certain ends? Worldwide, there seems to be a consensus on measuring health systems in terms of improving the health status, enhancing patient satisfaction and providing financial risk protection. In 2000, the World Health Organization (WHO) further expanded the definition to include a reduction in disparities for improving health status; being mindful of the patient's need for privacy and confidentiality and providing services promptly and with courtesy as characteristics of a responsive system; and sharing the financial burden in accordance with the ability to pay as being a fair form of health financing (World Health Report, WHO, 2000). There is, however, little consensus on what constitutes an ideal health system in universally acceptable terminology to enable better intercountry comparisons. This is because, unlike any other sector, health systems are highly contextualized and influenced by various exogenous factors such as societal values, epidemiology and disease burden, availability of financial resources, technical capacity, individual preferences and the nature of demand. Technological innovation in the health sector has improved the quality of life but has also increased costs. In countries that have no social insurance and where the role of the state is limited, people spend a substantial proportion of their incomes on seeking medical treatment, and in the process, get impoverished, thus widening disparities in the health status. To contain spiraling prices and distortions created by market failures such as moral hazard, asymmetry in information, induced demand etc., countries resort to multiple policy instruments. Health systems have five aspects or knobs that interact with each other and influence its basic nature and direction: (i) financial (tax, user fees, out-of-pocket expenditure, insurance), (ii) payment systems (how providers are paid: salary, per service rendered, capitation), (iii) organizational (manner in which the delivery systems are organized/structured), (iv) legal (regulatory frameworks) and (v) social (access to health information, advertising) (Hsiao 2000). The effectiveness with which these instruments of state policy are designed and used determines the extent to which the health system is equitable, appropriate or fair. The health system in India consists of a public sector, a private sector and an informal network of providers of care operating within an unregulated environment, with no controls on what services can be provided by whom, in what manner, and at what cost, and no standardized protocols to help measure the quality of care. There are wide disparities in access, further worsened by the poor functioning of the public health system. In this chapter, we diagnose the nature of the health system in India, in the public sector, analyse the problems that constrain it from achieving the stated goals, and identify issues that require to be addressed for overhauling the system of health care for meeting future challenges.
Part I
An overview of the evolution of the health system in India
The evolution of India's health system can be categorized into three distinct phases: Phase I (1947-83)-when the health policy was based on two principles: (i) that none should be denied care for want of ability to pay, and (ii) that it was the state's
43
SECTION II
responsibility to provide health care to the people. Phase II (1983-2000)-when the first National Health Policy of 1983 articulated the need to encourage private initiative in health care service delivery, while at the same time expanding access to publicly funded comprehensive primary health care. Phase III (post-2000)-which is witnessing a further shift that has the potential to profoundly affect the health sector in three important ways: (i) the desire to utilize private sector resources for addressing public health goals; (ii) liberalization of the insurance sector to provide new avenues for health financing; and (iii) redefining the role of the state from being only a provider to a financier of health services as well.
Phase II (1983-2000)
The National Health Policy of 1983
Despite the remarkable achievements in disease control, the failure to control the population, the lack of access to basic health facilities in rural areas, and the international commitment to focus on providing comprehensive primary care as envisioned by the Alma Ata Declaration in 1978, led to the formulation of the National Health Policy of 1983. Limited resources to meet the growing demand for health services led to the articulation for private sector to shoulder some part of the burden. An estimated Rs 6500 crore worth of subsidy in terms of exemptions in customs duty for import of equipment, subsidized inputs such as land, etc. were extended to stimulate private investment in health. Alongside, the focus of state policy shifted to primary health care to reduce the iniquitous urban-rural divide and expand access to the rural populations, particularly the poor. Lack of resources resulted in segmenting health into independent silos of disease control programmes rather than visualizing health care as a continuum of service. Such segmentation led to simplistic formulations of the role of state being confined to primary health care and a selected list of diseases and health interventions, rather than being responsible for the well-being and health of the people. This phase witnessed an expansion of health facilities for providing primary health care in rural areas and the implementation of national health programmes (NHPs) for disease control under vertically designed and centrally monitored structures. The adoption of this twin strategy had its advantages. With less than Rs 200 per capita investment (2000), prioritization of interventions that benefit the poor and entail wide externalities, provided a moral and technical justification. Besides the establishment of health facilities in accordance with a population norm, guinea worm was eradicated and the disease load due to infectious diseases reduced and deaths averted. During the 1990s, with assistance from the World Bank, NHPs were upscaled with impressive outcomes: the cure rate of tuberculosis (TB) under the Directly Observed Treatment, Short-course (DOTS) programme doubled and averted an estimated 50 lakh deaths, leprosy was eliminated except in 70 districts, the incidence of cataract as a cause of blindness reduced from 80% to less than 50% and the number of polio cases decreased drastically from 29,709 to about 100 (Table 1). Fiscal stress gave rise to innovation; various States attempted to improve the overall performance of public health facilities by a combination of policies-improved availability of inputs, greater flexibility in spending; defining responsibilities and rationalizing performance outputs; widening the scope for involvement of local bodies, non-governmental organizations (NGOs), etc. Table 2 gives a broad idea of the policy areas, the direction and nature of such innovation and names of the pioneer states. The initiatives taken and the outcomes are impressive when analysed in reference to wide disparities in income and socio cultural behaviour, a fast-changing economic scenario,
Phase I (1947-83)
At the time of Independence, malaria affected almost a quarter of India's population; virulent diseases such as smallpox, plague and cholera were rampant, maternal mortality was over 2000 per 100,000 live-births and longevity of life was less than 32 years (Bhore 1946). While the public sector consisted of a few city hospitals, the private sector consisted largely of individual practitioners of Indian systems of medicine and licentiates practicing in villages, as family doctors. With meagre resources, this period saw the effective containment of malaria, bringing down the incidence from an estimated 750 lakh to less than 20 lakh, eradication of smallpox and plague, halving of the maternal mortality rate (MMR), reduction of the infant mortality rate (IMR) from 160 per 1000 live-births to about 105, containing cholera and increasing longevity of life to almost 54 years. Institutes of excellence such as the All India Institute of Medical Sciences (AIIMS) were set up for research and quality training, making India an exporter of highly trained medical doctors. These gains were in no small measure due to the strong foundation of public health on which the health system was grounded and the highly professionalized cadre of public health specialists who provided leadership from the front, camping in villages in hostile environmental conditions, whether to eradicate smallpox or supervise the malaria worker. However, under the overarching influence of modernization that characterized the post-colonial phase of global development, the urge to be on par with the western norms of modern medicine proved to be too strong to resist. India, unlike China, missed the opportunity to launch public health campaigns to promote, at the community and individual household levels, healthy lifestyles alongside expanding public investment to assure universal access to water, sanitation, nutrition and education. Instead, and more particularly during the 1960s and 1970s, public health campaigns were focused only on promotion of the small family norm and family planning. India also failed to utilize the strengths of the traditionally used and accepted modes of medical treatment and gave undue emphasis to allopathy, gradually laying the base for an expanded market for western style curative services, which are urbanbased as well as costly.
44
SECTION II
epidemiological surveys to get a more accurate understanding of the changing Evaluation of World Bank-funded projects in four States profile of disease prevalence and incidence, under the State Health Systems Project which is necessary for measuring risk factors, designing interventions and launching information campaigns to reduce risky Programme Indicator before the project Current status behaviour; (iii) advantage was not taken TB control (cure rate) 25% (1997) 86% (2003) of the 73rd and 74th Constitutional Amendments for decentralizing proControl of cataract blindness/number of surgeries 21 lakh (1995) 42 lakh (2003) gramme implementation to the local bodControl of Leprosy (prevalence per 10,000) 24 (1992) 2.44 (2003) ies/community for increasing accountability in the system; (iv) neglected of Control of HIV - per 1,000,000 3.5 (1998) 5.1 (2005) research and development to promote Control of malaria in 8 project districts API 13.8 API (1999) 9.5 API (2002) In 32 technological innovation; and (v) provided out of 100 districts API inadequate investment in developing the fell below 2. critical mass of required skills and human Reduction in Polio cases 29,709 <100 resources. In other words, the governments ran public health programmes that would SOURCE: Ministry of Health and Family Welfare (MOHFW) have been more cost-effective for the communities and local bodies and in the process comparatively unstable political environment in several States neglected their more fundamental responsibility of goverand a near stagnant average per capita investment in primary nance-of laying down a framework, defining the rules of the health care of Rs 105. Despite the reduced health spending as game and monitoring systems to see that no player takes a result of fiscal pressures that States faced during this period, undue advantage in the health sector. most of them took advantage of available opportunities to achieve Phase III (post 2000) whatever they could, underscoring the fact that a limited National Health Policy II, 2002 level of investment can only give a commensurate level of outcome. By 2000, India had not achieved 13 out of the 17 goals laid Notwithstanding the above, five serious omissions occurred down in the first National Health Policy of 1983 (see Annexin the public health policy: (i) the private sector was encourure IV). Analysis of the 52nd Round National Sample Survey aged without provisions for regulations, standards and accred(NSS) on the utilization of health services showed that duritation processes; (ii) there was an absence of surveillance and
Table 1
Table 2 Innovation in the health sector by States 1995-2000

Area of Innovation Broad Direction of the innovation and innovators
Public-private partnerships Decentralization Human resources Financing
Handing over the management of public facilities to NGOs (Gujarat, Karnataka); Contracting private specialist services and outsourcing other services, such as diet, distribution of IEC materials, etc. (most States) Transfer of budgets to and involvement of local bodies (Kerala, Karnataka, Himachal Pradesh, Orissa); Management Boards of Health Facilities (Rajasthan, Madhya Pradesh, Andhra Pradesh) Contracting professionals for service delivery-ANMs, doctors, surveillance, auditing, etc. (all States); Multiskilling, pre-internship training, Mandatory pre-post graduate rural service (Orissa) User fees and financial autonomy to hospitals (Madhya Pradesh, Rajasthan, Andhra Pradesh, Karnataka, Punjab, West Bengal, Maharashtra); Health insurance (Andhra Pradesh, Karnataka, West Bengal); Direct transfer of funds from GOI to districts under NHPs; Financial delegation of powers to PHCs, CHCs and district CMO (Tamil Nadu. Gujarat) Delegation of powers to district-level officials (Gujarat, Tamil Nadu, rationalizing responsibilities for better accountability, performance-based monitoring (Andhra Pradesh, Gujarat) Link couple schemes (Gujarat, Rajasthan); Village Planning and Community Health Worker (Madhya Pradesh, Uttar Pradesh) Quality control circles (Gujarat); Blood transfusion standards (NACO); ISO certification (Karnataka, Himachal Pradesh) Ensuring the availability of essential drugs at health facilities under the Panch Byadhi Chikitsa scheme (Orissa); Centralized drug procurement (Tamil Nadu, Orissa, Andhra Pradesh, Rajasthan)
Accountability Community mobilization Regulation/standard setting
IEC: information, education and communication; GOI: Government of India; NHP: National Health Policy; PHC: primary health centre; CHC: community health centre; CMD: chief medical officer; NACO: National AIDS Control Organization. SOURCE: Initiatives from Nine States, MOHFW, GOI 2004
45
SECTION II
ing 1986-96, there was a decrease in the utilization of public facilities for outpatient care from 26% to 19%; a decrease in access to free care from 19% to 10% and an increase in the number of persons not seeking care due to financial incapacity (Table 3) see also Annexure I.
government brought forth the National Population Policy (2000), the National Health Policy (2002), and the AYUSH Policy (2000), reiterating its resolve and commitment to achieve a set of goals by 2010. The goals envisaged are to increase public investment in health from the current level of 0.9% to 2%-3%; to increase the utilization of primary care facilities from less than 19% Table 3 to over 75%; to reduce the MMR by threeUtilization of primary and community health centres quarters from the current level of over 540 for outpatient care in rural areas per 1000; to reduce the IMR from 62 per 1000 live-births to less than 30, eradicate polio, eliminate leprosy, reduce deaths State Utilization Utilization of Untreated Untreated Average total of PHC/CHC PHC/CHC for OP ailments ailments household on account of TB and malaria by over 50%, for OP care by the poorest out of total due to expenditure etc. Many of these objectives are in con(out of 2 quintiles number of financial for treatment total OP) (out of total ailments(%) reasons per ailment sonance with the Millennium Develop(%) PHC/CHC OP) (out of total (OP) (in Rs) ment Goals (MDGs) for 2015. The follow(%) number of ing section highlights the systemic issues ailments)(%) 1 2 3 4 5 that may constrain us from achieving these goals within the given time-frame unless Well performing States addressed on priority.
Kerala 5.4 Tamil Nadu 7.2 Andhra Pradesh 5.7 Maharashtra 6.4 Karnataka 11.0 Moderate performing States Gujarat 9.9 West Bengal 4.3 Punjab 1.8 Haryana 5.1 Poor performing States Rajasthan 10.2 Orissa 18.4 Madhya Pradesh 8.9 Uttar Pradesh 1.5 Assam and NEast 27.13 Bihar 2.0 All India 6.4 49 41.5 52.1 47.7 55.1 29.3 49.1 41.2 23.5 11.7 22.4 25.5 11.4 22.3 8 19.9 1 3 1.5 5.2 2.9 2.6 119 79 116 144 91 144 105 173 183
Part II
Organizational Structure of the Public Sector Delivery System
There has been a clear absence of any deliberate strategy to use the organiza4 tional tool for achieving public health 0.5 goals, except family planning, until the Sixth Five-year Plan when, under the Minimum Needs Programme, concerted 44.1 10.2 6.2 172 efforts were made to focus on expanding 30.2 32.3 14.6 99 access to primary care in rural areas. Thus, 27.6 16.3 1.7 129 built over the years, the public health deliv38.6 9.4 202 ery system consists of a large number of 44 9.02 83 dispensaries, primary health care institutions, 19.6 21.9 5.5 220 small hospitals providing some specialist 37.9 17.3 3.5 144 services, large hospitals providing PHC: primary health centre; CHC: community health centre; OP: outpatient tertiary care, medical colleges, paramedical NOTE: The total OP for a reference period of 15 days is 375.3 lakh.. The total number of ailments (rural) is 408 lakh yearly. The average total training institutions, laboratories, etc. expenditure for OP care is for the reference period of 15 days. Total expenditure includes medical expenditure and all expenses other than medical expense incurred by the household for availing the treatment. (Table 4). SOURCE: Mahal et al. 2002 The failure to improve the health status, be accountable and responsive to people's State-wise comparisons show that the poorest in the poorer needs or protect them from financial risk has brought into States of UP and Bihar had to pay substantial amounts for focus the functioning of the public health system, underoutpatient treatment and a low utilization of public facilities, scoring its failure in fulfilling such legitimate expectations. which indicates a virtual breakdown of the public health system. The focus of this section is to understand the causal factors On the other hand, in Assam and Orissa, a large proportion that have led to such a failure. These causal factors can be of persons did not avail of treatment at all. Read along with divided into three broad groups: the number of untreated ailments due to financial reasons, 1. Poor goal setting and lack of formulation of strategic the picture is dismal, as it further emphasizes the failure of interventions; the public health system in providing risk protection, since 2. Management Failures; the average cost of outpatient treatment for every episode of 3. Limited role of the State. illness is equivalent to three to five days' wage of one earning Goal-setting and Strategic Interventions member of the family. To reduce the disease burden affecting the poor and alarmed The public health system is inaccessible, disconnected to by the falling levels in the utilization of public facilities, the
46
SECTION II
Table 4 Public health infrastructure In India, 1951-2001

1951 1961 1971 1981 1991 1998 2000
Hospitals
Hospital/dispensary beds
Dispensaries
Total % Rural % Private Total % Rural % Private Total % Rural % Private
2,694 39 117,000 23 6,600 79 725
3,054 34 229,634 22 9,406 80 2,695
3,862 32 348,655 21 12,180 78 5,131 27,929
PHCs Subcentres CHCs
6,805 27 43 504,538 17 28 16,745 69 13 9,115 84,736 761
11,174 57 664,135 32 27,431 60 18,671 130,165 1910
NA
NA
NA
22,149 136,258 2,633
15,888 22 71.2 719,861 11.06 38.2 23,065 53 57 22,842 137,311 3,043
PHC: primary health centre; CHC: community health centre SOURCE: Health Statistics/Information of India, CBHI, GOI, various years; Rural Health Bulletin, GOI 2002; National Health Policy, MOHFW, GOI, 2002
public health goals and inadequately equipped to address people's expectations. For the majority of citizens, the public health system is out of their reach due to distance, lack of money, lack of confidence in the system or the availability of a cheaper alternative. The organizational structure requires a villager to travel an average distance of 2.2 km to reach the first health post for getting a paracetamol; over 6 km for a blood test and nearly 20 km for hospital care. Given the poor road connectivity, the unreliability of finding the provider at the health centre, the indirect costs for transport and wages foregone, the marginal cost of availing a public service outweighs that of getting some treatment from the local quack. Further, even when accessed, there is no continuity of care guaranteed. In other words, the segmentation of the health system into primary, secondary and tertiary, administered and monitored by different bodies, with none working in coordination, has resulted in the dilution of the concept of the integral nature of health where curative services are a continuum of the preventive and promotive health care. In 8 States, substantial investments were mobilized from the World Bank to upgrade, strengthen and establish hospitals at the district, sub-district and block levels. Under these projects, the comprehensive definition of the primary health infrastructure (Health for All Report of 1980) got a further distortion with the community health centres (CHCs) rechristened as first referral centres (FRUs), divorcing them from their contextual framework. In Andhra Pradesh, Karnataka, Punjab, etc. the World Bank-funded CHCs were brought under the administrative control of autonomous Directorates dealing with secondary level hospitals while those CHCs not covered under the project are continued to be administered by the Director of Health Services. An evaluation report of West Bengal, AP, Karnataka and Punjab showed that while these projects were successful in improving the quality of care in urban and semi-urban areas (Table 5), an expected outcome, such as, for example, an increase in institutional deliveries was
not realized. Had the focus been on establishing the referral system and linkages with the other World Bank-assisted disease control and Reproductive and Child Health (RCH) projects, investments made for strengthening the health systems would have had a measurable impact on reducing maternal, neonatal and infant deaths, or deaths due to malaria, TB which require hospitalization. This experience clearly demonstrates that mere increase in investments in infrastructure does not automatically translate into better public health outcomes. It also underscores the urgent need for conceptual clarity on the expectations of the organizational structures that have been established and the urgent need for standardization of facilities across the country. Shortage of funds has been primarily responsible for the non-availability of facilities in accordance with the norms set by the government; and inadequate provisioning of critical inputs such as drugs, equipment, facilities such as operation theatre, etc. Due to lack of budgets and the pressure to achieve targets, several States upgraded the two-roomed subcentres to PHCs. With no place for laboratory, examination, pharmacy, etc. most are non-functional. There are PHCs with over 33 subcentres and there are subcentres which cover over 200 habitations. It is estimated that 25% of people in Madhya Pradesh and Orissa, and 11% in Uttar Pradesh could not access medical care due to locational reasons (NSS-India Health Report, 2003). The question that then arises is to what extent is infrastructure an important determinant in health outcomes? Is there any association? Box 1 symbolizes the mockery we have made of the health care service delivery system by having subcentres function in non-standardized places denying dignity and privacy to women who visit the ANM for treatment and care. Annexure I gives the levels of utilization of the PHC facilities. Annexure II links outcomes with the infrastructure to examine if there is any such association. What emerges from the
47
SECTION II
involvement of the community and intersectorally linked to non-health determiEvaluation of World Bank-assisted projects nants such as water, sanitation, etc. Such for State health systems an approach if implemented would have helped avert an additional 15 lakh infant and 800,000 maternal deaths. Gains could State/year of project Increase Increase Increase Additional Increase Reduction in the in the in beds in bed in institutional have been impressive. However, as can be utilization utilization laboratory (% occupancy deliveries seen from Annexure III and Annexure of outpatient inpatient tests increase) (%) 1999-2003 care (%) care (%) (%) IV the NHP was hardly implemented. Instead, due largely to resource constraints, strategies contrary to what was stated in Karnataka (1996-2001) 72.2 83.3 290 29.3 10.8 From 55% to 33% the policy, were adopted (such as the selecWest Bengal (1996-2004) 44.6 29.3 54.4 12.9 71.6 From 77% to 74% tive primary health care approach). Punjab (1996-2003) 115.9 65 456.6 45.6 14.4 From 97% to 26% The adoption of the strategy of selective primary health care, running counter Andhra Pradesh (1995-2002) 102.2 100 67.3 From 35% to 33% to the vision of a comprehensive primary SOURCE: Implementation Completion Report, World Bank 2004 health care laid down in the NHP of 1983 was on account of resource constraints. data is that while in the poorer performing States, the ratio Compulsions to prioritize resulted in selecting interventions of facilities to 100,000 population are on par with the rest of based on the criteria of the extent to which the disease/conthe States, and even better than that in Andhra Pradesh and dition affected the poor disproportionately more, was techWest Bengal, the health outcomes are poor. This shows that nically feasible to implement and could be made available at it is not the mere establishment of a physical facility but a comparatively low cost, and to be implemented vertically from combination of factors such as distance, availability and qualthe centre. Evidence from community-based experiments and ity of skills, adequacy of infrastructure and access to altersurveys however tell another story. They conclusively show native sources of care that seem to influence health-seeking that people have other health needs and expectations from behaviour and determine outcomes which have been captheir health system which make integrated approaches more tured by a set of indicators such as complete immunization, effective, efficient and, in the long run, more sustainable. The percentage of those severely malnourished, full antenatal covexperiments also show that vertical programmes fail to inteerage, safe and institutional deliveries and finally, the IMR grate with the provisioning of general health services, weaken and the under 5 mortality rate (U5MR). the health system as a whole and, over a period of time, get While it is clear that infrastructure development had little disconnected from local health problems, priorities and the linkage to goal setting, it is also seen that policy intervencommunity itself. tions per se often lacked focus, were not based on hard eviThese observations find resonance in the experience gained dence, and had weak institutional capacity to translate polso far. A range of health needs such as treatment for debiliicy into action. tating fever that incur wage losses for the labourer, treatment for epilepsy, uterine prolapse, infertility or menstrual problems Lack of Focus, Evidence and capacity affecting women's ability to work are concerns that are ignored as public health systems narrowly focus on achieving proLack of focus: Vertical versus horizontal programmes gramme targets: sterilization, immunization, collection of The NHP 1983 made a strong policy commitment to estabblood smears in case of fever, providing drugs to sputumlish a comprehensive primary health care, based on the active positive persons etc. In fact, even under a programme such as the RCH, which is expected to be gender-sensitive, due to its vertical, target-oriented nature, the number of women receivBox 1 ing postpartum care was very low (NFHS II). Given the large The state of India's health delivery system number of domiciliary deliveries, the health workers visited an average of 5.1% mothers within one week of delivery and In one district, where the NCMH took up a facility survey, 16.5% mothers within 2 months of delivery. In Madhya Pradesh, officials stated that 90% of the 369 ANMs did not reside in the these figures were 1.8% and 10% and in Uttar Pradesh 2% area of their jurisdiction-a situation referred to in Rajasthan and 7.2%, respectively. This not only explains the reason for and Gujarat as 'up-down'-and that with just Rs 75 per month such high neonatal mortality but also the unattended moras rental most subcentres were functioning in verandahs. Now bidity which in these two States was reported to have affected the rent has increased to Rs. 250 but the 'verandahs cannot be nearly 17% women, while 10%-13% suffered heavy vaginal left as dues have to be paid'! Due to lack of any facility and privacy, bleeding (NFHS-2, 1998-99). Such postpartum morbidities go the ANM does not provide any maternal services. unmonitored, as they are not part of the programme targets to be achieved. Apart from such distortions, vertical programming with line item-wise budgeting provides little flex-
Table 5
48
SECTION II
ibility for front-line workers responsible for delivering care, making integration difficult as seen in the case of HIV with Family Welfare or providing treatment for malaria or TB to pregnant women. Another example of a narrow, programmatic approach is TB. While there is no doubt about the technical efficacy of DOTS for curing TB, there is some concern about the technomanagerial approach to a disease that is embedded in the biosocial determinants of poverty, poor housing, illiteracy, financial problems, migration, and low resilience to the initial side-effects of the drugs affecting the ability to work. UK and other countries that achieved successes in TB reduction and containment had no DOTS - indicating that addressing social determinants such as housing could have manifold dividends as witnessed in post industrial UK. The DOTS programme is a highly sophisticated one and very well designed, ensuring the availability of microscopes, trained manpower and drugs etc. but has little effort or budgetary resources for tackling the root cause of the disease, for spreading awareness about the programme, for social mobilization to see that people in need get the treatment. Inattention to the social causes or community involvement can result in dropouts or the very poor not being able to access or continue with the treatment, for example migratory labour. Besides, a legitimate concern expressed widely is the potential for increase in primary multidrug resistant (MDR) TB, which is currently estimated to be 2.8% in North Arcot near Chennai. This is largely on account of the existence of multidrug regimens being administered by doctors in the private sector and the tendency of shopping that patients resort to, on an average about 6-9 providers, before finally reaching the DOTS center. Such frequent switching of doctors by the patients is not only draining their financial base but also, with the irrational prescriptions given, could well be contributing to drug resistance. In Russia, it is reported that during 1997-99, MDR TB rose from 6% to 13% while among the chronic cases it was over 60%. Drug resistance happens due to inadequate treatment, use of sub-standard drugs, use of inappropriate preparation and non-compliance by the patient due to various reasons. MDR TB is not only far more expensive to treat but may also not be treatable. Yet, India barely has a surveillance network to closely monitor this aspect. The story of TB reiterates the need for social/community control on the process and the need for adopting a public health approach to the disease (Atre and Mistry, FRCH 2005). Weak Evidence Base for Interventions Neither the Ministry at the Centre nor at the State level has adequate in-house capability to design research studies, collate data and analyze research findings of the various health interventions to enable evidence-based policy-making. Substantial resources are being spent on programmes and interventions that have a poor evidence base. For example, there is no evidence to indicate the current burden of malaria, or maternal mortality. Similarly, hardly any studies are available to assess the efficacy of the use of a drug or of a treatment protocol in different settings and conditions for formulating
differential strategies to suit the diverse conditions prevailing in India. Such non-availability of good quality research for evidencebased policy formulation is one instance of the health delivery system missing the woods for the trees. For example, the principal goal of the National Reproductive Health Programme is to reduce maternal mortality. Over 100,000 women die every year due to pregnancy-related reasons that necessitate skilled attendance and some surgical interventions. The international definitions of skilled attendants disqualify either the traditional birth attendants (TBAs) or the 18 months' trained ANMs. Surgical interventions on the other hand require a minimum infrastructure such as access to blood, an operation theatre, access to personnel skilled in surgery and administration of anaesthesia, etc. It follows then that, as in Malaysia and Sri Lanka, public policy should in all these years have focused on making investments on development of infrastructure and building-up a professional and skilled cadre of attendants for facilitating safe and institutional deliveries. The failure to link intervention with evidence has resulted in poor outcomes (Table 6).
Table 6 Maternal mortality per 100,000 live-births

Country 1950 1963 1980 1996
Sri Lanka Malaysia China India
555 580 1500 1321 (1957)
245 280 (1958) 1000 (1960) 1195
58 78 (1976) 100 580
24 20 (1995) 61 440
SOURCES: Bhat M. India. In: Maternal mortality: An update. 2002. For other countries: World Bank, 2003
The clarity and consistency of their strategy helped Sri Lanka succeed in bringing down the MMR. The organizational strategy consisted of three concepts: (i) Village-level clinics conducted by a professional health team consisting of a medical doctor, a trained nurse, laboratory assistant, etc. to provide antenatal care (ANC) and examine other ailments, with the auxiliary nurse attending to mandatory registration of all pregnant women, other public health duties and promoting institutional deliveries, etc; (ii) Investment in establishing wellequipped maternal and child health (MCH) clinics/hospitals for delivery; and (iii) a strong health management information system (HMIS) and monitoring system including a regular medical audit of every maternal death for taking corrective action. Compared to the above, India for several years promoted training of village-based TBAs, consistently lowered the quality of training and competencies of the ANMs and neglected supervision and monitoring. Resorting to such low-cost solutions helped avoid committing resources required for the establishment of the requisite infrastructure and human resource development. Table 7 depicts the health care strategies followed and outcomes in Malaysia, Sri Lanka and India. The example of MMR is useful as it is a good proxy for
49
SECTION II
Table 7 Comparison of the healthcare strategies of Malayasia, Sri Lanka and India
Intervention Malaysia Sri Lanka India
1950-1957 Establishment of systems to train and supervise midwives, regulate midwifery practices Introduce accountability for results, systems for monitoring births and deaths Models for effective communication with women and communities Better obstetric techniques for those who already had access Introduction of modern medical advances into existing services-general health improvement including control of malaria, introduction of antibiotics
1950: 534 1957: 282
1947: 1056 1950: 486
1947: 2000 1957: 1321 Recognized the importance of maternal care; focused on socioeconomic development and access to nutrition and antenatal care
1957-1970 1957: 282 Improved access for rural population-the critical elements of obstetric care 1970: 148 were made available to the bulk of the rural population through development of a widespread rural network of trained skilled midwives as its backbone, along with hands on support from supervisory staff competent in basic obstetrics and a system for prompt access to facilities that could treat obstetric complications 1970 onwards 1976: 78 Use of strategies to increase the utilization of existing services through better 1985: 37 management, a focus on quality and systemic responsiveness to public needs 1991: 18 and expectations
1950: 486 1963: 245
1957: 1321 1970: 900 Created the post of an ANM but merged Maternal and Child Health (MCH) and Family Planning Programme; family planning gained priority 1970: 900 1980: 810 1990: 519 1995: 440 1998: 540(NFHS) 1983 NHP recognized high MMR and IMR but reiterated the need to train TBAs as the main strategy; In 1985, the technology Mission for UIP was launched. In 1990, the policy shifted to comprehensive CSSM programme Child Survival and Safe Motherhood with focus on providing EmOC in 1720 FRUs. However, only 600 were set up but not one had the full complement of inputs. Besides, the focus on FRU was misplaced as evidence showed that 85% of maternal complication could be handled at CHC/PHC with training in obstetrics and midwifery; and providing 6 days' training in skill improvement for expanding access to skilled attendance. In 1997, RCH-I designed with about 30 interventions, adding RTI treatment, RCH camps, contractual appointees, etc. without consolidating initiatives of the earlier project.
1973: 121 1981: 58 1992: 27
ABM: auxiliary nurse-midwife; NFHS: National Family Health Survey; NHP: National Health Policy; MMR: maternal mortality ratio; IMR: infant mortality rate; TBA: traditional birth attendant; UIP: Universal Immunization Programme; CSSM: Child Survival and Safe Motherhood; EmOC: emergency obstetric care; FRU: first referral unit; CHC: community health centre; PHC: primary health centre; RCH: Reproductive and Child Health; RTI: respiratory tract infection. SOURCE: World Bank, 2003
50
SECTION II
demonstrating the effectiveness of the health system. A similar mismatch between goal and strategic intervention is evident in the case of reducing the IMR. While 40% of deaths take place within one week of birth, and nearly 23% on account of upper respiratory tract injections and diarrhoeal diseases, strategies required to address these causal factors have been overshadowed by the immunization programmes, particularly the one for polio. The single-point pursuit of polio eradication has resulted in adversely affecting the routine immunization programme, which was initiated in 1986 as a Technology Mission for achieving full protection against all vaccine-preventable diseases by 2000. As per a household survey conducted in 1998 and again in 2003 (Indian Institute of Population Sciences 2004), the data for 220 districts showed that in the majority of the districts, there was either a declining performance or no improvement at all under the Universal Immunization Programme (UIP). Second, the high percentages of drop-outs for oral poliomyelitis virus (OPV3) indicated the wrong perception among mothers of the need to adhering to the immunization protocol (Table 8). Discussions with field staff seemed to suggest that this decline was largely on account of the emphasis given to polio, which not only commanded better resources and visibility in the media but also consumed nearly one-third of the time, 30 times the cost and exhausted the staff In 2003, the Government of India (GOI) had to dispatch half the departmental officers to oversee the Pulse Polio Initiative (PPI) Round due to resistance from the local staff which had got tired of participating in one campaign after another4 rounds of PPI with each round requiring one whole month of preparation, two family health awareness programmes camps of the National AIDS Control Organisation (NACO), health melas of the GOI, leprosy household rounds for identification of left-out cases, registration of patients with guinea worm infection, RCH camps, family planning targets, and so on. Such isolated programmatic approaches have made it impossible to allow the health system to develop. Therefore, even as we get set to achieving zero polio prevalence in India, the question remains as to whether vertically driven strategies implemented in a campaign mode, which are also resource intensive and neglect equally important public health
functions, are worthwhile.
Inadequate Capacity to Plan and Implement at the Centre, State and District levels
Failure to develop a public health cadre and widening the eligibility criteria to include clinicians, without making public health training a mandatory requirement for working in posts that need public health skills, have adversely affected the implementation of public health programmes. Non-reservation of posts or the absence of a dedicated public health cadre have also reduced the employability of persons trained in public health resulting in an accumulated shortage of the critical mass of epidemiologists, biostatisticians and other personnel. With radiographers, orthopaedicians, surgeons working as additional chief medical officers in charge of the RCH programme or programmes for malaria or TB, or IAS officers as project officers of HIV/AIDS, etc., the lack of technical capacity in providing the required level and quality of leadership at the State/district-level has been a serious handicap. Mavlankar (Mavlankar 1999), persuasively argues that one reason for the successful implementation of the maternal health strategies by Sri Lanka and Malaysia is the availability of technical capacity to design and monitor at all levels, from the village to the Central Government. While Sri Lanka with its small population of 180 lakh has a Family Health Bureau (basically dedicated to maternal care) and 3 technical officers and consultants exclusively for maternal health (MH) at the Central level, India with a billion population has one Director-level officer for MH in the Ministry of Health at the Centre. Besides the gross inadequacy of the number, technical posts in the Central Government are manned by personnel drawn from the Central Health Service with no fixed tenure nor any pre-qualifications. For example, a Director of MH should have knowledge of public health, obstetrics and midwifery and related fields. While so, unlike Thailand, the personnel of the Central Health Service have a distinct handicap of not only not having these technical qualifications but also no experience of working in a PHC or a CHC, made worse with no field training upon recruitment as is the case with IAS officers. Lack of technical expertise and non-availability of the critical mass or a minimal number at the Central and State levels are reasons for public health programmes lacking in focused designing, development of national treatment protocols and standards, the non-integration with other related sectors/programme such as TB with HIV, HIV with MH, MH with malaria, health with nutrition or water, etc.; or absense of technical leadership in States and districts on the operationalization of interventions based on technical norms; or assessing and building up of technical skills and human resources required by the programme. Most importantly, this absence of adequate technical skills have also been responsible for the near absence of operational research for obtaining the evidence base for designing better targeted programmes in keeping with the wide social and geographical disparities that characterize India. Instead, at the Central and State levels, almost
Table 8 Comparison of performance under the routine Universal Immunization Programme in 220 districts between 1998 and 2003 (%)
Vaccine Positive decline Stagnant Improved
BCG DTP3 OPV3 Measles Full immunization
13.2 40.4 54.1 30.0 48.2
72.3 53.8 43.6 57.7 43.2
14.5 5.8 2.3 12.3 8.6
BCG: bacille Calmette-Gurin; DTP: diphtheria, tetanus, pertussis; OPV: oral poliomyelitis vaccine SOURCE: IIPS, GOI
51
SECTION II
40% of the time of these ill-equipped officers in charge of complex programmes is spent in attending to administrative duties. The situation in the States is no better. A survey conducted in 6 States to assess the technical capacity of these States for maternal health (MH) programmes, (or for that matter malaria) showed that except one Deputy Director-level officer in Kerala, in none of the other 5-States of Tamil Nadu, Maharashtra, Rajasthan, Gujarat and Chhattisgarh was there even one officer exclusively earmarked for monitoring the maternal health programme (Mavlankar 1999). The situation in the districts is worse. The void in the unavailability of such capacity for surveillance and monitoring at district levels has temporarily been addressed under the TB control and Polio Pulse programmes by taking persons on a contract basis-many from the government itself, thus further weakening the already fragile technical capacity required for implementing the large number of government programmes. In addition is also the question of the State Governments ability to sustain these programme-based consultants after withdrawal of external support. The collection and review of data is hardly given any importance, leave alone analysing it for future planning. Monitoring is essentially confined to the bare minimum of NHP targets and now, polio pulse immunization targets. In the absence of any system of surveillance or epidemiological data gathering, planning interventions lack an evidence base and also make it impossible for the system to be responsive to felt needs. A study conducted in Zenana Hospital in Udaipur, Rajasthan found that during 1983-93 nothing had changed despite the improved road network and awareness levels (Pendse 1993). Table 9 compares the cause of deaths over the decade. The report further observes the failure of the system to provide ambulance services, which resulted in incurring expenditures on transport ranging between Rs 150 and 300, borrowed from moneylenders leaving the people poorer both materially and emotionally when despite their desperate efforts the woman's life could not be saved'. The study also showed that during this period while there was a drop in eclampsia, there was a 6-fold increase of deaths on account of malariainduced anaemia and abortions induced by unqualified prac-
titioners. Abortion and emergency obstetric services remain almost unavailable to the vast majority of the rural women.'
Inconsistent procedures
Rules and procedures do not synchronize with objectives of a programme or foster any accountability among the functionaries. For example, unsafe abortion is said to cause at least 8% of all maternal deaths. Yet field surveys showed that untrained and unqualified providers in the informal sector routinely conduct illegal abortions. This flourishing clandestine business is because of government procedures that take over 15 months for getting a centre certified the the conflicting provisions such as the requirements for a person trained in medical termination of pregnancy to be working at the centre, but then having no facilities to train such private providers, etc. It is for such reasons that a large State like Rajasthan has only 338 certified private facilities with 78% of them in 9 districts, 5 districts having no private facility and 6 having one (Iyengar 2002). With no effective intervention to ensure government facilities having all the required skills, equipment and drugs, the number of deaths due to unsafe abortions remains high.
Management failures
Management failure due to a combination of reasons such as low budgets, untimely and irregular supplies, corrupt practices and poor governance has adversely affected the functioning of the health system. The dispersed and disaggregated nature of responsibilities, and conflicting job profiles make accountability a difficult proposition. While the Secretary of the Department of Health has no control on when and how much money will be made available to implement programmes, the medical officer (MO) in the peripheral centre has no administrative powers over the front-line workers and other functionaries working under him. With most supplies such as vaccines and drugs being provided by the Centre for the NHPs, the States have little control to ensure outcomes, as in several instances procurement delays by the Centre can take as long as over one financial year, affecting the credibility of the system. All these factors have serious implications for the quality of management and efficiency. We now discuss the most frequently cited and widely accepted reasons for management failure.
Table 9 Comparison of causes of death in Zenana Hospital, Udaipur, 1983-93

Indicator 1983-84 (in %) 1994-95 (in %)
Performance-based monitoring
There is absence of accountability in the system. To this end, Andhra Pradesh introduced performance-based monitoring in 1998-99. Primary health facilities, where the maximum absenteeism among doctors and health workers were observed, were graded into four categories, and based on programme targets/achievement indicators, scores/grades were given. This was then the basis for review at the highest level. It enabled identification of the problems and corrective action to be taken.
Number of deaths Poor SC/ST ANC
7 55 45 28
Ambulance
12 68 77 50 only change was increase in tetanus toxoid (TT) 6
SC: scheduled caste; ST: scheduled tribe; ANC: antenatal care .Source: Dr. Pendse, HOD Gyneacology Department, Zenana Hospital, Udaipur
52
SECTION II
Under various programmes and in some States, such performance-based monitoring is done but is neither timely nor systematic, except under the donor-funded programmes of blindness, TB Control Programmes and the Pulse Polio Initiative. It is however pertinent to note that an estimated 1500 consultants were appointed by WHO at the field level to monitor the TB and polio control programmes. In addition, common to these three programmes is the extensive computerization of monitoring and review systems that provided access to information at the district level. Such systems need to be adopted by other programmes and also for other aspects of implementation.
unfriendly, rude, corrupt behaviour of the personnel working in these facilities, distance, inconvenient timings and lack of reliability in the availability or the skill of the provider, etc. reflecting management failure. The subcentres are never open as the single ANM is required to undertake village visits, attend to fixed day immunization schedules, domiciliary deliveries, disseminate health information, oversee the work of the TBA, coordinate with the anganwadi worker (AWW), conduct household survey, attend review meetings in PHCs maintain records, etc. With better rearrangement of these factors utilisation can be drastically improved.
Lack of policies for human resource development Absenteeism from place of work
A majority of doctors opt for specialization and/or urban practice. The reluctance to serve in rural areas has become a major impediment in the government's ability to provide health services to the rural population. Not surprisingly, absenteeism among doctors and front-line workers from their place of work is high. A study conducted by the World Bank (2004) and other studies (Mohan et al. 2003; Rao 2003) show absenteeism ranging from 40%-45% among doctors working in primary health centres. The World Bank Study based on a simple regression analysis showed the relationship between income and absenteeism, which suggested that higher income States have lower rate of absenteeism with point value at 0.001, meaning that every increase of Rs 1000 State per capita income is associated with a reduction in absence of 1% point, with p values on the co-efficient on income at 0.13. However, this is a crude analysis as, at another level, absenteeism is high in these rich States where doctors are also engaged in private practice. Punjab has the lowest utilization of public facilities only because of large-scale absenteeism of doctors. The recruitment policy is a contributory factor for the lack of motivation among doctors to provide services in rural areas. Quite often, postgraduate students are recruited by the governments and placed at PHCs where the skills acquired by them during postgraduation are of little relevance. This is made worse by the lack of equipment, drugs and adequate caseload. Similarly, there is almost always a mismatch of skillsa gynaecologist is posted at a CHC where there is no anaesthetist resulting in the underutilization of skills. Likewise, transfers are often arbitrary and without adherence to any norms, resulting in the low morale of doctors. Even the States that do have a transfer policy rarely adhere to it. Recently, there was an instance in a State where at a CHC all the 7 doctors were transferred out in one go, leaving behind a hapless lot of patients. Often, the skills needed or acquired in a training programme are not taken into consideration. Therefore, under the NHP, money may be spent in training a doctor in anaesthesia, intraocular lens (IOL) implantation surgery, or a manual vacuum aspiration (MVA), but fail to impact on the programme as, more often than not, on return from training, he or she is posted to a place where the acquired skills are not required or the required equipment is not available. The absence of transparent transfer policies, norms for deployment of personnel, and reward for merit, are some of the factors contributing to the deviant behaviour among providers.
Quality of service delivery - An imbalanced mix of inputs

Vehicles without POL budgets, beds without washing allowances, X-ray machines lying idle for the want of consumables or maintenance budgets, empty shelves in pharmacy counters, etc. also contribute to management failure. In addition, quality is also perceived to be low due to the often
Limited promotional avenues

In many States (such as Orissa, Bihar, Uttar Pradesh, Rajasthan) an MO often gets the first promotion after 15-20 years of service. There are many doctors who continue to remain MOs without promotion while their counterparts in civil services might have been promoted from the post of an SDM to Special Secretary or even Secretary and from Accounts Officer to Financial Advisor. Career stagnation affects morale. In Madhya Pradesh, the Departmental Promotion Committee (DPC) meeting has not been conducted in the past 20 years. In Chhattisgarh, all chief medical officers have been posted on an ad hoc basis.
Box 3 Management issues in the rural health Care

Doctors do not stay at PHCs and absenteeism among PHC staff is high. Training during MBBS is not geared to impart skills for providing service in rural areas. Doctors need to be provided financial and non-financial incentives for staying in rural areas. There is a need for increasing paramedicalization of primary health care services.
Poor payment systems and dual practice

To compensate for the relatively low salaries, doctors are
53
SECTION II
permitted private practice outside office hours or are given a non-practising allowance, often 25% of the basic pay. Lack of monitoring, effective supervision and, at times, collusive relationships are causes for the abuse of this facility affecting patient care in public facilities. Due to financial constraints, most States have now stopped recruiting MOs in the regular pay scales and instead are now offering contractual services for as small a remuneration as Rs 8000 per month, a strategy which has a high turnover with doctors joining services only for getting rural service experience for admission to Postgraduate Entrance Examination, or as a makeshift service for preparing for the PG entrance exams, or joining service and just lingering on to it in the hope that some day their services might get regularized. Time has come to review such arrangements keeping a long-term perspective in view. Doctors, particularly, specialists need to be paid better and there is a need to sanction posts of specialists and public health managers in hospitals at district and State levels. Low cost solutions or decisions based on present day contingencies cannot sustain the system which will develop fissures, and cost more to repair.
Box 4 The ANM first interface with the community

The ANM still continues to be the only worker for delivery of primary health care in rural areas in the public sector. She is presently working in isolation without a team and with no support or supervision from either the lady health visitor (LHV) or the Medical Officer. She is overloaded with too many functions and activities to be delivered at too many places to too many groups of clienteles. She is required to deliver health services, travel, educate communities, counsel clients and mobilize communities. She has to fill in several registers and submit several reports.The mean number of years of gap between her obtaining qualification and joining service is 4.2 years. Few subcentres operate from government-owned buildings which are poorly maintained and many are in rented buildings. The subcentre is a small area and cannot accommodate an examination or labour table, and the supplies are inadequate, irregular and erratic. About 40%-62% ANMs do not live at headquarters, the most common reason for their non-availability being security concerns. In about half the cases, the subcentre, are located far from the village.
Source: Rangarao, 2003, Mohan et al. 2003
Poor Facilities at work

The most demotivating factor is the lack of appropriate facilities and required inputs to enable a qualified doctor to do his best for his patient and derive job satisfaction. In addition, lack of decent housing facilities and educational facilities for their children are further contributory factors to the reluctance to work in rural and underserved areas. The working conditions of nurses / midwives is worse, ranging from the lack of basic amenities such as toilets to physical safety. Inadequate and unreliable supply of inputs, absence of supervision and technical guidance, limited opportunities for career advancement, absence of accommodation with over 60% of the subcentres functioning in rented places hired for about Rs 100-300 per month, and often doubling up as a part of her residential accommodation are other factors that contribute to sub-optimal outcomes. Initially, subcentres were envisaged to consist of a multipurpose worker (male) (MPWM) and one multipurpose worker (female) (MPW-F). However, 60% of the posts of MPW-M are lying vacant, thereby increasing the workload of the ANM and affecting the ANM's quality of services. In the community setting, female health functionaries face many problems with regard to transportation, accommodation, gender-based harassment and lack of security, in addition to lack of incentives, stagnation of career due to inadequate development opportunities and inadequate provision for living with the family and education of their children.
Corruption
This then brings us to the key issue of corruption. As per Transparency International India, health has the maximum public interaction and is the second most corrupt sector. The Karnataka Lok Ayukta has estimated that at least 25% of the
budget is siphoned off through corrupt practices. An analysis of the Lok Ayukta shows that all categories of government health functionaries-ayahs and ward boys to nurse, doctors and specialists-are involved. Corruption is in many areas ranging from indulging in unauthorized private practice to issuing medical certificates, transfers, postings, recruitment, in tolerating' absenteeism, etc. The most sensitive areas are in the procurement of drugs and licensing of blood banks, where unlicensed manufacturers have been recipients of orders and action on spurious drug suppliers tardy. The pervasive spread of corruption is not limited to the public sector. The private sector is also working under low thresholds of integrity. Patients are exploited by being made to undergo unnecessary tests only for making money. Providers in private practice are seen to own pharmacies and diagnostic centres. They get cuts' and commissions for referrals and such fee splitting is the mainstay of many doctors' monthly earnings. There are adequate studies that have shown the disproportionately large number of caesarean sections-66% of all deliveries in private hospitals in Kerala (Kutty 1995). The rate of hysterectomies being performed among young women is one example of the absence of ethical standards that need to be effectively countered by fostering transparency, widening participation, strictly enforcing inspections and, above all providing leadership, in technical, administrative and political organizations in reiterating and reasserting value systems. Enforcing good management and governance is then absolutely essential since the implication of bad practices in the health sector hurts persons who are poor and suffer the double tragedy of being sick. No market can function or sustain itself unless there is a minimal level of integrity, fair play and rule of law. Therefore, if insurance and contracting the
54
SECTION II
private sector are to be the new ways of expanding access and financing health, then it is essential that values of probity, nurturing of informed consumers and wider participation through good governance be ensured. Consumer forums, patient management committees, village health committees, patients /citizens' charter, Transparency Act, right to information, imparting of value systems and training in management practices, e-governance, redressal systems, etc. are some of the instruments that need to be employed by the government for counter-checking malpractice.
Urgent need for infusion of new skills

What emerges from the recounting of the several areas of management failure particularly at the point of service is the need to institute a class I All India cadre of Public Health Managers-directly recruited and trained in public health and posted at district levels, like the IAS officers. Over a period of time these young recruits will become the backbone for providing leadership in the public health area. Such persons need not necessarily be doctors-they could be from a wide variety of related disciplines such as a PG in microbiology etc but possess a Masters in Public Health. In such a system, those keen to specialize can gradually be veered to work in the hospitals and be provided career opportunities to work in teaching hospitals and super specialize etc. Such options for human resource management will be critical for steering the country from out of the veritable mess we are in presently.
Lack of Discipline and Work Ethic

In India, government employees often explain the omissions or commissions on lack of political will'. It is however, a fact that more often than not, there is large-scale abdication of responsibility at the field levels, say for example when a head of the department or a CMO does not undertake field visits, conduct review meetings, monitor the implementation of various activities, attend office on time and check attendance registers, listen to grievances, fill vacancies, promote people, punish the wrong, reward the good, then there is abdication of duty. When the CMO allows' doctors and other functionaries to absent themselves from duty, then it is collusion. No amount of funding or administrative reforms can help till there is an overall institutional discipline enforced at all levels and pride for good work instilled. Creating such an environment again carries the implication of having systems and tools that facilitate its emergence.
Dysfunctional structure-the role of the State

Though health is a State subject, the Central Government has certain powers and responsibilities related to the control of infectious diseases, family planning, education, drugs and research. Therefore, the departments dealing with health and family welfare, at the Central and State levels are large in terms of the human resources employed and the wide span of work covered. At both levels, there are several directorates headed by doctors and technical units dealing with the myriad issues in the health sector. For discharging their multiple functions of provider, regulator, facilitator, educator and promoter, the departments employ a large number of technical people-doctors, nurses, paramedical staff, etc. for running hospitals, dispensaries, health centres, medical colleges, nursing schools, and public health laboratories, for inspecting the quality of food and pharmaceutical products, for providing information on public health issues, production of vaccines, etc. Structurally, the administrative units do not take into their purview the functioning of the private sector, which is seen as an independent, autonomous entity. This disassociation is in part due to the fact that various ministries administer matters that directly effect health outcomes and have no mechanism to ensure coordination among them. For example, in the Central Government, the pharmaceutical industry is under the Ministry of Chemicals, policies related to import or export of drugs and technology are the responsibility of the Ministry of Commerce, drug regulation is under the Ministry of Health, programmes related to nutrition are part of the Department of Women & Child Welfare, while water and sanitation is looked after by the Ministry of Rural Development, research in medical diagnostics or vaccines by the Department of Biotechnology, health insurance by the Ministry of Finance, etc. Such intense fragmentation across departments and States is the single most important factor that confines the Ministry of Health to narrowly focus on the implementation of budgeted programmes and activities. The second structural mismatch is the fragmentation of
Use of IT for Better Decision-Making

Effective leadership rests on access to organized information which is increasingly becoming possible due to e-governance. Information about health inputs and outcomes, achievement of targets and goals are necessary for formulating policies and monitoring activities, be they related to technology, human resources or infrastructure. Since quality monitoring based on performance indicators on a concurrent basis is fundamental to curbing errant behaviour, the need for the use of IT cannot be overstated. IT should be used for record maintenance, monitoring supply and inventory control, tracking events and disseminating information to consumers. This would place a great amount of power in the hands of the government to guide, monitor and correct. Such data analysis also reduces subjectivism in transfer policies and personnel development, and ensures transparency in all transactions, the only check to abuse of discretionary power. Besides, even for patient care through the use of telemedicine, or establishing call centres for giving instant advice on coping with a small emergency or advising which hospital to check into etc. technology has the solution. Such a system development will become even more important with the government shifting its role as a financier of services rather than a provider; as a regulator of providers; and as the final protector of patient and consumer rights to medical practices that are safe and appropriate.
55
SECTION II
the Ministry itself: into the Departments of Health, Family Welfare and AYUSH. Such fragmentation that took place in the 1990s had negative downstream effects down to the implementation level, making interprogramme integration problematic, diluting the technical capacity to think holistically and duplicating resource use. For example, the Reproductive and Child Health (RCH) programme rarely addresses HIV/AIDS, malaria or tuberculosis (TB). Likewise, the programme for malaria control has no indicator focusing on pregnant women; or nutritional deficiencies in the child health programmes. In addition to the inadequate technical oversight, the departments also function more like casualty wards' where managing themselves rather than the system has taken centre stage (India Health Report 2003). The Department of Health, for example, spends over three-quarters of the time addressing VIP claims under the Central Government Health Scheme (CGHS); sanctioning medical colleges; procuring medical drugs and supplies, and transfering doctors and court cases. Lastly, the problem of governance, whether at the Centre or States, has also been compounded with the frequent transfers of ministers and officers. During 1998-2003, there were five ministers in the Central Government and as many Secretaries.
Restructuring of the Administrative Departments

The issues raised above have been felt for a long time. The Ministry of Health itself commissioned studies to restructure its organization to suit the emerging challenges. The three reports: Administrative College of India (1986); the Bajaj Committee (1996) and the Center for Policy Research (2000) made some important recommendations which are waiting to be implemented: Constitute Hospital Committees and delegate administration to them; Outsource and decentralize promotional and publicity functions; Convert the CGHS to an autonomous board; Constitute an Advisory Body to advise the ministry on policy issues; Decentralize planning and programme formulation to States, confining the Centre to monitoring adherence to national policy goals and providing technical support; Outsource procurement to an independent body; Establish a Federal Drug Authority and a Commission for medical education; Transfer all Delhi-based hospitals to the Delhi Government and make the Central hospitals autonomous; Merge all the three departments; Create a Indian Medical Service such as the Indian Administrative Service (IAS); Establish an institutional mechanism for interdepartmental coordination; Establish a manpower planning cell in the ministry. Implementation of the above recommendations would free' the Ministry of Health at the Central and State levels to address
the more important issues of governing the health system as a whole. In other words, the Ministry of Health is not only expected to be concerned with the implementation of its programmes but the functioning of the health system comprising both the public and private sector, by diligent oversight safeguarding the interests of the public in general and patients in particular. Such a change in understanding of the functional responsibilities would not only require space in terms of time but also capabilities and skills to address such a role. Organizational structures reflect the objectives and aims of a policy. For example, since RCH objectives emerged as a consequence of the failure of a family planning strategy, it was added on to the Family Planning Programme and renamed as Family Welfare (FW), explaining the anomalous position of the DGHS who does not have any role in the technical aspects of the RCH programme. In the districts, such disassociation of FW from the technical head, namely the Director of Health Services, has had a negative impact on the technical quality of the program. In States where the Health Department is divided into Health and FW, implementation of the FW programmes has been problematic due to nonalignment between authority and responsibility. Due to these factors, recently, the two departments have been merged at the Centre. While this is a positive step, there is still need to restructure the set-up on a functional basis all through the chain.
Part III
Case for systemic reforms: Restructuring Institutional Frameworks
The process for systemic reform will need to start from the Central Ministry of Health, looking at the big picture- setting standards and laying down rules and regulations to be followed by all stakeholders; mobilizing resources; providing leadership based on its knowledge and technical superiority; and facilitating and steering the health system to ensure that the goals of equity, efficiency and quality are met. Such a role would require the Central Ministry to restructure its work allocation based on functional homogeneity. The Ministry should also shift from micromanagement by divesting and delegating powers and authority to functional units. There is also an urgent need to establish new institutions, such as an autonomous institute for health information and disease surveillance, a food and drugs authority; a social health insurance corporation to take care of government employees and the labour in the organized sector by merging the CGHS and the Employees State Insurance Scheme (ESIS); enable the Indian Council of Medical Research (ICMR) to have more autonomy (such as the National Institutes of Health, USA) by generating its own resources; and outsourcing all procurement work to professional bodies. The manpower and time that would be available with the removal of this historical burden of functions would enable the Ministry to discharge its stewardship functions which require laying down standards on health infrastructure and quality, classification of diseases, costs and norms for monitoring utiliza-
56
SECTION II
tion levels, carrying out research to evaluate the cost-effectiveness of the various interventions being implemented, training, etc. The functions listed under the stewardship role are not simple, and entail mobilizing multidisciplinary groups and collecting and collating evidence for revising existing policy or formulating a new one. Standard setting is a tedious process and has cost implications. For example, setting a standard to include five ultrasound tests for an antenatal protocol would have substantial financial implications, besides driving investment to expand availability of this technology, though there is no evidence to establish its efficacy in assuring better outcomes of pregnancy. Likewise, it is through research that longterm consequences of policies need to be studied before taking decisions. For example, India's hasty decision to relax vigil in the 1970s and disbanding the malaria programme resulted in its resurgence in a form more serious and also more expensive. Such decisions therefore need inputs from public health specialists as well as economists to state which interventions work and which do not, and what policies should be adopted and why. If public policies fall short it is because such expertise is sadly lacking and in short supply in the country. Thus, good governance is not only dependent on political commitment but also having the appropriate tools, instruments and information. Bad policies need not only be the result of careless oversight or narrow sectional interests, but also due to lack of evidence and information. In addition to the above, the administrative departments of health, including those at State levels, need to achieve greater efficiency in reference to some aspects described below.
suade and convince them of the need to adhere to quality and patient safety. This calls for a different mindset to be cultivated through intensive training programmes and performance monitoring systems. Supportive supervision is a new skill that needs to be nurtured in the government sector. The key challenge to governance is the enforcement of regulations related to the quack' or the unqualified practitioner in the villages. In a setting where the public health system does not function and the private sector is too expensive, it is this quack who enjoys social consent. Rational arguments of quality or harmful practices, lack of qualification, etc. do not matter as, for the people, the quack is able to provide instant relief to a need at affordable cost. How then does the Government achieve its norms for quality and standards of patient care while allowing this clearly illegal and perhaps harmful practice to continue? Good governance would require a political will to resolutely enforce discipline and make the public health system work, besides educating the people on the rational use of medical practices or drug use.
Devolution of authorityThe district societies: A mechanism for better utilization of funds

A major problem being faced by the Department of Health was the untimely release of funds. Routinely, Central assistance meant for specific programmes would be diverted by the State finance departments for tiding over their ways and means position, resulting in delayed release of funds, stalling the implementation of health programme activities. Therefore, under the National Programme for the Control of Blindness, district societies for blindness control programmes were first constituted during the early 1990s. Under this arrangement funds were directly released to the district societies. This mechanism was subsequently used by all programmes resulting in the constitution of over 4-5 societies, one each for TB, Blindness, Malaria, RCH, and Leprosy. The experience has been a positive one as it has enabled better absorption of funds and quicker implementation. The experience of district societies is now being used to integrate them into District Health Societies so as to facilitate district-level health planning and monitoring activities to achieve health goals. A review conducted on the functioning of these different societies in the Pune district of Maharashtra brought forth some interesting suggestions from programme officers: Develop capacity for better management through training; Establish more rigorous monitoring and programme review systems to improve outcomes and ensure cost-effective utilization of funds; Standardize reporting and auditing formats; and Sensitize officers on programme goals and objectives, and increase the involvement of civil society to reduce the temptation to misuse or misallocate funds. Based on the above, training in data analysis and planning processes, developing indicators for performance review and
Regulation in the health sector: accreditation of facilities

The role of the government in the health sector is to look after patients' welfare. Canada and the US have some scores of regulations on or related to health. Drawing up legislation in a sector like health is complex and requires an understanding of the incentives or disincentives such a legislation may have on human behaviour and a balanced approach. For example, if the legislation is too inflexible and specific, putting all risks on the provider, then it may result in mindless litigation, increasing defensive medicine and higher costs for the patient, endanger the patient-doctor relationship which should be based on trust and entail harassment and outright corruption at the hands of the bureaucracy. If, on the other hand, it is too considerate to provider concerns, the patient may end up getting shortchanged. Besides, it is the enforcement of the laws that is more important. In other countries, inspectors and assessors sent to evaluate provider facilities for accreditation or licensing are trained, so that at all times the focus is on achieving the objective of increasing awareness and creating a sense of accountability among providers regarding the quality of patient care, and not the blind and mindless application of a standard or a rule. Thus, supervision requires to be supportive, not prescriptive or fault-finding, as the objective is not to drive away the providers but to per-
57
SECTION II
monitoring for corrective action will need to be accorded priority focus. The societies also need better expertise, persons trained in health economics, financial planning, statistics and data analysis, epidemiology etc. In the absence of such expertise and evidence-based planning, the tendency is to merely repeat what was being done earlier, nullifying the benefits of a bottom-up planning concept. Resources are not only financial. It is the government's responsibility to monitor the availability of human resources as well. What skills are needed, what are being produced, where and by whom are they being utilized, where are they concentrated, etc. are the sort of issues that should attract priority attention, as 5-8 years are needed before the required human resources are available. Past neglect of human resources is the cause for today's imbalanced skills mix, acute shortage of trained nurses. This function will gain even greater importance in future years as with the General Agreement on Trade in Services (GATS), more professionals from India will be able to find employment abroad. The government needs to establish mechanisms to know the migration flows of skill and identify areas of shortage so that corrective action can be taken in advance.
Local bodies
In the health sector in India, decentralization has to be viewed, not only in the context of devolving authority and power to States by the Centre, to districts and States but to the multilayered local bodies as well. Such devolution of authority has taken place only in Kerala. Kerala has invested both time and resources in systematically focusing on building capacity for governance among elected leaders. Leadership and governance means having the ability to plan, budget, implement, manage, monitor, review and accept responsibility for the decisions taken. The strategy of the big bang' approach adopted in Kerala where, in one sweep, functions, powers and responsibilities were transferred rather than the usual cautious approach, of training and building capacity before delegating responsibility, has proved to be successful when compared to the experience of other States where devolution has been incremental, halting and sporadic. Devolution of powers has, however, not been easy. The Kerala experience shows that despite the transfer of some proportion of the budgets and bringing all-district level institutions under the control of the local bodies, the benefits in terms of health indicators have not really been visible (Vijayanandan 2003). This is largely because of the lack of technical guidance at the panchayat level, lack of standardization of facilities laying down clearly the functions, duties, responsibilities and outcomes of health personnel working in facilities located at different levels, lack of clarity and clear delineation of what services ought to be available where, making it difficult for the local bodies to understand what exactly should be their priorities and areas of focus. Lack of integration between different systems of medicine, ego problems between the highly educated doctor, senior in rank, to functionaries of the local government, dual control, multiplicity
of bodies handling health budgets such as the chief medical officer (CMO), hospital superintendent, zila parishad, district societies for each national programme, hospital development committees, etc. are other reasons that were found to have complicated matters. Kerala is therefore now working towards evolving minimum standards of care and conduct, a citizens' charter and community-based monitoring of health programmes. Decentralization to local bodies has been under consideration for several years but was never implemented in true spirit due to various reasons. The attitude towards the involvement of local bodies has nearly always been to sensitize the representatives and use them in an advisory capacity or for execution of government works under the Rural Development Programme. In the health sector, utilization of the local bodies as agents of change or in social mobilization has been minimal and perfunctory. Experience shows that unless the local bodies are provided funds, specific responsibilities and powers, the benefits of decentralized systems cannot be fully realized. In this context, it would be useful to keep in mind the international experience in fiscal decentralization as they provide a few lessons to be learnt based on certain principles (Sethi 2004). For fiscal decentralization, all aspects and components need to be addressed such as: Assignment of expenditure responsibility to local governments to be followed by revenue responsibilities; Availability of a strong state ability to monitor and evaluate the intergovernmental fiscal system; Devolution of powers and responsibilities in keeping with capabilities; Linking of revenue-raising and expenditure decisions; The intergovernmental system should be designed to match a set of clearly specified objectives, kept simple and flexible, while at the same time be subject to the discipline of budget constraints. Applying these principles will mean having a clear-cut delineation of duties and functions to be carried out by the local bodies at different levels vis--vis the government departmental hierarchies; the financial implications of those functions and systems for utilization and reporting; and finally the kind of authority, powers, or control they have on the functionaries responsible for discharging those duties. Such delineation needs to be based on clear government orders or legislation as the case may be and backed by intensive training and guidelines provided in simple, easy-to-understand formats. Without such a systems approach merely orienting' locally elected representatives to be involved' in health activities is as valuable as the paper on which it is written. Given the vastness and diversity, India will find it difficult to reverse the trend on communicable diseases such as malaria and TB unless the local bodies and the wider community are also fully involved. However, such involvement needs to be formalized. For example, the local bodies should be made responsible and accountable for certain health actions, for example, registering births and deaths, carrying out all antimalarial activities such as plugging the breeding grounds of
58
SECTION II
mosquitoes, etc. In fact, later when social health insurance picks up, it will be necessary to have such a capability available at the local level for making the health insurance scheme function at minimal cost. Wider participation of the communities through village health committees working in coordination with management committees at higher-level facilities is the only way the health system can be made more accountable to the people they are meant to serve. More inclusive approaches and greater democratization is essential if health gains are to be achieved. The initiatives would remain platitudes unless there is close monitoring by the State and provisioning of technical advice. This would require having a team at primary health centres (PHCs) and community health centres (CHCs) to work exclusively on the development of the community-based strategies-the village health workers, village health teams, local bodies, etc. In the absence of such administrative restructuring to guide, facilitate and supervise the development of the demand side of the health system, decentralization may not really go beyond tokenism.
Conclusion
Technological advances, investment and good policies can be turned to naught in the presence of a system lacking in leadership, direction and a core sense of integrity pervading all levels of health care. Unless all stakeholders are motivated by a set of values-of compassion and human concern for the sick and ill, of not accepting a system which allows people to be denied care only because of circumstances beyond their control, of a minimal sense of equality and dignity among allthe health system will continue to reflect the cement and mortar issues of the expanding medical and drug industry, which can, in the absence of the guiding hand of the state, degrade human suffering into an opportunity for making profits. It then becomes critical to define the role of the State as the current utilitarian liberal approach of the health sector offers no acceptable solution. The issue is broader and needs to be examined within the context of the principles that underlie the concept of social contract of Rousseau or sense of justice of Rawls. If these principles enshrined in our Constitution are adhered to, then the State will need to intervene both intelligently and firmly.
59
SECTION II
References
Atre SR, Mistry NF. Multidrug resistant tuberculosis (MDR-TB): An attempt to link biosocial determinants. The Foundation for Research in Community Health (unpublished). Bhat Mari. Maternal mortality: An update. 2002 Bhore J. Report of the Health Survey and Development Committee. New Delhi, 1946. Hsaio W. Unmet health needs of 2 billion: Is community financing a solution. 2001 Government of India. National Sample Survey Organization (NSSO). Morbidity and utilization of medical services, 42nd Round, July 1986-June 1987. Report No 364, New Delhi: Department of Statistics, GOI. Government of India. National Sample Survey Organization (NSSO). Morbidity and utilization of medical services. 52nd Round, July 1995-June 1996, New Delhi: Department of Statistics, GOI. Government of India. Rural Health Bulletin, Ministry of Health & Family Welfare, GOI 2004. Government of India. Initiatives from Nine States, Ministry of Health & Family Welfare, GOI 2004. International Institute of Population Sciences (IIPS) and ORC Macro. National Family Health Survey 2, (NFHS2), Bombay, October 2000, (1998-1999). Kirti and Sharad Iyengar. Elective Abortion as a Primary health service in Rural India : Experience with MVA. Reproduction Health Matters, 2002, Vol. 10, No. 19, 5463. Kutty Raman, Panikar Impact of fiscal crisis on the public sector health care system in Kerala-A research project. Achutha Menon Centre for Health Science Studies, 1995. Kutty Raman. Historical development of health care in Kerala. Health Policy and Planning 2000. Mahal A, Singh J, Afridi F, Lamba V, Gumber A. Who benefits from public health spending in India-results of a benefit incidence analysis for India. National Council of Applied Economic Research, 2002. Mavalankar DV. Study of technical top management capacity for safe motherhood programme in India.Study commissioned by the World Bank, New Delhi (unpublished monograph). Misra R, Chatterjee R, Rao S. India Health Report. Delhi: OUP; 2003. Mohan P, Iyengar S, Mohan SB, Sen K. Daily up-down. Why should an auxiliary nurse-midwife (ANM) of Rajasthan prefer to reside within her work-area? Udaipur: Action Research and Training for Health; 2003. N.Chaudhury, Jeffrey Hammer, Halsey Rogers, in Teacher and Health Care Provider Absence: A multi country study Development Researh Group, World Bank, Washinton, June 2004. Pendse V. Maternal deaths in an Indian hospital: A decade of no change?, Udaipur, 1993. Rangarao AP. Report on role and efficiency of ANM and male worker in primary health care. Andhra Pradesh: A Qualitative Study funded by DFID. 2003 (Unpublished paper comissioned by DFID, New Delhi, India) Sen PD. Community control of health financing in India: A review of local experiences. October 1997. Sethi G. Fiscal decentralization to rural governments in India. Delhi: World Bank, Oxford University Press; 2004. Vijayanand S, Decentralization of Health Planning and Implementation - the Kerala Experience on the Role of Local Government Institutions in Population, presented at NIHFW Workshop 17-23rd. February, 2003 World Bank. Implementation Completion Report, Andhra Pradesh First Referral Health System Project, February, 2002 World Bank. Implementation completion report. State Health Systems Development Projects II. , Washington, USA: World Bank; September 2004. World Bank. Investing in maternal health-learning from Malaysia and Sri Lanka. Development in Practice Series. The World Bank, East Asia and the Pacific, 2003. World Health Organisation. World Health Report, Geneva, WHO; 2000.
60
SECTION II
Annexure I Utilisation and Expenditure Pattern of Health Services in India

% utilisation of PHC/CHC % utilisation of for OP care private facilities for OP (out of total OP) (out of total OP) 1 2 Rural % Untreated Ailments (out of total number of ailments) 3 Urban Rural Average total expenditure for treatment per ailment (out-patient) (in Rs.)* 4 Urban Rural Average total expenditure per hospitalised case (in Rs.)* 5 Urban
India 6.4 Andhra Pr. 5.7 Assam 27.13(for north-east) Bihar 2.0 Gujarat 9.9 Haryana 5.1 Karnataka 11.0 Kerala 5.4 Madhya Pradesh 8.9 Maharashtra 6.4 Orissa 18.4 Punjab 1.8 Rajasthan 10.2 Tamil Nadu 7.2 Uttar Pr. 1.5 West Bengal 4.3
82.0 85.2 58.1 (for north-east) 92.1 77.4 85.7 77.9 69.0 75.1 87.0 58.4 92.2 58.2 70.0 94.0 86.3
17.3 25.5 44 21.9 8 3 22.3 11.7 16.3 11.4 32.3 1 10.2 22.4 9.4 19.9
9.3 15 36.4 15.5 3.5 1.6 8.6 10.8 6.7 7.6 13.4 3.5 10.4 8 6.5 10.1
144 116 83 220 144 183 91 119 129 144 99 173 172 79 202 105
175 143 110 176 211 402 155 108 351 170 117 155 176 117 212 124
3202 6428 1945 3860 2663 3224 2997 2293 2191 3089 1641 6171 3971 4333 4521 4303
3921 4886 3790 3724 3327 6537 3593 1927 2774 3997 3868 5712 3149 3934 5896 3217
Note: 1 - Total OP for a reference period of 15 days: 375.3 lakh, 2 - Total OP for a reference period of 15 days: 375.3 lakh, 3 - Total number of ailments (rural): 408 lakh; Total number of ailments (urban): 154.5 lakh 4 and 5 - Total expenditure for outpatients is for the reference period of 15 days and for hospitalisation is for a reference period of 365 days. Total expenditure includes medical expenditure and all expenses other than medical expense incurred by the household for availing the treatment. SOURCE: NSSO, 52nd Round; Mahal et al. 2002
61
SECTION II
Annexure II Status of health infrastructure and outcome

Rural Number of Number of Percentage Percentage population PHCs/CHCs ANM/nurseof CHCs of CHCs (2001) per midwives with inadequately 100,000 per obstetrics/ equipped in population 100,000 gynaeinfrastrpopulation cology ucture Safe Full delivery immunization coverage (%) % of children under 3 years of age severely malnourished (below 3SD) Full ANC InstitutIMR ional (2002) deliveries U5 MR
Well performing States Kerala Tamil Nadu Andhra Pradesh Maharashtra Karnataka 23,574,449 34,921,681 55,401,067 55,777,647 34,889,033 4.4 4.3 2.9 4.2 5.5 31.0 30.7 21.5 23.5 32.2 100.0 100.0 94.7 71.9 21 61 38 3 19 96.5 80.0 67.9 60.8 62.0 91 92 72 85 81 4.7 10.6 10.3 17.6 16.5 64.3 20.0 35.2 23.8 29.5 96.4 76.0 56.0 53.0 52.9 10 44 62 45 55 18.8 63.3 85.5 58.1 69.8
Moderate performing States Gujarat West Bengal Punjab Haryana 31,740,767 57,748,946 16,096,488 15,029,260 3.9 2.4 3.7 3.1 26.2 19.5 28.9 14.6 63.2 33.3 61.1 18 20 43 20 59.1 42.3 61.3 44.1 68 78 74 56 16.2 16.3 8.8 10.1 22.1 11.7 13.6 9.9 51.1 40.2 37.7 28.5 60 49 51 62 85.1 67.6 72.1 76.8
Poor performing States Rajasthan 43,292,813 Orissa 31,287,422 Madhya Pradesh 44,380,878 Uttar Pradesh 131,658,339 Assam 23,216,288 Bihar 74,316,709 India Census, 2001 4.5 4.8 4.6 3.1 3.1 3.2 Rural Health Statistics, 2002 48.5 23.4 25.7 17.8 26.5 10.1 Rural Health Statistics, 2002 62.0 26.4 41.1 80.0 52.9 Facility Survey, 2004 25 79 74 21 75 Facility Survey, 1999 37.7 36.9 32.1 25.8 20.5 17.5 20 56 77 27 57 13 20.8 20.7 24.3 21.9 13.3 25.5 NFHS -2 3.6 11.3 5.6 3.9 4.5 RCH -2 26.3 25.9 20.5 17.9 13.8 13.3 78 87 85 80 70 61 114.9 104.4 137.6 122.5 89.5 105.1
MICS, CES-02, 2000 UNICEF
MICS SRS, NFHS -2000 2004 -2
PHC: primary health care; CHC: community health care; ANM: auxiliary nurse-midwife; ANC: antenatal care; U5MR: under-five mortality rate; NFHS: National Family Health Survey; RCH: Reproductive and Child Health; SRS: Sample Registration Survey
62
SECTION II
Annexure III Current Status of the Goals Laid Down in The National Health Policy of 1983
1. Indicators Status in 1983 Goals set for 2000 by NHP 1983 Current status
MMR IMR Leprosy Tuberculosis (% of disease arrested cases out of those detected) Blindness (%) Immunisation status TT (pregnant women) TT (school children) DPT (children below 3 years) Polio (infants) BCG (infants) DT (new school entrants 5-6 years) Typhoid (new school entrants 5-6 years) Pregnant mothers receiving ante-natal care (%) Deliveries by trained birth attendants (%)
4-5 (1976) 125 (1978) 20 (% of disease arrested cases out of those detected) 50 1.4 20 20 25 5 65 20 2 40-50 30-35
Below 2 Below 60 80(% of disease arrested cases out of those detected) 90 0.3 100 100 85 85 85 85 85 100 100
4.1 (1998, SRS) 63 (MoHFW, 2002)
86 1.03(M0HFW, 2003) 60.3 (MICS-2000) 46.6 (MICS-2000) 58.9 (MICS-2000) 67.7 (MICS-2000)
62% (MICS-2000) 42.5% ( Deliveries by Skilled birth attendant - MICS-2000)
SOURCE: NHP 1983; MQHFW various years; SRS, 1998 , We could have saved 14.3 lakh infants and 8 lakh mothers, if we had achieved the IMR and MMR goals set by NHP 1983.
63
SECTION II
Annexure IV Implementation of National Health Policy, 1983

2. Strategic interventions as per NHP 1983 Current status
Universal, comprehensive primary health care services Integration for all plans for health and human development i.e. agriculture, food production, water, sanitation, housing, education, drugs and pharmaceutical, prevention of food adulteration, conservation of environment. To formulate a National Medical and Health Education Policy
Not Done Not Done
Policy brought out in 2000 Primary health care system to be given importance and to be decentralized. Achieve a well-dispersed network of comprehensive care, Not done transfer of knowledge, simple skills and technologies to health volunteers, use of inexpensive interventions, and more community participation. The decentralization of services to be linked to a well worked out referral system. Not done Establish nation-wide chain of sanitary-cum-epidemiological stations at the primary or secondary levels depending on local situations. Not done beginning made recently under the Disease Surveillance Project funded by the World Bank, 2004 Location of curative centres should be related to the population they serve keeping in view, densities, distances, topography and Not done transport connections Establish a 'Health Team' approach and to phase out the system of private practice of government doctors Not done Dovetailing of the functioning of the practitioners of various systems of medicines Not done Device State-wise health insurance schemes Not done
SOURCES: NSSO, 52nd Round; Mahal et al. 2002 NOTE: The total OP for a reference period of 15 days: 375.3 lakh. The total number of ailments (rural) is 408 lakh. The total number of ailments (urban) is 154.5 lakh.Total expenditure for outpatient is for the reference period of 15 days and for hospitalization is for a reference period of 365 days. Total expenditure includes medical expenditure and all expenses other than medical expense incurred by the household for availing the treatment.
64
SECTION II
Training for effective delivery of health services
UMAN RESOURCES ARE CRITICAL FOR EFFECTIVE IMPLEMENTATION OF HEALTH programmes and delivery of quality health care to achieve the national health policy goals in India. The availability of an adequate number of health personnel to effectively and efficiently manage and implement health programmes cannot be overemphasized. However, numbers alone may not necessarily lead to the desired changes in the health status and outcomes. It needs high levels of concern, commitment and competence among the health personnel responsible for the management and delivery of health care, especially health care providers at the grassroots level. Human resource needs have been increasing, with new health programmes being added to the package of health services over the past few decades, along with the growth of health infrastructure and expanding scope of the health services. Several new health programmes have been introduced and the strategies of existing programmes have been revised. These changes in health services and strategies have led to an increased need for developing new competencies and skills among health personnel, in addition to the increasing need for more human resources at various levels. There has been a phenomenal growth in human resources in the health sector, especially peripheral health functionaries and supervisors, who are directly responsible for implementing the interventions aiming at reducing maternal mortality, infant and child mortality, as well as reducing morbidity and mortality due to communicable diseases. There have been major gains in India's health status since Independence. Life expectancy has gone up from 36 years in 1951 to 64 years in 2000. The infant mortality rate (IMR) has come down from 146 in 1951 to 70 in 1999. The crude birth rate has been reduced from 40.8 in 1951 to 26.1 in 1999, and crude death rate from 25 to 8.7 during the same period. One of the major reasons for these gains has been the development of an impressively vast, three-tiered system of rural health infrastructure, with a subcentre for 5000 population, a PHC for 30,000 population, and a CHC for about 100,000 population. Immunization for the control of communicable diseases has made a major contribution to these gains; success stories include smallpox eradication, the near elimination of leprosy, and the extraordinary social mobilization for polio eradication. Over the past few decades, the support of UN agencies and other multilateral agencies contributed to Indian health system significantly. However, their focus and initiatives were limited to a select few areas: Transition from the so-called 'Family Planning' to 'Reproductive Health' strategy Targeting a few communicable diseases (e.g. polio and leprosy) for eradication Controlling diseases such as malaria, tuberculosis (TB), HIV/AIDS Strengthening the State-level health delivery system (9 States) Facilitating the health sector reform process Prioritizing health interventions according to need and disease burden Achievements notwithstanding, much more improvement is required in the health status of Indians. This becomes evident when one focuses on the progress made in the past decade. The problem of major communicable diseases remains unsolved and there is no significant progress in controlling TB, water-borne diseases and respiratory infections. Concurrently, epidemiological transition has set in and the burden of non-communicable diseases is increasing. Most of the health outcomes remained stagnant during the past decade while new challenges, such as HIV/AIDS, have surfaced to further stress the overstretched health care system. The situation in many
S.D. GUPTA, N. RAVICHANDRAN, DHIRENDRA KUMAR

INDIAN INSTITUTE OF HEALTH MANAGEMENT RESEARCH 1, PRABHU DAYAL MARG, NEAR SANGANER AIRPORT JAIPUR 302011, INDIA E-MAIL: sdgupta@iihmr.org
65
SECTION II
Indian States is grave. Several States lag behind with respect to key indicators at the national level, especially regarding maternal mortality ratio (MMR) and IMR. It is also to be noted that there has been no significant increase in public expenditure on health in recent years. The per capita public investment on health is one of the lowest in the world. The continuing gaps in the health status of the people may be attributed to several factors, mainly related to systemic issues, such as poor access and availability, inequity of distribution of health care, poor financing and management of the health systems. The health system's performance is seriously affected by poor human resource development, especially the competence and skills. This paper focuses mainly on the issues related to human resource development with a focus on training. The current status of competency and skills raises several questions. Do health care givers at the peripheral level lack the competency and skills to implement interventions correctly? Does the system have the potentiality to build the capacity of these health functionaries? Is the quality of training appropriate? Is there a preparedness embedded in the system to develop the requisite skills in providing health services effectively?
decline and behaviour changes in health, nutrition and development that States have been experiencing. The differences and differentials in population health and development programmes prompted our attention to these States for the study. One district in each State, Khammam in Andhra Pradesh, and Udaipur in Rajasthan were identified for detailed discussion and fieldwork. A step-wise approach was adopted. The following were main steps:
Step I: Review of training documents and records

Published material and policy documents of the Government were reviewed. The materials available at the State Institutes of Health and Family Welfare (SIHFWs) and other institutions of the two districts were referred to and reviewed.
Step II: Rapid assessment and field visit to the SIHFW

The SIHFWs of Andhra Pradesh and Rajasthan were visited. The performances of the regional Health and Family Welfare Training Centres (HFWTC) and District Training Centres (DTC) were reviewed. The PHCs of two districts were surveyed. The following procedure was used to undertake rapid assessment: Personal interview with health functionaries (structured questionnaires were developed to record the information from the functionaries available at the health centre) Group discussions with health personnel using checklists Record/document scrutiny using checklists Observation All existing relevant documents and reports were consulted and field visits made to obtain first-hand knowledge of issues, problems and concerns. Deliberations were held at the national, state, district and block levels with current and former policymakers, health administrators, training coordinators, executives of training institutes and leading researchers. Interviews
Approach and methodology

We adopted systems framework approach to analyse and understand the various factors and constraints in training for capacity building and skills development in the health sector. The systems framework has typically three dimensionsinputs, process and outcomes, and a review and analysis of training has been undertaken in each of these dimensions (Fig. 1). A multipronged procedure was used to collect data. A triage in the form of a combination of asking questions, making observations and reviewing relevant records and reports was adopted. As the demographic transition is under way in most of the States in India, the case of Andhra Pradesh and Rajasthan have been of great interest. The level of socioeconomic development is not high enough to justify the kind of mortality
Fig. 1 The systems framework

Process Identification of training load Training needs assessment Training approaches Persons who attended the training Post-training follow-up Faculty development and growth Type of training conducted Training in quality assessment Training in impact assessment Enabling environment
Input Infrastructure Institution Manpower (Training faculty/trainees) Training material and aids Financial resources
Output Persons trained and able to achieve the goal Reduction in maternal and child morbidity and mortality Impact assessment
66
SECTION II
and group discussions with important officials of various multisectoral programmes with RCH components were conducted at the levels of the PHC, Mandal PHC and District with selected administrators, project officers, district medical and health officers and medical officers. In the process, nearly 150 individuals-programme functionaries, administrators, training coordinators, trainers, project officers, civil surgeons, medical officers, pharmacists, auxiliary nurse-midwives (ANMs), multipurpose health workers (MPHWs), supervisors, laboratory technicians and others-were engaged in lengthy sessions. The team members collated the relevant information generated in the field and the documents collected from the State headquarters in Jaipur and Hyderabad for finalizing the report. There is ambivalence, confusion and differences in points of view on various facets of training. However, the deliberations were guided by the utmost objectivity, avoiding any bias in our analysis and keeping in view the overall interest of the programme. The following health functionaries were interviewed to assess the perceived knowledge of their skills and actual gaps in their knowledge (Table 1). The effort was more on eliciting qual-
itative information rather than being distracted or overwhelmed by magical statistical significance.
Training policy, infrastructure and system

Training policy
Despite increasing realization of the importance of training in human resource development, no serious efforts have been made to develop an effective and comprehensive training policy at the National and State levels in the health sector. A training policy is needed that identifies priorities and training needs, types of training, processes and mechanisms, training institutions and cadre, quality assurance, and monitoring and evaluation of effectiveness. The States had no such training policy, mainly due to the low priority assigned to training and a wrong perception of training being a timeand money-wasting intervention that has failed to enhance performance and improve effectiveness of health care services in achieving the desired goals of reducing mortality, especially maternal and infant mortality. However, Rajasthan has made efforts to develop state training policies, although the implementation of these remains questionable. The training policy has been at the draft stage for the past eight years. The draft training policy could not see the logical end of acceptance and approval by the Government. Andhra Pradesh did not even start initiatives in this direction. The States do not have a manpower policy or training policy. The manpower planning and human resource development process is ad hoc and generally follows the national norms based on population ratio.
Table 1 Distribution of health functionaries interviewed (in numbers)

Health functionaries Rajasthan Andhra Pradesh
At PHCs Doctors 17 Clinical nurse ANMs 20 MPHW (M) Staff nurse 8 Laboratory technician 8 Pharmacist 4 Supervisor 8 At Mandal PHC Doctors ANMs Staff nurse Laboratory technician Pharmacist Supervisor At District Level Doctors 2 Staff nurses 4 Laboratory technician 2 CMHO 1 At private institution/programme personnel Project officer of District Training Team cum District Immunization Officer Senior assistant (clerical) NGOs/Institutes (course coordinators) 12
6 5 22 7 4 4 4 6 2 2 1 1 1 2 2 3 2 1
Training infrastructure and system

There is no separate manpower planning division in the State Directorate of Medical and Health Services in the study state of Rajasthan. However, Andhra Pradesh had a system in place. Both the States have an elaborate training infrastructure, but there is no well-functioning training system. There is now a vast training infrastructure in the States with significant growth in training capacity. There are well-established training institutions at various levels. These institutions mainly include the State Institutes of Health and Family Welfare (SIHFW), Health and Family Welfare Training Centres (HFWTC), District Training Centres (DTC) and ANM Training Centres (ANMTC). There are also some other types of training institutions in the states in addition to these.
State Institutes of Health and Family Welfare

These were envisaged as state-level institutes that would provide leadership to all other training institutions in the respective States. The administrative and technical control of all training institutions in the State would be vested with the SIHFW, which would perform a higher role by providing training of trainers, coordinating the entire training network and system, and organizing in-service training for senior health professionals.
1 2 16
67
SECTION II
In Rajasthan, the SIHFW has been created under IPP-IX to ensure autonomy and flexibility. While the SIHFW is expected to play a crucial role in planning, designing and coordinating training in the State, it is in a pathetic condition. There has been no regular director for about four years. There is no regular faculty available and most of the faculty positions are lying vacant. There is physical infrastructure but it is yet to be developed to the desired level. The SIHFW did not have its own field practice area for hands-on practical training and for undertaking operational research in the health systems as well as training interventions. As such, there is no training budget in the State and for the Institute. Funding from the World Bank has come to an end, thus bringing uncertainty in staff salary and continuation of training programmes. There was no training software development activity, such as designing new training programmes and curricula, developing materials and new training pedagogy. The present situation of the premier training institute reflects not only its apathy to training but also the level of priority accorded to capacity development in the State. In contrast, the Andhra Pradesh SIHFW has established itself as a national-level institution, named the Indian Institute of Health and Family Welfare (IIHFW) and has made an effective contribution to the capacity-building process in the State. While it has developed an excellent training infrastructure and trainers/faculty, it has also developed training software. Further, the AP SIHFW is financially self-sustaining through generating revenue from operational research, projects and programme evaluation and consultancy. With some exceptions, the SIHFWs in most States in India have not performed well, to the disappointment of policymakers, administrators, programme managers and funding organizations. These institutions are not prepared to undertake quality training and overall human resource development. The SIHFWs are struggling for funds; leadership is not regular; qualified and experienced faculty is not available; no training of trainers is conducted; and they are not involved in and entrusted with the planning and development process for the training of health personnel.
age, for two districts. These DTCs work under the supervision and control of the SIHFW. Though DTCs have their own building, these are used for other purposes. A large part of the DTC building houses the CMHO offices or stores and warehouses. They suffer from the chronic problems of lack of effective trainers, training software and equipment. The situation is better in Andhra Pradesh. ANMTCs conduct basic training for ANMs/Health Workers (female). They still follow the old curriculum prescribed by the Nursing Council of India. Efforts to revise the curriculum to meet the changing training needs have failed so far. The physical condition of the ANMTC buildings is pathetic and hostel facilities are severely limited. Training is usually conducted in district hospitals, and community-based training of ANMs is neglected. Skills to implement interventions that would reduce maternal and infant mortality are lacking.
Training process
The training processes are not streamlined and systematic. The process of organizing a training programme is as follows: Decisions are taken to provide training under the funded project or programme; training plans are laid down in the form of training load for various categories of health personnel and types of training, keeping in view the target number; and the calendar is prepared for conducting training. The first training course has to start without delay. The syllabus is developed in a hurry by the faculty. Opinion may be taken from other resource persons, which ensures that the subject 'gets covered'. A curriculum is finalized and resource persons may be identified. Circulars are issued to the district officials to nominate staff of the particular category (generally the circulars are not received in time and information often reaches the prospective participant after the start of the programme, resulting in their joining the programme late) (Box 1). Circulars generally contain instructions to nominate participants without explaining the purpose, objectives and contents of the programme. No criteria for selection are mentioned. Hence, anyone is nominated. Box 1 shows that some directives are simply too ambitious in their goal, and are often not feasible for field implementation. As observed in various Government orders/reports, lack of implementation of the existing directives-from record-keeping to motivation, service delivery, training, supervision, monitoring and evaluation-makes many strategies and planning exercises redundant as far as actual operations are concerned. In the absence of clear-cut mechanisms for efficient execution, one would be doubtful whether new strategies, if any, would produce better results than their predecessors. For operational purposes, it would be important to distinguish between problems at the policy level and those purely at the execution level and, with respect to the latter, intra-, as against interdepartmental/directorate levels. Issues of policy order do need the attention of State Administrators/policy-makers and perhaps it is time-consuming to solve. Interdepartmental/directorate coordination can, however, be worked out reasonably fast through clearly established mechanisms. As far as interdepartmental/directorate problems are concerned, there
Health and Family Welfare Training Centres

HFWTCs were established as per the standard norms of the Ministry of Health and Family Welfare, Government of India. These would conduct in-service training of medical officers and trainers of DTCs. These HFWTCs have their own field practice areas but scarcely visit and utilize them for training in the field. The HFWTCs suffer from gaps in infrastructure, training equipment and aids, training material, and lack of qualified and experienced trainers. The libraries are ill-equipped and are virtually non-functional. However, the HFWTCs are not starved of funds. Funding is done by the Government of India.
District Training Centres

The DTCs are responsible for organizing regular in-service training programmes for health workers as well as basic training programmes. Each DTC is responsible, on an aver-
68
SECTION II
Box 1 How training processes are carried out

For instance: A government directive No states to CM & HO, AP , '. It decided that under IPP (VI), we would like to train 30 ANMs and MPHWs in the forthcoming Integrated Skill Development Programmes at Hyderabad. Please nominate the staff within 10 days of this notice'. Counter-discussion with higher authority at the district level, revealed that 'we have received the letter (directives) only yesterday that means, we have literarily five days left to nominate the staff Secondly, it is not at all possible to send 30 staff from the district at one point of time We would be sending only 4 or 5 staff for training. Now we need training on dengue fever related issues as 126 cases were registered in the last two months This is what is happening'. Interaction with the authority who issued the directives to district officials revealed the following :'...they (district officials) always say like that only. They don't do the job and facilitate the process in time...' When a researcher posed a question to the authority, 'Why did you send the directive to district officials in the last minute to nominate the staff?' the authority replied, '...as per the government order, I directed. And now fund is available and the Government wanted to initiate the training programme'
Box 2
All those who expressed the need for training stated that the training programme should be practice-oriented. In other words, training should be provided with hands-on experience. More than four-fifths of the respondents emphasized the skills development aspects of training. One suggestion came from the participants: 'As the needs of each category of health and non-health functionaries are different, exclusive modules could be produced catering to each category. 'A supervisor strengthens the above case-notes by adding, 'complicated issues like conducting deliveries, high-risk pregnancies, concepts like supportive supervision, syndromic case managements, etc. which are to be highlighted in skills training can easily be understood and put to practice if they are taught with the help of video films and hands-on experience with field exposure. 'In contrast, most of the doctors do not like training or orientation or refresher training, as they are aware of all aspects. Laboratory technicians and pharmacists do not know whether they will be upgraded or not due to their specified nature of jobs. For instance, a laboratory technician said, 'I have completed 17 years as malaria lab technician and am able to diagnosis diseases (100%) successfully. I do not want to learn further as I have specialized. 'Another laboratory technician added, 'Even if we learn, our profile will not change now because it is too late. A pharmacist echoed, 'I learnt all aspects during my Diploma in Pharmacy Course and that is sufficient to handle the amount of the job I have at the PHC. 'This reflects that in general no need-based study has been conducted to organize refresher-training programmes. This further reinforces the need to strengthen training materials to improve the quality and maintain the uniformity and quality of training activities across PHCs. Review of records with the DM&HO office indicated that in a majority of the cases staff stayed in a particular PHC for more than a decade without any change in their job profile. It was observed that the level of participation of staff was not equal in the PHCs. Generally, just one or two were active and took a lead role in providing services, meeting the targets and getting involved in various issues. Most played a passive role. More than two-thirds participants belonged to this latter category.
is no acceptable reason why they should be allowed to remain unanswered and no initiatives taken to provide need-based training/human resources development. Generally the number of participants is low, some fail to report for various reasons. The method of 'training' is mostly dominated by lectures by the faculty or resource persons with little respect to continuum of the theme and the overall perspective of the programme (Box 2). Random simulation exercises are developed and used. Group work, which is now a commonly used approach, is unstructured and unguided, and is used as a time filler. The focus is on knowledge rather than on competencies for action. Trainers feel that the participants will at least be 'exposed' to the subject or 'oriented'. The programme is over within the stipulated time period. The new batch arrives and the same process is repeated. The above paragraphs succinctly summarize the whole process of training, and the implicit assumption that generating knowledge rather than building competencies may empower the health care provider to deliver services effectively. There is no effort to build an appropriate training environment that is conducive to learning, raising concerns and enhancing the commitment of health personnel. At the end, there is no behaviour change and participants leave for their respective places of work with the perception that it was yet another training of no use to them. There is no monitoring and follow-up to assess change in performance and effectiveness of the programme. The training programmes are
overwhelmed with the assumptions that participants' acquisition of knowledge means greater competence; learning is a simple function of the capacity of participants to take in and the ability of trainers to teach; and individual improvement leads to improvement in the organization.
Views on public-private partnership in training

Private institutions involved in providing training to health professionals and health care providers were contacted for their views on partnership between them and government
69
SECTION II
institutions for organizing training and capacity-building programmes. Institutional heads, course coordinators and other related officials in 16 institutions in AP and 10 in Rajasthan were interviewed. They unanimously stated that there was no organic link between private and government institutions for training. No systematic and innovative efforts have been made to understand the training needs of the clients. They stated that: The training programme at public health institutions is last leg work. Delay in nomination which further delayed the programme. One of the private institutions interviewed added, 'At the year end, the PHS requests us to conduct a training programme for about 9500 PHC staff within a two-month time period.' In addition to these, often due to paucity of funds, the Government pruned down the budget and reduced the total duration of training modules. The whole exercise of revisiting the list of topics, duration, and categories was redone after several rounds of discussions. This resulted in considerable loss of time and slackened the progress of the project. Because of these reasons, formal processes that solicit and identify the gaps between the current and required outputs were not worked out. Due to these, the following issues were not addressed at all (i) whether training is relevant, (ii) whether training will make a difference, (iii) whether focusing training needs of organizational problems should be done along with skills development issues, and (iv) whether an improved role should be linked with training goals and the bottom line. A private training institute stated categorically that 'we stopped doing training programmes for the Government of AP due to its attitudes in reducing funds, last minute patchwork and using pressure tactics on us'
Box 3
Almost all ANMs who participated in the study revealed that 'The training modules should be pre-tested in tune with each category of the staff job responsibilities and field situations. At the same time, the training approaches which failed to produce satisfactory results should be dropped... 'Interestingly, MPHWs (male) who are in less number, said, 'The training module for male health workers should be aimed at increasing male participation in family welfare programmes 'The above case studies reflect that the training modules should be based on actual field situations, and location and characters should, as far as possible, be close to the actual nature to project ground realities and objectivity. The training coordinator should involve the concerned trainer or expert team members at every stage of preparation and editing of training materials and during the training programme and review the material on a time-to-time basis. Based on the suggestion, the changes should be incorporated in the next training session itself.
Perceived knowledge and training needs

Are the training programmes designed to address the competency needs required to perform specific tasks? An analysis of the knowledge and training needs perceived by the key functionaries was highly revealing. A questionnaire was administered to all health care personnel to assess their knowledge on health and diseases of public health importance, their role, and related aspects. It revealed that the level of knowledge regarding national health programmes such as immunization was almost 100%. However, their knowledge of the other national programmes such as TB control, AIDS control, malaria eradication, leprosy eradication and others was very limited. Only medical officers had knowledge of these aspects. The ANMs and male MPWs, who are entrusted with the implementation of health programmes at the subcentre and village levels, did not posses adequate knowledge of national programmes that have been executed in their areas. Almost all pharmacists and laboratory technicians were not aware of the national programmes. Even knowledge of the basic antenatal care process was limited. Though medical officers and ANMs described the process
correctly, the majority of supervisors (LHVs) could not (Box 3). Only 60% of the ANMs and supervisors were confident of their skills in screening risk factors during pregnancy. Furthermore, interaction with health functionaries revealed that only doctors had adequate skills in recording blood pressure, while none of ANMs and LHVs had skills in measuring blood pressure, which is a very important procedure for assessment of risk (Table 2). Further, the majority of the PHC staff was not aware of their job responsibilities as compared to their counterparts who are at Mandal PHCs and District Hospitals. A further analysis of skills specifically needed to avert maternal and child deaths was also undertaken with key staff members. The skills needed and the current levels of knowledge are presented in the following matrix. Table 2 shows the main causes of maternal deaths and essential interventions to avert these. The next column shows the current levels of skills to undertake the recommended interventions. It clearly shows that the ANMs and LHVs did not possess these skills. How can one expect reduction in the maternal mortality rates in the country? ANMs and LHVs are expected to learn these skills during their nursing training. Given the quality of nursing training on the one hand and the poor emphasis on community-based obstetrics during the course on the other, one cannot expect them to perform. Practically no in-service training of these health care providers is conducted to develop their clinical skills (Box 4). The effectiveness of the recent in-service clinical training of ANMs and LHVs under the RCH Programme is questionable as no serious efforts were made to give on hands-on clinical practice. A similar analysis of skills was undertaken with respect to neonatal mortality, which accounts for almost two-thirds of infant mortality (Tables 3a and 3b). The causes of neonatal death and required interventions are well known. Neonatal death is closely associated with the obstetric process; therefore, it would require effective obstetric skills among ANMs and LHVs.
70
SECTION II
Table 2 Perceived knowledge about skills and actual gap

Area Staff interviewed Perceived awareness (%) AP Rajasthan
National Programme on Women and Child Malaria, TB, AIDS and leprosy, etc. Maternal Health Enumerate the process correctly for providing ANC
Doctors ANM MPHW/LHV (M &F) Staff nurse LT/Pharmacist Doctors ANM MPHW/LHV (M &F) Staff nurse LT/Pharmacist Doctors ANM MPHW/LHV (M &F) Staff nurse Doctors ANM MPHW/LHV (M &F) Staff nurse Doctors ANM MPHW/LHV (M &F) Staff nurse Doctors ANM MPHW/LHV (M &F) Staff nurse Doctors ANM MPHW/LHV (M &F) Staff nurse Doctors ANM MPHW/LHV (M &F) Staff nurse
50 5 16 37 10 93 94 72 100 25 100 55 72 100 86 55 0 100 64 67 32 100 71 55 0 100 100 100 0 100 43 67 68 62
47 10 25 13 8 94 90 62 87 100 60 62 100 88 0 0 100 80 0 0 100 88 70 75 87 100 100 0 100 70 90 87 87
What do you do in ANC? Screen for risk factors and medical conditions
Record BP*
Weight and height
Screen for anaemia
Give tetanus toxoid
of neonatal death, which can easily be avoided by efficient obstetric care and subsequent newborn care including aspiration of mucus and amniotic fluid. Simple interventions are available for acute respiratory infection (ARI), diarrhoea and neonatal tetanus. However, health functionaries and supervisors had limited skills in prevention and management. Hypothermia, an important cause of neonatal death, was not considered a priority by these functionaries. These findings are not new and this is not the first time they have been reported. The need for training was positively perceived by all the members interviewed, irrespective of their titles and level of functioning. They stated that the idea of training is good and expressed the need for training in essential and emergency obstetric care and essential newborn care. Need for training was also identified for communicable and infectious diseases, diagnosis of syndrome-based diseases, sanitation and control of epidemics, methods and measures of eliciting cooperation and coordination of the community, and sexually transmitted infection (STI) and HIV/AIDS counselling and their first-aid treatment. These findings also underscore the need for enhancing communication skills.
Provide education for nutrition
*Only weight taken
Furthermore, for improving neonatal outcomes, these functionaries are expected to possess skills of essential newborn care. Birth asphyxia and birth injuries are very important causes
All those who expressed the need for training stated that the training programme should be practice-oriented. In other words, training should be provided with handson experience with field exposure. The majority of respondents emphasized the skills development aspects of training.
Box 4
It was observed that the greatest problem for the MPHW (F) in Khammam was that while the approach lacked good communication abilities that are essential for dialogue delivery, the professional trainer lacked the required technical skills such as explaining complicated delivery. The researcher asked the Trainer a question, Researcher: 'What are the measures to be taken for a pregnant woman with malaria? ' Trainer: 'I suggest that not medicines should be given; please refer the pregnant woman immediately the to PHC. 'Researcher: 'As I am coming from a non-medical background, (to a pregnant woman with severe malaria and not able to move from her village) what type of first-aid measures would be given? 'Trainer: 'Give a paracetamol tablet and better refer her to a nearby medical centre and not take risk. 'This question was specifically asked by the researcher during his visit to the MPHW (F) training at Khammam district, AP as there were 13 pregnant women who had died due to malaria in September 2004. The investigation reports of CM & HO regarding these 13 deaths said that these were due to 'the negligence of PHC and District Hospital 'This revealed that the technical training sessions are conducted without the concerned specialist/expert who can provide better suggestions (than the trainer) to improve the services.
71
SECTION II
Table 3a Maternal deaths: Select causes, main interventions and skill levels
Causes of maternal death Interventions Current levels of skills of ANMs and LHVs
Antepartum haemorrhage (APH)
Postpartum haemorrhage (PPH)
Puerperal sepsis
Pregnancy-induced hypertension (PIH)Eclampsia/toxaemia
Obstructed labour
Complications of abortion
Early identification of bleeding during pregnancy Counselling Continued risk assessment Referral Prevent and treat anaemia in pregnancy (prophylactic and therapeutic) Early identification and risk assessment Skilled attendant at birth Manual removal of placenta Prevent/treat bleeding with appropriate drugs Replace fluid loss by IV drip/transfusion, if severe Early referral and transport Skills in aseptic delivery Clean practices during delivery Administration of antibiotics Early identification of risk in pregnancy Counselling Treat eclampsia with appropriate anticonvulsive drugs Urgent delivery-Caesarean section if needed Pelvic assessment Referral Assisted delivery or caesarean section as per indications Identify and diagnose complications Treat sepsis-antibiotics Fluid replacement if necessary Referral
Poor knowledge of APH Poor APH management skills
Poor knowledge of PPH Poor skills to diagnose and manage PPH including manual removal of placenta Poor skills of blood/IV transfusion
Poor knowledge of puerperal sepsis and its management Poor knowledge of PIH Poor counselling skills Poor management skills Poor pelvic assessment Poor management skills of obstructed labour Poor knowledge and skills in managing complications
Table 3b Infant (neonatal) deaths: Select causes, main interventions and skill levels
Cause of death Interventions Current levels of skills of ANMs and LHVs
Birth asphyxia Birth injury Prematurity
Congenital malformation
Neonatal jaundice Neonatal tetanus ARI-pneumonia Diarrhoea
Safe delivery practices Proper newborn care Safe delivery practices Newborn care Proper antenatal care Supplementary nutrition (IFA) Proper newborn Care Proper counseling Screening during ANC Newborn care Proper newborn care Aseptic delivery TT immunization of mother Proper management of ARI Proper diarrhoea management
Inadequate skills for obstetric care Lack of skills in newborn care Inadequate skills for obstetric care Lack of skills in newborn care Inadequate skills to assess foetal growth Inadequate newborn care skills Poor counselling skills Inadequate newborn care skills Inadequate newborn care skills Inadequate skills in aseptic delivery Poor diagnostic and assessment skills for severity of ARI Poor assessment skills for severity of diarrhoea
IFA: iron-folic acid; TT: tetanus toxoid; ANC: antenatal care; ARI acute respiratory infection
Constraints in training for better performance

The review process revealed glaring inadequacies in the human resource development process and training of the health personnel. Some salient observations are summarized here:
Training institutions and training have received a low priority. There is a generalized apathy towards training and capacity building. Training is not recognized as an intervention to improve performance. Owing to lack of nominations, programmes are frequently cancelled. The function of training is seen in isolation. There is no proper
72
SECTION II
planning and implementation of training programmes. In the development of training programmes, the training needs and expectations of participants are not considered. Most of the programmes are lecture-based and didactic in nature. There is no focus on practical skills development. Even in clinical skills development programmes for ANMs and LHVs, scant attention was paid to giving practice to the participants. The morale of trainers is low. There is no training cadre in the States. There is no system for appointing trainers. Normally, persons are posted or deputed to training institutions as trainers rather than regularly selected. There is no career stream in training. There are no facilities for regular professional development of trainers. The SIHFWs, HFWTCs and DTCs are poorly equipped with hostels, training infrastructure and libraries. The physical facilities at ANMTCs are appalling. Various training programmes are offered under various Programmes and a health worker is nominated more than once to attend different training programmes. The multiplicity of training was a constraint in work performance. Incompetent trainers and lack of technical guidance to training institutions has resulted in poor quality training, thus lowering the credibility of training institutions. Trainers of various training centres feel that there are no formal linkages among these institutions and they feel left out. There are financial constraints. The payment of TA/DA to participants, procedures/facilities for inviting guest faculty and lack of funds for developing good- quality training material are major problems. Training is not taken seriously by the trainees as it has no relationship with career development of health professionals. The current appraisal system does not take into account the training received in placement or promotion. There is no system of nomination for training. It is highly centralized and, more often than not, based upon personal fancy or preference of the concerned officers. There are no norms for in-service training. Some health personnel attend training programmes frequently irrespective of their utility in their job. The training is not seen as an intervention for improved job performance by most trainers. This is because there is a mismatch between organizational and personal goals. The need for management training is seldom felt by functionaries and health administrators. It is thrust upon them. There is no linkage between service providers and trainers. Training is viewed as a constraint in achieving programme objectives rather than facilitating them. There is no training or personnel information system in the States. As result, there is no proper planning. There is no thinking on operational research in training institutions.
ability of health manpower have improved, but the productivity and performance have remained poor. There is a need to reposition training, which should be accorded a high priority. There is a need for rethinking. Is it a knowledge-building process or skills development intervention for better performance, or both? The experience of the previous decades suggests that current training approaches have not yielded the desired changes in health status or performance of the organization. There is a need for better workforce management and improving the working conditions to enhance potential and improve performance. Training should be seen as a part of the overall process of human development. There is a need to consider changing the current training paradigm from knowledge- and competence-building to organizational transformation. The time has come to seriously consider training as an intervention. While training should emphasize skills development to perform tasks effectively, training designs should be re-oriented to ensure a change in the attitude and mindset of health care providers at all levels to achieve high organizational and professional commitment.
Develop HRD and training policy

There is an urgent need to develop a health manpower policy at the national level as well as in the States, clearly stating the priorities, future projections of manpower needs in different categories, policies for recruitment, transfer and promotions of the health cadres. The health manpower policy should also consider the creation of a public health cadre. The emphasis should be on creating a climate for independent and interdependent work rather than dependency. Adequate salary, good working conditions, job security, physical facilities, good human relations and the quality of supervision contribute to job satisfaction of the employees. Factors such as recognition of work done, opportunity for growth, nature of work, responsibility and the challenges of the task have been found to play an important role in creating motivation to work. These need to be considered to improve employee productivity. The roles and responsibilities at each level, from the Directorate to the subcentre level, should be clearly identified and documented. This is necessary for enhancing accountability and achieving the desired goals and outcomes. A clear training policy should be developed and implemented. The training policy should identify: Priorities and training needs Types of training Criteria for nomination for training Mechanism for linking training with promotion Mechanism of integration of various training activities Management of training institutions Training cadres, especially for trainers System for quality assurance in training Mechanism for monitoring and evaluation of training Support to the health system Emphasis should be given to develop an effective and
Way to go in the future

Over the past five decades, the emphasis has been more on the quantity, and the quality of human resources has taken a back seat. As a result, the indicators of access and avail-
73
SECTION II
functional training system, a system that works. Some the guiding principles include: Decentralization of planning, monitoring, evaluation and decision-making Autonomy to all training institutions Accountability in terms of training effectiveness and efficient utilization of resources Synergy with the client system Linking the training function with HRD, especially the career system Openness to continuous feedback, new ideas and developments
An elaborate training strategy and curriculum have been developed (and implemented!) for orientation of staff at various levels of the health system. These include managerial as well as clinical training. Medical officers, LHVs and ANMs undergo a three-day to one-week clinical training programme in district hospitals or medical college hospitals. The duration of clinical skills training is abysmally small and little effort is made to provide hands-on practical skills development opportunities. There is a need to re-design these training programmes for a longer duration focusing on clinical skills development. The trainees may be attached to hospitals for the appropriate time duration with clear objectives and tasks to be achieved.
Strategy for improving training

The following key strategies are suggested to strengthen training:
Develop learning resource materials

Availability of appropriate learning material (course material) is crucial. The training institute should develop the requisite training material and its uniformity should be ensured. Several training manuals have been developed which can be updated and modified. Unfortunately, the modules are not updated and adapted to the needs. Further, such material is not available to the participants.
Strengthen training institutions

Strengthening of the training institutions is quintessential to enhance the effectiveness of training. The SIHFWs should be seen as apex training institutions for planning, development and research. These institutes should be adequately supported for infrastructure and faculty development. Continued funding of SIHFWs should be ensured. SIHFWs should also be entrusted with coordination and control of other training institutions. These training institutions should develop close collaborative linkages with district hospitals and the district health administration.
Develop alternate training approaches

Distance leaning programmes should be designed for ongoing training programmes for various categories of health personnel with an accountability system and compulsory requirement. Recent developments in the IT sector must be harnessed. Recent developments in the use of satellite technology must be explored to deliver training programmes uniformly with high quality in a very short time. Pilot projects (GRAMSAT) undertaken in collaboration with Indian Space Research Organization (ISRO) have been found to be very successful in conducting training programmes effectively. More recently, ISRO has launched EDUSAT for education and training. Application of these technologies in developing training programmes in health care will prove to be cost-effective and have a high level of efficiency without displacing health personnel from their workplace.
Identify trainers and build their capacity

Qualified and experienced trainers are critical for quality training. A system of identification and recruitment of the faculty for training institutions should be developed. A search should be conducted to identify health professionals who are interested in training and wish to take it up as a career. They should be given good trainers' training and exposure to the training process. Additional financial incentives should be considered and career development opportunities should be created for trainers. A regular programme for training of trainers and refresher courses should be organized.
Develop a functional field practice area Develop and design need-based training programmes
Designing appropriate curricula and pedagogy for training are prerequisites for addressing the competency needs of health care providers. A needs assessment as perceived and expressed by the health care providers should be undertaken vis--vis programme goals and objectives and interventions thereof. Brainstorming sessions with trainees and programme managers may be carried out to identify needs. Further, evaluation of training programmes and participants' feedback would be very useful to improve the programme design. Practical sessions and hands-on experience would be very useful for improving skills. The SIHFW and HFWTC should develop a field practice area for the purposes of demonstration and exposure to field situations. These field practice areas could be used for testing new interventions and conducting operations research. It would need designated staff and required mobility. At present, field practice areas are adopted by the HFWTCs but these are nonfunctional and practically redundant for various reasons.
Develop monitoring and evaluation systems

Training should be accountable. A regular monitoring system should be developed and critical indicators should be identified with emphasis on measuring training effective-
74
SECTION II
ness in terms of performance and improved programme implementation. There should be a mechanism for regular interaction with the trainees, and providing feedback to them.
System's support for training

Training cannot be seen as a part of the overall system. There is a need to improve the working conditions of health workers and the facilities where services are delivered. Continuous supervisory support is critical for improving the performance and quality of services. While supervision should be regular with a feedback system, the supervisory skills and mechanisms must be strengthened. There should be serious thinking on absenteeism and appropriate interventions may be considered. One of the most important interventions is to increase the motivation levels and help develop ownership and accountability through effective OD interventions. Efforts should also be made to analyse the reasons for absenteeism and low performance.
Estimated cost
Training is a highly resource-intensive activity, though the costs are apparently not visible and realized. It would always be difficult, if not impossible, to justify the cost of training vis--vis results achieved. Generally, the visible part is the direct cost, i.e. cost incurred in developing, designing and implementing training programmes. The hidden costs such as participants' time away from work and salaries for those
days are taken into account. However, costing of a training programme is not without risk of being overestimated or underestimated. The scale of training, levels of the participants and duration of training determine the overall cost. The cost may be divided into two major heads: fixed cost, which would include money spent on programme development, faculty, venue, administration and logistics; and variable cost, which would include money spent on course material, lodging and boarding, and travel. A typical one-week programme for 20 mid-level professionals would cost about Rs 200,000 with the additional travel cost of Rs 50,000 within the State. This cost would meet the programme development cost, administration, logistics, board and lodging, course material and instructions. About 50% of this is the fixed cost and the remaining is the variable cost. The proportion of the fixed cost would decrease with an increasing number of course participants. Similarly, the cost of a one-week programme for 20 health care providers (at the district level) would cost about Rs 50,00060,000, all expenses inclusive. A detailed costing of strengthening training infrastructure, faculty development, distance learning programme and satellite-based programme needs to be done after extensive review of the system. The cost estimates provided are based on the norms proposed under various training programmes for donor-assisted projects. However, it would need a detailed costing study and resource-mapping exercise to arrive at an appropriate estimate.
75
SECTION II
Effective Integration of Indian Systems of Medicine in Health Care Delivery: People's Participation, Access and Choice in a Pluralistic Democracy
Overview of the Indian Systems of Medicine

NDIA HAS A RICH AND LIVING TRADITION OF HEALING. AS EARLY AS CIRCA 4000 BC, Sushrut, the father of surgery, stressed the need to integrate theory and practice. 'What is observed and demonstrated directly in practice and what is intuited by Shastra have to be mutually and judiciously integrated for the growth of knowledge.'1 India's strength has been this attitude of continuous creative assimilation of practical knowledge.
Vision of Health: Bharatiya Ayurvidya

For millennia, as per the Indian ethos, the major concerns and activities of life are four purusharthas-dharma, artha, kama and moksha, in that order.2 Dharma implies living ethically each day, fulfilling one's responsibilities and having faith. Artha implies wealth, in the true sense of prosperity and not mere money or currency notes. Karma covers fulfilment of desires, within the framework of dharma. Moksha implies freedom that transcends consciousness, the cramping identification with mere nameform.3 Knowledge that liberates the individual is called vidya. Vidya has two categories: apara and para, the knowledge essential for work, wealth and wishes on the one hand and wisdom and faith, leading to enlightenment and liberation, on the other. In a way, Ayurveda was a bridge between these two domains of knowledge. Ayurveda was defined uniquely by Charaka:4 'Wherein the beneficial and adverse influences leading, respectively, to happiness and misery and to life healthy or ill are described, besides the respective helpful and harmful measures are described and quantified that system is called Ayurveda.' It is an integral vision that , retains fidelity to the fundamental principles of gunas, doshas, dhatus and malas.5 The evidence-based practices and products of other Indian systems of medicine (ISM)-Siddha, Unani, Yoga, Homeopathy, etc.-also offer unique opportunities to fulfil unmet medical needs. The Unani system of medicine has been active in India for hundreds of years. Handbooks of simple Unani remedies for common ailments have been published by the Council and can easily be referred to for integrative medicine. Yoga in daily life offers advantages of health, equanimity and longevity, which are tangible, safe and economically viable.
Global and Local Attitudes to Indian Systems of Medicine

The spectrum of global attitudes to ISM varies from derisive ridicule to unconditional reverence.6 At one extreme, there is an organized tirade against Ayurveda as comprising toxic metal therapy7 and, at the other, a fundamentalism raising Ayurveda to a religious dogma, capable of solving all health problems. Instead of these fixed stances, we need a balanced, scientifically open and curiosity-driven mindset.8
ASHOK D.B.VAIDYA
E-MAIL: bhaspa@bom5.vsnl.net.in
Ayurveda, Yoga and Naturopathy, Unani, Siddha and Homeopathy (AYUSH): Current Status
The current official status of ISM needs a quantum jump in terms of (i) the quality of professionals, (ii) academic excellence in teaching, (iii) path-breaking research, and (iv) development of high performance in clinical services.
77
SECTION II
Effective Integration of Indian Systems of Medicine in Health Care Delivery
Several reports exist on the manpower, number of colleges, hospitals and dispensaries, specializations, etc. in Ayurveda, Siddha, etc. As per the provisional State-wise distribution provided by the Department of ISM and H,9 there are more practitioners of Ayurveda, Yoga and Naturopathy, Unani, Siddha and Homeopathy (AYUSH) than of allopathy in India. This widespread resource needs to be strengthened, retrained and effectively utilized in the national health care delivery system. Table 1 shows the profile of Ayurveda, Unani, Siddha, Yoga, Naturopathy and Homeopathy as per the Department of AYUSH Annual Report (2003-04). The number of practitioners of Ayurveda and Homeopathy are far more than those of Unani, Siddha and Naturopathy. This is obviously due to a much smaller number of educational facilities for these three ISM (Figs 1 and 2). It is recommended that a task force be urgently commissioned to establish more educational institutions for the Naturopathy, Unani and Siddha systems, with appropriate learning modules of integrative medicine (IM) useful for primary health care (PHC), including emergency obstetrics.
Fig 2 Fig. 2. Medical education facilities for ISM in India (number)
Table 1 Manpower and institutional profile of AYUSH in India

Facilities Ayurveda Unani Siddha Yoga Naturopathy Homeopathy
Registered medical practitioners 432,625 42,833 17,550 0 Dispensaries 13,925 881 399 70 Hospitals 2,253 255 276 8 Bed strength 43,803 5,031 2,386 115 Teaching institutions (undergraduate) 209 36 6 Upgraded postgraduate departments 59 8 2 Specialization/ postgraduation 16 7 6 -
532 52 17 922 8 -
201,484 5,398 290 14,087 180 27 3
Currently, ISM incorporate several diagnostic and other modalities of modern medicine (MM), a practice that needs to be encouraged. Ayurveda teaching hospitals must have excellent diagnostic facilities, including imaging and endoscopy. Operation theatres and obstetric units must be well equipped and functional. All ISM practitioners need in-depth training in emergency obstetrics and first-aid care, as well as in cardiopulmonary resuscitation. Life-saving drugs, of any system, must be understood and their rational usage taught to all practitioners.
Fig 1 Registered medical practitioners of AYUSH in India (number)
Strategy, Role and Knowledge Skills for Integration

Integrative medicine: Definition and scope
A suggestion gaining ground is for medical courses to include Yoga and the ISM so that students are able to get the best of all aspects of medical knowledge. Another proposal before the Government is to have a medical facility combining western medicine and ISM to provide comprehensive medical treatment as it is not possible to open hospitals having facilities for only traditional medicine (TM).10 Similarly, the World Health Organization (WHO) has been emphasizing that inclusion of TM in health care would have a positive impact. It would be particularly so if the practitioners of TM are also trained in certain key elements of allopathy and vice versa. IM is a new paradigm in health care that focuses on the synergy and deployment of the best aspects of diverse systems of medicine, in the best interest of the patients and the community. Attention to clinical evidence, long-term usage
78
SECTION II
and safety, accessibility, economic considerations and culture compatibility constitute the key elements of IM. Fulder stated that the line separating TM and MM has become fuzzy. The blending of TM-MM is more active.11 The scope of IM in the promotion of health and prevention of diseases is immense. Particularly for chronic and degenerative diseases, ISM have much to offer to MM. At all levels of health care-the home, school, community, dispensaries, nursing homes and hospitals-the integrative elements of ISM have to be identified, procured and deployed. Table 2 lists some of the common conditions wherein ISM can play a significant role.12 This list is by no means complete. There is an urgent need to carry out Ayurvedic pharmacoepidemiological studies to identify drug and non-drug modalities widely used in the field.13 Studies have been initiated for diabetes mellitus14 and arthritis.15
Table 2 Scope of Indian systems of medicine

Acute Subacute Chronic
Wounds and burns Sprains/swellings Pyrexia Colds and influenza Diarrhoea Conjunctivitis syndrome Malaria Urticaria Dysmenorrhoea
Eczema Indigestion Menorrhagia Sinusitis Constipation Herpes Splenomegaly Urinary infection Leucorrhoea
Diabetes Arthritis Asthma Cancer Parkinson disease Irritable bowel Acid peptic disease Memory-cognition Chronic fatigue
Horizons and Deliverables: Programmes and Performance

The process of integration of ISM in national health care can be strategically perceived at three horizons:
Horizon I: Ayurvedic/Unani (ISM) physicians have to be incorporated into all PHC teams, with adequate therapeutic resources at their disposal. They should be trained in basic PHC skills and emergency obstetrics (Desai, Sadhana, FOGSI, 2005). [Au? Pls provide the complete ref] Horizon II: All municipal and district-level hospitals would have full-fledged ISM outpatient departments (OPDs) and wards, panchakarma and dispensaries with adequate resources. The hospital management should be sensitized and trained to give due importance to ISM in patient care. Horizon III: Tertiary medical centres should have advanced ISM centres, incorporating education, research and sophisticated services. Private-public partnerships will have to be encouraged. IT-80 G and charitable status will be granted to hospitals depending on the size of the infrastructure and usage of ISM, defined on a case-by-case basis, such as general hospitals, specialty hospitals, etc. The deliverables for each horizon will have to be congruent with the goals envisaged. Table 3 lists the deliverables under each horizon. The programmes for the integration of ISM in health care will have to be drawn up at the grassroots level based on a model and then worked upwards. People's participation is vital for the process of integration to succeed. Health being a State subject, there is an urgent need to appoint a Director of Integrative Medicine (DIM) in each State, empowered with resources, personnel and reporting relationships (directly to the Chief Minister [CM]). As statewide needs are different and will have to be differentiated even further at the district levels, the DIM will have a Board of IM (BIM) to draw up the programme, with projects according to horizon priorities. Budgetary needs and resource-raising have to be the responsibility of the BIM, with a sizeable but accountable allocation from the Central Government. The current allocation of Rs 150 crore to the Department of AYUSH is miniscule and has to be raised at least ten-fold, as soon as possible, to assist the DIM. States will also have to evolve public-private resources.
Table 3 Deliverables as per horizons

Horizon I Horizon II Horizon III
Knowledge, attitudes, practices survey: PHC/ISM PHC learning internship Survey of facilities at PHC and needs Current number of ISM personnel CME for PHC/ISM Panchayat participation Health needs and ISM ISM in medical colleges Complementarity Records and documentation
Local self-government health officials: ISM Ambulant ISM care in private-public domains Resources for ISM: Central, State and local levels Linkages with PHCs: Aushadhis Computers and Ayusoft (C-DAC) Mobiles and tele-ISM Communications/education Panchakarma units Management synergy ISM dispensaries
Centres of excellence and model spread Ordinance of IT-80 G and ISM Private-public enterprises, ISM priority Bridges with ISM hospitals CME workshops for ISM Panchakarma units Rehabilitation ISM Computer networks Specialty integration Research and education
CME: continuing medical education; PHC: primary health care; ISM: Indian systems of medicine; C-DAC: Centre for Development of Advanced Computing
79
SECTION II
The performance of the State DIM and BIM, based on targets and deliverables as per the time schedule, will be accountable both to the public as well as to the Central Director General of IM (directly reporting to the Prime Minister [PM]). Total transparency and right to information of citizens should be ensured. The State DIMs will have websites displaying the progress, problems of and solutions to the programmes. Networking with State health services has to be harmonious. The existing State health infrastructure has to be strengthened in a phased manner, at both district and local levels.
Table 5 Recommendations for health and longevity

Government Community Individuals
Paradigm Shift in Health Care: Emphasis on Prevention

The promotion of health and prevention of diseases, as stressed by ISM, are currently considered to be of great importancegenomics and proteomics help predict a propensity for specific diseases, years or decades before they manifest. ISM can play a vital role in this changing scene of global health wisdom. Ayurgenomics and prakriti genomics can play a major role in the integration of ISM for long-term health-promotive and disease-preventive management.16,17 ISM have modalities in ahara, vihar and aushadhi, which can effectively fulfil the need for promotion of health and prevention of diseases. These need to be integrated even in school health programmes and in all community health projects at the grassroots level. Table 4 gives a list of certain nondrug modalities of ISM that deserve attention. Experts in the system will have to evolve incremental modules for schools and appropriate programmes for communities. There is thus a need to call a meeting of experts to prioritize ISM nondrug modalities.
Sanitation Safe water Clean air Adequate food Health education Encourage Ayurveda Avoid pollution Sports and games Production: rasayanas Respect for prevention Respect for seniors Monitor health care
Hygiene Guard the supply Gardens and parks Supply free Health centres Community vaidya Activist groups Group events Ensure supply Obesity clinics Award longevity Health indices
Personal cleanliness Purifiers Fresh air Pathya miatahar Health diary Griha Ayurveda Home/work milieu Walking/yoga Individualize intake Family weight control Revere the aged Commitment to health
General and specific prevention of the disease burden in India

Certain general conditions that are widespread in India can be addressed by the integration of ISM with MM. Iron deficiency is highly prevalent in India, both in the rural and the urban populations. It is proposed that ISM practitioners initially focus on the iron deficiency problem at the PHC level. The baseline survey, haemoglobin values and the response to Ayurvedic iron preparations, e.g. varitara loha bhasma or punarnava mandoor, will help gain community support for practitioners of ISM. Quantitative methods will also sensitize practitioners of ISM to other major endemic problems. A national task force will evolve the approaches, treatment modalities, etc. The programme would be evolved with appropriate software, data management and evaluation, with advice from the Centre for Development of Advanced Computing (C-DAC). Diarrhoea in children being a major disease burden, practitioners of ISM and MM should jointly work on preventive measures-pure water supply and precautions, and link these with laja manda of ISM and WHO-oral rehydration therapy (ORT) recommendations. Baseline prevalence and improvement due to interactions should be monitored to assess the performance of the team. Educational programmes in schools and for mothers should be integrated with diarrhoea prevention programmes. In the Gadchiroli district of Maharashtra, laypersons have been trained in health and basic elements of ISM, to be practised at home. This can go a long way in preventing infections, diseases, nutritional deficiencies and allergy in children. Emphasis has to be on: (i) intestinal helminths, (ii) vitamin A/D deficiencies, (iii) tuberculosis, (iv) malaria, (v) upper and lower respiratory tract infections, (vi) otitis media, (vii) poliomyelitis, (viii) whooping cough, (ix) malnutrition, etc. Women's infections and sexually transmitted diseases, including AIDS, can also benefit from inputs from ISM. In a project sponsored by the Department of AYUSH, panchavalkal has been shown to be safe and effective in leucorrhoea. The
Table 4 Non-drug modalities of therapy in ISM

ISM Modality Indications
Yoga Ayurveda
Asana-pranayama Pragnya-vivek Panchakarma
Stress, anxiety, asthma Pragnya-aparadha Dosha homeostasis Sama Srotavarodha Aggravation or relief Ageing problems
Naturopathy Unani Siddha Homeopathy
Fasting Water therapy Mizaz Kalpas
Avoidance of precipitation Migraine, allergy, etc.
Health and longevity

Ritucharya, Dinacharya and Swasthavritta have to be developed in health care.18Knowledge and suggestions for healthy ageing and a long life are available in ISM and must be adopted. Table 5 lists some Ayurvedic recommendations and other aspects of healthy ageing.
80
SECTION II
Mahatma Gandhi Institute of Medical Sciences (MGIMS), Sevagram can be developed as a centre for evolving models for the extension of ISM to the rural areas.
ISM modalities for the management of the identified diseases

Almost 25 years back, a meeting was held by the Indian Council of Medical Research (ICMR) to consider appropriate technology for PHC.19 At that time, a large number of medicinal plants were included in the Indian Pharmacopoeia (IP). These were later dropped. Now the IP is reconsidering the inclusion of selected Indian plants. But for integration of ISM at the PHC level, the Ayurvedic Pharmacopoeia and the Herbal Pharmacopoeia (IDMA-RRL) can also be utilized. Monitoring of the quality of integrated health care has to be transparent. A drug-utilization survey of Ayurvedic teaching hospitals and practising vaidyas would help to assess the usage frequency of Ayurvedic drugs. Prescribing habits in Ayurveda vary significantly according to the geographical location, vaidya's background, pharmaceutical advertisements, etc. Hence, the BIM has to play a pivotal role in each State.
The Botanical Survey of India (BSI) and National Botanical Research Institute (NBRI) can coordinate this effort. These data should be available on the Internet. A record of the list of GAVs and their status have to be kept at every district medical centre as well. Wherever villages are close to forests, special projects can be initiated for socioeconomic growth through medicinal plants. Ethnobotanical studies also have to be encouraged. Tribal belts must receive sizeable funds and expertise to conserve healing plants and preserve knowledge. Cooperative medicinal plant farms have to be established on the Amul Cooperative Model.
Gram aushadhi nirmana (GAN)

Villages with a population of more than 5000 persons must evolve a rural pharmacy of ISM remedies. The raw materials of the GAV as well as from other sources have to be properly stored and used. The GAN has to have minimal pharmaceutical facilities, viz. capsule-filling, granulation and tablet making, quath preparation, and ointments and creams. The village 'pharmacist' should be trained in simple manufacturing with periodic inspections for hygiene, operations, etc. Pharmacy colleges also need to be involved in ISM efforts for GAVs at the district level.
Primary health care and ISM services and products

The family and village communities should be the targets for and active participants in absorption of ISM in health care. The Gram Panchayat must have a Gram swasthya rakshak (GSR) as one of its members. The GSR must be trained as a basic primary health worker with additional training in basic ISM, first aid, core nursing and mother-child care. The GSR and Sarpanch should provide regular reports to the District Medical Officer (DMO) on the prevalence of diseases, unusual cases, epidemics, health statistics, health education activities, ISM data, hygiene, sanitation, problems of integration, etc. The office of the DMO will enter the data on a village-wise basis. The consolidated data analysis will be provided regularly to the State DIM, who will consolidate State data for the Central DGIM. A precise but simplified format will be evolved at each level, with help from C-DAC and the University of Pune.
Gram swasthya samvad (GSS)

The health of women, children and the elderly is often neglected. Gram Panchayats must hold quarterly GSS to review problems, obstacles and solutions for IM. The report of each GSS must be sent quarterly to the DMO. Referrals of serious cases and feedback on management must be reviewed. The village can invite practitioners of ISM from outside to conduct a camp along with the GSS. The Panchayat should take advantage of the expertise of those who have moved out of the village by informing them about the GSS. Their assistancefinancial, technical and professional-should enhance ISM integration at the village level. ISM health education mobile vans should be commissioned at the district level for audiovisual and other modes of learning.
Gram aushadhi vatika (GAV)

All village communities should preserve and enhance already existing groves. The Gram Panchayat and social forestry will harmonize efforts to create new groves and GAV, for a village or a group of villages. The list of herb plants and trees to be grown should be as per the climatic zone, water supply, soil and local medicinal requirements. The National Medicinal Plants Board (NMPB), Ayurvedic and agricultural institutes and universities should facilitate the process by providing seeds, planting materials, advice for cultivation and post-harvest practices, etc. Local religious, social, political and business leaders must be sensitized to the need for GAV. They have to be the champions of the cause. A database of existing medicinal plant nurseries, farms and large cultivated tracts must be created both state-wise and on a national basis.
Bridge model for secondary and tertiary care

Bhavan's SPARC and the Nandigram Trust have evolved a novel rural health programme-top medical experts and vaidyas from the city visit the rural centres (primary health centres).Those who need secondary and tertiary health care are referred to cities close by or a metropolis such as Mumbai to identified and empathetic specialists. ISM health care mobile vans can serve the important purpose of such bridge models. ISM colleges also have to be involved in this effort.
Quality of health care delivery in ISM

Historically, ISM were based on patient-physician relationships and trust. Hence, hardly any effort was made in the field of research in health care delivery. Even in ISM educaFinancing and Delivery of Health Care Services in India
81
SECTION II
tion, there is little emphasis on how to assess the quality of health care delivery. A task force needs to be established to evolve a module for the quality of care in ISM. Experts in ISM, health administrators and community medicine specialists have to identify criteria and health indices, which would be impacted by good ISM practices (GISMP). Each component of ISM has to evolve a set of guidelines for Good Clinical Service Practices (GCSP). For example, for diabetes mellitus, the New Millennium Indian Technology Leadership Initiative (NMITLI) national group has evolved GCSP guidelines. Such efforts would assist the process of effective integration of ISM for health care delivery.
3. Monitoring mechanisms and accountability should be introduced at the district, State and Central levels via data transparency, performance appraisals and quality assurance. 4. Private ISM practitioners should be encouraged to assist the integration process by financial, professional and social inducements. 5. Intersystem case presentation fora must be encouraged to enhance the quality of pluralism and emphasize the need for integrative care.
Pharmacoepidemiology and Reverse Pharmacology in ISM

A major drawback of the practice of ISM is the paucity of documentation of clinical records. The clinical notes are poor in quality or non-existent. Also, ISM education does not cultivate the habit of detailed clinical records. It will be some time before these habits can be rectified. Hence, to start with, Ayurvedic pharmacoepidemiology can be initiated at the PHC level. The major emphasis should be on drug utilization, prevalence of diseases and safety of drugs. Special training should be provided in epidemiology even to allopathic doctors.13 Reverse pharmacology or observational therapeutics, emerging at the interface of MM and ISM has to be actively encouraged with utilization projects and centres for excellence and research at all tiers of health care. This approach may convince the world too about the evidence-based nature of ISM. Many leads for drugs have been obtained by astute clinical observations in the field. Table 6 lists the drugs obtained by the reverse pharmacology path, which could be an economical and effective drug development path among diverse R&D paths for natural products (Fig. 3).
Dynamic Learning Model and Deployment

ISM Education and Relevant Life Sciences
The International Association recently held an Intellectual Conclave for the Study of Traditional Asian Medicine (IASTAM) at Pune. Vaidya Vilas Nanal20 presented the results his survey of Ayurvedic students. These are as follows: A revamping of the syllabus of ISM education is urgently needed. Allopathy metamorphosed into MM because the basic sciences of chemistry, physics, biology, etc. were incorporated. There was resistance from the protagonists of shuddha Ayurveda against any change. However, without compromising on the fundamental principles of Ayurveda, modern life sciences should be introduced in ISM education at the earliest. Further, following subjects also need to be considered for incorporation: (i) immunology, (ii) clinical biochemistry, (iii) genetics and molecular biology, (iv) pathological physiology and (v) obstetrics and gynaecology. These changes would facilitate the integration of ISM in health care delivery. ISM undergraduate training must include the needs of essential rural medical practice.
Table 6 Drugs obtained by the reverse pharmacology path

Indian Medicinal plant Disease Other Medicinal plant Disease
Pluralistic Health Care: India's Leadership Role

India is the only nation in the world with officially recognized multiple systems of medicine.19 But the absence of functional bridges across the systems is a major lacuna of our pluralistic health care. With the integration of ISM, this lacuna will be filled. Then India will offer truly global leadership in IM. This will depend on how efficiently and effectively ISM can be integrated into national health care. Some recommendations to enrich our pluralistic health care system by ISM are: 1. Sanitation, hygiene, clean water and nutrition should receive maximal attention to minimize infectious disease. Positive ISM practices of personal hygiene should be ingrained. 2. Family, schools, workplaces and communities must be actively involved in following the healthy lifestyles recommended by ISM.
Rauwolfia serpentina Hypertension Catharanthus roseus Cancer Commiphora wightii Hyperlipidaemia Cinchona officinalis Malaria Mucuna pruriens Parkinson disease Digitalis purpurea Heart failure Picrorrhiza kurroa Hepatitis Salix alba Fever Curcuma longa Oral cancer Ephedra sinensis Asthma
Evidence-based IM and ISM

The age-old experiential healing wisdom of ISM constitutes a different kind of evidence from randomized controlled trials of drugs. ISM even have different conceptualization of pathogenesis based on prakruti, ahar, vihar, ritu, pragnyaparadh, etc. These cannot be lightly brushed aside due to the dominant paradigms of MM. Only massive clinical data, col-
82
SECTION II
Fig. 3 R&D paths for natural products

Field use Ayurvedic usage Standardize
Standardize
Whole formulations
Reverse pharmacology Experiential documentation (observations) Exploratory studies
Ethnomedicine
AYURVEDIC THERAPEUTICS
Medicinal plants
Standard extract
Screening of extracts
Experimental research
Clinical trials
Isolation of active principle
Clinical phase II
Herbal remedy
Modern drug
Natural drugs
lected from multiple locations, can provide evidence on the safety, efficacy and quality of ISM. Information technology and telecommunications must be efficiently utilized to create massive databases from thousands of villages. The diseases responsible for high morbidity and mortality should receive priority. Such data, when analysed and interpreted, will expedite the growth of IM for health care. Evidence-based ISM requires strong support for research at all levels of biological organization, in a nationwide R&D network such as the New Millennium Indian Technology Leadership Initiative (NMITLI) project. There is an urgent need to build into ISM education and service strong elements of clinical pharmacology and research methods. The vital elements are: pharmacovigilance, rational drug therapy, adverse drug reporting (ADR), experimental design and epidemiology. ISM practitioners trained in this manner would enhance the process of ISM-MM harmonization. Later, specialization with dominant ISM or MM categories could be undertaken for postgraduate courses. Integration of ISM would be automatic if such a change in medical education takes place. The new model of integrative health care that India can evolve would need to be: (i) pluralistic and patient need-based; (ii) accessible and economically viable; (iii) evidence-based, in a broad sense, with experiential data; (iv) learning and dynamic in terms of emergent diseases; (v) environment-friendly, with the growth and use of plants; (vi) people-driven at the grassroots levels of democracy; (vii) research and education-oriented for national needs; and (viii) transparent and accountable. Figures 4a and b outline the proposed new model and priorities of integrative health care for India.
Fig. 4a New model of integrative medicine

ORGANOGRAM
PM and PMO Central BIM MOH/ISM Central DGIM State BIM State DIM State CM/HM
DMO DMO Panchayats Panchayats DMO Panchayats
Summary and conclusions

In this paper an attempt has been made to think in an outof-the-box manner and suggest radical ideas to revamp health care by a three-horizon strategic approach for ISM integration. This will involve identifiable deliverables and grassrootsFinancing and Delivery of Health Care Services in India
83
SECTION II
Fig. 4b Priorities of integration of ISM

Quality of life and longevity Learning and dynamic model
Reduction in morbidity and mortality
Integrative disease management
Promotion of health
Health pluralism, patient's needs and accessibility
Prevention of diseases
and State-level 'empowered' DIM. The approach has to be from bottom-up and not top-down. Certain proposals such as a village-level ISM herbal garden, pharmacy and health meetings, if well implemented, would make India a global leader in health matters. The emphasis has to be on the promotion of health and prevention of diseases by synergy of ISM-MM. The use of information technology, telecommunication and computers in IM is proposed at the PHC level. The ICMR-ISSR document of 1981 has recommendations that are still very relevant. Certain precise and prompt actions need to be taken to emphasize the seriousness of the resolve for IM. The following are summarized for programmes in education, service and research: 1. ISM modules in the fundamentals of Ayurveda and common non-drug/drug modalities of health care must be included in all medical colleges within a year. Similarly, ISM students must be trained in rural health practice as per MM and in essential drugs of MM. 2. ISM and MM practitioners should be deployed for health care delivery, on an equal footing, in States with the poorest health statistics. They have to undergo training in rural health practice (cf Sri Lanka) 3. The list of drug and non-drug modalities of ISM suggested by consensus has to be widely circulated and a for-
mulary published as soon as possible (within 6 months). A composite medicine kit needs to be evolved for villages to address common household ailments. Ayurvedic pharmacoepidemiology and reverse pharmacology centres must be established in each State to identify currently used, safe, effective and quality ISM remedies. 4. Gram aushadhi udyan cooperative farms and Gram aushadhi nirmana must be developed in at least 10,000 villages already being served by competent NGOs. Village healers have to be identified, and their skills assessed, enhanced and utilized in the integrative model. The FRHLT can be a catalytic agency. 5. Ayusoft and computer-friendly case-record forms must be deployed in these 10,000 villages (C-DAC has already initiated a project). An interactive website needs to be created for inputs and suggestions in the process of integration of ISM. 6. The Central Directorate of Integrative Medicine (CDIM) must be created within six months to steer the course. 7. Targets, transparency, accountability for ISM/IM performance, etc. have to be in the public domain and monitored by local citizens' groups for quality and deliverables. The infrastructure and adequacy of supplies of AYUSH drugs have to be actively ensured. Modern management techniques are to be used. 8. The community must be involved for maximum support to the local IM personnel and infrastructure. Financial inputs have to be at the Panchayat/PHC level, need-based and accountable. Health education at school level must include ISM lifestyles and disease-prevention modalities.
Acknowledgements
The paper used diverse contributions-written, verbal and scientific-from several authentic experts and colleagues. Some of the communications have been part of long-term dialogues on the subject of health care and ISM. I crave indulgence for any errors of opinions. I thank all the following for their inputs: Dr Ranjit Roy Chaudhury, Dr Shailaja Chandra, Dr Narendra Bhatt, Dr Urmila Thatte, Dr Darshan Shankar, Late Professor Sharadini Dahanukar, Dr Navneet Fozdar, Dr D.B. Ananthnarayana, Dr Abhay Bang. I wish to thank Drs Pradnya Talawadekar and Shridhar Anishetty for technical and Ms Anupama Bhaskaran for secretarial assistance.
84
SECTION II
References
Chandrashekhar S. Hindu dharma-the universal way of life. 3rd edn. Mumbai: Bharatiya Vidya Bhavan; 1996:911. Charaka Samhita-Sutrasthana, Adhyaya 1, shloka 41. Dietman AJ. Sufism and mental health science. In: Walsh R, Shapiro DH (eds). Beyond health and nomrality. New York: Van Nostrand and Reinhold; 1983:275. Fulder S. The handbook of alternative and complementary medicine. 3rd edn. Oxford, New York: Oxford University Press; 1996. Joshi NH. Personal communication. 2005. Nabar N. Ayurvedic pharmacoepidemiology of diabetes mellitus and metabolic disorders. Ph.D thesis. University of Pune (ongoing). Nanal V. Is education of Ayurveda delivering what it should? Curriculum and syllabus policy issues. Background paper-IASTAM Conclave 2005;I:4:1-I:4:5. New vistas in therapeutics, from drug design to gene therapy. Skarlatos SI, Velletri P, Morris M (eds). Ann NY Acad Sci 2001;953. Palep HS. Scientific foundation of Ayurveda. Delhi: Chaukhambha; 2004. Pelletier KR. The best of alternative medicine. What works? What does not? New York: Simon and Schuster; 2000:231-50. Saper RB, Kales SN, Paquin J, et al. Heavy metal content of Ayurvedic herbal medicine products. JAMA 2004;292:2868-73. Strom BL (ed). Pharmacoepidemiology. New York: Churchill Livingstone; 1989. Sushruta Samhita-Sharirsthana, Adhyaya 5, shloka 48. The Times of India, 28 October, 2001. Tillu G. Pharmacoepidemiology of Ayurvedic medicines. Ph.D thesis. University of Pune (ongoing). Joshi K, Patwardhan B, Raut AA, et al. Unpublished observations. [Au? Personal communication?] Vaidya AB. Therapeutic potential of medicinal plants-A global perspective. Supplement to cultivation and utilization of medicinal plants. Handa SS, Kaul MK (eds). [Au? Place?] RRL-CSIR. 1996:1-12. Vaidya AB, Antarkar DS. The use of scientifically validated herbs and plants in primary health care. New Delhi: ICMR; 1981:79-86. Vaidya AD. Ayurveda: Revivalism or renaissance? Keynote Address: AIMD Symposium. New Delhi. National Policy on ISM. Department of ISM and H; 2001.
85
SECTION II
Bibliography
Transforming traditions for tomorrow's health, IASTAM Conclave, Pune, Background Papers- Discussion guide and framework, January 2003. White KL. The task of medicine. Messlo Park Kaiser, 1988. Gautam V. Road beyond boundaries (The case of select Indian healthcare systems). Mumbai: EXIM Bank; 2003. Gautam V, Raman RMV, Kumar A. Exporting Indian healthcare (expert potential of Ayurveda and Siddha products and services). Mumbai: EXIM Bank; 2003. Draft National Policy on ISM. Department of ISM and H. New Delhi: Ministry of Health, Government of India; 2001. Commission on Macroeconomics and Health. Protection of traditional medicines; Working paper series No.WG 4:40; 2001. Commemorative Issue: IASTAM, Silver Jubilee; 2005. Shastri GK. Rasoddhara tantra. Gondal: Rasashala; 1931. Warrier PS. Chikitsa Sangraha; 1990. [Au: Details?] Ornish D. Program for recovering [Au? from?] coronary artery disease. New York: Random House; 1990. Benson H. The relaxation response. New York: Ason, 1976 Rafer UM, Chaudhury RR (eds). Traditional medicine in Asia. New Delhi: WHO- SEARO; 2002 Svoboda RE. Ayurveda: Life, health and longevity. Penguin Books; 1993. Kenji C. Forty years of integrated medicine in China. Chin J Integr Trad Western Med 1997;3:162-6. Heyn, B. Ayurvedic medicine: The gentle strength of Indian healing. New Delhi: Harper Collins; 1992. Dash B. Ayurvedic treatment for common diseases. New Delhi: Delhi Diary; 1979. Sharma S. Realms of Ayurveda. New Delhi: Arnold Heinemann; 1979. Sairam TV. Home remedies. New Delhi: Penguin; 1998. Dahanukar SA, Thatte UM. Ayurveda revisited. Mumbai: Popular; 2000. Sigerist H. Medicine and health in the Soviet Union. Mumbai: Jaico; 1947. Chopra D. Quantum healing. New York: Bantan; 1989. Dossey L. Meaning and [Au? matter missing?] Hahnemann S. Organon of medicine. New Delhi: B. Jain; 1974. Vaid NK. Who cares for tribal development? New Delhi: Mittal (2004). New York, Bantan; 1991. [Au? Pls check the year of publication and publisher] Palep HS. Scientific foundation of Ayurveda. Delhi: Chankhamba; 2004. Lele RD. Ayurveda and modern medicine. Mumbai: Bharatiya Vidya Bhavan, Zna [Au?] Edn; 2001. Anjaria J, Parabhia M, Dwivedi S. Ethnovet heritage, Indian ethnoveterinary medicine-An overview. Ahmedabad: Pathik; 2002. Panchakarma. Ayurveda's mantra for rejuvenation. Kerala Tourism; 2003. Chopra D. Ageless body timeless mind. New York: Harmony Books; 1992. Siddha Vaidyam. Thiruvanathapuram: Edisons; 2003. Thirunarayanan T. An introduction to Siddha medicine. Thiruchendur, Thirukumaran. Valiathan MS. The legacy of Charaka. Hyderabad: Orient Longman; 2003. Gogte VM. Ayurvedic pharmacology and therapeutic use of medicinal plants. Mumbai: Bhavan's SPARC; 2000. Fazal U, Razzack MA. A handbook of common remedies in unani system of medicine. [Au? Details] Hamdard. Development of Unani drugs from herbal sources and the role of elements in their mechanism of action. New Delhi: Hamdard; 1985. Parsons DF, Fleisher CM, Greenes RA. Extended clinical consulting by hospital computer networks Edu. Ann NY Acad Sci 1992:670.
86
SECTION I
Essential Ayurvedic drugs for dispensaries and hospitals. New Delhi: AYUSH; 2000. Government of India. Ayurvedic pharmacopoeia of India (Parts 1 & 2). New Delhi: Ministry of Health, Government of India; 1986 and 2000. Bhavan's SPARC. Selected medicinal plants of India. Mumbai: CHEMEXCIL; 1992. Chopra RN. Indigenous drugs of India. Kolkata: Academic; 1982. Kulkarni RD. Principles of pharmacology in Ayurveda. Mumbai: Kulkarni; 1977. Sharma PV. Dravyagunavignan. Varanasi; Chankhamba; 1978. Handa SS, Kaul MK (eds). Supplement to cultivation and utilization of medicinal plants. Jammu: RRL; 1996. Fernando S. Herbal food and medicines in Sri Lanka. Maharagama: Sampath; 1999. Payyapilly C, et al. Holistic health work book. Pune: Sahaj; 1989. Tsarong TJ. Handbook of traditional Tibetan drugs. Kalimpong; Tibetan Books; 1986.
Rgyu A-Bzhi. Fundamentals of Tibetan medicine. Dharamsala; Men-Tsee-Khang; 1981. Ravishankar J. Physician for the needy: Easy and economical remedies. Ahmedabad; JL Trived [Au? ?]; 1914. Achaya KT. Indian foods. New Delhi; Oxford University Press; 1994. Shah B, et al. Medicinal plants for primary health care (WHO-Gujarati-English Transl). Vadodara: Samanvaya; 2002. Wilcox M, et al. (eds). Traditional medicinal plants and malaria. Boca Raton: CRC Press; 2004. Foundation for Revitalization of Local Health Traditions (FRHLT). Clinically important plants of Ayurveda (CDROM). Bangalore: FRHLT; 2002. Shankar D, Manohar R. Ayurvedic medicine today: Ayurveda at cross roads. In: State of India's health. New Delhi: VHAI; 1995. Kurup PNV, et al. Handbook of medicinal plants. New Delhi: CCRAS; 1979.
87
SECTION II
Delivery of health services in the private sector
P
K. SUJATHA RAO
SECRETARY, NATIONAL COMMISSION ON MACROECONOMICS AND HEALTH, GOVERNMENT OF INDIA, NEW DELHI E-MAIL: ksujatharao@hotmail.com
MADHURIMA NUNDY
CENTRE OF SOCIAL MEDICINE AND COMMUNITY HEALTH, JAWAHARLAL NEHRU UNIVERSITY, NEW DELHI E-MAIL: madhurima.nundy@gmail.com
AVTAR SINGH DUA

MEMBER, SUB-COMMISSION, NATIONAL COMMISSION ON MACROECONOMICS AND HEALTH, GOVERNMENT OF INDIA, NEW DELHI E-MAIL: avtarsinghdua@yahoo.co.in
RIVATE HEALTH MARKETS ARE COMPLEX AND PROFOUNDLY AFFECTED BY several factors. The nature of health financing and payment systems, type of technology, cost of initial education and training, public expectations and perceptions, regulatory framework, and social values are some of the factors that interact with one another to determine how equitable, efficient, safe and accessible the private sector could be. International experience shows that the private sector tends to focus on profit maximization and is hardly concerned with public health goals (Bennett 1997), making state intervention essential. In India, largely due to lack of understanding of the implications, the state's role has been negligible in articulating a strategic vision of the health system that we should have and would be appropriate for us, given our levels of development, wide disparities, huge diversity and poverty. In the absence of such a vision, the growth and development of the private sector in the delivery of health services has been relatively autonomous and independent, and not a consequence of any deliberate state policy. Over the years, the private sector in health care has gained a dominant presence in all the submarketsmedical education and training, medical technology and diagnostics, manufacture and sale of pharmaceuticals, hospital construction and ancillary services and finally, the provision of medical services. Three-quarters of the human resources and advanced medical technology, 68% of a total of over 15,097 hospitals and 37% of over 623,819 beds in the country are in the private sector (Directory of Health Services, GOI 1996). Another estimate showed that the private sector provided almost 81% of all outpatient (OP) and 46% of inpatient (IP) care (52nd Round of the National Sample Survey [NSS]). Analysis of the 57th Round of the NSS covering 30,000 health providers shows that there are an estimated 13 lakh private health care provider enterprises employing 22 lakh people. Over one-third of them have no registration of any kind and 25% are AYUSH practitioners. An important subset of providers are the large number of informal providersquacks (almost one in every village), bonesetters, traditional healers, traditional birth attendants (TBAs), etc. (See Annexure I for note on analysis and detailed tables of the 57th Round NSS). A survey of 'quacks' in 3 districts of Andhra Pradesh showed that there was one for every 2000 population (Rao et al. 1997) Private expenditure as percentage of the gross domestic product (GDP) is estimated at 72%, and the private health market is over Rs 71,000 crore. If the pharmaceutical industry is added, it would be another Rs 31,000 crore. The CII-McKinsey Report of 2004 has estimated the private sector in India to be worth Rs 69,000 crore and expects it to double to Rs 156,000 crore by 2012, besides an additional Rs 39,000 crore if health insurance picks up. The study also estimates that the proportion of IP care will go up to 47% largely due to lifestyle diseases, namely cancer and cardiovascular diseases. This growth is likely to require an additional 750,000 beds, 520,000 doctors and an overall investment of Rs 100,000-150,000 crore, of which 80% has been projected as the share of the private sector. Coinciding with falling public health investment, emergence of non-communicable diseases and a spiralling demand, there has been a steady growth in the corporatization of medical care. In no small measure is this development a result of the process of liberalization since the early 1990s. Several NRIs and industrial/pharmaceutical companies are investing money in setting up superspecialty hospitals such as Medinova, CDR, Mediciti, L.V. Prasad Eye Institute in Hyderabad, Hindujas and Wockhardt in Mumbai, Max and Escorts in Delhi, etc. Apollo Hospitals raise a substantial proFinancing and Delivery of Health Care Services in India
89
SECTION II
management, more in the nature of a joint partnership. Under this scheme, about 35 primary health centres (PHCs) (31 in Karnataka and 4 in Gujarat) have been handed over to NGOs to manage. In Karnataka, the superspecialty hospital constructed in Belgaum under the OPEC Assistance programme has been handed over to the Apollo Group for management with some further grants for meeting a part of the recurring costs during the first year. In Chhattisgarh, the State Government provided Escorts a grant of Rs 12 crore to build and operate the cardiac specialty centre, subject to earmarking 15% of patients identified by the Government to be treated at discounted rates. (2) A major initiative and the first, serious, large-scale experience of contracting the for-profit sector has been the Central Government Health Scheme (CGHS), which has a contract with over 200 private providers for the medical treatment of Public Policy Response: its members as per pre-fixed rates. Other public sector underPublic-Private Partnership takings (PSUs) have private doctor/hospital panels for their employees for OP and IP care, and some organizations such In response to this enormous resource, since the past five years, as the Railways have their own health facilities, contract out the Government has been attempting to engage the private to the private sector for specified superspecialty services. sector in providing services under the National Health Pro(3) Similarly, State Governments are also innovating with grammes (NHPs). The primary objective of such an attempt various forms of such public-private partnerships for specific has been to expand access to health care. As can be seen services ranging from ancillary services, such as, laundry and from Table 1, the experience has been far from satisfactory security to diagnostics, drug management, etc. In West Benand even the little success achieved is more due to the partgal, diagnostic services for high-end equipment such as magnership with the not-for-profit sector and non-governmennetic resonance imaging (MRI) are outsourced to private tal organizations (NGOs). The huge for-profit sector continproviders who work within the medical college campus subues to be a parallel development that public policy has yet to ject to covering a certain number of cases referred by them take cognizance of. at pre-fixed rates. In Rajasthan, drug stores established in all Apart from the limited engagement of the government with health facilities are managed by the private sector in lieu of the private sector for achieving public health goals, other a commission over the rates fixed. This has been found to be forms of public-private partnerships can be categorized into beneficial as there are no stockouts or non-availability of three types: essential drugs at any point of time. Other forms of contracting (1) Handing over public facilities to the private sector for are for specific ancillary services such as security, canteens, sanitary services, landscaping, etc. Efforts of the Government to collaboTable 1 rate with the private sector have been proPublic-private partnership in health care gramme-based, sporadic, disjointed and tentative, and not the result of a well Name of the programme Nature of collaboration Outcome thought-out strategy aimed at achieving Malaria Nil national health goals. Despite the mixed TB IEC, provision of diagnostic services Negligibleproblem of conflict and varied experience, it is clear that coland laboratory support of interest laboration with the private sector could Blindness due to cataract IEC and cataract surgeries Positive30% of cataract surgeries enable expansion of access. The problem contributed by the private sector is a lack of clarity as to the financial, legal Leprosy IEC Substantial and institutional arrangements that govRCH Contracting specialists, conducting Poorhuge amounts allocated for ernments need to possess to ensure that RCH camps, IEC contracting services of specialists such partnerships result in social gain. The left unutilized due to the nonexperience of giving incentives to private availability of specialists where needed and amount offered found hospitals, such as excise duty exemptions, not worthwhile by private free land, etc. in lieu of treating 10% of practitioners. IPs and 40% of OPs free has not been HIV IEC, care and support Positivemainly NGOs. favourable. Such adverse experiences of non-compliance with the conditions TB: tuberculosis; IEC: information, education and communication; RCH: Reproductive and Child Health; NGO: non-governmental organization Source: Authors' assessment imposed were observed by the Legislature Committee of Andhra Pradesh as well as
portion of their resources from the stock market. These institutions are capable of providing world-class care at a fraction of the cost compared to the West. There is, thus, enormous potential for India to become a hub for medical tourism, with trade-offs in terms of welfare implications such as raising the overall cost of health care in the country (Purohit 2001). For example, a substantial proportion of insurance claims are for treatment in corporate hospitals. If this trend increases then we can soon be talking about a high-cost medical system for those who are insured and able to afford such health care. Keeping pace with this kind of hi-tech, inappropriate, supplier-induced system, which keeps an eye on profit maximization and good dividends for their shareholders, could well mean higher budget outlays for government hospitals if they are to stay and be counted.
90
SECTION II
the Delhi High Court. Demographic shifts, the dual burden of pre- and posttransition diseases, constrained public resources and the growing demand for services that require resource-intensive treatment are factors that cannot be ignored. The private sector provides services only to those who are able to pay for them, leaving out a majority of the poor and lower middle-income groups, who become impoverished when they do avail of such services. While efforts to strengthen and optimize existing public facilities with more investment and better management should receive priority, collaborating with the private sector will still be required due to the government's limitations in mobilizing the required capital for meeting the growing demand and, more importantly, the expertise and skill base that the private sector possesses. Adoption of a more holistic and pragmatic approachof contracting with the private sectorwill need to be considered for expanding access. Any such policy will have to be within the context of a regulatory framework and insurance system to ensure that there is no adverse selection and risk-sharing is facilitated. In other words, supporting the private sector without accompanying policies to provide financial risk protection could be ruinous as is being witnessed today. We now discuss the current status of the private health sector in India based on a qualified provider facility survey conducted by the National Commission on Macroeconomics and Health (NCMH) in eight districts and briefly analyses, with the help of available data, issues that public policy needs to consider before collaborating with the private sector. The section is divided into three parts. Part I gives a brief overview of the salient points regarding the current status of the private sector; Part II details some key findings of the eight-district facility survey of all qualified providers; and Part III discusses policy issues arising from the study.
engaged, technology used, etc. by the private sector. Some studies have, however, documented the abysmal quality of care given by private providers at the rural periphery. Baru (2002) found that the private sector appointed persons at low wages, and discharged patients earlier than medically advisable to keep a quick turnover of patients. In a survey of 24 hospitals in Mumbai, half were found to be operating from sheds and lofts, congested spaces, with leaking operation theatres (OTs) and over 90% of unqualified nurses and doctors with degrees in alternative medicine providing care in allopathic medicine. Care included unnecessary tests, consultation and surgery, without providing any information on diagnosis or treatment (Nandraj 1994). Thankappan's study in 1999 of 9 hospitals in the Thiruvananathapuram district of Kerala showed that the private sector tended to keep patients in their hospitals for a shorter time and order more diagnostic tests as compared to public sector hospitals. These studies also suggest that the macroeconomic context provided the main stimulus for the proliferation of the private sectorthe National Health Policy of 1983, advocating the private sector to offer specialty services provided a policy context for extending a spate of subsidies in the form of excise duty exemptions, subsidized land, bank credit, etc. The corresponding decline in public expenditure on health, particularly capital investment, constrained the public sector from keeping pace with the private sector, in terms of financial, human and technological resources; the shifting demand for services as a result of demographic and epidemiological changes; rising incomes that increased the willingness to pay for health services that did not entail long queues and were aligned with their perceptions of quality (saline injections, or going through needless tests, for example); an unfettered regulatory environment, etc.
Public Policy Challenges

The irrefutable evidence brought forth in all these studies often raises questions of what the future direction of public policy should be. Should it, as argued by some (National Health Policy 1983; World Development Report1993), confine itself to providing only public goods and primary care and leave all curative care to market forces; or regulate the private sector, selectively contract its services to achieve public health goals and compete with it; or should the Government over time become a purchaser and regulator of service delivery and divest itself from the responsibility of service provisioning? What would be the correct approach keeping in view the state's obligations to the poor on the one hand, and severe limitation on public finances on the other? Current research on the private sector falls short in providing the financial and policy implications for the Government on how to contract the private sector, where, for what and why, and what would be the most cost-effective option based on a clear understanding of the functioning of the health system at the district level. A survey conducted of all qualified providers in 8 districts is a part of the effort to get a closer appreciation of the ground-level reality so that based on such evidence policy implications can be examined at the local level.
Part I Private Sector in IndiaCurrent Status

Since the past 15 years, several researchers have sought to study the nature, spread and functioning of the private sector in India (Baru, Bhat, Nandraj, Dugal, Mahapatra, Jesani). Those studies have brought out the diversity in the composition of the private sector, which ranges from voluntary, not-for-profit, for-profit, corporate, trusts, stand-alone specialist services, diagnostic laboratories, pharmacy shops, unqualified providers (quacks); each addressing different market segments. Most characterization of typology suggests that the private sector consists largely of sole practitioners or small nursing homes having 1-20 beds, serving the urban and semi-urban clientele and focused on curative care. In the absence of regulations governing location, standards, pricing, to name a few, private facilities run in marketplaces, residential colonies, pharmacy shops, with the freedom to provide any kind of services, of whatever quality and at exorbitant cost, which varies from facility to facility. However, due to a weak legal framework, only a few studies based on small localized samples reveal the type of services being provided, human resources
91
SECTION II
Part II District-based Facility Survey of Qualified Providers

Since any public contracting or social insurance arrangements can be implemented only with qualified providers and accredited facilities, it was felt necessary to understand the supply markets in the qualified provider subsector. Accordingly, the subcommission of the NCMH undertook a facility survey in 8 districts to obtain information for filling existing information gaps. The key research questions for the survey were whether the private sector is actually available in underserved areas where the functioning of the public sector is poor or unavailable. What is the organizational structure of the private sector? What services does the private sector actually provide and what skills and technology do they possess? What is the level of utilization and what prices do they charge for basic services? Is the private sector really much more efficient than the public sector? Is it complementing or substituting public services, and if so, to what extent? Is it adding value or merely duplicating what the public sector is doing resulting in an overall waste of resources? The motivation for undertaking this survey was to assess the 'supply gaps' that need to be addressed for designing demand-side financing such as social health insurance (a key concern in the context of the increasing costs of medical treatment resulting in the impoverishment of about 3% of the population). The failure of the Universal Health Insurance Scheme to reach out to the poor despite its attractiveness is in large part due to the lack of availability of services within reach, provider markets being fragmented, dysfunctional and weak.
While the 8 districts have 93 blocks in all, due to the nonavailability of the coordinates for Varanasi and Kozhikode, the tabulation was in terms of tehsils. Both these districts have 3 tehsils and 7 and 12 blocks, respectively. Due to this, the total number of blocks is expressed as 80, instead of 93, as they include 6 tehsils instead of 19 blocks of Kozhikode and Varanasi. The surveyed districts covered a population of 210 lakh9987 villages and 83 towns.
Results of the Survey

Size, growth and organizational structure of the private sector
In the surveyed districts, there are a total of 9457 facilities run by qualified providers. Of these, 61% are private. Since the past decade, the number of private hospitals has increased. Twothirds of the corporate hospitals and 50% of the hospitals owned by partnership firms were established after 1995. The survey also showed a high concentration of these facilities, clustered as they are in 5.5% of the villages and about 73 towns. The ratio of the public-private sector is 60:40 in rural areas as compared to 10:90 in urban areas. The higher percentage of public facilities in rural areas is on account of the subcentres established for every 5000 population. On analysis we found that the presence of the private sector in the poorest 15 blocks is negligible (See Annexure II for detailed tables of the survey and also the website of the Ministry of Health, GOI, for Block wise data). Organizationally, the private sector is fragmented, with 91% of the facilities being run by sole proprietors. The survey, however, indicated a falling trend under this category and an increase of partnerships. There was also an increase of corporate hospitals from 5 before 1980 to 36 in 2000-04 (Fig. 1).
Methodology
The survey was taken up in 8 median districts identified in the 1991 Centre for Monitoring Indian Economy (CMIE) index: Nadia (West Bengal), Jalna (Maharashtra), Khammam (AP), Kozhikode (Kerala), Vaishali (Bihar), Varanasi (UP), Ujjain (MP) and Udaipur (Rajasthan). It covered only the qualified provider sector of both the disciplinesallopathy and AYUSH. The survey was undertaken by well-known and reputed agencies and the data validated by random inspections of facilities as well as detailed discussion with district authorities. The survey obtained information on the ownership pattern, services provided, utilization levels, human resources appointed, equipment used, prices charged for some services, etc. Such a database is useful for micro-level planning for ensuring an equitable geographic distribution of facilities. Geographical information systems (GIS) mapping has helped us measure the distances people have to travel for accessing services, duplication or overlap between the public and private facilities, identifying the underserved areas to quantify the level of additional investment required; relocating of existing facilities for achieving better efficiency, etc. The block-wise survey data for key indicators was tabulated.
Fig 1 Cumulative establishment of facilities over the years

n = 7319, public facilities = 1605, private = 5714; Year of establishment not available for 1967 facilities
Source: Faculty survey, NCMH 2004
92
SECTION II
Fig 2 Distribution of beds in rural areas
Fig 3 Distribution of beds in urban areas
Infrastructure of the private sector: Beds, human resources, and diagnostic services
Almost 86% of the facilities are small, OP clinics with one or two beds. Of the 1281 facilities having 35,349 beds, 49% beds are in 931 private facilities (Figs 2 and 3). Two-thirds of the facilities and 79% of beds are in urban areas. An intra-district analysis based on measuring the adequacy of beds by the norm of 100 beds for every 100,000 population shows a disturbing picture. Considering a block has, on an average, over one lakh population, nearly 75% of the blocks have barely 3 beds, and one block in Nadia with a population of 1.9 lakh has no beds at all.
Human resources in the private sector

In the 8 districts there are a total of 2746 MBBS doctors and 4466 specialists. Seventy-five per cent of these specialists are in the private sector: 61% anaesthetists, 78% cardiologists, 85% general physicians and 73% gynaecologists and surgeons (Tables 2 and 3). The number of specialists working per facility increased in accordance with the number of beds in the facility, with the majority in the above 30-bed category hospitals. The ratio of doctors per 1000 population is disturbingly low, except in Kerala. The total average for all these districts is 0.40/1000, far lesser than 0.59/1000 for all India. The position is particularly severe in Jalna, Nadia, Khammam and Vaishali. This is despite Khammam having a private medical college.
Table 2 Availability of full-time and part-time MBBS doctors in public and private sectors
Urban District Public Full time Part time Rural Private Full time Part time Public Full time Part time Private Full time Part time Full-time Public Private % of full-time doctors in the Total public sector
Jalna Khammam Kozhikode Nadia Udaipur Ujjain Vaishali Varanasi Total
9 34 89 123 311 74 24 52 716
0 0 0 0 0 0 0 0 0
52 56 139 71 68 90 100 288 864
2 17 15 207 20 13 6 64 344
44 75 117 85 99 20 40 38 518
0 0 0 0 0 0 0 0 0
10 35 214 56 57 106 76 94 648
2 1 11 109 0 9 16 14 162
53 109 206 208 410 94 64 90 1234
62 91 353 127 125 196 176 382 1512
115 200 559 335 535 290 240 472 2746
46.1 54.5 36.9 62.1 76.6 32.4 26.7 19.1 44.9
Note: A part-time doctor could serve part-time in more than one facility and hence there are chances of double-counting Source: Faculty survey, NCMH 2004
93
SECTION II
Table 3 Availability of full-time allopathic specialists in public and private sectors

Urban Specialist Public % public Private % Private Total Rural Public % public Private % private Total Total Public % public Private % private Total
Anaesthesist Cardiologist Dentist Endocrinologist ENT specialist Gastroenterologist Gynaecologist Ophthalmologist Orthopaedician Paediatrician General physician Psychiatrist Skin and VD specialist General surgeon Urologist Total
96 22 28 6 47 9 138 54 56 102 99 25 24 136 10 830
40.85 23.91 10.04 35.29 29.01 27.27 24.38 23.68 25.00 28.25 15.02 37.88 23.76 32.30 20.00
139 70 251 11 115 24 428 174 168 259 560 41 77 285 40 2572
59.15 76.09 89.96 64.71 70.99 72.73 75.62 76.32 75.00 71.75 84.98 62.12 76.24 67.70 80.00
235 92 279 17 162 33 566 228 224 361 659 66 101 421 50 3402
8 0 14 0 5 0 32 10 4 26 38 0 2 26 0 165
28.57 0.00 6.83 0.00 16.67 0.00 22.86 17.86 10.53 30.59 15.83 0.00 11.76 22.22 0.00
20 6 191 0 25 2 108 46 34 59 202 7 15 91 1 801
71.43 100.00 93.17 0.00 83.33 100.00 77.14 82.14 89.47 69.41 84.17 100.00 88.24 77.78 100.00
28 6 205 0 30 2 140 56 38 85 240 7 17 117 1 966
104 22 42 6 52 9 170 64 60 128 137 25 26 162 10 995
39.54 159 22.45 76 8.68 442 35.29 11 27.08 140 25.71 26 24.08 536 22.54 220 22.90 202 28.70 318 15.24 762 34.25 48 22.03 92 30.11 376 19.61 41 22.28 3373
60.46 263 77.55 98 91.32 484 64.71 17 72.92 192 74.29 35 75.92 706 77.46 284 77.10 262 71.30 446 84.76 899 65.75 73 77.97 118 69.89 538 80.39 51 75.53 4466
The position is equally dismal in relation to nurses. The average for all districts is 0.32 per 1000 population while it is 0.23 in Vaishali and only 0.16 in Khammam. Of the 80 blocks, 32 have less than 0.10/1000. The position of other paramedical professionals is equally dismal.
Access to emergency obstetric care

Of particular interest for us was to assess the access to skills and emergency obstetric care (EmOC) as Madhya Pradesh, Uttar Pradesh, Bihar and Rajasthan are States with unac-
ceptably high levels of maternal and infant mortality, and where institutional deliveries are particularly low. To measure access, facilities having a full-time gynaecologist and anaesthetist or a full-time surgeon and anaesthetist, as proxy for providing EmOC facilities, were listed. The norm adopted was 1 facility with a combination of these specialists for every 100,000 population, where, as per government norms, a community health centre (CHC) consisting of 4 specialists is expected to be located. As per this norm, 70% of the blocks have no EmOC facilities, while the remaining 23 have a clustering of 113 facilities accounting for 50% of the caesarean sections (Table 4).
Fig 4 Availability of equipment/investigations in public and private facilities
Access to diagnostic laboratories and technology

Data were collected for 21 items of equipment and 17 investigations/tests. Almost half of the haematology and urine tests and one-third of angiographies are being done in the public sector, while the rest of the tests are mainly done in the private sector. With regard to diagnostic equipment, the dominance of the private sector is total; 90% of the expensive equipment is concentrated in a few urban areas. Of all tests and investigations, haematology is considered the most basic, while electrocardiogram (ECG) and X-ray are now considered fairly routine. Computerized tomography (CT) scan and MRI require high-end equipment. There is substantial infusion of technology, brought in largely by the private sector. Almost 20% of the blocks do not have even one X-ray machine and in a backward district, such as Vaishali, there are 69 X-ray machines, all in the private sector. Taking the norm of 1/10 lakh population for a CT scan or an MRI, analy-
ECG: electrocardiography; CT: computerized tomography, MRI: magnetic resonance imaging
94
SECTION II
Table 4 Block-wise availability of emergency obstetric care facility

Number of facilities with full-time specialists Gynaecologist and anaesthetist District Blocks Population Public Private Surgeon and anaesthetist Public Private Total* EmOC per 100,000 population
Jalna
Khammam
Kozhikode
Nadia
Udaipur Ujjain
Vaishali
Varanasi
Ambad Badnapur Jalna Bhadrachalam Khammam Kothagudem Kozhikode Quilandy Vadakara Krishnaganj Krishnagar-I Municipal Areas Ranaghat-I Santipur Girwa Salumbar Ghatiya Nagda Ujjain Bhagwanpur Lalganj Vaishali Varanasi
207142 131362 432129 158625 690728 346472 1100582 621623 575344 133359 280386 749705 207394 217318 740863 212492 90828 84929 968693 162213 173856 146364 2334190 10766597
1 0 1 1 1 1 2 1 2 1 0 3 1 1 3 0 0 0 3 0 1 0 5 28
0 1 4 0 4 2 10 0 5 0 1 3 0 0 6 0 1 1 5 1 2 0 20 66
1 0 2 1 1 0 3 1 1 0 0 4 1 0 3 1 0 0 2 0 1 0 6 28
0 0 4 0 1 1 9 0 4 0 0 3 0 0 7 0 1 1 5 0 2 1 20 59
1 1 8 1 5 4 13 1 7 1 1 7 1 1 13 1 1 1 8 1 4 1 31 113
0.5 0.8 1.9 0.6 0.7 1.2 1.2 0.2 1.2 0.7 0.4 0.9 0.5 0.5 1.8 0.5 1.1 1.2 0.8 0.6 2.3 0.7 1.3 1.0
EmOC: emergency obstetric care Only 23 blocks have full-time gynaecologists and anaesthetists/general surgeons and anaesthetists *Total facilities providing EmOC is less than the total of previous four columns because one facility could have all three-gynaecologist, surgeon and anaesthetist
Fig 5 Public-private share in National Health Programmes
TB: tuberculosis; ARI: acute respiratory infection; MTP: medical termination of pregnancy
sis showed that there were a total of 30 CT scan and 24 MRI equipment, indicating an excess (Fig. 4). The study also found a proliferation of other technologyover 300 ultrasound and 106 Doppler machines.
Nature of Services
The share of the private sector in the provision of dental, ENT, orthopaedic services, and for all non-communicable diseases such as myocardial infarction, cancer chemotherapy, mental health, medical termination of pregnancy (MTP), hysterectomies is very high, accounting for almost three-fourths of the total caseload. The survey also showed that the private sector seemed to be actively engaged in providing treatment for acute care as welltuberculosis (TB), deliveries, childhood diseases (Fig. 5).
95
SECTION II
Table 5 Number of outpatients in the past 7 days

Grand total District Total public Share of the public sector Total private. Share of the private sector Total
Jalna Khammam Kozhikode Nadia Udaipur Ujjain Vaishali Varanasi Total %
17228 85165 139262 81381 75255 31893 5839 27166 463189
19.9 60.0 58.2 59.0 65.3 40.7 13.5 22.2 48.0
69508 56884 100153 56584 39946 46521 37534 95341 502471
80.1 39.8 41.8 41.0 34.7 59.3 86.5 77.8 52.0
* Information on subcentres has not been included in calculations for these tables Source: Faculty survey, NCMH 2004
Utilization of Facilities
Outpatient care
Overall, the private sector share of OP cases is estimated to be 52% (Table 5), comparatively lower than the estimate of 81% indicated in the 52nd Round of NSS. Besides, on extrapolating the OP cases over 7 days to the year, data showed that the number per thousand was double that of NSSO estimations at 2171/1000, with Kerala at a high of 4204 and Vaishali at a low of 775 per 1000 population (Table 6). This indicates a gross underestimation of the NSS data as the survey covers unqualified quacks as well. Another caveat to be kept in mind while interpreting these data is the timing of the survey. This survey was taken up in 2004, following a three-year economic downturn in the rural areas, which could have made people opt for public sector facilities on grounds of affordability. Such an association between
The survey provided information on the use of facilities for IP care and the bed occupancy rate. Of the total IP cases, 47% utilized private facilities. In Udaipur and Nadia, utilization of the private sector was low. The IP data also showed that the rate of IP per 1000 populaFig 6 tion was 45.5 (Table Average number of inpatients 7) against 16/1000 per facility in the past 30 days estimated in 52nd Round of NSS. by availability of beds However, rough calculations suggest that the NSS may be a gross underestimation. Bed occupancy rate is a good indicator of the efficiencies in the system. It also indicates the urbanrural differentials. The bed occuSource: Faculty survey, NCMH 2004
86736 142049 239415 137965 115201 78414 43373 122507 965660 100.0
economic hardship and preference for public sector care has been observed in other countries such as Indonesia, Korea and Thailand during the economic recession of the late 1990s. While private facilities had higher OP cases, the average turnover per facility was almost four-fold higher in the public sector facilities, as also the number of patients seen per doctor, in three districtsKhammam, Jalna and Nadia.
Inpatient services
Table 6 Number of outpatients per 1000 population per year

OP District Government Private Government
No.of OPs per year Private Total
OPs per 1000 per year Government Private Total
17228 85165 139262 81381 75255 31893 5839 27166 463189
69508 56884 100153 56584 39946 46521 37534 95341 502471
895856 4428580 7241624 4231812 3913260 1658436 303628 1412632 24085828
3614416 2957968 5207956 2942368 2077192 2419092 1951768 4957732 26128492
4510272 7386548 12449580 7174180 5990452 4077528 2255396 6370364 50214320
528.5 1651.8 2445.4 871.3 1382.6 910.2 104.3 419.2 1041.5
2132.4 1103.3 1758.7 605.8 733.9 1327.7 670.7 1471.1 1129.8
2660.9 2755.1 4204.1 1477.1 2116.4 2237.9 775.1 1890.2 2171.3
OP: outpatient *Information on subcentres has not been included in calculations for these tables Source: Faculty survey, NCMH 2004
96
SECTION II
Table 7 Inpatients per 1000 population per year

Total Facilities District Government Private IP for 30 days Government Private No. of IPs per year Government Private Total IPs per 1000 per year Government Private Total
54 559 203 436 630 178 314 276 2650
466 300 964 1452 361 307 579 1360 5789
2287 6054 7229 16863 6901 2962 426 4071 46793
4304 8572 7695 1821 3994 2880 4579 7139 40984
27444 72648 86748 202356 82812 35544 5112 48852 561516
51648 102864 92340 21852 47928 34560 54948 85668 491808
79092 175512 179088 224208 130740 70104 60060 134520 1053324
16.2 27.1 29.3 41.7 29.3 19.5 1.8 14.5 24.3
30.5 38.4 31.2 4.5 16.9 19.0 18.9 25.4 21.3
46.7 65.5 60.5 46.2 46.2 38.5 20.6 39.9 45.5
IP: inpatient Source: Faculty survey, NCMH 2004
pancy rate in rural areas is 47% and 55% in urban areas. On an average, the bed occupancy in the private sector is 44% as compared to 62% in the public sector. Nearly 50% of all the private facilities having beds had less than 50% bed occupancy rates. Though the utilization of public facilities is more than the private (Fig. 6), in terms of per facility number of cases and also in the bigger range hospitals, the total number of IPs is higher in the private sector because there are a larger number of small units. This clearly corroborates with other evidence on this subjectthe failing public sector in rural areas due to absenteeism and inadequate facilities being made up by small-sized nursing homes and the overflowing city hospitals. In cities, public hospitals are able to compete with the private sector hospitals, despite relatively lesser access to resources (Fig. 7).
Cost of care
The payment system in the private sector is predominantly based on fee for service. Due to the absence of any system of provider control, there is a huge variation in the prices charged for similar services. An IOL surgery can cost anywhere between Rs 2000 and Rs 80,000, a caesarean section can cost between Rs 3500 and Rs 50,000. To get an idea of the prices in these median districts for some common medical interventions, data were obtained through exit interviews of patients and rate cards where available. The average charges for some common procedures are listed in Table 8. The pricing structures indicate an interesting pattern: prices are lower in rural areas for the same procedure/service/investigation; prices are far higher in Kozhikode than the 3 north-
Fig 7 Distribution of inpatients by availability of beds
Table 8 Charges for some common medical procedures (in Rupees)

Services Urbanprivate Ruralprivate Cost in private facilities Cost in Private: public facilities Public ratio
Normal delivery Caesarean section Major surgery ECG X-ray Blood test OP per episode
582-1925
275-1350
472-1573
0-128
18.3 24.3 20.8 3.6 2.2 5.1 10.7 2.6
2291-5385 1843-3910 1792-4647 50-250 2314-6950 2021-5000 1638-5975 0-711 60-119 72-111 36-86 24-64 56-121 62-140 20-43 12-31 20-59 56-115 68-123 30-59 18-37 38-84 0-55 0-143 0-19 0-12 0-80
IP per 38-117 hospitalization

Source: Faculty survey, NCMH 2004 Source: Faculty survey, NCMH 2004
97
SECTION II
ern States, which is clearly indicative of the cost of inputs and paying capacity, as it is known that Kerala is more expensivelabour costs are higher as also the people's willingness and ability to pay. Thirdly, government pricing is three times lower than market prices. This is one reason for the general perception that physicians take payments under the table making it equally expensive for the poor. However, it is clear that the prices are unaffordable for the poor, particularly those who earn less than Rs 50 a day. Since most of these services are also clustered in specific urban areas, accessing them also entails substantial indirect expenditures. This is one factor why an increasing proportion of persons do not avail of treatment on grounds of 'not being serious'. Therefore, in the absence of subsidized care in public facilities or insurance coverage, the poor either resort to low cost solutions such as the village quack, do not use services, or get impoverished when they do.
Supply gaps and distributional inequities

No insurance policy, no matter how attractive, can have any value for the poor if the provider supplying the services is located at a distance that would entail huge indirect expenses in terms of loss of wages, transport costs, etc. This is more relevant for women and the elderly who may have to depend on another person to accompany them to the facility. The survey clearly showed that in the poorest districts, the distribution of facilities is highly skewed. In half the rural areas, the only alternative is the ill-equipped and under-funded public sector which, in some States such as Bihar, is 'mostly on paper'. Figures 8 and 9 show the long distances that women have to travel for accessing basic EmOC, so necessary for saving lives, in the Jalna district of Maharashtra and Kozhikode in Kerala. While Maharashtra has a maternal mortality rate of 4/1000, Kozhikode has less than 1 with 98% institutional deliveries. The figures also indicate the concentration and duplication of facilities which then provide possibilities for a more equitable spread and greater access. Jalna has 10 facilities all clustered in 3 towns. If these could be relocated through a set of policy incentives, the distance to be travelled could be halved (Figs 8 and 9).
Government initiatives for public-private partnerships

Given the huge spread of the private sector, it is expected that public policy would have assiduously sought their services to extend the reach of public health programmes. Synergizing the efforts of the two sectors in mutually beneficial partnerships to provide a package of services could have the potential to enhance and upscale the implementation of several programmatic interventions, such as institutional deliveries, cataract surgeries, sterilizations, provision of Directly Observed Treatment, Short-course (DOTS) under the TB Control Programmes, etc. The survey showed that such partnerships were few, not exceeding 4.6% of the total private facilities and that too largely in Jalna, Khammam and Udaipur.
Fig 8 Distance to emergency obstetric care facilities in Kozhikode, Kerala (in km)
What can we learn from the evidence?

The facility survey was undertaken in the context of assessing the supply-side issues of the health system in a scenario that seeks to introduce a universal social health insurance scheme. Such an evaluation was considered necessary in the light of the prevailing perception at policy level that the private sector could address the health needs of the poor where the public sector has failed, by addressing the financial barrier through health insurance. The survey has clearly brought out three important issues: (i) that there is a very highly skewed distribution pattern of facilities, beds and specialists; (ii) a very large number of providers working as sole practitioners are providing substandard treatment; and (iii) the cost of the private sector is high. These then are the aspects that a public policy will have to address while designing an appropriate strategy. Addressing these issues from the standpoint of a Universal Health Insurance scheme that will help bring down the financial barrier and enable people to access health care services would require clear policies; these are elaborated below.
98
SECTION II
Fig 9 Distance to emergency obstetric care facilities in Jalna, Maharashtra (in km)
Such distributional inequity raises three issues that need to be addressed. The first is the duplication of public and private facilities almost everywhere. The private sector, in most places, is located where the public sector is already established. Does this show an unmet demand that the private sector is meeting or are many of these facilities established and sustained by government doctors doubling up to work in private hospitals, creating a win-win situation where the government doctor makes additional income and the private facility gets respectability and a ready market? Government doctors are generally believed to be more experienced and also carry the stamp of being certified'a major psychological need for patients in the absence of accreditation processes. While one study showed that two-thirds of private hospitals employed government doctors, it is believed that 30% of the private sector consists of government doctors working in small, rural nursing homes or large city hospitals. This plurality needs to be more closely studied as in the event of the government doctor 'owning' the private facility, chances of relocation could be initially problematic, calling for hard policy decisions to address such issues of conflicts of interest. Besides, having multiple facilities in the same location does provide choice. However, such choice in a small market space could also lead to creating small unviable units, each adopting unhealthy practices and cutting corners on quality to stay competitive, as well as the suboptimal utilization of public facilities which have been designed to cope with the whole caseload for a population of 100,000. Given the smallness of the population units, i.e. the CHC area, there is not enough room for a full-fledged, 30-bed CHC as well as a variety of private facilities to function. How is the decision to be taken to relocate the CHC to a more needy area which, in a rationalization of the infrastructure matrix, would make economic sense? Policy choices in such a situation could range from reducing the norms of public hospitals to cater to half
the patient load or, through a range of financial incentives, motivate the private sector to relocate in underserved areas. Secondly, non-standardization of facility location as per norms or needs, creates problems for considering other alternative payment systems such as capitation, a better system for containing cost than the existing fee-for-service system. Capitation systems, such as those in the UK, function if there is an assured population base per provider whose health needs are to be taken care of by the provider. Such a system also requires the provider to be within easy distance and accessible at all times. Thus, while there are an 'adequate' number of qualified providers available, they need to be de-concentrated to get a more equitable spread where populations can then be 'attached' for providing basic care. Third, if the design for collaborating with the private sector is based on a basket of services to be provided and not a single, one-time activity such as a delivery or a sterilization, then it would require a multiskilled health team rather than a sole practitioner who can at best provide some OP services and refer. Such an approach would require the sole practitioner to expand his facility to conform to the standards. Such expansion would require investment that a practitioner will be willing to make only if he is certain that he will get the contract for that period of time till he breaks even. In a competitive system, where providers would perhaps be selected on the basis of open tenders, such assurances become problematic, requiring a new way of doing business. At the same time, if competition is eliminated, the provider can also charge monopoly prices. What emerges from the data is the need to undertake detailed microplanning of facilities and, based on regulations, financial incentives and a process of negotiation, undertake the task of redistribution in the manner required. Addressing this issue will therefore call for a policy package that will help stimulate such reorganization and restructuring of the public-private sectors so as to have an equitable spread of facilities in accordance with viability norms and functional needs. This then becomes the first requirement for any health insurance scheme to work.
Concentration of specialists and technology

The NCMH survey showed that more than three-quarters of specialists and technology are in the private sector, all located in a few towns. In the absence of insurance and given the huge burden of non-communicable diseases, there is a need to bridge this divide. As per estimates from the Registrar General of India (RGI), 1998, 275 per 100,000 persons died of cardiovascular disease (CVD) in Rajasthan compared with 187 in Kerala. Could some of the deaths have been avoided in Rajasthan if there was better access to timely treatment? The policy options before the Government to address the problem of timely access could be to increase public investment. In China, county hospitals equivalent to our 30-bed CHCs have CT scans. The second is to strengthen the capacity of public facilities to force down the prices in the private sector to reasonable levels. Another option could be to get
99
SECTION II
into public-private partnerships, as in Chhattisgarh and Belgaum, for joint venture approaches to high-end care. Public health goals can be assured by having public representatives on the board and the power to fix rates based on a costing. Any such collaboration requires basic treatment protocols and standards that would form the basis for costing of services. In the absence of such standards and protocols, there is an element of arbitrariness in fixing prices, often stretching these to the maximum that the market can bear. With treatment costs reportedly increasing at the rate of about 22% every year, no government or insurance system can sustain such inflation over a period of time.
Need for Standards and Treatment Protocols

Engaging the private sector and controlling health markets will need to have a basisa framework of rules, regulations and transparency. This is because in the ultimate analysis, service delivery is based on discretionary judgement of the provider and this can (and does) change from case to case, since no single case is similar to another. Balancing the dual role of protecting the interest of the patient and his own creates a grey area where the provider can abuse his power by getting the patient to undergo unnecessary tests and procedures, stay longer in the hospital, or resort to irrational prescribing, etc. The single most effective way of countering such perverse incentives and speeding the restructuring process of provider markets to offer multiskilled quality care under one roof is through standards and treatment protocols and having a system for enforcing them. Standards-based payment systems do help in enforcing provider accountability, and also check unethical practices and conflict of interest issues. These are very critical as, for example, the survey found that in some places there was a clear nexus between private medical practitioners and pharmacy shopsin one district it was learnt that most pharmacy shops were 'owned' by the doctors; most private doctors depended on referrals from quacks who acted as 'procuring' agents for getting patients to their facilities for which a certain commission was paid; fee-splitting between diagnostic centres and referring doctors, AYUSH practitioners practising allopathy; etc. Such practices contribute to increasing costs on account of over-prescription of drugs, over-diagnosis of tests and over-treatment, or subjecting the patient to unnecessary investigations and procedures. The indiscriminate proliferation of technology is a clear pointer of such tendencies.
Part III Issues for Policy

Equity: Cost of care
Most literature on the private sector has found it to be 'exploitative' and three to four times more expensive when compared to the public sector, making it inaccessible to the poor and the chronically ill. The 52nd Round of the NSS showed that 35% of those hospitalized in Bihar (compared to 16% in Ker-
ala) got pushed below the poverty line on account of meeting the cost of medical treatment. This needs to be read within the context of what the survey showedover 90% of service delivery in Bihar is in the private sector compared to about 60% in Kerala. The question that is often raised in the context of differentials in public-private sector pricing is that the private sector rates are unreasonably high. The issue to examine then is whether they could be lesser. The second issue is whether the comparison is fair. Are we comparing apples and oranges or different types of apples? For, after all, in the pricing of the All India Institute of Medical Sciences (AIIMS), the salary structure or the cost of maintenance of building and land, etc. are not factored in but only the cost of consumables and drugs, etc. and therefore, government rates do not really reflect true costs. The question of what goes into the pricing of services in the private sector is an important issue that needs to be understood in a comprehensive manner. In the private sector (barring some faith-based institutions) pricing is influenced to a large extent by the market prices of inputsland, building, equipment, provider payments, etc. Based on this understanding, the Government extended subsidy to the private sectorthe logic being that in subsidizing the actual cost of inputs by giving land free or excise waivers on import of equipment it would enable lowering of prices. Such 'lowered prices' were then seen as a social gain, justifying the public subsidy. Time has shown this logic to be faulty. Bhat (1999) carried out a survey of 108 private practitioners of allopathy to examine the factors that influence healthseeking behaviour and growth of private practice. The survey showed that three factors provided the competitive edgethe experience of the treating physician, technology and location, which also acted as barriers to entry. Of these, experience of the treating physician carried the highest score. Since experience comes with repetitive practice and is built over the years, the value addition commands a price. Similarly, specialists are high consumers of technology as it enables better diagnosis and a psychological edge as they are more knowledgeable and well trained. Finally, distance and ease of access is again an important factor; facilities located in commercial areas do get higher clientele though such areas are more costly. The study found that competition pushed most doctors to locate their practice in residential areas so as to bring down the prices. Pricing by the private sector is then determined by the cost of capital and operational expenses. Input prices such as the cost of labour, rentals, level of technology and specialization, and source of capital and interest rates determine the cost, explaining for the rural-urban differentials in prices. However, it is also observed that due to the generally low occupancy of beds, the private sector does attempt to stay competitive by appointing unqualified nurses and AYUSH doctors at far lower wages, combine sale of drugs and earning commissions from diagnostic laboratories for every case referred, etc. Muraleedharan (1999) in a study of private sector pricing in Chennai identified three forms of payment systems: fixed-
100
SECTION II
fee schedule, flexible-fee schedule and fee-sharing system. Of these, the most prevalent was the flexible-fee schedule system under which the physician and the hospital charged their rates separately. In none of the systems is there any incentive for cost control since the payment system is based on fee for service. Public policies in fixing rates do not take these factors into account. For example, under the Central Government Health Scheme (CGHS) for its employees, rates for reimbursement of services availed of in private hospitals are based on an average of the rates quoted by all the tendering hospitals. In such a system, higher than market rates are paid to facilities located in smaller towns where the input prices are lower, and lower rates paid for city hospitals where the input prices are higher. If the CGHS is an important source of revenue for the town hospital, then it generates an overall increase in the price structure in those areas, while patients in the city hospitals are forced to pay the differential amounts out of pocket. Similarly, in public hospitals, the pricing of diagnostic tests is lower than the variable costs (Purohit 1995) making it attractive for the richer sections to avail of these services. These aspects need to be looked into for addressing the issues related to resource efficiency. In the light of the above, it is necessary for the Government to undertake the unit costing of services. This is an important and useful exercise even for itself as it gives the benchmark with which to compare the extent to which the private pricing structures are unreasonable. Such pricing also would help instil some consciousness of costs and prices, which is very necessary, as there is nothing called 'free health care' since someone does pay for itdirectly through user fees or indirectly through taxes. Therefore, when the Government provides 'free care' it reflects the principle of solidarity where the richer sections through taxes enable the poor or all sections of society obtain free or subsidized care depending on the value society attaches to health. This then makes it unjustifiable to allow wasteful use of these resources which happens as a consequence of not being 'cost conscious'. For example, no private company would waste its money in funding subcentres in a State such as Kerala where road connectivity, health-seeking behaviour, 98% institutional deliveries, easy access to specialists and different health needs from what auxiliary nurse-midwives (ANMs) are trained for, together make more sense to go to the PHC or the city hospital than the subcentre. Yet these are not only continued but further money is also being spent in having the already overworked medical doctor of the PHC conduct OP clinics at the subcentre once a week. These arrangements continue unchanged because we have no idea of the costs involved and the fact that the same money could be better spent on services that the people desperately need.
referred to as the third sector, which could be nurtured to be a credible alternative to a cash-strapped and poorly managed public health system and an expensive private system. Though scattered, isolated and small in scale, there are examples of NGOs which have conclusively demonstrated that they have the capability of providing reasonably good-quality care at affordable rates to the poor. Besides, contrary to our experience with the for-profit health sector, public subsidies extended to NGOs have shown substantial social gain, as experienced, for example, under the Blindness Control Programme. Under this Programme, almost 30 organizations, located in underserved areas, were provided a non-recurring grant of Rs 18 lakh for construction of an operation theatre or a ward, purchase of a microscope or vehicle, etc. in return for doing a certain number of IOL surgeries free of cost and later at low rates against a subsidy of Rs 600 from the Government. These 30 NGOs in association with other not-for-profit private sector bodies perform almost 30% of the total 40 lakh cataract surgeries in a year. The experience under the National Leprosy Programme is similar. Due to the rising cost of inputs and uncertainity of grants, both foreign and domestic, the proportion of user fees is increasing and free care reducing. This is corroborated by the results of the 52nd Round of the NSS that showed a decrease in access to free care from 19% to 10% during 1986-96. However, to ensure that the poor are not denied care for want of ability to pay, the not-for-profit institutions follow the system of differential pricinghigher amounts for well-off patients and free or subsidized rates for the poor. An enquiry by the NCMH with three Delhi-based hospitalsone each to represent a charitable hospital, a not-for profit-private/trust hospital and one for-profit corporate hospital showed that the charitable hospitals charge a nominal fee for registration and provide treatment at differential ratesa private room for a person willing to pay and a general ward for free care. However, neither the drugs nor the expensive diagnostic tests are subsidized. The for-profit hospitals, on the other hand, have uniform rates for all sections. A comparison of costs on some common procedures in these three hospitals showed that the charges of the not-for-profit hospital were almost one-third to one-half of the corporate, for-profit hospital (Table 9).
Table 9 Rates of some common procedures (in Rs)*

Procedure Charitable Trust I (Tertiary) Corporate
Caesarean section 2500-5850-7475 Removal of cataract 4500-6500 Appendectomy 2970-4290
5750-11,500 7700-25,800 6500-20,000 8400-28,000 4750-9500 6500-21,500
Pricing in the Not-for-Profit Sector: Is the third sector an option?

Given the high costs of the for-profit private sector, we examined the pricing system in the not-for-profit sector,
*The rates vary according to the graded bed charges. The table gives the lower and the upper limits Source: NCMH 2004
A survey that included community-based organizations in rural areas (Berman and Dave 1994) showed that the notFinancing and Delivery of Health Care Services in India
101
SECTION II
for-profit hospitals are able to achieve cost efficiencies due to the following factors: Low wages of employees, using contract workers, thereby fixing the wage bill not to exceed 30% of the total; Utilization of specialist services on an honorary basis; Use of generic and essential drugs manufactured by notfor-profit organizations such as Low-Cost Standardized Therapeutics (LOCOST) Emphasizing referrals and stringent use of expensive technology. Analysis seems to suggest that the not-for-profit sector, particularly community-based organizations, seem to have had a beneficial impact on access, equity and quality of services in rural and backward areas. For public policy this provides an option to examine whether by providing financial and technical support this sector can be strengthened to engage in providing health services, particularly in rural areas.
Quality of Care
The question under quality of care is whether people receive value for their money. No one knows, as there are no norms or yardsticks with which to measure good quality against inferior. All that is known is that while the private sector has expanded access and been responsive to patient needs, competitive pressures have set off a 'technology race', making quality a concern. While on the one hand, there is the private sector getting known to overtreat, undertake unnecessary and expensive investigations; on the other hand is the rapid mushrooming of substandard facilities indulging in malpractice with impunity. The private sector, particularly at the lower end of the spectrum, is seen to have a poor knowledge base and tends to follow irrational, ineffective and sometimes even harmful practices for treating minor ailments (Nandraj 1994). A study showed that 74% of hospitals in Delhi did not meet the standards of the Delhi Nursing Homes Act. In the absence of a nationally accepted set of standards and quality assurance mechanisms, there is a disturbing perception that equates the use of sophisticated technology with 'good' quality and good value for money. In the health sector, the patient's perceptions determine health-seeking behaviour, which have important implications in a system where money follows from the patient. Quality is perceived to be expensive. This is true as quality of care requires conforming to certain minimum standards of patient care. Such standards range from physical standards to the type and proportion of personnel that should be appointed, the equipment that ought to be available and how they have to be maintained, the records that have to be filed, the reports that need to be furnished, etc. In other words, adherence to such standards makes the provider more accountable to the system and the patient in particular. However, as adhering to a minimum level of standards requires investments, there is a cost pressure both on capital as well as revenue. As the Government is constantly poorly funded and private providers seek to save on costs to maximize profits, low quality is an issue for both. In other words, low prices give low
quality. In Nigeria, a community-based movement raised the slogan of 'free care is free death,' to draw the government's attention to the rotten state of the public facilities, which existed only on papera situation similar to Vaishali. Therefore, there is a need to focus on quality of care rather than on whether services are priced or not, since public policies can always protect the poor and those unable to pay through various means, such as exemptions. It is pertinent in this context to revisit the pervasive influence of the untrained practitioner at the village level. In a discussion with private sector doctors in Khammam, it became clear that their practice was entirely dependent on this informal provider to get them patients for a fee. They reportedly earned almost Rs 10,000-15,000 just from referrals. Similarly, drug companies provided commissions for the sale of drugs. Their unanimous opinion that the primary health care services would totally collapse if these informal providers were removed as required by law was the most disturbing. This sentiment has been echoed elsewhere too. A study conducted by Bhat (1999) of 49 unqualified private medical practitioners in 4 blocks, spread over 3 districts in West Bengal, showed that unqualified practitioners enjoyed close rapport with and the trust of the local community as they were a part of the community, were accessible at all times, provide treatment for several types of ailments, including antibiotics that gave quick relief, etc. However, the study observed that while such rapport earned them yearly incomes ranging from Rs 8400 to Rs. 65,000, their poor knowledge and lack of training did result in substantial morbidity, as many would go 'beyond their level of expertise in providing inappropriate treatment to retain patients'. Government doctors are reported to have stated that several women suffering from toxaemia, eclampsia, septic abortions, fever, retained placenta, abdominal pain, incomplete abortions, etc. examined in the PHC were all traced to the administering of incorrect dosage of medicines, intramuscular administration of cintocenon, manhandling, lack of knowledge and the use of roots, hormones, twigs, etc. for abortion, faulty handling of labour, unscientific delivery, etc. by these providers. In the long run, quality reduces morbidity and mortality, which entail huge costs to the society and family when a breadwinner dies. However, the motivation to institute quality assurance systems for enhancing patient safety will be a low priority so long as the payments are based on fee for service for, in such a system, every visit and every additional investigation brings revenue to the provider. Therefore, nondevelopment of standards and non-establishment of quality assurance systems either by law or professional bodies is a barrier for expanding social insurance.
Improving efficiencies: Is market segmentation practical?

In Canada, the private sector can provide only those services that the country's national insurance policy does not cover, namely, physician fees for OP services and all hospitalized care. Therefore, the private sector provides all home care and any
102
SECTION II
treatment that does not need hospitalization. This avoids duplication of resources, better system efficiencies and unhealthy competition due to which the public sector suffers. The feasibility of such segmentation should be explored so that the two sectors complement and supplement each other rather than substitute and duplicate efforts. Such segmentation, for example, may imply that all public health programmes under the National Health Policy alongside a set of regulations such as the Certificate of Need for installation of equipment that cost beyond a certain amount, could be in the public domain, where the Government should pay for the services rendered, whether in the public or private sector. Such a policy will strengthen the demand side and may well facilitate the private sector to develop capacity and invest in skills that are not being provided by the public sector. The second advantage would be the standardization of treatment and reduced probability of drug resistance induced by irrational use of drugs and medicines for treating diarrhoeoa, TB, malaria, etc.
skills and technical capacity is the other and needs immediate attention. Without such capacity, institutionalizing publicprivate participation is difficult to sustain.
Conclusion
It is to be realized that, as elsewhere in the world, the private sector in India too has been shown to maximize profits; fail to address public health goals; lack integration with government health services; draw professionals from the public sector instead of supplementing it; and in this unregulated environment provide inappropriate or poor quality care (Bennett 1994). It is clear that the need of the hour is to regulate provider markets and correct distortions that have created an inequitable, inefficient and expensive system. The regulations will need to address all market failures that give rise to malpractices such as fee-splitting, overmedication, low adherence to quality standards. They also need to ensure ethical practices, transparency and dissemination of information on prices and quality to consumers, impose requirements for licensing and accreditation of hospitals, protocols and prices. If these market failures are not urgently and decisively addressed, the health care system will be unsustainable. It is important to acknowledge that considerable resources have been invested by the private sector. It makes no economic sense for the Government to duplicate investments, when these resources can be directed towards underserved areas and achieving public goals. Therefore, fresh investment should be need-based. Second, expansion of access to health care should now be through innovative financing strategies such as universal social insurance or subsidized community financing options. Mechanisms that separate the role of the state from being the provider and financier will facilitate contracting private health services and, with public facilities also enabled to improve quality, create a healthy competitive environment to the advantage of the Government and the consumer. However, the success of such a system will be dependent on having comprehensive regulations, the consensus of professional organizations, consumer advocacy forums, institutionalization of quality assurance mechanisms, a responsive grievance redressal mechanism, an administrative capacity and the will to enforce them.
Regulatory Capacity of the state

Collaboration with the private sector carries the implication of substituting in many ways the role of the state by market forces to regulate several aspects of provider and patient behaviour. By and large, regulations are not known to have sorted out many of the issues as can be seen in the case of the US, which has several regulations and yet an inequitable health system. Yet, since health markets do entail several failures, the state has to be vigilant and control and guide through its power as a regulator. Therefore, regulations on all aspects of health care service provisioning need to be formulated and certifying institutions established for laying down benchmarks for excellence and accreditation of facilities. The development of the capacity to enforce these regulations is also important. Enforcement is expensive; extensive computerization and people to monitor, inspect, verify and correct are required. The capacity to contract and enforce the contractual obligations is an important set of monitoring skills. Data collection, collation, analysis and research require medical doctors as well as economists and biostatisticians. Microplanning, restructuring of the health provider markets, price-setting, etc. require patience, negotiating skills and dialogue with various provider associations. Money and making laws is thus only one part of the solution, having the required human
103
SECTION II
References
Baru R. Private health care in India: Social charecteristics and trends. New Delhi: Sage Publications; 1998. Berman and Khan, Paying for Indias Health Care, Sage Publications, New Delhi, 1993. Bennett S, McPake B, Mills A (eds). Private health providers in developing countries: Serving the public interest? London: Zed Books Ltd.; 1997. Berman and Dave, Experiences in Paying for Health Care in Indias Voluntary Sector in Saroj Pachauri (ed.) Reaching Indias Poor: Non-governmental Approaches to Community Health, Sage Publications, New Delhi, 1994. Bhat R. Characteristics of private medical practice in India: A provider perspective. In: Health policy and planning. London: Oxford University Press; 1999. Government of India. National Health Policy, 1983. Government of India. National Health Policy, 2001. Government of India. Directory of Health Service, 1996. Government of India. (1998), Survey of Morbidity and Utilization of Medical Services July 1995- June 96, National Sample Survey, 52nd Round, GOI Government of India. (2004), Unorganised Service Sector Enterprises in India National Sample Survey, 57th Round, 2001-02 Indian Health Care Federation. Healthcare in India: The Road Ahead (CII-Mckinsey Report), supported by the Indian Health Care Federation, October, 2004. Purohit, B. Private initiatives and policy options: recent health system experience in India by in Health Policy and Planning, 16 (1), 87-97, 2001. Rao KS, Ramana GNV, Murthy HVV. Financing of primary health care in Andhra Pradesh. WHO, New Delhi. 1997. Rick K Homan and K.R. Thankappan, Achuthe Menon Centre for Health Services, Trivandrum, Kerala,1999. World Development Report. Investing in Health OUP, New York, 1993. Mavalankar DV. Management constraints for operationalization of Reproductive Health Program Interventions in PHC systems in India. Prepared for the National Consultation toward Comprehensive Womens' Health Policy and Programs by VHAI-WAHI-DSE. Conference held at the Voluntary Health Association of India, New Delhi, 18-19 February 1999. Muraleedharan, V.R. (1999), Charecteristics and structure of Private Hospital Sector in Urban India: A Study of Madras city submitted to Abt Associates Inc., Maryland, USA, 1999. Nandraj S, Muraleedharan VR, Baru RV, Qadeer I, Priya R. Private health sector in India: Review and annotated bibliography. Mumbai: CEHAT; 2001. Nandraj S: Committee for regulating private nursing homes and hospitals, report Submitted to the High Court, Mumbai, 1992. Purohit B.C. and Siddiqui T.A. Cost Recovery in Diagnostic Facilities Economic and Political Weekly, July, 1995.
104
SECTION II
Annexure I Size and Structure of the Private Health Sector in India

Compiled by: S Sakthivel,
Technical Consultant, NCMH, 2005 The dominance of the private sector in the health care delivery system in India is well known. Large-scale household surveys have clearly established the supremacy of the private sector in the provision of health care. On the financing front as well, private expenditure (mostly households) on health care is substantially higher. Although the dominance of the private sector is established from the demand side, with people preferring to seek health care whether it is outpatient or inpatient facilities, evidence from the supply side is weak and almost negligible. This section intends to fill this void from the recently conducted large-scale National Sample Survey Organization (NSSO) survey. On the broader aspect of the size and structure of the private health sector, we draw largely from the 57th Round of NSS on Unorganised Service Sector Enterprises in India, conducted during 2001-02.
the overall service sector workforce and around 6.5% of the total workforce engaged in community, social and personal services. Since OAEs sometimes employ workers on a temporary basis, the number of workers in OAEs is slightly more than the number of enterprises per se. While the number of OAEs was 10.77 lakh in 2001-02, the number of workers involved in OAEs was 11.56 lakh, accounting for roughly 56% of the total workforce in the health sector. The remaining 44% were engaged in establishments. A casual glance at size-class distribution of workers reveal that of the 10.77 lakh OAEs, 10.04 lakh are single person-run health facilities and the remaining 73,204 OAEs hire one or two workers on a temporary basis. In rural India, 80% of the private sector health workforce is engaged in 90% of health OAEs. In urban areas, however, around 38% of health establishments employs close to 70% of the urban workforce. Such a pattern is visible across most States, with few exceptions.
Structure of Private Health Providers in India

If we consider all the 13 lakh private health providers, both OAEs and establishments taken together, a little over half of them are allopathy-practising physicians and specialists. Tables 1.5 and 1.6 reveal that the remaining 50% of health facilities, among others, are equally divided between (i) nurses, paramedicals, physiotherapists, (ii) Ayurveda practitioners, and (iii) Homeopathy practitioners. Thus, over one-fourth of the enterprises (involving both OAEs and establishments) belong to the category of ISM practitioners/health facilities. Further, the 57th Round of the NSS shows that diagnostic/pathology laboratories account for less than 3% of the health facilities in India.
Size of Private Health Enterprises in India

The 57th round of NSS studied all health practitioners, from sole practitioners to the largest hospitals, from Indian Systems of Medicine (ISM) to allopathy, from qualified doctors to unqualified quacks in the private sector. According to the survey, in 2001-02, all these practitioners and facilities put together were approximately 13 lakh enterprises providing health services in the country, excluding public facilities. Tables 1.1 and 1.2 show that the majority of these enterprises are own-account enterprises (OAEs), which accounted for over 80% of the total health facility in the country. OAEs are typically run by an individual or are a household business providing health services without hiring a worker on a fairly regular basis. OAEs are also those enterprises where there are temporary labourers working in such enterprises not on a regular basis. On the other hand, the number of health establishments in the country was roughly around 2.3 lakh, which accounted for less than 20%. Establishments are those that hire at least one worker on a regular basis. The predominance of OAEs and the lack of establishments in rural areas as compared to urban India are quite stark, with over 92% of OAEs and around 7% of establishments in rural areas. In contrast, in the urban areas, establishments accounted for roughly 38% and the remaining 62% facilities were OAEs. Except a few, these trends are more or less the same in most of the States.
Status of Regulation in Private Health Enterprises

The need for regulating a lifeline sector such as health care hardly needs to be emphasized. Although beer bars and pan shops require a licence for establishing and running these stores in India, health facilities-whether consultation chambers run by doctors or a big private hospital-do not require a licence. The mushrooming of the private health sector without a regulatory structure is a cause for concern. Since health is essentially a State subject, regulatory mechanisms needed for uniform and transparent application of various legislations are lacking. Poor quality of medical care, medical negligence, self-seeking behaviour, etc. have been the bane of the health sector in India causing untold misery particularly to the needy, as well as the rich. In the country as a whole, the 57th Round of the NSS reveals that only a little over half of the enterprises are actually registered under the Medical Practitioners Act and another 8% are registered under other Acts (mostly Societies Act, Shops & Establishments Act, and Local Bodies Act). More than one-third of the health enterprises do not have any form of registration. State-wise analysis shows that in Assam, only one-fourth of the health facilities appear to be registered, with a paltry 12% of the facilities registered under the Medical Practitioners Act (Tables 1.7 and 1.8). This is followed by Orissa, wherein over half the enterprises do not have registration. On the other end of the spectrum are smaller
Magnitude of the Private Health Workforce in India

Of the 13 lakh are engaged enterprises in India, roughly 20 lakh health providers, including skilled, semi-skilled and unskilled ones, are involved. These 20 lakh individuals range from dais, quacks, paramedicals to specialized doctors, etc. in the private sector (Tables 1.3 and 1.4). They account for less than 1% of the total workforce in India, around 2.5% of
105
SECTION II
States such as Goa and Uttaranchal where only 1.5% and 7.25% of enterprises are not registered, respectively. Maharashtra is the only big State in which roughly three-fourths of the health facilities are registered under the Medical Practitioners Act.
Non-Profit Health Institutions in India

With the State abdicating its responsibility in the provision of health facilities in India ever since the liberalization process started in the early 1990s, health NGOs are reportedly mushrooming rapidly. Presently, however, evidence on the size of non-profit institutions (NPIs)/non-governmental organisations (NGOs) involved in the health sector in India is virtually nonexistent or is at best, inadequate. The 57th Round of NSS captures the magnitude of NPIs in India (Tables 1.9 and 1.10). Compared to for-profit institutions, NPIs account for a minis-
cule 1.32% of the total enterprises. However, the break-up of NPIs among OAEs and establishments show that the former accounted for a paltry 0.85% and the latter 3.54%. Since OAEs are practically run by a single person, a non-profit motive is unlikely. Therefore, it would be worth assessing the establishments. Among the establishments that employ more than one worker on a fairly regular basis, 3.54% of the total establishments belong to the category of NPIs in India. However, the spread of NGOs is quite erratic in different States. For instance, Uttaranchal has a substantial number of NGO health establishments followed by Punjab. The respective shares of NGOs in the total health establishments in these two States are roughly 43% and 15%. States such as Bihar, Goa, Jharkand and Karnataka have a negligible presence of NGOs, accounting for less than 1% of the total health establishments in these States.
106
SECTION II
Table 1.1 Number of enterprises in the unorganized health services by States 2001-02
Rural State OAE Establishments Total Urban OAE Establishments Total Aggregate OAE Establishments Total
Andhra Pradesh Assam Bihar Chhattisgarh Delhi Goa Gujarat Himachal Pradesh Haryana Jammu and Kashmir Jharkhand Karnataka Kerala Madhya Pradesh Maharashtra Orissa Punjab Rajasthan Tamil Nadu Uttar Pradesh Uttaranchal West Bengal Others Total
48,306 38,802 105,563 7,222 170 246 14,235 4,449 10,591 7,876 56,702 12,181 15,132 26,547 23,409 46,064 20,298 16,935 11,350 253,989 5,404 78,519 8,699 812,689
2,865 2,239 13,850 724 419 16 957 228 2,354 340 1,055 2,717 5,940 644 3,389 884 2,794 1,035 3,508 13,565 1,570 3,781 174 65,048
51,171 41,041 119,413 7,946 589 262 15,192 4,677 12,945 8,216 57,757 14,898 21,072 27,191 26,798 46,948 23,092 17,970 14,858 267,554 6,974 82,300 8,873 877,737
17,951 4,431 14,504 1,475 7,226 125 6,680 509 6,791 1,492 6,553 10,126 6,359 15,749 32,664 3,197 10,349 13,041 9,380 49,678 1,845 42,332 2,201 264,658
8,705 26,656 1,071 5,502 6,281 20,785 3,766 5,241 8,256 15,482 465 590 12,687 19,367 344 853 5,305 12,096 741 2,233 1,566 8,119 14,037 24,163 3,541 9,900 7,687 23,436 34,064 66,728 1,489 4,686 6,370 16,719 4,208 17,249 10,566 19,946 21,618 71,296 1,137 2,982 9,966 52,298 1,386 3,587 165,256 429,914
66,257 43,233 120,067 8,697 7,396 371 20,915 4,958 17,382 9,368 63,255 22,307 21,491 42,296 56,073 49,261 30,647 29,976 20,730 303,667 7,249 120,851 10,900 1,077,347
11,570 77,827 3,310 46,543 20,131 140,198 4,490 13,187 8,675 16,071 481 852 13,644 34,559 572 5,530 7,659 25,041 1,081 10,449 2,621 65,876 16,754 39,061 9,481 30,972 8,331 50,627 37,453 93,526 2,373 51,634 9,164 39,811 5,243 35,219 14,074 34,804 35,183 338,850 2,707 9,956 13,747 134,598 1,560 12,460 230,304 1,307,651
Source: Extracted from the unit-level record data of the 57th Round, Survey of Unorganised Services, NSSO Note: (i) OAEs indicate own-account enterprises wherein an undertaking is run by a household, usually without any hired labour working on a fairly regular basis. ii) Establishments are the ones that employ atleast one hired worker on a fairly regular basis.iii) Others include all minor States and Union Territories
Table 1.2 Percentage distribution of enterprises in the unorganized health services by States 2001-02
Rural State OAE Establishments OAE Urban Establishments OAE Aggregate Establishments
Andhra Pradesh Assam Bihar Chhattisgarh Delhi Goa Gujarat Himachal Pradesh Haryana Jammu and Kashmir Jharkhand Karnataka Kerala Madhya Pradesh Maharashtra Orissa Punjab Rajasthan Tamil Nadu Uttar Pradesh Uttaranchal West Bengal Others Total [Au? Pls check the total]
94.40 94.54 88.40 90.89 28.86 93.89 93.70 95.13 81.82 95.86 98.17 81.76 71.81 97.63 87.35 98.12 87.90 94.24 76.39 94.93 77.49 95.41 98.04 92.59
5.60 5.46 11.60 9.11 71.14 6.11 6.30 4.87 18.18 4.14 1.83 18.24 28.19 2.37 12.65 1.88 12.10 5.76 23.61 5.07 22.51 4.59 1.96 7.41
67.34 80.53 69.78 28.14 46.67 21.19 34.49 59.67 56.14 66.82 80.71 41.91 64.23 67.20 48.95 68.22 61.90 75.60 47.03 69.68 61.87 80.94 61.36 61.56
32.66 19.47 30.22 71.86 53.33 78.81 65.51 40.33 43.86 33.18 19.29 58.09 35.77 32.80 51.05 31.78 38.10 24.40 52.97 30.32 38.13 19.06 38.64 38.44
85.13 92.89 85.64 65.95 46.02 43.54 60.52 89.66 69.41 89.65 96.02 57.11 69.39 83.54 59.95 95.40 76.98 85.11 59.56 89.62 72.81 89.79 87.48 82.39
14.87 7.11 14.36 34.05 53.98 56.46 39.48 10.34 30.59 10.35 3.98 42.89 30.61 16.46 40.05 4.60 23.02 14.89 40.44 10.38 27.19 10.21 12.52 17.61
Source: Extracted from the unit-level record data of the 57th Round, Survey of Unorganised Services, NSSO Note: (i) OAEs indicate own-account enterprises wherein an undertaking is run by a household, usually without any hired labour working on a fairly regular basis. ii) Establishments are the ones that employ atleast one hired worker on a fairly regular basis.iii) Others include all minor States and Union Territories
107
SECTION II
Table 1.3 Number of workers in the unorganized health services by States 2001-02
Rural State OAE Establishments Total Urban OAE Establishments Total Aggregate OAE Establishments Total
51,084 39,177 108,201 7,686 170 257 14,588 4,725 10,778 8,825 57,007 13,686 17,786 26,920 25,178 47,321 21,323 17,511 15,281 276,581 5,466 79,298 8,912 857,761
8,461 59,545 5,669 44,846 28,330 136,531 1,698 9,384 1,256 1,426 32 289 4,622 19,210 692 5,417 4,708 15,486 740 9,565 2,392 59,399 5,938 19,624 31,355 49,141 1,725 28,645 8,063 33,241 2,042 49,363 8,220 29,543 2,978 20,489 16,059 31,340 32,823 309,404 26,322 31,788 11,130 90,428 622 9,534 205,877 1,063,638
20,862 4,500 15,498 1,528 8,515 125 7,884 636 7,781 1,675 6,996 12,316 7,077 17,564 37,590 4,058 10,877 14,054 11,259 58,938 1,951 44,279 2,351 298,314
38,794 59,656 5,041 9,541 21,302 36,800 13,257 14,785 33,058 41,573 1,532 1,657 48,587 56,471 1,601 2,237 21,347 29,128 2,184 3,859 4,181 11,177 49,590 61,906 30,005 37,082 24,543 42,107 151,708 189,298 5,968 10,026 28,130 39,007 16,272 30,326 54,252 65,511 79,998 138,936 5,001 6,952 50,912 95,191 5,513 7,864 692,776 991,090
71,946 43,677 123,699 9,214 8,685 382 22,472 5,361 18,559 10,500 64,003 26,002 24,863 44,484 62,768 51,379 32,200 31,565 26,540 335,519 7,417 123,577 11,263 1,156,075
47,255 119,201 10,710 54,387 49,632 173,331 14,955 24,169 34,314 42,999 1,564 1,946 53,209 75,681 2,293 7,654 26,055 44,614 2,924 13,424 6,573 70,576 55,528 81,530 61,360 86,223 26,268 70,752 159,771 222,539 8,010 59,389 36,350 68,550 19,250 50,815 70,311 96,851 112,821 448,340 31,323 38,740 62,042 185,619 6,135 17,398 898,653 2,054,728
Source: Extracted from the unit-level record data of the 57th Round, Survey of Unorganised Services, NSSO Note(i) OAEs indicate own-account enterprises wherein an undertaking is run by a household, usually without any hired labour working on a fairly regular basis.(ii) Establishments are the ones that employ at least one hired worker on a fairly regular basis.(ii) Others include all minor States and Union Territories.
Table 1.4 Percentage of workers in the unorganized health services by States 2001-02
Rural State OAE Establishments OAE Urban Establishments OAE Aggregate Establishments
85.79 87.36 79.25 81.91 11.92 88.93 75.94 87.23 69.60 92.26 95.97 69.74 36.19 93.98 75.74 95.86 72.18 85.47 48.76 89.39 17.20 87.69 93.48 80.64
14.21 12.64 20.75 18.09 88.08 11.07 24.06 12.77 30.40 7.74 4.03 30.26 63.81 6.02 24.26 4.14 27.82 14.53 51.24 10.61 82.80 12.31 6.52 19.36
34.97 47.16 42.11 10.33 20.48 7.54 13.96 28.43 26.71 43.41 62.59 19.89 19.08 41.71 19.86 40.47 27.88 46.34 17.19 42.42 28.06 46.52 29.90 30.10
65.03 52.84 57.89 89.67 79.52 92.46 86.04 71.57 73.29 56.59 37.41 80.11 80.92 58.29 80.14 59.53 72.12 53.66 82.81 57.58 71.94 53.48 70.10 69.90
60.36 80.31 71.37 38.12 20.20 19.63 29.69 70.04 41.60 78.22 90.69 31.89 28.84 62.87 28.21 86.51 46.97 62.12 27.40 74.84 19.15 66.58 64.74 56.26
39.64 19.69 28.63 61.88 79.80 80.37 70.31 29.96 58.40 21.78 9.31 68.11 71.16 37.13 71.79 13.49 53.03 37.88 72.60 25.16 80.85 33.42 35.26 43.74
Source: Extracted from the unit-level record data of the 57th round. Survey of Unorganised Services, NSSO Note(i) OAEs indicate own-account enterprises wherein an undertaking is run by a household, usually without any hired labour working on a fairly regular basis.(ii) Establishments are the ones that employ at least one hired worker on a fairly regular basis.(iii) Others include all minor States and Union Territories.
108
SECTION II
Table 1.5 Magnitude of various private health providers in Indian States 2001-02
State Hospital gen./spl. Physicians Nurse, physio-therapist and specialists and para medical Ayurveda Unani Homeopathy Diagnostic/ Blood Others Aggregate pathology lab banks (ambulance)
Andhra Pradesh 2,579 Assam 329 Bihar 595 Chhattisgarh 192 Delhi 240 Goa 0 Gujarat 5,507 Himachal Pradesh 103 Haryana 3,004 Jammu and Kashmir 25 Jharkhand 61 Karnataka 4,087 Kerala 2,419 Madhya Pradesh 1,239 Maharashtra 6,621 Orissa 431 Punjab 1,313 Rajasthan 1,153 Tamil Nadu 5,188 Uttar Pradesh 4,424 Uttaranchal 1,527 West Bengal 1,286 Others 310 Total 42,633
53,775 3,614 88,996 7,017 9,796 600 14,063 2,450 11,876 5,717 12,713 25,713 9,679 26,230 53,918 15,670 21,712 16,945 13,294 214,127 6,310 61,859 2,846 678,920
7,357 18,831 20,982 1,146 8 0 1,569 872 893 3,428 19,703 268 324 2,727 3,144 5,490 1,602 7,701 530 53,280 0 9,665 3,982 163,502
6,982 167 5,645 7529 1,864 2964 2,017 8 1,790 836 53 0 4,203 0 718 34 5,198 192 661 155 24,658 881 5,734 432 8,131 349 7,636 46 12,157 748 22,933 330 9,225 390 4,207 1437 4,174 513 27,042 5192 689 291 3,669 586 780 1861 160,166 24941
2,599 9,024 12,648 595 801 12 8,020 187 3,110 35 2,588 1,326 7,817 2,191 11,228 4,417 2,534 1,187 5,106 19,861 253 48,779 1,218 145,536
3,925 599 1,719 1,352 1,965 187 1,109 259 678 336 737 1,460 1,828 1,305 3,645 1,257 1,466 1,149 1,691 5,957 279 2,344 310 35,557
4 49 0 0 0 0 37 0 0 0 0 0 0 0 134 17 0 0 23 6 0 3 0 273
439 77,827 923 46,543 10,430 140,198 860 13,187 62 15,498 0 852 51 34,559 907 5,530 90 25,041 92 10,449 4,535 65,876 41 39,061 425 30,972 9,253 50,627 1,931 93,526 1,089 51,634 1,569 39,811 1,440 35,219 4,285 34,804 8,961 338,850 607 9,956 6,407 134,598 1726 13,033 56,1231,307,651
109
SECTION II
Table 1.6 Percentage of various private health providers in Indian States 2001-02
State Hospital Physicians Nurse, physio-therapist Ayurveda and para medical Unani Homeopathy Diagnostic/ pathology lab Blood banks Others (ambulance) gen./spl. and specialists
Andhra Pradesh 3.31 Assam 0.71 Bihar 0.42 Chhattisgarh 1.46 Delhi 1.55 Goa 0.00 Gujarat 15.94 Himachal Pradesh 1.86 Haryana 12.00 Jammu and Kashmir 0.24 Jharkhand 0.09 Karnataka 10.46 Kerala 7.81 Madhya Pradesh 2.45 Maharashtra 7.08 Orissa 0.83 Punjab 3.30 Rajasthan 3.27 Tamil Nadu 14.91 Uttar Pradesh 1.31 Uttaranchal 15.34 West Bengal 0.96 Others 2.38 Total [Au? Pls check the total] 3.26
69.10 7.76 63.48 53.21 63.21 70.42 40.69 44.30 47.43 54.71 19.30 65.83 31.25 51.81 57.65 30.35 54.54 48.11 38.20 63.19 63.38 45.96 21.84 51.92
9.45 40.46 14.97 8.69 0.05 0.00 4.54 15.77 3.57 32.81 29.91 0.69 1.05 5.39 3.36 10.63 4.02 21.87 1.52 15.72 0.00 7.18 30.55 12.50
8.97 12.13 1.33 15.30 11.55 6.22 12.16 12.98 20.76 6.33 37.43 14.68 26.25 15.08 13.00 44.41 23.17 11.95 11.99 7.98 6.92 2.73 5.98 12.25
0.21 16.18 2.11 0.06 5.39 0.00 0.00 0.61 0.77 1.48 1.34 1.11 1.13 0.09 0.80 0.64 0.98 4.08 1.47 1.53 2.92 0.44 14.28 1.91
3.34 19.39 9.02 4.51 5.17 1.41 23.21 3.38 12.42 0.33 3.93 3.39 25.24 4.33 12.01 8.55 6.37 3.37 14.67 5.86 2.54 36.24 9.35 11.13
5.04 1.29 1.23 10.25 12.68 21.95 3.21 4.68 2.71 3.22 1.12 3.74 5.90 2.58 3.90 2.43 3.68 3.26 4.86 1.76 2.80 1.74 2.38 2.72
0.01 0.11 0.00 0.00 0.00 0.00 0.11 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.14 0.03 0.00 0.00 0.07 0.00 0.00 0.00 0.00 0.02
0.56 1.98 7.44 6.52 0.40 0.00 0.15 16.40 0.36 0.88 6.88 0.10 1.37 18.28 2.06 2.11 3.94 4.09 12.31 2.64 6.10 4.76 13.24 4.29
110
SECTION II
Table 1.7 Status of registration in the private health sector in Indian States (200102)
State Medical Practioners Act Other Act Not registered Total
39,749 5,509 77,981 10,753 12,421 763 26,651 3,303 16,303 6,308 14,085 30,568 15,565 33,641 68,855 13,356 22,862 14,567 22,864 189,720 7,500 75,809 3,739 712,872
6,147 5,649 6,947 1,282 1,251 77 5,497 403 2,574 865 648 3,883 9,705 2,561 17,034 10,656 1,999 4,395 3,995 8,525 1,734 18,696 665 115,188
31,907 32,545 54,844 1,152 2,399 12 2,411 1,824 6,164 3,276 51,143 4,610 5,517 14,395 7,271 27,622 14,831 16,257 7,945 140,429 722 38,612 7,799 473,687
77,803 43,703 139,772 13,187 16,071 852 34,559 5,530 25,041 10,449 65,876 39,061 30,787 50,597 93,160 51,634 39,692 35,219 34,804 338,674 9,956 133,117 12,203 1,301,747
Source: Extracted from the unit-level record data of the 57th Round. Survey of Unorganised Services, NSSONote(i) OAEs indicate own-account enterprises wherein an undertaking is run by a household, usually without any hired labour working on a fairly regular basis.(ii) Establishments are the ones that employ at least one hired worker on a fairly regular basis.(iii) Others include all minor States and Union Territories(iv) Other Acts include Local Bodies Act, Shops and Estabilishments Act, Societies Act etc.
Table 1.8 Status of registration in the private health sector in Indian States (200102) (in %)
State Medical Practioners Act Other Act Not registered
51.09 12.61 55.79 81.54 77.29 89.55 77.12 59.73 65.11 60.37 21.38 78.26 50.56 66.49 73.91 25.87 57.60 41.36 65.69 56.02 75.33 56.95 30.64 54.76
7.90 12.93 4.97 9.72 7.78 9.04 15.91 7.29 10.28 8.28 0.98 9.94 31.52 5.06 18.28 20.64 5.04 12.48 11.48 2.52 17.42 14.04 5.45 8.85
41.01 74.47 39.24 8.74 14.93 1.41 6.98 32.98 24.62 31.35 77.64 11.80 17.92 28.45 7.80 53.50 37.37 46.16 22.83 41.46 7.25 29.01 63.91 36.39
Source: Extracted from Unit Level Record Data of 57th Round, Survey of Unorganised Services, NSSO Note: Other Acts include Local Bodies Act, Shops and Estabilishments Act, Societies Act etc.
111
SECTION II
Table 1.9 For-profit and non-profit institutions in Indian States

Non-profit institution State OAE Establishments Total For-profit institution OAE Establishments Total All institutions OAE Establishments Total
Andhra Pradesh Assam Bihar Chhattisgarh Delhi Goa Gujarat Himachal Pradesh Haryana Jammu and Kashmir Jharkhand Karnataka Kerala Madhya Pradesh 50,627 Maharashtra Orissa Punjab Rajasthan Tamil Nadu Uttar Pradesh Uttaranchal West Bengal Others Total
53 1861 120 16 61 0 39 793 446 9 15 384 69 66 1595 57 379 45 132 1013 0 1794 142 9089
150 122 130 65 279 0 224 49 530 16 15 28 450 173 648 55 1,399 469 879 551 1,162 654 96 8,144
203 1,983 250 81 340 0 263 842 976 25 30 412 519 239 2,243 112 1,778 514 1,011 1,564 1,162 2,448 238 17,233
66,204 41,352 119,947 8,681 7,335 371 20,876 4,165 16,936 9,359 63,240 21,923 21,422 42,230 54,478 46,069 30,268 29,931 20,598 302,654 7,249 119,057 10,758 1,065,103
11,420 77,624 3,188 44,540 20,001 139,948 4,425 13,106 8,396 15,731 481 852 13,420 34,296 523 4,688 7,129 24,065 1,065 10,424 2,606 65,846 16,726 38,649 9,031 30,453 8,158 50,388 36,805 91,283 2,318 48,387 7,765 38,033 4,774 34,705 13,195 33,793 34,632 337,286 1,545 8,794 13,092 132,149 1,464 12,222 222,159 1,287,262
66,257 43,233 120,067 8,697 7,396 371 20,915 4,958 17,382 9,368 63,255 22,307 21,491 56,073 46,126 30,647 29,976 20,730 303,667 7,249 120,851 10,900 1,074,212
11,570 3,310 20,131 4,490 8,675 481 13,644 572 7,659 1,081 2,621 16,754 9,481 42,296
77,827 46,543 140,198 13,187 16,071 852 34,559 5,530 25,041 10,449 65,876 39,061 30,972 8,331
37,453 93,526 2,373 48,499 9,164 39,811 5,243 35,219 14,074 34,804 35,183 338,850 2,707 9,956 13,746 134,597 1,560 12,460 230,303 1,304,515
Source: Extracted from Unit Level Record Data of 57th Round, Survey of Unorganised Services, NSSO Note:(i) OAEs indicate own-account enterprises wherein an undertaking is run by a household, usually without any hired labour working on a fairly regular basis.(ii) Establishments are the ones that employ at least one hired worker on a fairly regular basis.(iii) Others include all minor States and Union Territories
Table 1.10 For-profit and non-profit institutions in Indian States (in %)

Non-profit institution State OAE Establishments Total OAE For-profit institution Establishments Total
0.08 4.30 0.10 0.18 0.82 0.00 0.19 15.99 2.57 0.10 0.02 1.72 0.32 0.16 2.84 0.12 1.24 0.15 0.64 0.33 0.00 1.48 1.30 0.85
1.30 3.69 0.65 1.45 3.22 0.00 1.64 8.57 6.92 1.48 0.57 0.17 4.75 2.08 1.73 2.32 15.27 8.95 6.25 1.57 42.93 4.76 6.15 3.54
0.26 4.26 0.18 0.61 2.12 0.00 0.76 15.23 3.90 0.24 0.05 1.05 1.68 0.47 2.40 0.23 4.47 1.46 2.90 0.46 11.67 1.82 1.91 1.32
99.92 95.65 99.90 99.82 99.18 100.00 99.81 84.01 97.43 99.90 99.98 98.28 99.68 99.84 97.16 99.88 98.76 99.85 99.36 99.67 100.00 98.52 98.70 99.15
98.70 96.31 99.35 98.55 96.78 100.00 98.36 91.43 93.08 98.52 99.43 99.83 95.25 97.92 98.27 97.68 84.73 91.05 93.75 98.43 57.07 95.24 93.85 96.46
99.74 95.70 99.82 99.39 97.88 100.00 99.24 84.77 96.10 99.76 99.95 98.95 98.32 99.53 97.60 99.77 95.53 98.54 97.10 99.54 88.33 98.18 98.09 98.68
Source: Extracted from the unit-level record data of the 57th Round, Survey of Unorganised Services, NSSO Note(i) OAEs indicate own-account enterprises wherein an undertaking is run by a household, usually without any hired labour working on a fairly regular basis.(ii) Establishments are the ones that employ at least one hired worker on a fairly regular basis.(iii) Others include all minor States and Union Territories.
112
SECTION II
Annexure II Results of facility survey of eight districts in India

Table II-1 Distribution of public and private sector facilities by year of establishment
Info NA Pre-1980 1980-84 1985-89 1990-94 1995-99 2000-04 Total
Public sector Cumulative total % increase Private sector Cumulative total % increase Total Cumulative total % increase
1893
593 593 677 677 1270 1270
74
1967
243 836 41.0 428 1105 63.2 671 1941 52.8
330 1166 39.5 447 1552 40.5 777 2718 40.0
222 1388 19.0 844 2396 54.4 1066 3784 39.2
120 1508 8.6 1083 3479 45.2 1203 4987 31.8
97 1605 6.4 2235 5714 64.2 2332 7319 46.8
3498
5788
9286
Note: The table does not include information on subcentres from Jalna, Maharashtra which were not surveyed
Table II-2 Distribution of establishment of private facilities over the years by ownership
Info NA Pre 1980 % 1980-84 % 1985-89 % 1990-94 % 1995-99 % 2000-04 % Total %
Corporate Partnership Sole proprietorship Trust Total Cumulative total % increase
2 3 64 5 74
5 0.7 26 3.8 603 89.1 43 6.4 677 100.0 677
2 0.5 8 1.9 409 95.6 9 2.1 428 100.0 1105 63.2
3 0.7 13 2.9 420 94.0 11 2.5 447 100.0 1552 40.5
3 0.4 30 3.6 790 93.6 21 2.5 844 100.0 2396 54.4
9 0.8 48 4.4 997 92.1 29 2.7 1083 100.0 3479 45.2
12 0.5 157 7.0 2005 89.7 61 2.7 2235 100.0 5714 64.2
36 0.6 285 4.9 5288 91.4 179 3.1 5788 100.0
Table II-3 Geographical spread of facilities in towns

Towns District No. of Towns public facility % of towns with public facility Towns with any private facility % of with any private Towns Towns No. of villages Village with any public facility % of villages with public facility towns with any towns covered without by a facility any (%) Villages Village % of Villages Villages by any facility (%) with villages covered without any with any private facility private
facility facility*
facility facility*
Jalna Khammam Kozhikode Nadia Udaipur Ujjain Varanasi Vaishali Total
4 9 13 25 10 8 11 3 83
4 8 8 18 8 7 4 3 60
100.0 88.9 61.5 72.0 80.0 87.5 36.4 100.0 72.3
4 7 9 23 8 6 6 3 66
100.0 77.8 69.2 92.0 80.0 75.0 54.5 100.0 79.5
4 8 9 25 8 7 9 3 73
0.0 11.1 30.8 0.0 20.0 12.5 18.2 0.0 12.0
971 1229 87 1346 2351 1107 1327 1569 9987
38 356 68 376 713 188 248 190 2177
3.9 29.0 78.2 27.9 30.3 17.0 18.7 12.1 21.8
56 18 67 105 96 22 116 73 553
5.8 1.5 77.0 7.8 4.1 2.0 8.7 4.7 5.5
65 358 75 408 728 194 306 230 2364
93.3 70.9 13.8 69.7 69.0 82.5 76.9 85.3 76.3
Note: There are some towns and villages with both public and private facilities and hence this figure is less than the total number of towns or villages with any public or private facility
113
SECTION II
Table II-4 Distribution of facilities by availability of infrastructure

Public Rural Urban Rural Private Urban Rural Urban Total Total* %
Floor space (in sq. ft) 1-500 500-1000 1000+ Total
1315 236 283 1834
186 33 126 345
1863 163 163 2189
2611 305 534 3450
3178 399 446 4023
2797 338 660 3795
5975 737 1106 7818
76.4 9.4 14.1 100.0
Note: Information on floor space was not available for 1468 facilities
Table II-5 Distribution of facilities by availability of beds

Rural District Facilities with 1-30 beds Public Private Facilities with 31-75 beds Public Facilities with >75 beds Total Facilities with 1-30 beds Urban Facilities with 31-75 beds Private Facilities with >75 beds Public Private Total Public/ private Public Private With 1-30 beds Grand Total No. of facilities With With 31-75 >75 Total
Private Public Private
Public Private Public
beds beds
34 56 8 29 30 5 22 39 223
33 34 46 8 9 4 51 65 250
0 1 3 2 1 0 0 0 7
1 1 7 1 0 1 1 2 14
0 1 2 1 1 0 0 0 5
0 0 2 0 0 1 0 1 4
68 93 68 41 41 11 74 107 503
3 9 4 2 9 10 7 13 57
83 123 34 31 58 33 67 156 585
2 0 0 3 4 4 0 4 17
4 13 8 0 8 4 1 14 52
1 3 11 8 4 4 1 9 41
3 2 9 1 1 3 0 7 26
96 150 66 45 84 58 76 203 778
40 70 28 45 49 23 30 65 350
124 173 106 41 76 46 120 245 931
153 222 92 70 106 52 147 273 1115
7 15 18 6 13 9 2 20 90
4 6 24 10 6 8 1 17
164 243 134 86 125 69 150 310
76 1281
(27.3%) (72.7%) (87.1%) (7.0%)(5.9%)
Table II-6 Availability of beds by year of establishment of facilities

Pre 1980 1980-84 1985-89 1990-94 1995-99 2000-04
Public
Private
Total
Facilities Beds Average bed per facility Facilities Beds Average bed per facility Facilities Beds Average bed per facility
593 11151 18.8 677 3900 5.8 1270 15051 11.9
243 771 3.2 428 832 1.9 671 1603 2.4
330 483 1.5 447 1992 4.5 777 2475 3.2
222 387 1.7 844 2681 3.2 1066 3068 2.9
120 759 6.3 1083 2734 2.5 1203 3493 2.9
97 588 6.1 2235 4909 2.2 2332 5497 2.4
114
SECTION II
Table II-7 Number of institutions with beds in rural and urban areas
Districts Public Rural Private Public Urban Private Total Facilities (%) Public Private
34 58 13 32 31 5 22 39 234
34 35 55 9 9 6 52 68 268
6 12 15 13 17 18 8 26 115
90 138 52 32 67 40 68 177 664
164 243 135 86 124 69 150 310 1281
24.4 28.8 20.7 52.3 38.7 33.3 20.0 21.0 27.2
75.6 71.2 79.3 47.7 61.3 66.7 80.0 79.0 72.8
Table II-8 Distribution of specialists according to number of beds in facilities

Without beds Functional category Rural Urban 1-30 beds Rural Urban 31-75 beds Rural Urban 76+ beds Rural Urban Rural Total Urban Grand Total
Anesthetist Cardiologist Dentist Endocrinologist ENT Specialist Gastroenterologist Gynaecologist Ophthalmologist Orthopaedician Paediatrician Physician Psychiatrist Skin and VD Specialist Surgeons Urologist Total Number of facilities Average per facility
4 4 83 0 13 2 42 34 9 42 137 2 11 20 0 403 1955 0.21
13 28 201 4 56 9 108 77 43 90 337 13 48 56 6 1089 3193 0.34
11 2 108 0 4 0 68 13 20 22 71 2 2 54 1 378 477 0.79
72 19 16 3 21 6 230 44 75 104 150 14 6 141 12 913 642 1.42
1 0 6 0 5 0 14 2 5 10 13 0 0 7 0 63 21 3.00
20 7 17 2 11 0 61 18 17 36 38 3 6 41 6 283 69 4.10
12 0 8 0 8 0 16 7 4 11 19 3 4 36 0 128 9 14.22
130 38 45 8 74 18 167 89 89 131 134 36 41 183 26 1209 67 18.04
28 6 205 0 30 2 140 56 38 85 240 7 17 117 1 972 2462 0.39
235 92 279 17 162 33 566 228 224 361 659 66 101 421 50 3494 3971 0.88
263 98 484 17 192 35 706 284 262 446 899 73 118 538 51 4466 6433 0.69
Note: The table excludes number of subcentres in calculation of availability of specialists
Table II-9 No. of CS operations by 24-hour functional emergency obstetric care (EmOC) facilities
Gynaecologist and anaesthetist Public Private Total Surgeon and anaesthetist Public Private Total Public In all facilities Private Total
Total No. of EmOC facilities Average no. of CS operations per facility

CS: caesarean section
1432 28 51.1
675 66 10.2
2107 94 22.4
1268 28 45.3
509 59 8.6
1777 87 20.4
1432 35 40.9
678 78 8.7
2110 113 18.7
115
SECTION II
Table II-10 No. of cases of different diseases/health conditions seen in facilities in the past 30 days
Name of disease/health condition Urban Rural Total public % public Total private % private Grand total
Acute
Diarrhoea Acute respiratory infection Dental extractions Malaria New cases of TB Total TB cases on rolls Orthopaedic surgery under GA Closed fracture cases managed Open fracture cases managed Dislocations managed New cerebrovascular accidents (CVA) cases treated Acute myocardial infarction Coronary angiography Deliveries Caesarean sections Hysterectomy Accidents/injuries General medicine Chronic Total cases of STD managed Tonsillectomy Ear surgery Eye care/cataract surgery Leprosy Adult diabetics on insulin Child diabetics on insulin Hypertension Asthma and COPD Psychiatric illness Coma cases managed Kidney/ureter surgery Lithotripsy Major surgery Minor surgery Nasal surgery New cases of cancer on screening Patients for chemotherapy Patients for radiotherapy Prostrate surgery Root canal treatment Others Immunization Antenatal care Sterilization female MTPs Sterilization (male)
67931 73734 25377 37815 7686 17591 1550 6454 6720 3462 2339 5486 1138 15368 4184 1725 16889 273077 10153 2021 6433 51148 2462 14502 1190 49440 49076 21340 1049 740 410 7010 22993 955 7259 5252 2927 4364 4909 68798 75378 4126 5537 664
73861 70733 15347 39711 5218 10515 166 731 715 2278 1258 913 53 8517 509 302 12574 461144 6571 483 3009 18086 1944 11530 423 60172 62057 2173 57 78 52 890 10808 113 229 137 9 534 2625 163503 55983 3407 1441 369
57218 67959 8000 34875 5865 14844 349 2018 2264 2738 1184 1868 309 14922 1912 504 23255 406510 8271 402 4123 17310 2725 10783 750 53301 54031 8499 181 253 120 3383 21369 382 6376 1640 2840 3576 538 211590 73859 5742 1567 748
40.35 47.04 19.64 44.98 45.45 52.81 20.34 28.09 30.45 47.70 32.92 29.19 25.94 62.47 40.74 24.86 78.93 55.37 49.46 16.05 43.67 25.00 61.85 41.42 46.50 48.63 48.62 36.15 16.37 30.93 25.97 42.82 63.22 35.77 85.15 30.43 96.73 73.01 7.14 91.08 56.23 76.22 22.46 72.41
84574 76508 32724 42651 7039 13262 1367 5167 5171 3002 2413 4531 882 8963 2781 1523 6208 327711 8453 2102 5319 51924 1681 15249 863 56311 57102 15014 925 565 342 4517 12432 686 1112 3749 96 1322 6996 20711 57502 1791 5411 285
59.65 52.96 80.36 55.02 54.55 47.19 79.66 71.91 69.55 52.30 67.08 70.81 74.06 37.53 59.26 75.14 21.07 44.63 50.54 83.95 56.33 75.00 38.15 58.58 53.50 51.37 51.38 63.85 83.63 69.07 74.03 57.18 36.78 64.23 14.85 69.57 3.27 26.99 92.86 8.92 43.77 23.78 77.54 27.59
141792 144467 40724 77526 12904 28106 1716 7185 7435 5740 3597 6399 1191 23885 4693 2027 29463 734221 16724 2504 9442 69234 4406 26032 1613 109612 111133 23513 1106 818 462 7900 33801 1068 7488 5389 2936 4898 7534 232301 131361 7533 6978 1033
TB: tuberculosis; GA: general anaesthesia; STD: sexually transmitted disease; COPD: chronic obstructive pulmonary disease; MTP: medical termination of pregnancy
116
SECTION II
Jalna (Maharasthra)
117
SECTION II
Khammam (Andhra Pradesh)
118
SECTION II
Nadia (West Bengal)
119
SECTION II
Kozhikode (Kerala)
120
SECTION II
Udaipur (Rajasthan)
121
SECTION II
Ujjain (Madhya Pradesh)
122
SECTION II
Vaishali (Bihar)
123
SECTION II
Varanasi (Uttar Pradesh)
124
SECTION II
The not-for-profit sector in medical care
N INDIA, HEALTH SERVICES ARE PROVIDED BY THE PUBLIC AND PRIVATE SECTORS. The public sector provides health services through the Central and State Governments, municipal corporations and other local bodies. The private health sector consists of the 'not-for-profit' and the 'for-profit' organizations. Individual practitioners from various systems of medicine provide the bulk of medical care in the for-profit health sector. The not-for-profit sector is heterogeneous, with varying objectives, sizes and the areas they cater to. Their activities could be multifunctional and include welfare programmes such as health, education, nutrition, family planning, water supply and housing, agriculture-related development programmes, livelihood programmes, etc. The objective of this paper is to understand the nature and character of not-forprofit organizations in the delivery of curative health services in India. It seeks to gain insights into the spread of this sector within the country in the field of providing medical care, utilization of services, funding patterns and costs of care for the notfor-profit sector. In doing so, it seeks to define the role of these organizations in delivering curative health services and the way they can intervene to positively impact the health of the people.
Defining 'not-for-profit'
The not-for-profit sector is generally said to comprise non-governmental organizations (NGOs), the third sector, and the voluntary or charitable sector. There is no clear definition as to what precisely constitutes a not-for-profit organization. However, it is important for the purpose of setting internal government policy that at least within a country a workable definition is adhered to. In India, one of the criteria for a not-for-profit organization/NGO given in the Seventh Plan document is that the organization should have a legal entity. The Planning Commission considers societies, associations, trusts or companies registered under the Societies Registration Act, 1860; Indian Trust Act, 1882; the Charitable and Religious Trusts Act, 1920 or Section 25 of the Companies Act, 1956 as NGOs (Planning Commission 2002). Religious trusts and missionaries are usually governed by the Charitable and Religious Trusts Act, 1920. 'Charitable purpose' includes relief for the poor, education, medical relief and the advancement of any other object of general public utility but does not include a purpose that relates exclusively to religious teaching or worship. The Societies Registration Act, 1860 defines society as any seven or more persons associated for any literary, scientific or charitable purpose. Both trusts and societies are exempted from income tax. At present, almost every State has adapted its own Societies Act and Charitable Trust Act. For example, Maharashtra registers not-for-profit medical care providers under the Bombay Public Trust Act, 1950. Some States have retained the original Act. Public Trusts are constituted for the benefit of the public at large but the author of a Public Trust may restrict the benefit to a particular group or section of society, on the basis of caste, class, creed, sex, age, etc. Ideologically, the development potential of the not-for-profit sector originates from the currently dominant neo-liberal perception that State organizations are inefficient. It is commonly argued that non-profit organizations constitute the 'third sector' located between the State and the market. Organizations grouped in the third sector are bound by an appeal to voluntarism. According to the proponents of the 'third sector', the organizations in this sector share distinct characteristics: they possess an internal organizational structure, they are structurally separate from the Government,
MADHURIMA NUNDY
CENTRE OF SOCIAL MEDICINE AND COMMUNITY HEALTH, JAWAHARLAL NEHRU UNIVERSITY, NEW DELHI, INDIA E-MAIL: madhurima.nundy@gmail.com
125
SECTION II
and they do not generate profits that are distributed to members (Robinson and White 1997). During the 1980s, this sector grew worldwide in terms of size, scope, number and volume. The term NGO, which began to be used during this period, includes a kaleidoscopic collection of organizations differing in size, form, orientation, resources, target groups and ideological affinity (Valhans 1990).
Methodology
Since there are no comprehensive documentation and databases to assess the spread of all the non-governmental bodies working on issues relating to health, the focus of this paper has been narrowed down to those not-for-profit organizations that are providing medical care. The data for the present paper have been compiled from various sources. A questionnaire survey through mail was undertaken for the present study to get a brief overview of the nature of the not-for profit sector in health care. The objective of the survey was to gather insight into the nature and presence of not-for profit organizations providing curative services. The questionnaire was brief and responses were sought on the location of the organization, number of beds, number of inpatients and outpatients, nature of funding, dispensing of drugs and presence of medical personnel (doctors, nurses, paramedics). The sampling was purposive as there was no comprehensive database available of not-for-profit organizations providing curative services. The questionnaires were sent to organizations whose addresses were available. The Voluntary Health Association of India (VHAI) is one of the major national networks of more than 4000 NGOs spread across the country. It is an association of voluntary agencies working in the area of health and development. Questionnaires were sent to the 27 State Voluntary Health Associations and they were requested to provide names of the organizations in their network that provide medical care. Other than these, many organizations were directly contacted (community-based NGOs/faith-based NGOs) and questionnaires sent to them and also to those organizations/individuals who could give us any further leads in the form of names of not-for-profit health providers. In all, 173 letters were sent and 86 institutions providing services at the primary, secondary and tertiary levels responded to the questionnaire. Various Christian groups (Catholic Health Association of India, Catholic Bishops Conference of India and Christian Medical Association of India) were contacted as they form the largest network of health services in the not-for-profit sector. Secondary sources such as articles, books and various websites were accessed to gain more data on the spread of this sector in health care. An eight-district survey to map all health facilities/providers was conducted as part of the larger study. The eight-district health facility survey was conducted for the National Commission on Macroeconomics and Health. The survey mapped the entire universe of health facilities available in each of the districts. The districts were Jalna in Maharashtra, Kozhikode in Kerala, Khammam in Andhra Pradesh, Ujjain in Madhya Pradesh, Vaishali in Bihar, Nadia in West
Bengal, Varanasi in Uttar Pradesh and Udaipur in Rajasthan. Data were obtained on the not-for-profit sector in these districts. Data on the spread of not-for-profit health care services in the unorganized sector have been obtained from the 57th Round of the National Sample Survey (NSS). Data from all the above sources have been consolidated in this report and are attached as Annexure 1. For the utilization of medical services and expenditure patterns in charitable institutions, the NSS data for the 42nd and 52nd round were studied. Data on foreign funding of the not-for-profit sector were available from the Foreign Contribution Regulatory Act (FCRA) handbook.
Limitations of the study

The exact number of health institutions in the not-for-profit sector in India could not be consolidated through the questionnaire survey due to difficulty in collating data by contacting all organizations within a short period of time. Many organizations did not respond to the questionnaire in spite of several reminders. Since the sampling for the survey was purposive, more responses were received from some States such as Madhya Pradesh, Karnataka and Gujarat. Several faith-based organizations were directly contacted and therefore, more responses were received from them. Some organizations gave the names of institutions providing medical care but no details. Due to lack of time, the questionnaire could not be sent to all these institutions to obtain further details about their institutions. However, they have been integrated into the table on the spread of the not-for-profit sector as providers. The results from various sources reiterate the fact that a more comprehensive documentation of this sector is essential.
Role of the not-for-profit sector in India

In India, the role, activities and functions of the third sector, which originated outside the State structure, were previously performed by local governments and local voluntary efforts. The efforts and initiatives towards welfare and developmental activities came into prominence during the colonial period. The agents of these activities were called voluntary organizations. Until the mid-1960s, the not-for-profit health sector was hospital-based but later expanded to include community health in developmental projects. Financial issues needed to be considered; setting up institutions for medical care meant involvement of large amounts of funds and subsequently, the question of their sustainability. A number of NGOs, therefore, took up health education as part of their community programmes. Often, health was used by NGOs as the entry point to communities. Many professionals moved towards the NGO sector in response to the growing disillusionment with the public sector. The characteristics of an NGO also underwent a change when these professionals entered the field. The projects now called for more involvement from the community and aimed at making them more self-reliant than adopting a paternalistic attitude as in charity (Sundar 1994). Structural adjustment policies also resulted in restructuring the
126
SECTION II
provision of health services, resulting in support of the private for-profit and not-for-profit sectors (Baru 1998). The efforts of the not-for-profit sector in health care today covers a wide range of activities and can be classified broadly into: advocacy, awareness and education, research, and actual provisioning of services. Several NGOs in India work on varied issues such as livelihood and poverty alleviation, women's empowerment, health awareness and education, improving water supply and sanitation, etc. other than those providing medical care. These wider developmental issues addressed through social mobilization, a more holistic approach to improving the health status of the population, are known to impact individual, family and community health.
Response of the State to the role of voluntary organizations/NGOs in health care

The State has attempted to define the role of voluntary organizations/NGOs through the Five-Year Plans, national health policies and international commitments. After Independence, the Five-Year Plans were started to prioritize and allocate resources to developmental programmes. In the first two Plans, the emphasis was on the role of the State to provide welfare services. In addition, the Five-Year Plans have constantly recognized the role of the voluntary sector. The First Five-Year Plan stated that private efforts should be utilized for the promotion of social welfare. Voluntary organizations were recognized for their contribution to the tuberculosis and leprosy programmes from the First Plan itself. From the 1960s, the Government offered subsidies and grants-in-aid to various NGOs to assist the State in National Health Programmes such as the tuberculosis, leprosy and family planning programmes. From the Fifth Plan onwards, the Government encouraged NGOs to take over some health programmes. There were a number of initiatives at the governmental level to establish consultative groups of voluntary agencies in each State from the Sixth Plan onwards but they were not very successful. In the Seventh Plan, NGOs were given the freedom to plan their own schemes and follow the methodology they thought best to tackle social and economic problems. Individuals and NGOs working on rural development activities were appointed members of the governing body of the Council for the Advancement of People's Action and Rural Technology (CAPART). The Seventh Plan assigned an important role to voluntary agencies and sought their active participation in realizing the goals and objectives of the Plan, especially in the field of community participation and in the delivery of health services, as stated in the Indian Council for Social Science Research (ICSSR)/Indian Council of Medical Research (ICMR) report on Health for All (ICMR/ICSSR 1980). In the Eighth Five-Year Plan, it was proposed that grants-in-aid would be given to the NGO sector for experimental schemes. They were expected to help raise awareness of the small family norm, provide antenatal and postnatal care, etc. In the Ninth Plan, the Government, recognizing that NGOs were complementary in nature, handed over a number of primary health centres to NGOs. Each plan channelled a greater amount to NGOs for developmen-
tal programmes. The objective was to incorporate various elements of the approaches that NGOs adopted in government programmes and to utilize NGO efforts to implement government programmes. In the field of health care, the Government has used NGOs to train functionaries and has also given them the responsibility of delivering health services in their area. In the latter part of the 1980s, there was emphasis on NGOs playing a greater role, especially in delivering the national health programmes of leprosy, tuberculosis, blindness control, reproductive and child health (RCH) and later HIV/AIDS. India was a signatory to the Alma Ata Declaration on Primary Health Care. The NGO report presented at Alma Ata in 1978 defined the role of the third sector more clearly and indicated that NGOs could: Provide assistance to develop and/or strengthen local NGO capabilities and activities with particular attention to local community development groups Conduct reviews and assessment of existing health and developmental programmes and assist communities in the exercise of their own role in such reviews Place primary health care in the context of comprehensive human development Ensure that their existing programmes and new initiatives promoted full participation by individuals and communities in the planning, implementation and control of these programmes Expand training efforts as in training of health workers, supervisors, administrators, planners and other development workers Extend their efforts to develop locally sustainable and appropriate health technologies, and the use of resources with particular attention to energy, water, sanitation and medical care. Among the two National Health Policy (NHP) reports, the NHP of 1982 stated that: 'with a view to reducing government expenditure and fully utilising untapped resources, planned programmes may be devised, related to local requirements and potentials, to encourage the establishment of practice by non-governmental agencies establishing curative centres and by offering organised logistical, financial and technical support to voluntary agencies active in the health field' (Government of India [GOI] 1983). The NHP of 2002 suggests policy instruments for the implementation of public health programmes through individuals and institutions of civil society. The State will encourage the handing over of public health service outlets at any level for management by NGOs and other institutions of civil society. The Policy highlights the expected roles of different participating groups in the health sector. Further, it recognizes that, despite all that may be guaranteed by the Central Government for assisting public health programmes, public health services would actually need to be delivered by the State administration, NGOs and other institutions of civil society. The attainment of improved health levels would be significantly dependent on population stabilization, as also on complementary efforts from other areas of the social sectors-such as improved drinking water supply, basic sanitation, minimum nutrition, etc.-to minimize health risks for the population (GOI 2002).
127
SECTION II
Spread of not-for profit organizations in providing curative services

Not-for-profit organizations that are presently delivering curative services range from faith-based to community-based organizations working at the primary and secondary levels, and also a few at the tertiary level. In addition, big business groups have also established hospitals as trusts or societies, which qualify them for tax exemptions. Due to the heterogeneity and plurality of providers, the nature of services also varies across the providers in the not-for-profit sector; within this wide spectrum we have community-based organizations in rural areas that provide health services at the primary level. At the other end of the spectrum we have trust hospitals located mostly in urban centres providing secondary or tertiary care. The dominant system of medicine in most notfor-profit hospitals is allopathic. General health services are provided by almost all the institutions but very few provide only specialized and superspecialized services such as cardiology and neurology. From the 86 responses received through the questionnaire survey, 30% were dispensaries and health centres, and the rest were hospitals; 84% of the organizations provide general health services, 54% maternal health services and almost 30% paediatric services. Other special-
Box 1
Faith-based organizations include missionary organizations that have a large network of health facilities (hospitals and dispensaries). The two biggest networks are the Catholic Health Association of India and the Christian Medical Association of India. Two smaller networks are the Emmanuel Hospital Association and the Seventh Day Adventist Hospitals. Other faith-based trusts have also emerged as health care providers at the primary, secondary and tertiary levels. The main providers of this kind include the Ramakrishna Math and Mission, Mata Amritanandamayi Trust, Sathya Sai Central Trust, Sri Chaitanya Trust, Swaminarayan Sanstha, the Aga Khan Health Service network, Chinmaya Mission, etc. Several faith-based charitable hospitals were also established after Independence by local philanthropists. Several community-based NGOs provide services at the primary level. Due to their immense contribution and commitment to serve rural communities, some have gained credibility and recognition, and have been given the responsibility of running primary health centres so as to ensure better functioning. Examples are the Karuna Trust in Karnataka, SEWA-Rural in Gujarat, King Edward Memorial Hospital (KEM), Pune districts; Rural Unit for Health and Social Affairs (RUHSA), Kilvayattanan Kuppam Block in Tamil Nadu, Voluntary Health Services (VHS) in Tamil Nadu and the Kasturba Hospital in Sewagram, Maharashtra. Health services are integrated into preventive, promotive and curative services, and are just one aspect of their developmental and outreach activities. The most vital health personnel in these projects are the community health workers, and the delivery system follows a proper system of referral. These organizations cater to the rural and poor populations, and make health services accessible to them.
ized services include ENT, urology and dental, to very specialized services at the tertiary level such as cardiac surgery and neurosurgery. Of the institutions, 6% provided only specialized services at the tertiary level. Almost 50% of the institutions provide more than three services. Therefore, we conclude that multispecialty services are provided by several of these institutions. Many charitable institutions venture into providing specialized services for communicable diseases at the primary level. For example, in the Catholic network, there are 165 institutions providing services only for leprosy and 62 institutions for HIV/AIDS and TB, other than their hospitals and dispensaries. They also provide curative and rehabilitative services through 188 facilities for the disabled and 416 health institutions for the aged. Various studies show that most not-for-profit health institutions are located in semi-urban/urban areas. Most of these organizations establish themselves in places where infrastructure is already present (Jesani et al. 1986; Baru 1993 and 1996). The questionnaire survey showed that 43% of the institutions were in rural areas, and the remaining in semiurban and urban areas. According to a study, the percentage of villages with any kind of NGO presence ranges from 1.4% in Uttar Pradesh to 34.4% in Maharashtra. For India as a whole, it is estimated that 10.6% of the villages have the presence of some type of NGO (Mahal et al. 2000) (Table 1). According to another estimate by the Independent Commission on Health in India (VHAI 1997), more than 7000 NGOs are working in the field of health care. The only official source that exists on the total number of hospitals and beds in the not-for-profit sector is the data from the Directory of Hospitals, published last in 1988. The num-
Table 1 Percentage of villages with non-governmental organizations (NGOs)

State Villages with NGOs (%)
Andhra Pradesh Bihar Gujarat Haryana Himachal Pradesh Karnataka Kerala Maharashtra Madhya Pradesh Orissa Punjab Rajasthan Tamil Nadu Uttar Pradesh West Bengal Aggregate
Source: Mahal et al. 2000
21.2 2.6 9.1 7.8 6.4 11.1 8.0 34.4 8.8 9.7 12.9 4.7 14.5 1.4 6.4 10.6
128
SECTION II
ber of hospitals in this sector was estimated to be 937 (10% of all hospitals) and the total number of beds 74,498, comprising 13% of all beds in India (GOI 1988). It showed that 17% of all the private hospitals were not-for-profit and 42% of all the private beds were in this sector (Table 2). Data available for the present study (the questionnaire survey on not-for-profit organizations alongside other existing data sources on the not-for-profit sector) show that 11% of all hospitals and dispensaries, and 18% of all beds are in the not-for profit sector (Table 2). They also show that 17% of all private hospitals and dispensaries are not-for-profit and 47% of all private beds are in this sector. This corroborates the data for 1987. The more developed States of Kerala, Tamil Nadu, Karnataka, Maharashtra, Andhra Pradesh and Gujarat account for 50% of such institutions with 64% of beds of a total of 132,907 beds (Annexure 1). Christian missionary organizations lead in providing health services. States with a larger number of these organizations are Tamil Nadu and Kerala, and account for almost 30% of the total institutions; 42% of the beds in this network are in Kerala alone. The eight-district facility survey conducted in 2004 shows that just 3% of all private providers (including sole proprietorships, corporate, partnerships and trusts) in these districts were not-for-profit (Annexure 1); 45% of these not-for-profit organizations were established between 1995 and 2004. The data on economic and operational characteristics of unorganized enterprises in the service sector besides consumer expenditure, and employment and unemployment, collected by the NSS were also analysed. It took into account own-account enterprises (OAEs) and establishments and provided a differentiation between those that are for-profit and those that are not-for-profit. An own-account enterprise is an undertaking run by household labour, usually without any hired worker employed on a 'fairly regular basis'. The data show that there are 17,233 health providers in the not-forprofit unorganized sector (comprising only 1.32% of all health providers in the unorganized sector). They further show that only 3% of all health establishments in the unorganized sector are not-for-profit. State-wise data show that there is a concentration of these not-for-profit providers in West Bengal, Maharashtra, Assam, Punjab, Uttar Pradesh, Uttaranchal and Tamil Nadu (Annexure 1).
Utilization of the not-for-profit sector in medical care

It is important to ascertain what constitutes the definition of a 'charitable institution' by the NSSO. However, due to the absence of a definition in any of the NSS reports it was difficult to reach any definite conclusion. The perception of a charitable institution is one that provides free care or provides care at an extremely nominal rate. But if, by definition, we say that it includes all those institutions that are registered as charitable trusts, then it would include a wide range of hospitals that may otherwise be perceived as for-profit private hospitals. Hence the data on utilization rates might be an underestimate. Some data from the 42nd and 52nd round of the NSSO on the utilization of health services in charitable institutions elicits the following: Both the 42nd and 52nd rounds of the NSSO showed low utilization of charitable institutions for outpatient care (Table 3). This might be because the spread of charitable institutions, especially in rural areas, is negligible compared to the availability of private doctors. For outpatient care in rural areas, most people access a private practitioner. For the lowest income group, the primary health centre (PHC) is the next frequently accessed source of treatment after private providers. For outpatient services, the private doctor seems most accessible even in urban areas (NSSO 1998). For inpatient care (Table 3), the 52nd round of NSS shows that there is an increase in the utilization of charitable institutions in rural and urban areas; but this does not mean that free care has increased. Only 2.8% of the rural population receive free inpatient care in the private sector (one assumes that this is due to some cross subsidization in the not-forprofit sector). For inpatient care in urban areas, only 3.5% receive free care in the private sector. For every 1000 inpatients treated in charitable institutions in urban and rural areas, more patients are from the higher income groups. Most of the poor seem to access public facilities when it comes to hospitalization (NSSO 1998).
Financing of not-for-profit organizations

Not-for-profit organizations draw on a wide variety of sources for finance. These include donations, government funding
Table 2 Share and growth of voluntary and not-for-profit hospitals and beds
1987 Hospitals (%) Beds (%) Hospitals and dispensaries 2004 (%) Beds (%)
Government and local Voluntary/not-for-profit Private Total
4180 935 4488 9603
43 10 47 100
395,062 74,498 1,04,018 5,73,578
69 13 18 100
14160 3979 19419 37928
37 11 52 100
4,36,208 1,32,907 1,47,093 7,16,208
61 18 21 100
Source: Directory of Hospitals in India, 1988, Misra et al. 2003 Data for 2004: Survey data, data from missionary organizations; secondary sources: CBHI, GOI 2002
129
SECTION II
Table 3 Percentage distribution for outpatient/inpatient care by source of treatment

Outpatient (%) Source of treatment 1986-87 Rural 1995-96 1986-87 Urban 1995-96 Rural 1986-87 1995-96 Inpatient (%) Urban 1986-87 1995-96
Share of the public sector Private hospitals Nursing homes Charitable institutions Private doctors Others Share of the private sector Total
Source: Sen et al. 2002.
25.6 15.2 0.8 0.4 53.0 5.2 74.5 100.1
19 12 3.0 0.0 55.0 10.0 80.0 99
27.2 16.2 1.2 0.8 51.8 2.9 72.9 100
19 16 2.0 1.0 55.0 7.0 81.0 100
59.7 32 4.9 1.7 1.7 40.3 100
45.2 41.9 8.0 4.0 0.8 54.7 99.9
60.3 29.6 7.0 1.9 1.2 39.7 100
43.1 41 11.1 4.2 0.6 56.9 100
as grants-in-aid, funding from foreign donors, corporate funding, and user fees. Only 65 of the 86 institutions that filled the questionnaire in the survey responded to the query on the source of income. Most of the organizations have more than one source of funding; 17 organizations get some funds from the Government, 21 receive foreign charity, 29 receive private donations and 44 stated user fees as a major source. Along with private and foreign donations, most hospitals charge user fees.
Rs 542 crore (11.1%) was for the purpose of health and family welfare. FCRA data also show that the States of Delhi, Tamil Nadu, Andhra Pradesh, Karnataka, Kerala and Maharashtra received 74% of the total funds for the year 19992000 (Table 4). Of the 13,983 NGOs that received funds, 60% were located in these States. Among the top donors, the majority were Christian or Church-based organizations. The US, Germany and the UK were the largest fund givers (Account Aid 2002).
Foreign funding
Patterns of foreign funding have shifted. During the 1970s, the single largest funding for health came directly from the US Government constituting 57% of the entire funds, followed by UNICEF with 15.6%, the World Bank with 10.7% and the WHO with 6.3%. The European bilateral government funded 3.2% and American foundations 1.9%. These funds were channelled through the Government for a variety of disease control programmes and family planning. A few of the active bilateral agencies were United States Agency for International Development (USAID), Danish International Development Agency (DANIDA), Swedish International Development Agency (SIDA) and Norwegian Agency for Development Cooperation (NORAD). During the 1990s, there was an increase in funds from multilateral agencies and a decrease in funds from bilateral agencies. In fact, with the World Bank's growing presence, bilateral agencies have become secondary sources and have linked their funds to the Bank's programmes (Baru 1998). Private individuals, business organizations and NGOs can receive funds under the Foreign Contribution Regulation Act (FCRA), 1976. Every dollar that comes into India as a grant does not automatically become foreign. There are numerous exceptions to this-the FCRA does not include aid received from the UN organizations and the World Bank. However, it is one of the formal sources that gives an estimate of the kinds of funds that flow through it. Foreign funds for the year 2001-02 through the FCRA were Rs 4872 crore, of which
Table 4 State-wise distribution of foreign contribution (1999-2000) (Rupees in crore)

State 1999-2000 (Rs in crore) Number of organizations receiving foreign funds
Andhra Pradesh Assam Bihar Delhi Gujarat Himachal Pradesh Jammu and Kashmir Karnataka Kerala Madhya Pradesh Maharashtra Orissa Punjab Rajasthan Tamil Nadu Uttar Pradesh West Bengal Others* Total
536.99 24.35 104.75 636.11 126.95 68.20 13.64 411.34 361.70 84.57 350.23 111.65 35.22 37.26 572.51 128.10 233.99 87.08 3924.64
1616 163 723 735 551 77 34 1154 1483 432 1198 714 70 220 2143 802 1212 656 13983
Source: Accountable handbook: FCRA, 2002 * Includes north-eastern States and other Union Territories
130
SECTION II
Role of not-for profit organizations in National Health Programmes

Grants-in-aid released by the Central Government
The Central Government releases grants-in-aid to NGOs across the States for national programmes-tuberculosis, Reproductive and Child Health (RCH) Programme, leprosy, blindness and HIV/AIDS. Under the National Blindness Control Programme, approximately Rs 22 crore was dispersed to NGOs for 2001-02. Under RCH, Rs 18 crore was released for the year 2003-04. Comparatively, grants to NGOs for the Leprosy Programme and Revised National Tuberculosis Control Programme (RNTCP) were low at Rs 1 crore and Rs 61 lakh, respectively (Table 5). Some funds from foreign funding agencies are also channelled through the Government. For example, for HIV/AIDS, 40% of the funds allotted under the National AIDS Control Programme were from bilateral agencies (NACO 2004).
Table 5 State-wise grants-in-aid to NGOs under the National Health Programmes (Rs in lakh)
Leprosy (2002-03) RNTCP (grants and honorarium) (2001-02) Blindness Control (2001-02) RCH (2003-04)
tion of services by the poor, thereby undermining their principal objectives of serving the poor. An interesting insight in the not-for-profit financing system is cross-subsidization. It is observed that in some secondary and tertiary hospitals where services are provided at a cost, not-for-profit organizations tend to provide free services or charge lower rates for inpatient care from the poor and cross-subsidize them by charging higher rates from those who can afford it. The results of the survey show that only 7 of the 86 providers who had responded supplied drugs free of cost; 16 responses stated that drugs were supplied free to the poor. In some cases, some essential outpatient drugs are provided free of cost while the rest are either dispensed at a cost or need to be purchased on prescription. NGOs have to buy medicines like any other commercial establishment directly from the market, which restricts access to drugs for a large section. To make drugs easily accessible to people by dispensing them for free or at a subsidized cost, Locost, a not-for-profit pharmaceutical company in Baroda, manufactures essential drugs at a low cost and distributes them only to NGOs which provide curative services to the needy. They have a distribution centre in Karnataka for the NGOs in south India (Locost 2004).
Costs of care at not-for-profit facilities

According to the 52nd round of the NSS (1998) (Table 6), the average total expenditure per hospitalization in a charitable institution is less than in for-profit hospitals but higher than in public sector hospitals. Further, the 52nd round of the NSS also brings out the fact that there was a decrease in access to free care from 19% to 10% between 1986 and 1996. This reflects the fact that user fees have been introduced in several public and not-for-profit health institutions during this period. Studies have attempted to analyse whether not-for-profit health programmes are more expensive than government or the for-profit private sector. It is noted that the kind of aggregate figures presented for hospitals can be misleading. For instance, variation in the size and quality of services in hospitals is not considered here. A study among the not-for profit hospitals, government hospitals and private hospitals
Well-performing States Kerala 9.25 Tamil Nadu Andhra Pradesh 14.6 Maharashtra 10.96 Karnataka 4.09 Moderate-performing States Gujarat West Bengal 28.13 Punjab Haryana Poor-performing States Rajasthan Orissa Madhya Pradesh Uttar Pradesh 12.65 Assam 1.24 Bihar India 99.39
3.57 2.5 3.73 7.75 3.39 10.49 18.8 21.94 0.57 1.8 0.52 0.8 0.02 2.76 61.8
59.03 617.81 379.4 58.53 148.57 50.88 85.57 20.41 109.89 44.29 255.3 165.82 9.38 12.75 2151.88
104.34 55.38 135.52 145.95 30 23.64 159.84 13 47.3 5 199.29 208.12 179.14 36.64 180.73 1807.17
Table 6 Average total expenditure per hospitalization (in Rs) by source of treatment
Type of hospital Rural Urban
RCH: Reproductive and Child Health; RNTCP: Revised National Tuberculosis Control Programme Source: Annual reports, MoHFW, various years
User fees
As the responses to the questionnaire survey showed, user fees is one major source of revenue, particularly due to the gradual decrease in foreign funding. However, there has been concern that the levy of user fees may be affecting the utiliza-
Public sector hospital Private hospital Nursing home Charitable institution Others Private sector hospital All hospitals
Source: 52nd round, NSSO 1998
2080 4394 4185 3808 3015 4300 3202
2195 5524 5749 3078 1630 5344 3921
131
SECTION II
reported that, in general, the cost per hospital bed per day in the not-for-profit sector was very low compared to others. But this study included only community-based organizations in rural areas (Berman and Dave 1994). These not-for-profit institutions are able to achieve substantial 'cost savings' due to the following reasons: They give low wages. In some cases, they use the services of honorary physicians and these lead to a lower wage bill and lower overall cost. For example, the results of the survey show that 73% of the not-for-profit organizations had part-time medical personnel. Another example of cost control is observed where community-based organizations serving the poor purchase generic and essential drugs manufactured and distributed at a low cost by a not-for-profit organization (Locost 2004). Such community-based organizations focus on rational care by emphasizing on referrals and lessening the role of unnecessary technological interventions. A study of four private hospitals was conducted in Delhi for the National Commission on Macroeconomics and Health to understand the financing patterns. A study of two charitable hospitals and one for-profit hospital in Delhi (2004) showed that the rates of some common procedures are lower in the charitable hospitals when compared to for-profit hospitals (Table 7).
Table 7 Rates of some common procedures (in Rs)*

Procedure Trust I Trust II Private hospitals
Caesarean section Cataract removal Appendicectomy
5850-7475 4500-6500 2970-4290
5750-11,500 6500-20,000 4750-9500
7700-25,800 8400-28,000 6500-21,500
*The rates vary according to the graded bed charges. The table gives the lower and upper limits.
Recommendations and Conclusion

The objective of this study was to examine the characteristics, structure and spread of the not-for-profit sector in delivering medical care in India. The study shows that it is not possible to put the not-for-profit sector into one typology because of its heterogeneity in terms of organizational structure, pattern of funding, ownership, nature of services and its changing character. It is also scattered and disorganized. The not-for-profit sector has its own constraints and limitations. Even if one wants to establish a charitable institution, for most of these organizations, especially those functioning in rural areas, the question of sustainability is central
to their existence. In order to achieve appreciable and sustainable results, NGOs have to make long-term commitments to the community. They frequently face difficulties such as shortage of trained staff, high turnover of middlelevel workers, and dependency on donor agencies. For example, funding from foreign churches of the Christian network has reduced. User fees have therefore been introduced to take care of recurrent costs. It is difficult to study each and every not-for-profit provider to assess the kind of services or cross-subsidies they provide to the poor. It is evident that in the wake of increased privatization and corporatization of health services, not-for-profit institutions have also faced demands and competition to improve their services by introducing technology and specialized services. Numerous trust hospitals have become more commercial in their operations, hence altering their character from a charitable institution to a private for-profit/corporate image. There needs to be greater transparency to see if they are adhering to the conditions of cross-subsidizing prescribed by the law, such as 20% free admissions and free outpatient services for the poor. Curative services are just one aspect of the health services, which also include preventive, promotive and rehabilitative services. A study of community-based NGOs in West Bengal showed that simply health education and awareness do not improve the health status of people. In the absence of a proper referral service system, awareness and health education have little impact (Sarkar 2003). Several NGOs have made efforts to work and coordinate with the State Government, and have been successfully providing and managing primary health services. The experiences of these NGOs show that health services need to integrate preventive, promotive, curative and rehabilitative services and the issue of health has to be addressed as a part of the broader developmental goals with the involvement of the communities. A 1990 World Bank study showed that limited numbers of NGOs are involved in health and family welfare in rural areas and they mostly had weak financial management and technical capacity (Misra et al. 2003. The macro picture shows that the not-for-profit sector is not present universally and, therefore, cannot be seen as taking over the responsibilities of the public sector. As the National Health Policy 2002 calls for expanding the coverage of services and strengthening primary-level health services through the third sector, the challenge is for the Government to intervene effectively and formulate strategies to assist those community-based NGOs/notfor-profit organizations committed to working in backward areas by providing adequate support and engaging them in not just implementing and managing National Health Programmes but in various health and developmental activities in the context of comprehensive primary health care.
132
SECTION II
References
Account Aid. Accountable Handbook: FCRA. New Delhi: Account Aid India; 2002. Berman P and Dave P. Experiences in paying for health care in India's voluntary sector. In: Pachauri S (ed). Reaching India's poor: Non-governmental approaches to community health. New Delhi: Sage Publications; 1994. Baru R. Missionaries in health care. Economic and Political Weekly 1999. Available from URL: http://www.epw.org.in/showArticles.php?root=1999&leaf= 02&filename=110&filetype=html. Baru R. Inter-regional variations in health services in Andhra Pradesh. Economic and Political Weekly 1993;28:963-7. Baru R. Private and voluntary health services: An analysis of inter-regional variations. Report submitted to UNDP. 1996. Baru R. Structural adjustment and health: Changing role of NGOs. Paper presented at International Seminar on Global Governance and Social Policy, Baltic Sea Centre: Kellokeski, Finland 1998. Baru R. Privatisation and corporatisation. Seminar 2000. Available from URL: http://www.india-seminar.com/semframe.htm. Catholic Bishops Conference of India (CBCI), Directory of catholic health facilities in India. New Delhi: CBCI Commission for Health Care; 2003. Duggal R. Do charitable hospitals deserve tax benefits? Express Healthcare Management 16-30 September 2003. Government of India. Five-Year Plans (First to Tenth Plan). New Delhi: Planning Commission of India. 1951-2002. Government of India. National Health Policy. New Delhi: Ministry of Health and Family Welfare; 1983 Government of India. Directory of Hospitals. New Delhi: GOI; 1988. Government of India. Health Information of India. New Delhi: Central Bureau of Health Intelligence, Ministry of Health and Family Welfare; 2002. Government of India. National Health Policy. New Delhi: Ministry of Health and Family Welfare; 2002. Government of India. National Sample Survey Organization (NSSO). Fifty-second round on morbidity and treatment of ailments, 1995-1996; 1998. Government of India. NSSO. Fifty-seventh round on unorganised sector, 2000-2001; 2002. ICSSR/ICMR. Health for all: An alternative strategy. Report of a Study Group. New Delhi: ICSSR 1980. Jesani A, Duggal R, Gupte M. NGOs in rural health care; Bombay: FRCH; 1986. Locost. Impoverishing the poor: Pharmaceuticals and drug pricing in India; Vadodra: Locost; 2004. Mahal A, Srivastava V, Sanan D. Decentralisation and its impact on public service provision in the health and education sectors: The case of India. In: Dethier J (ed). Governance, decentralisation and reform in China, India and Russia. London: Kluwer Academic Publishers; 2000. Misra, R, Rao S, Chatterjee R. India Health Report Delhi: Oxford University Press; 2003. Pachauri, S (ed). Reaching India's poor: Non-governmental approaches to community health. New Delhi: Sage Publications; 1994. Robinson M, White G. The role of civic organisations in the provision of social services. New York: United Nations University, World Institute for Development Economics Research; 1997. Sarkar AK. Non-governmental organisations in health care: A study of West Bengal. Unpublished PhD Thesis. JNU, New Delhi: Centre of Social Medicine and Community Health; 2003. Sen G. Iyer A, George A. Structural reforms and health equity: A comparison of NSS Surveys, 1986-87 and 1995-96 in Economic and Political Weekly; 2002; 37:1342-52. Sundar P. NGO experience in health: An overview. In: Pachauri S (ed.) Reaching India's poor: Non-governmental approaches to community health. New Delhi: Sage Publications; 1994. Valhans M. The new popularity of NGOs. Development and Corporation 1990; 3:20-2. Voluntary Health Association of India (VHAI). Report of the Independent Commission on Health in India. New Delhi: VHAI; 1997.
133
SECTION II
Annexure 1 Spread of not-for-profit organizations derived from various sources

States Facilities by missionaries Hospitals Dispensaries No. of Others Other Beds facilities 1 Christian missionaries and others Total beds Total Own-account Establishments facilities enterprises (OAE) 2 beds facilities in these 57th round, NSSO Health facility survey Total No. of notfor-profit facilities 3 beds No. of
Well-performing States Kerala 257 Tamil Nadu 100 Andhra Pradesh 93 Maharashtra 36 Karnataka 53 Moderate-performing States Gujarat 10 West Bengal 9 Punjab 6 Haryana 2 Poor-performing States Rajasthan 23 Orissa 9 Madhya Pradesh 26 Uttaranchal 5 Uttar Pradesh 41 Assam 10 Bihar 15 Others* 66 India 764
288 35,832 361 6598 177 7159 152 3878 168 5691 64 114 27 16 1503 2465 511 122
28 14 15 24 14 87 4 -
2413 38,245 8790 15,388 1327 8486 2712 6590 1992 7683 7856 824 9759 3289 511 122
573 475 285 212 235 161 127 33 18 84 128 178 22 207 107 138 683 3979
17% of all private hospitals/ dispensaries
69 132 53 1595 384 39 1794 379 446 45 57 66 0 1013 1861 120 1036 9089
0.8% of all OAEs providing health (n= 1,074,212)
450 879 150 648 28 224 654 1399 530
519 1011 203 2243 412 263 2448 1778 976
27 10 12 41 34 14 37 4 179
3% of all
1451 83 474 274 813 1147 75 18% of all
29 993 118 2420 130 1324 12 43 121 4073 93 939 114 2311 589 7712 2575 83,598
32 264 1257 1 147 2567 22 1596 2935 5 181 224 45 1809 5882 4 195 1134 9 420 2731 60 2031 9743 332 32,557 132,907
47% of all beds in private facilities
469 514 55 112 173 239 1,162 1162 551 1564 122 1983 130 250 520 1556 8144 17,233
3% of all 1.32% of all
private beds in the health private facilities
health unorganized establishments (n= 230303)
health providers/facilities services
(n= 2,76,000) (n= 23398)

* Includes north-eastern States, Union Territories and five States
(n= 5788) (n= 24,241)
Sources: 1. Data from the Directory of institutions, Catholic Bishops Conference of India 2003; Baru R. Missionaries in Charity 1996; Not-for-profit sector questionnaire survey; Directory of Hospitals in Delhi 2002; CBHI 2002 and various websites. 2.Data obtained from CD-ROM of 57th round of unorganized services NSSO, 2002 (Note: OAE: own-account enterprises are run by the household without any hired worker, n = 1,074,212 Establishments that have at least one hired worker n = 230,303) 3. Eight-district health facility survey conducted by NCMH (2004) in Kozhikode (Kerala), Khammam (Andhra Pradesh), Jalna (Maharashtra), Nadia (West Bengal), Udaipur (Rajasthan), Ujjain (Madhya Pradesh), Varanasi (Uttar Pradesh), Vaishali (Bihar)
134
SECTION II
Peoples Partnership for Health Towards a Healthy Public in India
Community Health Needs
LALIT M NATH
FORMER DEAN, AIIMS E-MAIL: lalitnath@vsnl.com
HE AIM OF THE HEALTH SYSTEM IS TO HELP EVERY MEMBER OF THE COMMUNITY to be and to remain as healthy as possible given the resources available. It has become increasingly clear that this can not be done through a paternalistic approach of delivering health to the people; the people must be partners in achieving health, the governments role being that of facilitator and catalyst. Active participation by the people should be an essential element of all (public sector) health systems and in this paper it is assumed that this component will be taken as a first step. All other aspects discussed can only achieve optimal efficiency along side an active partnership with the community. To achieve this aim of a healthy population, the community has to adopt a healthy life style, live in a life sustaining environment (especially quantitatively and qualitatively adequate water and effective sanitation), have access to preventive care and establish a system for the early detection and prompt response to potential and actual disease outbreaks. In addition, and not instead, the community requires access to curative care at an appropriate level. Though curative care is needed as there will always be breakdowns in health including injuries and disease, it is most important that the system plan to have a major focus on health promotion, preventive measures and water and sanitation. Health decision makers must resist the temptation to be swayed by the glamour of high-technology tertiary curative initiatives to neglecting health related measures in favour of disease linked interventions. Curative care does not lead to the health of the community; it merely helps to manage the breakdowns in health. To maintain health the system has to be geared to provide preventive and promotive health and subsequently any deviations from health can be tackled by the curative system at the individual level and by public health measures at the community level. In all interventions the role of the people themselves must remain at the fore. Effective health care, whether delivered in the community or in the most advanced tertiary care institution ultimately depends upon the active participation of the recipients of the care. Unless the people are actively involved, they and the community can not be healthy. Another look is suggested to re-align Indias health priorities. Today only 0.9% of the GDP is allocated to health, a proportion far less than many less developed and poorer countries than India. To make the situation even more stark, it must be remembered that of even this grossly inadequate allocation, a large proportion is spent on urban tertiary care and only a minor share is allocated to health care for the rural areas. Of this latter proportion the majority is dedicated to curative care leaving preventive and promotive interventions at the tail end of the budget allocation. The increasing role of the private corporate sector has further accentuated the disparity; the disparity between rich and poor, urban versus the rural, tertiary care versus primary care and the disparity between curative versus preventive / promotive interventions. The opening of many large deluxe hospitals has resulted in the affluent having many additional options for health care but the poor in slums and the rural community seem to have been forgotten. It is worth noting that many if not most of these deluxe institutions have been directly or indirectly heavily subsidised from the public exchequer. Improving the health status of the community is desirable not only because healths itself is an objective worth striving for, but also for very sound economic and social
135
SECTION II
reasons. A healthy population is not only a happy and contented population with a good quality of life, but it is a more productive population. Keeping people healthy also is cost effective because sickness is very expensive not only in terms of direct costs but also because of loss of productivity. Therefore not only the individual and family but the State also incurs expenditure in managing illness. Preventing disease and promoting health pays many dividends. Various evaluations of the state of health in India have confirmed that even though there have been very significant improvements in many health parameters, the state of health of large segments of the population are very far from satisfactory. India can be proud of the fact that the expectation of life at birth has doubled since India became Independent. Other parameters that have shown marked improvement include the infant mortality rate, child mortality rate, maternal mortality etc. However these same parameters show that the benefits of improvement in health status are not uniformly available in all parts of the country, particularly to the poor. In health and in health care the divide between urban and rural, between rich and poor is very real and marked. In the 57 years since Independence the laudable and farsighted recommendations of the Sir Joseph Bhore committee have largely been forgotten and health care in India has become dangerously skewed towards tertiary level curative medicine. Even the Alma Ata Declaration that India was a signatory to, did not serve as a catalyst to bring about a midcourse correction in emphasis. While some communicable diseases have been eradicated, most notably small pox and Guinea Worm, others are still far too common and play an important role in the morbidity and mortality experience in the country. While the usual communicable diseases still contribute a great deal of morbidity and even mortality, especially for women and children, the position has actually got more severe with the onslaught of new and re-emerging diseases. NACO estimates that India already has 5.1 million HIV infected persons. The position of Malaria and other vector borne diseases such as dengue, filarial and Kala Azar are also alarming. India even today remains one of the most important holdouts in the battle to eradicate polio and has more Leprosy and Tuberculosis cases than any other country in the world. At the same time the augmented longevity of Indians has allowed increasing numbers of people to reach an age where non communicable diseases become an important cause of morbidity and even mortality. The demographic transition has resulted there being a larger cohort of older people and therefore more diseases associated with that age group. In absolute numbers there is an increase in the numbers of people with non-communicable diseases. The demographic change has acted in consonance with the ill-effects on health brought about by a changing life-style and pattern of nutrition. Obesity and a largely sedentary life is showing its affects; cigarette smoking, alcohol and drug use have become more acceptable and all these factors taken together have resulted in a manifold increase in the life-style associated diseases. The prevailing view appears to be that health is a part of
the largesse that can be doled out by the government. Unfortunately this opinion is not limited to health care decision makers but has now come to be accepted even by the community. The government has tried with out success to provide health to the people for many decades. An urgent need is for the health care decision makers to come to terms with the fact that health is not a commodity that can be doled out to a passive community. People are not currently treated as important partners in the general plan for improving health in the country. To be healthy the community has to take an active or even pro-active role and the governments major role should be as a partner in health care that plays a catalytic role and in addition helps by providing infra-structure and human resources. Ill-health is increasingly realised to be a manifestation of the interaction of a multiplicity of factors - biological, nutritional, socio-cultural, environmental. The multiplicity and complexity of causal factors explains why ill-health as not amenable to simple technical fixes that the doctor can administer within his ward or hospital. Medicines, injections, and even vaccines can not ensure health; to have a healthy population requires an active role by the people themselves with attention to a multitude of correctional interventions. Even in the case of bacterial or communicable disease, the presence of the bacteria is only one of the factors that are required before disease manifests. Without the specific disease causing organism a communicable disease does not occur; but the presence of the concerned organism does not necessarily result in disease. Tuberculosis is an obvious example. The mere presence of the tuberculosis bacteria in the body is not automatically followed by the disease tuberculosis. Tubercular disease results when a multitude of factors come into play and thus permit the disease to gain a foothold in the body. The prevention of the disease tuberculosis therefore offers several options. Obviously there can be no disease if the bacillus does not enter the body, but this difficult to attain at the community level. It is also possible to prevent the disease by increasing the bodys resistance to the organism either specifically (by immunisation) or in general (by a healthy life-style). We do not have specific vaccines for many diseases. HIV is one such example as there is no vaccine available as yet. But the disease can be totally prevented by adopting a life-style that does not expose one to risk of infection. Malaria that affects so many millions of persons all over the world is another example. We do not have a specific vaccine against malaria yet but it can be controlled by preventing the breeding of the mosquitoes that act as the vector for transmitting the plasmodium from an infected person to a susceptible person by doing away with breeding places, or by introducing larvivorous fish such as the Guppy or Gambusia. Adult mosquitoes can be killed by residual insecticide spraying on their resting places or even by space sprays. The disease can also be prevented by not allowing mosquitoes to bite susceptible people by using bed-nets or insect repellents. There are many options available to prevent malaria but in almost
136
SECTION II
all of them the active cooperation by the community itself can make the difference between a successful effort and an expensive exercise in futility. Even a widespread and efficient curative facility for the treatment of those suffering from malaria, will have little impact on the prevalence of Malaria in the community! These examples are there not only for communicable disease. With the demographic change now manifesting India is starting to have a serious problem of non-communicable diseases such as coronary artery disease. The solution is not to focus exclusively or even largely on developing intervention facilities for angioplasty and bye-pass surgery, but to make a serious effort to initiate life-style changes at an early age. Exercise and physical fitness must be encouraged, obesity treated as a disease and a diet conducive to sensible lipid levels adopted. Together with avoiding smoking these interventions will do more to reduce the burden of coronary artery heart disease for the people than the setting up expensive tertiary care facilities for cardiac care and bye-pass surgery! A Columbia University project estimated the consequences of Cardiovascular disease in developing countries. India was one of the five countries studied (A Race against Time - KS Reddy editor). They reported that in 2000 India was loosing 9,221,165 person years of life due to cardio-vascular diseases. This figure was estimated to rise to 17,937,070 by 2030! The economic consequences of loosing over nine million person years of life can be imagined. Another way of looking at the same data is by expressing the burden of disease in terms of DALYs or disability adjusted life years lost due to cardiovascular disease. The Global Burden of Disease study estimated that India lost 28.6 million DALYs due to CVD in 1990. Initiating preventive promotive interventions to reduce the burden of disease due to CVD is likely to pay much greater dividends than establishing only treatment facilities to cater to the consequence of heart disease. What we need to address community health needs is a greater emphasis on areas such as preventive and promotive health, nutrition, water and sanitation, a system of prompt detection of and response to disease outbreaks and factors predisposing to ill health. At he same time a healthy life-style including curbs on tobacco and drug use, judicious exercise will have to be promoted. These are all factors that fall under the rubric of Public Health measures. It is not contended that tertiary care hospitals are unnecessary but that preventive and promotive health care interventions properly implemented can not only improve health but prevent conditions that would need heroic interventions to save life. Preventing heart attacks is better than doing coronary bye-pass surgery. Because there will always be instances when prevention efforts do not succeed, tertiary care institutions will be needed to provide care for those persons who have developed severe disease. However it is the balance between curative care and preventive care that has to be determined. It is manifestly wrong to ignore the prevention of illhealth and focus ones energy almost entirely on institutions with expensive technology to deal with the long term conse-
quences of preventable morbidity. Hospitals are needed, however the system can not ignore the fact that all too often hospitals cater to the consequences arising from our failure to keep people healthy. The treatment of ill-health can not only begin in the health care facility and neither is the treatment completed within the boundary walls of the hospital. Unfortunately too many practitioners of curative care are both physically, and more important mentally, confined to the hospital walls. Disease starts in the community and its management and treatments also comes to fruition in the community. Hospitals are monuments to disease and not temples of health; every sick individual is a reminder of the failure of the system to provide health care to the people It is not suggested for a moment that the sick do not need hospitals and health care facilities. However until more effective measures are taken to actively promote health and prevent ill-health, we will continue to cater to our failures. Nor is it being suggested that curative care is unnecessary. Curative care is essential, but in too many cases it is essential only because we have failed to prevent sickness. Comprehensive health care is a spectrum of activities that ranges from providing usable and understandable information on promoting health all the way to the most technologically advanced intervention at centres of excellence. Establishing a facility to do cardiac surgery and heart transplants is necessary and good, but is it not more important to set in place a system that will prevent persons from deteriorating to the extant that they need heroic interventions? Both types of investments are needed but a balance must be struck and it is manifestly both morally wrong and cost-inefficient to build curative infra-structure without putting in an even greater effort to prevent disease or to stop its progression as soon as possible. There have recently been some statements from the highest level of decision makers promising an increased allocation for health. That is good and a step in the right direction. Care must be taken however to ensure that the increased budget is not spent in the same proportion as the existing outlays and thus further accentuate the gap between tertiary curative infrastructure and Public Health services for urban slums and rural communitys. It is essential that urgent action is taken to correct the gross discrepancy between the allocations for curative versus that for preventive/promotive care. There are powerful lobbies that suggest that the governments role is limited to primary care and public health and that all curative care, certainly tertiary level care, should be left to the private sector. Perhaps the advocates of such an arrangement forget that some 37% of our population live at or below the poverty line. Private sector health care would be completely out of their reach. This tenet also completely ignores the right to health care ensured by our constitution. In a country like India it will be unthinkable to invest in primary care for the poor and disadvantaged without making tertiary care available to those that need such interventions. Morally a focus on primary care facilities for the rural and slum areas can only be justified if a system is developed
137
SECTION II
alongside to provide higher levels of care to those that need it. Of course the other side of the coin is that without an efficient and functioning primary care set-up, India will not be able to afford the cost of tertiary care. Tertiary care must be available to those whose condition needs specialised care, rich or poor. In such cases tertiary level care can be provided to the poor either at heavily subsidised public sector supported institutions or at private sector hospitals through a mechanism of State support or health insurance. But a health system that does not provide tertiary level care to those poor and needy persons who need it and yet caters to the affluent is wrong and should be unthinkable. The provision by design of one level of care for the poor and another for the rich urban dweller is something that can not be condoned in any country. The gap is not so much between the urban and rural but between the affluent privileged persons and the poor. The urban poor, the slum dwellers, are perhaps most discriminated against group in India; practically no health infra-structure is available for the slum dweller, especially the millions living in unrecognised slums. These people build our cities, clean them and provide the domestic service that ensures our comfort. Yet they have virtually no access to the civic services most urban people take as a matter of right. This is an issue demanding urgent action. Focusing on providing preventive and promotive care does not require a massive addition to the budget, but it does require funds. When a model Primary Health Centre was evaluated, Anand et al (Natl. Medical J India, 1995) reported the division of costs of running a properly functioning Primary Health Centre with a well balanced emphasis on preventive and promotive care. They have reported that curative care cost 32%, communicable disease control cost 17%, child care 17%, maternal care 11% and family welfare 10% of the total cost.
Fig 1 Northern Thailand
Condom use cannot take all the credit for the change in HIV prevalence in army recruits, though behaviour change played a major role in the reduction. The figure below is a composite representation of the interventions in northern Thailand. While it is clear the HIV prevalence fell after 1993 and that condom use increased, the concomitant reduction in the visits to sex workers clearly points to a change in behaviour in the community. Condom use probably played a role, but reduced access to sex workers was obviously a major factor.
Fig 2 Uganda
Preventive and promotive Care: Need, Efficacy and Cost

There is ample evidence that many if not most health problems of concern to the community are more amenable to management that includes community level interventions. Even that most intractable of public health problems HIV/AIDS, responds to action by the people rather than due to technical interventions. There are just a few countries that have rolled back or even arrested the progression of HIV/AIDS in the community. Thailand, Cambodia and Uganda are examples of programmes that have showed results. The example of Uganda is particularly relevant to India. It is important to note the analysis of two of the well-known success stories in HIV. Both Thailand and Uganda have demonstrated a reduction in HIV prevalence in the general population. The 100% condom programme in Thailand was more than condom promotion; education was an integral component. The Thai data showed a remarkable fall in HIV prevalence in young army recruits. However this fall in prevalence was accompanied not only by an increase in condom use but, most importantly by a reduction in visits to sex workers.
The data from Uganda is even more striking. There is a marked and consistent fall in HIV prevalence in pregnant women starting from a peak in 1992. Similarly detailed analysis of the data from Uganda has showed that the major fall in HIV prevalence occurred before condoms became available to the programme and before they were advocated nationally for protection against HIV infection. The change appeared to have come about largely because of behaviour change. Rates of partner exchange came down significantly and
138
SECTION II
more young persons were abstinent for longer. A matter of HIV/AIDS is of course a special case as there is currently no great interest is the fact that condom use became a part of vaccine and no cure. But it highlights the importance of prethe national programme only in 1996, long after the prevaventive and promotive care. The HIV epidemic can be conlence had started to fall. Ugandas success is attributed to an trolled, Uganda has shown us how. The key is making the cominitiative by the President who decided to control the rapidly munity partners in the venture and behaviour change through increasing prevalence of HIV/AIDS by starting a peoples movepreventive and promotive interventions. ment to reduce casual sex. To take another example, diarrhoea, acute respiratory infecIncidentally a sharp contrast is provided by a bordering countions (ARI) and meningitis are three of the most common try - Botswana - that started at the same level but followed causes of morbidity and mortality in early childhood. Twoadvice to focus on condom use. The HIV prevalence in Botswana thirds of all early childhood deaths are attributable to these has continued to rise until it is one of the highest levels in three conditions. The health care system manages these conworld, and the budget has remained comparable! ditions in the community by treatment at the primary, secBoth these examples have clearly demonstrated the great ondary and tertiary care institutions both in the public and role of promotive interventions. The lives saved and the benthe private sectors. A recent study by Anand et al from the efits to the economy of the country are immense. Of course AIIMS estimated the cost of treating these conditions. It will in the example of HIV/AIDS the benefits are even starker be seen from the Table1 given below that considerable expenbecause there is no cure for HIV/AIDS. In a rough estimate diture is involved, especially if one keeps the frequency of made In 1999 the cost and possible consequences of HIV/AIDS these episodes in mind. in India was estimated The following paragraph is taken from chapter 2 of the book Table 1 edited by Peter Godwin. Cost of inpatient treatment of diarrhoea, acute respiratory Estimates about the average cost of one infection (ARI) and meningitis in various settings bed day vary naturally from hospital to hospital. In the Government sector, one bed Level of care Private sector Public sector day at the All India Institute of Medical Sci- Disease 1 Rs (Mean; 95% CI) Rs (mean; 95% CI) ences costs Re600 per day . Recently the medicine costs of one hospital bed per Secondary 5672 (4436-6908) 1315 (1115-1514) day was calculated to be Re 250 in a gov- Diarrohea 2 Tertiary 3155 (2503-3807) 8580 (5918-11252) ernment tertiary care hospital in Bombay . 8261 (5886-10636) 2229 (1961-2497) This comes to Rupees 160 million annu- Acute respiratory infections Secondary Tertiary 4506 (3489-5522) 7598 (6143-9053) ally. In a district hospital on the other hand Secondary 7428 (3976-10881) 1842 (460-3203) the cost of one bed for one day comes to Meningitis Tertiary 6991 (3916-10067) 17844 (15407-20281) only Re2003. The costs in the private sector similarly vary greatly but they are in These data include the identification and measurement of direct medical costs, valuation of hospital costs, out-of-pocket expenditure, productivity losses (Anand et al.) general many times more expensive than a similar bed / facility in the government sector Even if we take a figure at the lower end of the range, It also must be realised that it is not the State alone that the 57 million bed days would cost Rupees 11400 million incurs this expenditure; in India it is now estimated that on It is worth noting that these estimates of the cost of care an average the patient bears 83% of the cost of health care. were made using an estimated 1.75 million infected and did The sad thing is that all three conditions can largely be prenot include the cost of anti-retrovirals. Today India estivented by action at the community level by a functional mates that we have 5.1 million persons living with the virus, Public Health system coupled with active community action the cost of hospital beds has gone up and India is commitfor health. Those cases of diarrhoea and ARI that do occur ted to the provision of anti-retroviral drugs. can be treated effectively at the early stages in the commuIn a paper published in Health Policy (47 (1999) 195-205) nity setting itself. we estimated that the cost of HIV to the nation would range Tobacco provides another example. It has been estimated from 6.73 to 59.19 billion Rupees annually. It is noteworthy that the cost of tobacco related disease in India in 1999 that the upper estimate was made on an assumption of 4.5 (Rath and Chowdhury, quoted in Tobacco Control in India million infections in India. Today NACO estimates 5.1 milp135) was estimated to be Rs. 277.611 billion (Table 2). The lion infections and the cost of treatment has increased mancost in 2002-2003 the cost was estimated to have risen to ifold with the availability of anti-retroviral drugs. The sound Rs. 308.33 billion, an increment of 11% in two years. The direct economic rationale for a focus on preventive and promotive costs of caring for patients with Coronary Artery Disease and interventions rather than on exclusively building more and Chronic Obstructive Lung Disease gives an idea of what these more care facilities needs no elaboration. conditions mean to the community.
1. Choubey PC. Add. Professor of Hospital Administration, All India Institute of Medical Sciences, New Delhi. 1997. Personal Communication. 2. Salunke, SR., 1997Director Health Services, Maharashtra. Personal communication. 3. Choubey PC. Add. Professor of Hospital Administration, All India Institute of Medical Sciences, New Delhi. 1997. Personal Communication
139
SECTION II
Table 2 Estimates of the cost of three major tobaccorelated diseases for the year 2001-2002
Total cost Rs 25,478 Cancers Direct cost Rs 49,980 (+) 1% Rs 4998 Indirect cost Rs 300,020 (+) 9.2 Rs 27,602 Total cost Rs 382,600 Total cost of CAD: 4.61 million x Rs 30,310 = Rs 139.7 billion Total cost of COLD: 40.65 million x Rs 25478 = Rs 103.57 billion Total cost of cancers: 0.17 million x Rs 382600 = Rs 65.04 billion Total cost of the three = Rs 308.33 billion Major tobacco-related diseases in 2001-2002 = Rs 308.33 billion
CAD: Coronary artery disease; COLD: Chronic obstructive lung disease Source: Reddy and Gupta 2004
nication to be effective needs to be structured correctly and that too requires skills and experience. Focusing on providing preventive and promotive care does not require a massive addition to the total health budget, though the health budget itself does require a major increment. What is urgently needed is that the priority change from a largely curative focus to a balanced programme of providing preventive, promotive public health care rather than an over-riding allocation to urban curative care. Even when a model Primary Health Centre was evaluated, Anand et al (Natl. Medical J India) reported the division of costs of running a properly functioning Primary Health Centre with a well balanced emphasis on preventive and promotive care. They have reported that curative care cost 32%, communicable disease control cost 17%, child care 17%, maternal care 11% and family welfare 10%. In other words at the peripheral level curative care needs only about a third of the non-salary budget, preventive, promotive interventions demand at least 2/3 of the available resources.
Even if we do not consider the opportunity costs due to loss of manpower and reduced work efficiency, in terms of direct payments patients with CAD and COLD involves a direct loss of Rs. 8520.30 and Rs. 2257.60 to patients and their caregivers and when the losses borne by the State/Employers are considered a total of Rs. 14,909 and Rs. 11,952 is involved per patient of CAD and COLD respectively. The extant of the morbidity related to cancers, heart disease and lung disease is phenomenal and to a very significant extant preventable by a combination of taxation and vigorous behaviour change communication. A study was carried out by Pandav of the cost benefits of the salt iodisation programme proposed for Sikkim. Iodine deficiency disease is common in Sikkim as it is in other parts of India. Iodised salt has been shown to be effective in preventing this condition. The benefits of iodised salt programme in Sikkim alone showed a total resource saving of Rs.24,406,000 with an investment of Rs.17,669,000. It is clear that preventive/promotive interventions make sense in many ways and pay real economic dividends even if we disregard the priceless benefits of eliminating the parents anguish over a cretin in the family!. One must not make the mistake of assuming that preventive and promotive interventions do not require any funds or technical expertise. The delivery of effective health promotion requires manpower, resources and technical expertise. When Thailand was rolling back HIV infection it was spending the equivalent of US$ 45 million a year on air time alone to propagate the concept of HIV prevention (. To some extant some staff is already in place, but because of the low priority given to health promotion, even this limited resource is misused or not utilised for Behaviour Change Communication. Technical expertise is required to plan and implement cost efficient and effective interventions, only the technology required is not establishing more advanced curative facilities without investing in mechanisms to effectively deliver preventive and promotive health care. Even health commu-
Health in the Community

Having discussed earlier in the document that preventive measures are effective and that though Prevention costs money the returns far exceed the investment, we will now consider possible ways to improve the health of the community. Of course this discussion will remain in the context of the basic premise that effective improvement in health and health care in the community is contingent upon involving the community itself as partners in the health system. To reiterate what has been said earlier in the document, improving the health of the community, especially of the poor in both rural and urban settings demands resources in the form of skills, a suitable structure and finances. First and foremost the system must be built around the concept of the centrality of the communitys involvement in its own health care. A health promoting life style in essential for health as lifestyle diseases are becoming increasingly important. Factors such as water and sanitation play a very important role Disease outbreaks, and risk factors for disease must be detected as soon as possible and corrective action initiated without delay Easily accessible essential basic care must be available to all with referral facilities for complicated conditions
Mechanisms for Health in the Community

To meet the health needs of the community interventions are required at a minimum of five levels. Peoples participation in health care is the basic prerequisite. A system of education and behaviour change communication has to be put in place with an objective of promoting a healthy life style
140
SECTION II
An effective mechanism for dealing with disease outbreaks has to be incorporated into the public sector health system Suitably trained staff with an expertise in Public Health measures must be become an essential part of the public sector health care team. Basic curative care and access to secondary and tertiary level care as needed
Peoples Participation or Community involvement.

To be effective and self sustaining the community itself must be involved in health care. This can take many forms, through the Panchayati Raj Institutions, the involvement of community based and non-government organisations and the involvement of people at all levels of health intervention. This must be the first change brought about in the functioning of the public sector health system. Only when the people are involved as equal partners in health care can the health system even begin to meet the health needs of the community. Doctors can prescribe but the medicine will be effective only when it is taken. Unless the people are involved in their own health care, not as passive recipients of a dole from the authorities but as active players the fullest benefits are not likely to reach the community on a sustainable basis. The active role of the community is more important now than ever before. The so-called lifestyle diseases both communicable and non-communicable are now contributing a major share of the morbidity and mortality experience of the community. These diseases, whether HIV, cardi-vascular disease, diabetes or even accidents are not amenable to prevention by conventional medical interventions such as medicines or vaccines. The key to their prevention is behaviour change and this can only be achieved by the people themselves, the health system increasingly needs to assume a catalytic and technical resource role.
dedicated staff for communication in the form of the district Media Officer and the block extension educators are already in place as a part of the health team; they need direction and supervision to do their own jobs rather than being utilised as odd bodies. A especial case must be made to use the school setting for inculcating the habits of healthful living right from the school age. It is these members of our society that are going to the citizens of the future and there is increasing evidence to suggest that school age children are powerful communicators who can change the way the family behaves and thinks. Similarly college going students need information on healthy living, not only because if the need to protect themselves from many life threatening conditions but also to act as change agents and information repositories in their future lives. The sudden freedom from parental and school supervision when a student joins college, coupled with peer pressure leads many youth to experiment with tobacco, alcohol and other drugs and with sex. In the current environment of the spread of HIV and hepatitis virus in the community such innocent experimentation may lead to serious or even fatal illnesses. Some Universities such as Delhi University have a system of compulsory subsidiary subjects that have to be taken by some groups. Though compulsory the marks do not count for the final result. A Health subsidiary subject is badly needed not only so that students are empowered to protect themselves for diseases such as HIV/AIDS but also to give them the facts to enable them to influence others with preventive and promotive information. In addition to a health subsidiary, a University Talks AIDS type activity can be expanded to provide preventive and promotive messages.
Early Detection and Prompt Response: Disease Surveillance

Disease outbreaks are still a feature of the health scenario in India. The major epidemics that were a feature in British India are now fortunately a thing of the past but smaller scale outbreaks are still a fact that has to be addressed by the health system. The peoples felt needs can not be said to have been met unless a system is put into place to detect actual and potential outbreaks very early and by responding promptly minimize suffering in the community. Outbreaks of communicable disease are still not uncommon, too frequently the health system learns about them from the media or the political system. It is rare to hear of an impending outbreak detected and tackled, even more uncommon to learn of the health system taking proactive steps to abort disease epidemics. Every disease outbreak that occurs in the community and is not tackled fast not only increases the avoidable morbidity and mortality, but also produces dissatisfaction with the health system. This in turn makes the community less likely to accept preventive and health promotive messages from the health system. There is an urgent need to have a system in place to detect
Education and Behaviour Change Communication

More and more the diseases that affect people are the group collectively described as Life Style Diseases. Most other diseases that plague the community though not strictly speaking in the above category can also be prevented or modified by behaviour change. It is essential therefore that the public sector health team include expertise in health education and behaviour change communication. Besides mass media efforts, the role of both the doctor and all other members of the Public Health team in one to one and small group communication can not be overestimated. As was clearly demonstrated by IEC based project in 68 districts of Bihar, Madhya Pradesh, Utter Pradesh and Rajasthan simple health messages communicated by health workers to village key persons (link persons) are effectively communicated to the community with clearly demonstrated health benefits to the community. Such measures require improved management rather than extensive infusion of funds. Some
141
SECTION II
the first signs of an impending or potential outbreak and then initiate steps to control the problem. This process needs two pre-requisites, a system in place and the expertise available to the community. Unfortunately the skills of Public Health are not a part of the public sector health infrastructure A system that follows the usual medical practise of diagnosis first and then an appropriate response is not a model that meets community needs where the first priority is control rather than diagnosis. This may seem contradictory but one must remember that John Snow controlled the epidemic of Cholera in London long before the causative organism had been identified and the more recent example where successful control measures were put in place for SARS much before the organism had been identified. It is not implied that a proper diagnosis is unnecessary but merely that response can not and should not be delayed until the exact organism is identified and typed. Much can be done to ameliorate the morbidity and mortality in the community on the basis of general public health disease control measures once an outbreak is identified. Early detection can best be achieved by the peripheral health staff and the community (including PRI) itself acting in consonance to bring actual and potential health problems to the attention of the health system. The current all too frequent scenario where the health system at the headquarters learns about a disease outbreak from the media or political system and then informs the peripheral health staff is obviously unsatisfactory. Information from the community should alert the local health functionaries and initiate a response mechanism without waiting for instruction from above. The etiological diagnosis is important, but response can not wait for it to be established because so much can be done to minimise morbidity and mortality even before the exact aetiology is established. To establish the diagnosis a laboratory back-up is required. It would seem logical to involve existing health institutions in the process rather than to establish a completely fresh chain of public health laboratories. India has over two hundred medical colleges and about 600 districts. It would seem logical to give the responsibility for microbiology support and tertiary referral for 3 districts to each Medical College. After all every such teaching institution has a complete microbiology department with laboratories, staff and specialist expertise. The clinical departments of these institutions can also provide expert advise for clinical management of persons affected by the outbreak. This process would minimise expenditure, make experts available to the community and provide valuable training opportunities for the students and junior doctors. Medical colleges should be persuaded to accept their corporate responsibility to the community they serve. Once a system is put into place for the early detection and reporting of disease outbreaks it can be expanded to also become aware of risk factors that predispose to disease outbreaks (increased mosquito breeding may well presage an outbreak of malaria, indeed an increase of potential breeding sites after an unusual monsoon may foretell of increased mosquito breeding). In such situations timely action can prevent
an outbreak. An extension of this concept would be awareness about risk factors for Non Communicable Diseases (NCD).
Skilled Manpower -Public Health Expertise

If the public sector health care system is to be efficient and responsive to the needs of the community, especially in view of the resource constraints that are an integral and continuing part of the health care scenario in India, several steps need to be taken. Micro-level planning is essential so that the activities and priorities of the health system match the actual and felt needs of the community that the system serves. The health interventions must be tailored to meet the particular health needs of the community concerned and can not be addressed by an Nation-wide or even State-wide plan. The people and the public sector health providers must together evolve a plan to meet the needs of the community. For appropriate planning, information is needed, for information to be generated from data, and data to be collected meaningfully from the system and community, skills are needed - the skills of Public Health. Lacking these skills it is no wonder that all attempts to initiate micro-level planning have met with resistance and failure. Micro-planning has been recommended repeatedly both by the Central Council for Health and Family Welfare and by the Planning Commission. If the health system does not address the felt health needs of the community, and if top down dictated health interventions are all that the community experiences, it is no wonder that the community looks at the government health structure as an expensive and redundant imposition by the government. The people feel that the health infra-structure serves the governments needs and does not cater to the needs of the community. India can not afford the luxury of developing a health care programme by trial and error. We neither have the resources nor the time to delay making the benefits of existing knowledge in health available to all our people. Of course advances in knowledge and technology are needed and will take place, but we must never forget that we already have the technical knowledge to tackle most of the health problems plaguing our citizens in the community. Can we justify the delaying the application of available science with the excuse that better science lays ahead. Staff skilled in modern Public Health will be in a position to bring the advantages of current scientific knowledge to finding solutions and selecting the best option for health interventions in the community. In most cases the effective options will include a partnership with the community. It is a peculiarity of the public sector health care system in India that in the field of health care in and for the community, there is no emphasis on the skills and knowledge of modern discipline of Public Health. The Medical Council of India (MCI) has mandated that a major portion of the undergraduate medical curriculum be devoted to Public Health by any name. Preventive and Social Medicine, Community Medicine, Public Health are all different names for essentially the same discipline and every medical student in India has to study
142
SECTION II
this subject. Many medical colleges also train post-graduate students in this subject and award the MD degree. Some Institutions also confer a doctorate in the subject. Yet the decision makers in all but two States in India have decided that the skills of Public Health and related disciplines are not essential in those responsible for looking after the health of a district or even that of a State or the Nation. They feel any medical degree, any specialisation is adequate qualification for planning and implementing health care plans in the community. A neurosurgeon or anatomist is supposed to be able to implement Public Health programmes by virtue of the wisdom that goes with seniority! This decision seems to defy logic and appears contrary to the basic objectives of health care. Not only the highest planning and health decision making bodies in the country have emphasised the importance of the discipline of Public Health (by any name) but even the premier health body in the UN system, the World Health Organisation has stressed the importance of the discipline of Public Health for the health of the community.
site for implementing a community centric health focus with an emphasis on preventive and promotive measures. The WHO has also proposed a model for Comprehensive Community and Home-based Health Care (SEARO Regional Publication No 40) which puts the community at the centre of the care paradigm and stresses the multi-facetted integrated care is needed to meet community needs for health care.
Public Health - Expertise to Meet Community Needs.

Except in the case of two states in India (Tamilnadu and Gujarat) formal qualification in Public Health or its allied disciplines is not a requirement for any of the health related positions. The District Medical Officer of Health can be a person trained in any discipline appointed only on the basis of seniority with no regard for job requirements as long as he or she has a basis MBBS degree. He or she can be an eye specialist by training and be expected to guide the provision of preventive and promotive health services, to tackle communicable disease outbreaks or to carry out an epidemiologic investigation. This mismatch between training and job responsibilities is unfortunately the norm. After all the Union Ministry of Health and Family Welfare appointed a professor of Anatomy from the Maulana Azad Medical College in Delhi to be the final technical authority in the Directorate, Union Ministry of Health and Family Welfare for determining the health related aspects for women and children! Public Health is well defined discipline with its own expertise and skills; a person from another branch of medicine, no matter how qualified in his or her own field can not take the position of a specialist in Public Health just as a specialist in Public Health can not take the place of another specialist. The body of knowledge and skills are not related to the seniority or influence of the person but to the training and experience. The fact that there is a special body of knowledge and skills that deals with the provision of health care to the community is not seriously questioned. The Medical Council of India has given the subject, by whatever name, a position of great prominence in medical education. All the over 215 medical colleges in India teach the subject - this is mandated by the MCI. In almost every one of those Medical Colleges the subject is taught badly and as an abstract discipline. The students in turn neither care nor are interested as they have imbibed the prevailing attitude that the subject is of no real value to their ultimate objectives. Unfortunately as things stand at present they are correct. The MCI has also made provisions for postgraduate degrees in the subject. At this point of time those students who opt for a MD in a Public Health related subject are generally those who have not succeeded in making it into a more prestigious subject. This is not surprising. Unlike other specialties, those qualified in Public Health are not uniquely qualified for any particular positions except perhaps to join faculties in medical schools to produce more misfits like themselves. There is no strong cadre of Public Health Specialists
World Health Organisation and Public Health

The WHO has given the development of Public Health a very high priority in its agenda. An idea of its stand on the need for better Public Health for the country , with all its implications of preventive and promotive health care, better disease surveillance and prompt response to disease outbreaks and the great importance of water, sanitation and sound affordable nutrition can be gauged from the fact that they organised in the end of 1999 an important international meeting with high level expert participants from many parts of the world to deliberate on Public Health in the 21st Century. The focus of this meeting was to a very large extent on India.
Calcutta Meeting
This meeting held in Kolkata (then Calcutta) resulted in what came to be called the Calcutta Declaration. The Declaration was presented at the final plenary chaired by Mr Jyoti Basu and endorsed not only by the Public Health community in India but by the international bodies and experts attending the meeting. The Declaration was again endorsed by the Indian Public Health Association in Agra and more recently during a follow-up meeting organised by WHO. The first clause of the Declaration reads as follow: 1. Promote public health as a discipline and as an essential requirement for health development in the region. In addition to addressing the challenges posed by ill-health and promoting positive health, public health should also address issues related to poverty, equity, ethics, quality, social justice, environment, community development and globalisation. The entire text of the Calcutta Declaration is given as appendix 1 at the end of the Document together with the recommendations of the follow-up meeting. However the role of the discipline of Public Health is emphasised as a pre-requi-
143
SECTION II
even though a very large part of the health care provider workforce in the country is engaged in delivering health care to the community rather than exclusively to individual patients. Our founding fathers who wrote the Constitution of India were a farsighted and wise group. There are many aspects of the provisions made in the Indian Constitution that have been acclaimed as establishing a high standard for other fledgling nations to emulate, or for that matter set an example for several important nations that had been free for a long time. Amongst the provisions is the fact that the Constitution establishes Public Health as a fundamental right and therefore enjoins upon the Government the responsibility to put into place health care systems that provides every citizen the right of access to health care. This does not mean that the State has to ensure that everyone is healthy, and obviously it can not do that, but merely that the best achievable health care provision is provided with equitable access being guaranteed. This provision is spelt out in the Directive Principles of State Policy which states:The State shall regard the raising of the level of nutrition and standard of living of its people and the improvement of public health as among its primary duties. -The Constitution of India; Part IV India has provided the infrastructure for providing health care, an extensive network of Primary Health Centres and Subcentres is in place; the vast majority, but by no means all, has staff in place. Unfortunately the government health care facilities do not meet the health expectations of the community and their functioning as nodes for preventive and promotive health care is largely reactive rather than proactive. The community identifies these facilities as the governments Family Planning Centre rather than their own health care facility. The efficiency of the peripheral health care facilitys functioning is reflected in the less than satisfactory state of various indicators of health prevailing in the community. Maternal mortality is still unacceptably high is much of the country, the infant mortality rate is even now very high in large parts of the nation. Immunization rates are unacceptable in many states and even with all the effort being put into it by many agencies, even polio immunization goals are not met. Outbreaks of communicable disease are not a thing of the past. CEA Wilson in 1920 defined Public Health and this was accepted with minor adaptations by the WHO in 1982 as:The science and art of preventing disease, prolonging life, and promoting health and efficiency through organized community efforts for the sanitation of the environment, the control of communicable infections, the education of the individual in personal hygiene, the organization of medical and nursing services for the early diagnosis and preventive treatment of disease, and the development of social machinery to ensure for every individual a standard of living adequate for the maintenance of health, so organizing these benefits as to enable every citizen to realize his birthright of health and longevity. Subsequently the organization of health care, both pre-
ventive and curative came to be included in the mandate of the specialist in Public Health. In modern usage the term has come to be used synonymously with various allied disciplines such a Preventive and Social Medicine, Community Medicine, Community Health, Public Health Administration etc and during discussions it is understood to include related skills such as epidemiology, behaviour change communication etc. It is in this larger context, incidentally very like the concept of comprehensive health care as proposed in the Bhore Committee report, that the term is used in this paper. Unfortunately the term Public Health is used to mean several different things. As defined above it refers to those special skills and the body of knowledge that deals with the organization and delivery of efficient health care to the community. Unfortunately the term is also used to talk about the state of health of the public. Another usage is synonymous with Public Sector Health Care or health care activities delivered by the state and includes both curative and preventive aspects. This unfortunate ambiguity in language prevents clarity of thought and action. Public Health the discipline has much to offer the community. Evidence based planning and proactive interventions to prevent or abort disease outbreaks are only some of the benefits. Currently reactive rather than proactive decisions are taken about health interventions in the community and even these decisions are to frequently being taken hurriedly in several ways in response to pressure from above. Very often the mandate comes down from above as if India is a homogenous entity and local priorities are identical across the country. Another way to take decisions is based on it has always been done this way. All too frequently interventions are based on a guess and the pressure to do something. The intervention is justified with the implicit understanding that if the intervention does not work something else can be tried. If the intervention does not produce results most often no body will know - the community does not expect results and the system does not have the information or skills to determine if an effect was produced. What matters is that the crisis has passed, the system met the need to do something and in any case the system is not accountable for inefficiency in results or cost. How long can we justify denying the people the fundamental rights guaranteed by the Constitution? By denying the people the benefits of modern Public Health, we are denying large segments of the population of the best options for health care. Having outlined some of the reasons for the sorry state of health in India, and also having argued that improved Public Health services are absolutely essential if the benefits of modern health care are to reach the community at large, this section briefly outlines a vision for the structure and organisation of health care in India in the context of optimising public sector health care in the country with the ultimate ben-
144
SECTION II
efits to the community. As described earlier, Public Health in India has entered a vicious cycle where the lack of professional avenues to use Public Health expertise has removed the incentive for the development of expertise in Public Health and has almost eliminated research and development in that field. New entrants into Public Health are almost always those persons who have failed to enter more highly regarded disciplines and have entered the discipline as a last resort. Poor material at point of entry means that the average product at the end of the training is still very limited in intellectual capacity, expertise and especially initiative. This further lowers the already low status of the discipline and in turn additionally reduces the ability of the discipline to attract talent. Other than teaching positions in medical colleges, specialists in Public Health (and its allied disciplines such as SPM, Community Medicine etc.) have almost no avenues for employment that demands their specialisation as a mandated specification. In a country of well over a billion population, the number of Public Health jobs (other than teaching positions) is still in two digits! Why should anyone good enter a discipline that gives very few avenues for employment and carrys the additional stigma of low status in the eyes of their professional peers, health decision-makers and even the public. The remuneration their specialisation earns is only a fraction of that taken home by their clinical colleagues. Those that opt for Public Health face the fact that their colleagues look at them as professional failures and deserving of pity rather than respect. Naturally the specialty has not made an impact on decision makers and decision makers see no reason to give the specialty a priority. A lot needs to be done. Better human material needs to be drawn into the discipline, the discipline itself needs to establish that Public Health specialists offer what no others can, and peer acceptance needs to be enhanced. It is unlikely that mere pious platitudes by health decision makers are likely to materially influence the situation as it exists today. Improving the quality of teaching and training, and even increasing the number of seats available in teaching institutions will ultimately make little difference to the overall picture; we will just get more of the same, students interested not in the practice of Public Health but desperate students who could not get admission in more coveted disciplines and who are therefore willing to settle for the postponement of the need to earn a living for another three years and the magic letters MD after their names to impress the unsuspecting public. The proposed Institutes of Public Health can not be considered as solutions to the problem, they are only one step. The limiting factor at this time is the lack of recognition of the discipline. Unless there are dedicated job opportunities for specialists in Public Health, good material will not be drawn into the discipline and the Institutes of Public Health will not affect the health situation in the country. If the objective is to bring the advantages of the skills and knowledge of the modern science of Public Health to the com-
munity and the health care delivery system then persons trained in Public Health must be available at all levels from the CHC to the Directorate of Health. This will improve the quality of health care to the public and by creating job opportunities will draw talent into the discipline. One can not expect a trained cardiac surgeon to fill the position where neurosurgery is to be performed. It does not matter that he too is a super-specialist and a surgeon. Why should the health decision makers not demand that every person who deals with Public Health have the expertise and qualification in a related discipline? This suggestion does not depend on a significant enhancement of money spent on salary and related staff costs. What are needed are the political will and the decision by Government. What is proposed is that the benefits of the knowledge and skills of modern Public Health be made available at all levels from the District to the Ministry of Health. This does not entail the creation of many new posts, what is needed is merely the division of existing positions into either the clinical stream or the Public Health stream. All those members of the health team, including doctors and nurses, whose primary job description requires the treatment of individual patients should fall into the clinical stream. All those health care providers, including doctors and nurses whose primary job description entails the provision of health care to the community, preventive and promotive interventions and first contact physicians can constitute the Public Health cadre. Public Health, just as clinical medicine, needs a team effort to function. The Public Health team needs to draw upon expertise from medicine, social sciences, communication, engineering and environmental sciences to provide the broad canvas that is required to sketch the scope of interventions required to provide a health promoting ambiance for the community. The difference can be illustrated by spelling out the implication at the district level. The clinical stream at the district would be headed by a appropriately qualified person in a suitably named position equivalent to what used to be called the Civil Surgeon of the district. The Civil Surgeon was generally also the chief of the district hospital. In the proposed arrangement, the Public Health stream at the district would be headed by the Chief Medical Officer of Health, while the curative care wing would be looked after by the Civil Surgeon. The other health department staff including doctors would be divided between the two depending upon their role. Perhaps one additional position would be required. The Civil Surgeon and CMOH should be of the same seniority. The most under-utilised part of the health infrastructure at present is the Community Health Centre. The CHC is supposed to be first level of referral for a network of PHCs in its hinterland. They are staffed by 4 specialists (clinical) and one general duty medical officer. This latter post could be designated for the Public Health stream and provide leadership for public health related activities in the PHCs. The principle of differentiation and parity should extend all the way up to the top. It would be necessary to have two Director Generals of Health, one for Clinical Services and the other for Public Health. Incidentally this was the practice
145
SECTION II
earlier when one position was designated the Inspector General of Civil Hospitals and the other as the Director of Public Health. It also has precedent in that the Ministry of Health has two Secretaries - one for Health and the other for Family Welfare so there should be no grave administrative reason why there can not be two directors of Health Services. Unless the Public Health stream has an equivalent career path it will remain a second choice, lower in status and esteem than the clinical group. The principle of parity must extend to a realistic assessment of remuneration and take home pay between the two streams, with an eye to the desirability of attracting talent to positions not normally open to medical practice for fees. The proposed scheme can only work if Public Health qualifications are mandated for all Public Health jobs. This will need an interim arrangement and relaxation until sufficient trained staff is available. Perhaps an example of what may work will involve the following decisions: A decision at the policy level to re-organise the public sector health care system in the country into two parallel and equal streams - clinical care and Public Health care The two streams to be equal in terms of real remuneration and seniority, with two positions of equal seniority at the level of Director General of Health Services and lower down two positions of equal rank at the district level Essential academic qualifications should be defined for each stream with Public Health and related disciplines being mandated for all positions in the Public Health stream. An interim relaxation for persons opting for the Public Health stream, perhaps all persons being required to get certification by the end of the second year, and only persons with diploma (DPH/MPH) / MD (PSM) / Dr.PH or PhD being eligible for posts after 5 years.
tification in aspects of Public Health for in-service candidates. These certificates would permit officers to continue to serve while they work towards more advanced academic qualifications. A similar arrangement would be required to strengthen the training of other team members for Public Health. In the long term the number of schools of Public Health would have to be increased from the current one so as to establish a training facility in each region of the country. There is already one at Kolkata and others would be needed for the south, west, central and north zones.
Public Health - a team effort.

Like modern curative care, Public Health too is a team effort. There is an urgent need to draw social scientists, nurses, specialists in communication into the discipline and the Public Health service. Public Health training facilities must open their doors to related disciplines and not confine their instruction only to doctors. Such a mix of knowledge and experience can only enrich the discipline and benefit the community.
Other Systems of Medicine

Various systems of medicine other than allopathic are present in the community and play an important role in providing health care to the community. Setting a system in place that does not take this large army of qualified and unqualified health care providers into partnership is wasteful of a valuable resource. While it is neither practical or even legal to train them in certain aspects of allopathic health care, much good can be achieved by using this valuable resource for preventive and promotive health care. A simple measure such as hand washing after using the toilet, before cooking or handling food and before eating has been demonstrated to have a significant effect on preventing diarrhoeal disease. Measures such as these can be as or more effectively propagated by practitioners of Indian Systems of Medicine as by the allopathic team. Most health promotive measures are likely to strike a sympathetic chord with our fellow providers of health care.
Producing the trained staff required.

If the scheme is implemented, provisions will have to be made to strengthen facilities for producing enough trained persons to meet the need for Public Health manpower. I feel this can be done fairly easily without too large an investment in training facilities or establishing many new Schools of Public Health. What is needed is a reorganisation of the postgraduate training present in many medical schools. At present the MD degree in PSM is a 3 year course. With a little readjustment, the intake can be increase manifold for the first year. The first year should be devoted to a conventional DPH/MPH type programme. At the end of the first year, those who qualify should be awarded a DPH or MPH. The best few in each class, not exceeding the usual intake for MD, can be offered the opportunity to go on for another two years where they would learn academic Public Health including advanced epidemiology, research methodology and do their research and write a thesis. A short term measure to tide over the immediate requirement of trained persons would be to use one or more of the Open Universitys to run distant learning courses offering cer-
Curative Care
Curative care is also required alongside preventive and promotive health interventions. The issue is not one or the other but the appropriate balance between the two wings. While basis or primary health care is needed at the periphery and bye and large is provided for in the rural areas, there is a marked unfilled need to cater to the requirements of urban slum populations. This must be considered a priority health requirement. Another shortcoming of the existing health system is the almost complete absence of a working referral system. An effective system of both upward and downward referral will do much not only to improve the care available to peripheral populations but also increase community satisfaction with
146
SECTION II
the health system. In summary to make sure that health care reaches out to every citizen of India, the system has to be re-organised. It must be ensured that every person, whether living in the poorest unrecognised slum or the most distanced small village gets the benefits of health care. The following steps are unavoidable if the intention is to improve the health status of all our people. People must be treated as active partners in their own health care The messages of health sustaining lifestyles must be incorporated into the formal educational curricula and also imparted to the public at large The focus must change to emphasize preventive and promotive health care ie Public Health Potable water and sanitation requirements must be addressed
Positions dealing with health care to the community must be occupied exclusively by persons trained in Public Health while those doctors and staff engaged largely in providing curative care to individual patients should form the curative wing of the health service. This differentiation must extend from the PHC up to the Directorate of Health Services An appropriate level of curative care must be provided to all those who need it. The financing mechanisms can be worked out The Constitution of India has given every citizen the right to expect good health care. This is being denied, not because of cost but because of the reluctance to accept that Public Health can make health care more efficient. How long can we avoid accepting what is self-evident.
147
SECTION II
Appendix 1
The Calcutta Declaration
We, the participants in this Regional Conference on public health in South-East Asia in 21st Century, appreciate the substantial achievements made in improving the health status of the people in the countries of the South-East Asia Region in the past decades. However, we enter the 21st century with an unfinished agenda of existing health concerns and new and complex challenges that demand innovative solutions. We uphold the centrality of meeting the health needs of the community and our responsibility to preserve, protect and promote the health of the people. We commit ourselves to the goals of poverty alleviation, equity and social justice, gender equality and universal primary education, which are all essential elements in the pursuit of health for all. We recognise that expertise in public health and capacity building, as well as experience are essential for sustaining partnerships in designing, developing and providing health for the community. We emphasise the importance of public health as a multidisciplinary endeavour to meet the health needs of people. Having noted the progress in public health practice, education and training, and research in the countries of SouthEast Asia Region, and having reviewed the lessons from public health-related policies and programmes, we endorse the following strategies and directions for enhancing health development in South-East Asia Region in the 21st century: 1. Promote public health as a discipline and as an essential
requirement for health development in the region. In addition to addressing the challenges posed by ill-health and promoting positive health, public health should also address issues related to poverty, equity, ethics, quality, social justice, environment, community development and globalisation 2. Recognise the leadership role of public health in formulating and implementing evidence-based healthy public policies; creating supportive environments; enhancing social responsibility by involving communities; and increasing allocation of human and financial resources 3. Strengthen public health by creating career structures at national, state, provincial and district levels, and by establishing policies to mandate competent background and relevant expertise for persons responsible for the health of populations, and 4. Strengthen and reform public health education and training, and research, as supported by the networking of institutions and the use of information technology, for improving human resource development We urge all Member Countries as well as WHO to continue to provide leadership and technical cooperation in building partnerships between governments and UN and bilateral development agencies; academia; NGOs; the private sector; the media, and other organs of civil society, and to jointly advocate and actively follow-up on all aspects of this Calcutta Declaration on public health Subsequently The WHO organized a follow up informal consultation in December 2003 in Delhi entitled Future Directions in Public Health-Calcutta and Beyond.
148
People's Partnership for Health Towards a Healthy Public in India
SECTION II
References
Anand et al. National Medical Journal of India. Government of India. Report on tobacco control in India. New Delhi: Ministry of Health & Family Welfare, Centres for Disease Control and Prevention and World Health Organization; 2004. University of Sydney, Earth Institute, Columbia University and IC Health. A race against time. New Delhi: Secretariat of IC Health; 2004.
149
SECTION III Drivers of Health Care Costs
SECTION III
Human Resources for Health
M
SHIV CHANDRA MATHUR
MBBS, MD DIRECTOR, STATE INSTITUTE OF HEALTH AND FAMILY WELFARE GOVERNMENT OF RAJASTHAN JAIPUR E-MAIL: shiv_mathur@hotmail.com
ODERN TECHNOLOGY HAS THE POWER TO PREVENT SICKNESS AND ASSURE EARLY cure. However, its delivery requires a vibrant health system based on the strong foundation of a welltrained, motivated and professional human infrastructure, which includes a wide array of community-based workers, nurses and other paramedics and doctors-persons who actually deliver care and transform inputs into dynamic outcomes. Human resources are the critical variable for the effective implementation of health programmes and delivery of quality health care to achieve the national health policy goals in India. The availability of an adequate number of health personnel to effectively and efficiently manage and implement health programmes cannot be overemphasized. However, numbers alone may not necessarily lead to the desired change in health status and outcomes. In keeping with the growth of the health infrastructure and the expanding scope of the health services human resource needs have been increasing. Several new health programmes have been introduced or strategies of existing programmes revised. The changing constellation of health services and strategies have led to an urgent need to develop new competencies and skills among the health personnel, in addition to increasing the critical mass of human resources at various levels. This section deals with the training, skills, competencies and professional development of the five critical categories of human resources for healththe village health worker, nurses, paramedics and finally, doctors and specialists.
The community
Historically, the health of communities in India were in the hands of local healers who practised holistic medicine. The vaids and hakims combined healing skills with counselling and concern for the well-being of the family. Modern technology has changed this and led to the increasing institutionalization and urbanization of health care. The resultant vacuum, barriers of cost and distance, combined with the gradual shift in perception towards allopathy as symbolizing good quality, resulted in the proliferation of a large number of unqualified or unregistered practitioners of health care, estimated to be 36% as per the 57th Round of the National Sample Survey Organization (NSSO). Individuals and communities play an important role in their own health. Experience, knowledge of their environment and traditional practices form a strong foundation for most communities to address minor ailments and short episodes of sickness. The low cost of medication for such treatment makes travelling long distances unaffordable and expensive. Communities also need to be provided information on health risks and guided on healthy lifestyles. Such information dissemination aimed at behaviour change strategies have been demonstrated to have a long-term impact on avoiding diseases and well-being. The need to restore the community base for the health system has been recognized for a long time (Bhore Committee, Srivastava Committee, NHPI, etc.). These concerns that require active participation of the people, led to the institution of the Village Health Guide (VHG) Scheme in 1977. The VHG was to be a person from the village, who was imparted a short training and provided Rs 50 per month for medicines. The experiment failed as no follow-up training was imparted and the scheme has since been abandoned. After the VHG Scheme, the second intervention for community mobilization on a countrywide scale was carried out in 1993. The Department of Family Welfare introFinancing and Delivery of Health Care Services in India
AVTAR SINGH DUA

MD, MBA MEMBER, SUB-COMMISSION, NATIONAL COMMISSION ON MACROECONOMICS AND HEALTH, GOVERNMENT OF INDIA, NEW DELHI E-MAIL: avtarsinghdua@yahoo.co.in
153
SECTION III
duced the Mahila Swasthya Sangh (MSS Scheme), under which women representatives from 20 households were made into a women's health group and the auxiliary nurse midwife (ANM) was provided with seed money for local health education activities. MSS have been constituted since 1990-91 in villages with a population of more than 1000, or 200 households in the plains, and for a population of 500 or more in hilly areas. The MSS comprises five grassroots-level functionaries and 10 prominent women from the village community. The MSS helps the ANM in educating and motivating the community for the welfare of women and children. The results have been mixed across the country and in several places defunct. Such initiatives of village-level committees have to be viewed not from the narrow perspective of provision of health services, but as part of a wider social mobilization process. Some attempts towards community mobilization have also been made in different states-Jan Swasthya Rakshaks (JSR) in Madhya Pradesh (MP), Mitanins in Chhatisgarh, Jan Mangal Couples in Rajasthan, community-based workers under SIFPSA in Uttar Pradesh (UP) etc. (Annexure I). There have been other isolated experiments by NGOs to deliver health services to communities through village-based workers. Under the National Rural Health Mission there is now a renewed commitment to provide in every locality a trained health worker-a married woman, preferably educated, trained to promote good health behaviour, recognize early signs of the onset of disease (for treatment if minor or referral if serious), run a drug depot to provide essential medicines for minor ailments and help communities access health care services. The vision is for her to be from the community, responsible and accountable to them. Evaluations have, however, indicated that community-based health workers need to be periodically trained, closely supervised and integrally linked to the organized health system to ensure sustainability and credibility. As shown in the JSR initiative of MP, absence of such a nurturing framework resulted in the worker losing focus and being reduced to another quack providing some curative care to make money. Similarly, in areas with a high maternal and infant mortality and where the system is unable to provide effective access to professional services, it is essential to focus on improving the skills of traditional birth attendants (TBAs). Such training, however, needs to be not for a single six-day loop but intermittent and closely supervised. Such intensively monitored upgradation of skills among community health workers in selected areas through special interventions would require substantially more resources than are now provided but will have greater dividends. However, in the ultimate analysis the success of such interventions, so vital to the people particularly those living in remote areas, is dependent on the support and sustenance received from the health providers in the health system.
The first health posts: Multi-purpose health workers

The first rung of the professionalized cadre of health services
are the 2 multipurpose workers (MPWs) manning a subcentre. The male worker is given 6 months' training in public health. With virtually no scope for in-service training, low motivation, high absenteeism and over 60% of posts lying vacant, this cadre is the most neglected. Of equal importance is the female multipurpose worker, or ANM. Though originally conceived to address maternal and child health care in the community, over the years, there has been a systematic dilution of skills and functions. With several states (Rajasthan) and categories (STs) opting to reduce the educational qualification to standard VIII pass and no back-up training in human anatomy and basic sciences, the initial handicap was worsened by reducing the training period from 2 years to 18 months. Training schools were established under political compulsions, without adherence to staffing norms or quality. Over the years, ANMs lost their clinical expertise. Finally, as a multipurpose functionary, her nursing and midwifery skills got eroded, affecting her credibility with the communitya factor influencing the increased level of absenteeism. The low competencies and poor skills among these frontline workers is largely the result of the consistently low priority that was accorded to training, both pre- and in-service. For example, the ANM training centres (ANMTCs) conduct basic training for ANMs/health workers (female). The syllabus is outdated and the physical condition of the buildings pathetic. The training is usually conducted in the district hospitals and does not make the ANM skilled enough to handle a delivery on her own in a house located in a remote village-a reality in most of rural India. To understand the knowledge and skill gaps of ANMs, a survey was undertaken among ANMs and Lady Health Visitors (LHVs) in the cadre of supervisors. Results of the survey revealed a grim picture of poor knowledge and wide skill gaps (Table 1). It is clear from the above that there is practically no inservice training of these health care providers. The training is of low quality and inadequate to provide them with the required skill base. It also reflects on the effectiveness of the recent in-service clinical training of ANMs and LHVs under the Reproductive and Child Health (RCH) Programme, which made no serious efforts to provide hands-on clinical practice. Given the very high burden of neonatal mortality and the urgency to bring down the levels to make a breakthrough in the IMR, a similar analysis of skills was undertaken with respect to neonatal mortality. As can be seen from Table 2, the situation is as disturbing. Neonatal deaths are closely associated with the obstetric process and require effective obstetric and essential newborn care skills among ANMs and LHVs. Birth asphyxia and birth injuries, important causes of neonatal death, can be easily avoided by efficient obstetric care and subsequent newborn care including aspiration of mucus and amniotic fluid. Easy interventions are available for acute respiratory infections ((ARI), diarrhoea and neonatal tetanus; however, health functionaries and supervisors had limited skills in preventing and managing these. Hypothermia, an important cause of neonatal death, was not considered a
154
SECTION III
Table 1 Maternal deaths: Select causes, main interventions and skill levels
Causes of maternal deaths Interventions Current levels of skills of ANMs and LHVs
Antepartum haemorrhage (APH)
Postpartum haemorrhage (PPH) manual removal of the placenta
Early identification of bleeding during pregnancy Counselling Continued risk assessment Referral Prevent and treat anaemia in pregnancy (prophylactic and therapeutic) Early identification and risk assessment
Poor knowledge of APH Poor APH management skills
Poor knowledge of PPH Poor skills to diagnose and manage PPH including Poor skills to give blood/IV transfusion
Skilled attendant at birth Manual removal of placenta Prevent/treat bleeding with appropriate drugs Replace fluid loss by IV drip/transfusion, if severe Early referral and transport Puerperal sepsis Skills in aseptic delivery Clean practices during delivery Administration of antibiotics Pregnancy-induced hypertension (PIH) Early identification of risk in pregnancy Eclampsia/toxemia Counselling Treat eclampsia with the appropriate anticonvulsive drug Urgent deliverycaesarean section if needed Obstructed labour Pelvic assessment Referral Assisted delivery or caesarean section as per indications Complications of abortion Identify and diagnose complications Treat sepsis with antibiotics Fluid replacement if necessary Referral
Poor knowledge of puerperal sepsis and its management
Poor knowledge of PIH Poor counselling skills Poor management skills Poor pelvic assessment Poor management skills of obstructed labour Poor knowledge and skills in managing complications
Note: Number of teachers estimated on basis of norms from Medical Council of India (1) This is the number of subject-wise faculty members required for imparting undergraduate training in the medical colleges where undergraduate medical education is being imparted (2) There are at least 105 medical colleges imparting undergraduate medical education that also impart postgraduate training. As per MCI's Postgraduate Medical Education Regulations, 2000 for conducting postgraduate courses, extra staff is required in the departments of Anatomy, Physiology, Biochemistry, Pharmacology, Pathology, Microbiology, Community Medicine, Radiodiagnosis, Radiotherapy, Anaesthesia and Forensic Medicine (four additional teaching faculty in each of these departments, in addition to those prescribed for undergraduate education). (3) There are also some institutions that impart only postgraduate training, and information on the faculty requirement of such institutions has not been incorporated in this table.
priority by these functionaries. Is the system prepared to meet the challenges of developing human resources with the requisite competencies and skills, based on an effective training policy? Is there an adequate and effective training infrastructure? Is there a functioning and efficient training system and process? The answer is No. A survey conducted by the IIHMR, Jaipur for the NCMH, in two states of Andhra Pradesh and Rajasthan and a review of training policy in India brought out several disturbing facts: No state has a separate manpower planning division in the State Directorate of Medical and Health Services though most have an elaborate training infrastructure-state Institutes of Health and Family Welfare (SIHFW), Health and Family Welfare Training Centres (HFWTC), District Training Centres (DTC) and ANM Training Centres (ANMTC), etc. However, most of these institutions suffer for want of good faculty or adequate budgets for any meaningful training. For example, the Rajasthan SIHFW, created under IPP-IX as a Society to ensure autonomy and flexibility for planning, designing and coordinating training in the state, has had no regular director for about four years. There is no regular faculty available and
most of the faculty positions are vacant. The physical infrastructure has been created but yet to be developed to the desired level. The SIHFW does not have its own field practice area to provide hands-on training and undertake operational research in health systems as well as training interventions. There is no training budget in the state and for the Institute. Funding from the World Bank has come to an end thus bringing uncertainty in staff salary and continuation of training programmes. There is no software development activity, such as designing new training programmes and curricula, learning materials development and new training pedagogy. The present situation of the premier training institute reflects not only the apathy to training but also the level of priority accorded to capacity development in the state. Most SIHFWs are in a similar position. Male workers get their initial training the HFWTCs, established and funded by the Government of India. The HFWTCs conduct in-service training of medical officers besides training of trainers of DTCs. These HFWTCs have their own field practice areas but these are scarcely visited or utilized. The HFWTCs are neglected, suffer from gaps in infrastructure,
155
SECTION III
Table 2 Infant (neonatal) deaths: Select causes, main interventions and skill levels
Cause of death Interventions Current levels of Skills of ANMs and LHVs
Birth asphyxia Birth injury Prematurity
Safe delivery practices Proper newborn care Safe delivery practices Newborn care Proper antenatal care Supplementary nutrition (IFA) Proper newborn care Proper counselling Screening during antenatal care (ANC) Newborn care Proper newborn care Aseptic delivery Immunization of mother with TT Proper management of ARI Proper management of diarrhoea
Inadequate skills for obstetric care Lack of skills in newborn care Inadequate skills for obstetric care Lack of skills in newborn care Inadequate skills to assess foetal growth Inadequate newborn care skills Poor counselling skills Inadequate newborn care skills Inadequate newborn care skills Inadequate skills in aseptic delivery Poor diagnostic and assessment skills for severity of ARI Poor assessment skills for severity of diarrhoea
Congenital malformation
Neonatal jaundice Neonatal tetanus ARI-pneumonia Diarrhoea

IFA: iron-folic acid; ARI: acute respiratory infection; TT: tetanus toxoid
training equipment and aids, training material, and lack of qualified and experienced trainers. The libraries are not equipped and virtually non-functional. Training carried out so far functions on the implicit assumption that generating knowledge, rather than building competencies for action, would empower the health care provider to deliver high-quality services effectively. No effort has been made to build an appropriate training environment which is conducive to learning, raising concerns and developing commitment of health personnel towards health care. There is no behaviour change. There is no monitoring and follow-up to assess changes in performance and effectiveness of programmes. Training programmes are overwhelmed with the assumption that participants' acquisition of knowledge means greater competence; learning is a simple capacity of participants to understand and the ability of trainers to teach; and individual improvement leads to improvement in the organization. An assessment of the knowledge and training needs as perceived by key functionaries was highly revealing. A questionnaire was administered to all health care personnel to assess their knowledge on health and diseases of public health importance; their role, and other related aspects (Table 3). The study revealed that while the level of knowledge regarding immunization was almost 100%, it was deficient regarding important public health programmes such as tuberculosis (TB), HIV/AIDS, malaria eradication, leprosy, etc. Only medical officers had knowledge of these aspects. The ANMs, the MPW (M), and almost all pharmacists and laboratory technicians did not posses adequate knowledge of national programmes. Further, the majority of PHC staff was not aware of their job responsibilities. The review revealed glaring inadequacies in the human
resource development process and training of health personnel. Some salient observations are summarized here: Training institutions and training receive a low priority. There is a generalized apathy towards training and capacity building. Training is not recognized as an intervention to improve performance. Owing to lack of nominations, training programmes are frequently cancelled. The training function is seen in isolation. There is no proper planning and implementation of training programmes. Training is organized as thrust upon by the Central Government or donor agencies. The training needs and expectations of the participants are not considered. Most of the programmes are lecture-based and didactic in nature. There is no focus on practical skills' development. Even in clinical skills' development programmes for ANMs and LHVs, there was scant attention on practice to the participants. The morale of trainers is low. There is no training cadre in the states, or system for appointing trainers. Persons are posted or deputed to training institutions as trainers rather than regularly selected. There is no career stream in training. There are no facilities for the regular professional development of trainers. The SIHFWs, HFWTCs and DTCs are poorly equipped with hostels, training infrastructure and libraries. The physical facilities at ANMTCs are appalling. Trainings of various types is offered under different programmes and a health worker is nominated more than once to attend different training programmes. This multiplicity of training has been a constraint in work performance. Incompetent trainers and lack of technical guidance to training institutions has resulted in poor quality training, thus
156
SECTION III
lowering the credibility to training institutions. Trainers of various training centres feel that there are no formal linkages amongst these institutions and they feel left out. There are financial constraints. The payment of TA/DA to participants, procedures/facilities for inviting guest faculty and lack of funds for developing good quality training material are major problems. The training is not taken seriously by the trainees as it has no relationship with the career development of health professionals; the current appraisal system does not take into account the training received for placement or promotion. There is no system of nomination for training. It is highly centralized and, more often than not, based on personal fancy or preference of the concerned officer. There are no norms for in-service training. Some health personnel attend training programmes irrespective of their utility in their job. The training is not seen as an intervention for improved job performance by most trainers. This is because there is a mismatch between organizational and personal goals. The need for management training is seldom felt by functionaries and health administrators. It is thrust upon them. There is no linkage between service providers and trainers. Training is viewed as a constraint in achieving programme objectives rather than facilitating them. There is no training or personnel information system in the states. As result, there is no proper planning. Operational research is not carried out in training institutions.
Table 3 Perceived knowledge about skill and actual gap

Area Staff interviewed Perceived awareness (%) Andhra Pradesh Rajasthan
National Programme on Women and Child Malaria, TB, AIDS, leprosy, etc. Maternal health Enumerate the correct process for providing anterated care (ANC)
Doctors ANM MPHW/LHV (M &F) Staff Nurse LT/Pharmacist Doctors ANM MPHW/LHV (M &F) Staff Nurse LT/Pharmacist
50 5 16 37 10 93 94 72 100 25 100 55 72 100 86 55 0 100 64 67 32 100 71 55 0 100 100 100 0 100 43 67 68 62
47 10 25 13 8 94 90 62 87 100 60 62 100 88 0 0 100 80 0 0 100 88 70 75 87 100 100 0 100 70 90 87 87
What do you do in ANC? Screen for risk factors Doctors and medical conditions ANM MPHW/LHV (M &F) Staff Nurse Record BP Doctors ANM MPHW/LHV (M &F) Staff Nurse Doctors ANM MPHW/LHV (M &F) Staff Nurse Doctors ANM MPHW/LHV (M &F) Staff Nurse Doctors ANM MPHW/LHV (M &F) Staff Nurse Doctors ANM MPHW/LHV (M &F) Staff Nurse
*Weight and height
Paramedical personnel
Two critical paramedical functionaries in the primary health care system are laboratory technicians and pharmacists. In the absence of a separate council, the training of most categories of paramedical personnel has been unregulated except for pharmacists, whose functioning is governed by the Pharmacy Council of India. Training of most categories of paramedicals has been unregulated as there is no council for regulation of training except the Pharmacy Council of India. The quality of training of most of these categories of personnel is poor.
Screen for anaemia
Give Tetanus Toxoid
Provide education on nutrition
Laboratory technicians
*Only weight taken
Laboratory technicians (LTs) are an important human resource. Although some institutions offer graduate (BSc) courses for Laboratory Technology Technicians, most institutions continue to impart a nine-month diploma course. However, in the absence of a regulatory body, there is no information on the numbers of diploma and graduate LTs. Any XII-standard pass student can take up this course, even students with an Arts/Humanities background in the short duration of nine months, the student, especially one with an Arts background, will not be able to acquire the skills required of him/her.
There is a need, therefore, to upgrade the training courses for LTs to graduate level-BSc (Laboratory Technology). There are a large number for LTs at PHCs and CHCs although as per norms, every PHC and CHC should have one. There is a shortfall of 48.9% in the number of sanctioned posts for LTs, out of a requirement of 25,885 LTs for PHCs and CHCs. Of the sanctioned posts, 15.2% were vacant in 2002. Under the RCH Programme, funds were provided to states for hiring the services of LTs on contract, due to which the gap was
157
SECTION III
filled to some extent. In the public sector, once an LT is recruited and placed in an institution, there is usually no in-service training and no system of continuing education for him/her, despite the fact that in-service training and continuing education is required due to rapid advances in the field of medicine, e.g. automation of laboratory investigations and procedures. There is no supervision of the work of the LT. Under the RNTCP, LTs at many CHCs have been trained in standard procedures for examination of sputum slides for AFB, and there is a supervisor to supervise the quality of work. Most institutions for the training of Laboratory Technicians conduct a diploma course of nine months. Students from the Arts or Humanities stream can also take up this course. There are large number for vacancies of Laboratory Technicians in the public sector and there is no system of continuing education for them. At the PHC and CHC levels, the LT usually performs basic investigations such as routine blood examination, urine examination, examination of a blood slide for malarial parasite, sputum examination for AFB, etc. However, at the district hospital and in medical colleges, there are a number of departments, and for better patient care or research, the LT needs to be posted to different departments by rotation. This helps to enhance his/her skills in different areas. However, there is usually no such system of posting and a LT posted in a particular department continues to be in that department for many years, in many instances throughout his service.
developing countries (Ghana, Fiji, Nigeria, etc.) but perform a variety of tasks normally reserved for registered pharmacists. The knowledge and expertise of diploma holders (and those with degrees as well) is inadequate for community practice. Thus, while the number of pharmacists may appear adequate, their quality requires urgent attention and upgradation. The pharmacist working in the pharmacy should have a diploma or preferably a degree in pharmacy, registered with the Pharmacy Council of the States where he/she is practising, have adequate practical training in community pharmacy and should have communication skills and capabilities to advise regarding proper use of medicines. The pharmacist must have the competence to assess prescriptions, advise patients on appropriate selection and use of over-the-counter medicines, appropriate use of prescribed medicines, advise on drugdrug and drug-food interactions, anticipate adverse drug reactions, comprehend the client's condition and advise on the proper use of the prescribed medication and diet, and decide when to refer to a doctor, etc. In the public sector in rural areas, every PHC and CHC should have a pharmacist. Out of a requirement of 25,885 pharmacists for PHCs and CHCs, there is a shortfall of 25.8% in sanctioned posts at these levels; 10.7% of the sanctioned posts lay vacant in 2002. The major reasons for inadequacies in the quality of and services provided by the pharmacist could be Inappropriate education and training at the college level Lack of facilities for continuing education Inadequate remuneration Unhealthy competition among pharmacies because in most places they are too many and too close to one another, which compels the owners of the pharmacies (often non-pharmacists) to treat pharmacy as a trade (and not a profession). Lack of implementation of existing drug laws, which make it mandatory for medicines to be sold under the personal supervision of a pharmacist, and prescription medicines only against a valid prescription.
Pharmacists
The Pharmacy Council of India (PCI) regulates the education and training of pharmacists under the provision of the Pharmacy Act. The present education regulations framed by it prescribe a curriculum of 2 years after the 10+2 (entry) stage followed by practical training of 500 hours over a period of not less than three months for obtaining the minimum registrable qualification-Diploma in Pharmacy. Consequently, training of most of these categories of personnel has been unregulated and many centres for training these categories have opened up all over India, with permission from the states. There are over 5.5 lakh registered pharmacists in India giving a ratio of one pharmacist for 1840 population, with wide inter-state variations ranging from 1:567 in Pondicherry to 1:43,000 in Madhya Pradesh (Table 4). These pharmacists could work as community pharmacists or as retail pharmacists in retail pharmacy outlets. According to the World Health Organization (WHO), the average ratio of pharmacist to the population in industrialized countries is 1:2300. The average ratio of registered pharmacists to the population compares favourably with that in developed countries. However, these diploma-trained pharmacists are at best equivalent to pharmacy assistants or technicians in developed and many
Nursing Services*
Shortage of nurses
Nurses and midwives are major health care providers. Overall, there is a shortage of nurses and midwives in India. In 2004, the nurse to population ratio in India was 1:1264 while in Europe the nurse to population ratio is 1:100-200. The nurse to doctor ratio is about 1.3:1 compared to a ratio of 3:1 in most developed countries (Table 5). In most states, there is no system of re-registration of nurses. As of March 2003, there were 839,862 nurses registered with State Nursing Councils. However, only about 40% of registered nurses are active because of the small number of sanctioned posts, poor working conditions, low pay scales and migration, retirement or death. At the community level there
* Based on background paper on Nursing Services
158
SECTION III
Table 4 State-wise number of registered pharmacists

State/Union Territory Number of registered pharmacists Population 2001 Pharmacist: population ratio
Andhra Pradesh Arunachal Pradesh Assam Bihar Chhattisgarh Goa Gujarat Haryana Himachal Pradesh Jammu and Kashmir Jharkhand Karnataka Kerala Madhya Pradesh Manipur Maharashtra Meghalaya Mizoram Nagaland Orissa Punjab Rajasthan Sikkim Tamil Nadu Tripura Uttaranchal Uttar Pradesh West Bengal Andaman and Nicobar Island Chandigarh Dadar and Nagar Haveli Daman and Diu Delhi Lakshadweep Pondicherry Total
NA. not available Source of number of pharmacists: Pharmacy Council of India
33,938 347 2429 4163 NA 255 20,948 874 2818 NA NA 71,736 7531 1381 NA 99,614 150 382 NA 12,159 35,290 18,214 NA 101,240 257 NA 30,276 89,630 NA NA NA NA 20,978 3082 1716 559,408
75,727,541 1,091,117 26,638,407 82,878,796 20,795,956 1,343,998 50,596,992 21,082,989 6,077,248 10,069,917 26,909,428 52,733,958 31,838,619 60,385,118 2,388,634 96,752,247 2,306,069 891,058 1,988,636 36,706,920 24,289,296 56,473,122 540,493 62,110,839 3,191,168 8,479,562 166,052,859 80,221,171 356,265 900,914 220,451 158,059 13,782,976 60,595 973,829 1,027,015,247
1:2231 1:3144 1:10966 1:19908 NA 1:5270 1:2415 1:24122 1:2156 NA NA 1:735 1:4227 1:43725 NA 1:971 1:15373 1:2332 NA 1:3019 1:688 1:3100 NA 1:613 1:12417 NA 1:5484 1:895 NA NA NA NA 1:657 1:19 1:567 1:1840
Table 5 Health manpower (per 100,000 population) across some countries

Country Physicians Nurses Midwives Pharmacists Nurse: doctor ratio
are no positions for nurses and health services are delivered mostly by ANMs. There are 502,503 registered ANMs and 40,536 registered LHVs in India. The optimum nurse:patient ratio norms recommended by various committees (Annexures II, III, IV and V) for better patient care have not been implemented, thereby resulting in overload on the existing nurses, affecting the quality of patient care. In many instances, even sanctioned posts are not regularly filled.
Australia Canada China Cuba India Sri Lanka Thailand United Kingdom United States of America
249.1 774.8 209.5 1009.9 164.2 104.2 590.6 744.2 59.7 79.1 42.8 79.1 30.1 161.7 166.5 496.6 548.9 772.6
60.2 1.2 INA INA 47.4 41.9 INA 43.3 INA
72.1 79.7 29 INA 52.7 4.5 INA 58.9 68.8
3.1:1 4.8:1 0.6:1 1.3:1 1.3:1 1.8:1 5.0:1 3.0:1 1.4:1
Roles and responsibilities

There are small categories of nurses and midwives with overlapping roles and responsibilities. Nurses in a hospital setting spend most of their time in non-nursing tasks such as inventory control, record maintenance, etc. Most nurses in service are diploma holders; some are graduate nurses. There are no specialist nurses in clinical areas in India. In other countries, besides the professional nurse who is a graduate, there is the advanced practice nurse (APN) who is a postgraduate. APNs are further categorized into Clinical Nurse Specialist (CNS),
Source: WHO website www.who.int, updated figures obtained from MCI and INC used for India Reference year: 2004
159
SECTION III
Nurse Practitioner (NP), Nurse Anaesthetist, and Midwife. The APN has many roles-a clinician expert, educator, researcher, consultant, and manager, with competency of clinical judgement and leadership qualities. The APN is an agent of change, and can collaborate and communicate. In the United States, the post of APN has been in existence for more than 20 years and it has been found that the APN can make an early diagnosis so that the patient can receive proper treatment in time, have a shorter length of hospital stay, fewer complications, and satisfaction with the care provided. As part of the health care reform in the US, the production of NPs has increased because of a shortage of doctors in general practice. In Australia, NPs work at the community level as case managers as well as independent practitioners. In Australia and New Zealand, there are midwives whose training is at the postgraduate level.
Nurse leaders in Thailand submitted a 10-year nursing and midwifery plan to the government in the early 1970s as a result of which 331 scholarships were granted to the nursing faculty for doctoral study outside the country and sufficient funds were also granted for libraries and computers.
Quality of nursing services and research evidence in nursing and midwifery

In the absence of proper nursing standards the quality of nursing services in India varies from hospital to hospital. There is no system of accreditation of hospitals contrary to what happens in countries such as Thailand where hospitals are accredited by an autonomous organization. The nursing component is included in the assessment criteria, which focuses on nursing activity, nurse's notes, participation of nurses in the patient care team, and nursing activity in infection control. Nurses are members of the hospital surveyor team. This accreditation activity stimulates nurses for quality improvement. In 2004, the INC conducted a workshop to develop a quality assurance model for the nursing services but this is yet to be implemented. The model focuses on the code of ethics and professional conduct of nurses, nursing standards, the nursing process and nursing care plan, patient teaching, management techniques, continuing education, research and the nurse's role during disaster. There is, in general, no emphasis on conducting research or creating evidence that could be cost-effective, or improve the nursing and midwifery services. As a result, there are no advocacy efforts for increasing the scope of work of nursing personnel and empowering them. Although there is a Nursing Research Society of India to promote research and there are a number of Indian nursing journals, the number of nursing research studies and publications is small. The reasons for the nursing staff not conducting research are lack of capacity to do research, heavy workload and lack of time, inadequate resources, lack of support from administrators, and because they are not aware of the significance of research.
Nursing and midwifery education: Inadequacy and poor quality

India has 635 nursing schools and 165 nursing colleges. Some nursing colleges are attached to medical colleges. The Indian Nursing Council (DNC) has set standards for education by identifying curriculum structure and syllabi for all educational programmes and conducts inspection of nursing education institutions every 3-5 years. The quality of nurse training is affected by an inadequate number of nurse teacher specialists, nonadherence of the INC teacher:student norm, inadequate infrastructure, insufficient budget, lack of commitment and accountability among educators for clinical supervision and guidance, inadequate and improper clinical facilities and inadequate exposure to hands-on experience for students. In 2004, it was found that 61.2% of nursing schools/colleges were unsuitable for teaching. De-recognition by the INC has no effect on any institution as it continues to function with permission of the State Nursing Council. The result is the production of nurses and midwives with inadequate skills and who later work in an environment of ineffective regulation. Thus, the INC does not have effective control over the nursing services. Many private health care institutions train their own health workers on the job instead of hiring qualified and trained nurses due to a shortage of trained nursing personnel and because hiring the services of non-qualified persons is less expensive. The INC has no control over such practices, as this issue is not addressed in the INC Act.
Continuing education for nurses and midwives

There is no formal continuing education system for the training of nurses and midwives to keep them abreast of the latest developments in the field of nursing and public health. There is no system whereby clinical nurse specialists can be produced in India. In India, there is also no quality assurance (QA) system for nursing education as opposed to Australia, Thailand and the UK, which have a quality assurance system for the quality of input, process, output and outcome.
Teaching faculty in nursing schools and colleges and higher education

There is a shortage of teachers with master or doctoral degrees and the postgraduate curriculum in nursing is inappropriate. There is limited research with regard to nursing services and nursing education. In Thailand, 30 years ago, the master's programme in Nursing aimed to produce nurse educators and nurse administrators and later nurse specialists. A master's programme for APNs is offered in many faculties of nursing to respond to the growing demand for more competent nurses.
Leadership in the nursing profession and empowerment of nurses

Nursing personnel lack leadership and negotiating skills and are rarely, if at all, involved in planning and policy formulation for nursing services, education, etc. at all levels even though they are vital members of the health care team. Nurses
160
SECTION III
and midwives are not easily accepted and recognized as leaders and administrators. In most Indian states, the Director of Health Services, a doctor, is the President of the State Nursing Council, and the highest rank to which a nursing person can be promoted is up to the level of a Deputy Director. This is ample evidence of suboptimal empowerment of health personnel from the nursing profession. By contrast, in 1988, the New Zealand Government invested heavily in multidisciplinary, experiential leadership and management development programmes to create a number of nurse leaders in the changing health care system. Bangladesh, Myanmar, Nepal, Sri Lanka and Thailand have also provided a leadership and management programme for nurses and midwives. In India, midwives working in the community setting are not allowed to administer injectable drugs even in an emergency, although they are allowed to inject vaccines to children. The roles of NPs include health promotion, disease prevention, therapeutic nursing interventions and rehabilitation. Nurses perform some tasks independently and some under supervision. Independent Tasks include all procedures that do not penetrate the body such as providing comfort, pain relief without medication, suction, education, counselling, health assessment, developmental assessment, primary care, midwifery and measurement of vital signs. Dependent tasks include giving medications, injections, immunization and withdrawing blood.
the State Nursing Council Acts. These are major hindrances to the maintenance of uniform standards by the Indian Nursing Council. In Canada, the US, the UK, Australia, New Zealand, Thailand, Korea, and Norway, nurses and midwives have either been elected or appointed to Parliament. The President and members of the Thailand Nursing Council sit on many national health committees, such as the National Universal Health Coverage Committee to set policies on health care services and reimbursement, subcommittee on quality control of health care services, subcommittee on health manpower development, etc.
Doctors
Availability of doctors
The Mudaliar Committee (1961) recommended a doctor:population ratio of 1:3000. Till September 2004, 633,108 doctors had been registered with different State Medical Councils in India (Table 6). This gives a doctor to population ratio of one doctor for every 1676 population in India (or 59.7 physicians for 100,000 population). In comparison, the number of physicians per lakh population in Australia, Canada, the UK, the US and Sri Lanka was 249.1, 209.5, 166.5, 548.9 and 42.8, respectively (Table 5). The doctor-population ratio in India is, however, skewed, with rural, tribal and hilly areas being underserved as compared to urban areas. However, the Medical Council of India (MCI) and State Medical Councils do not maintain a live register with updated figures taking into account attrition due to death, migration outside the country, or nonpractising of medicine by qualified doctors. Various committees set up by the government from time to time have recommended that data related to health manpower should be made available to facilitate health manpower planning.
Regulations in nursing and midwifery

The INC was constituted by Indian Nursing Council Act, 1947 to set a uniform standard of regulation for minimum requirements of courses in nursing education, inspection and accreditation of institutions for quality of education, and maintaining information on nurses, midwives and health visitors by compiling data from the State Nursing Councils. It has an important role in accepting and recognizing qualifications or certificates awarded by universities within and outside India. There are 22 State Nursing Councils whose functions are to inspect and accredit schools of nursing in their state, conduct examinations, prescribe rules of conduct, take disciplinary action and maintain a register of nurses, midwives, ANMs and health visitors in the state. In some states the Examining Body and the Registering Authority are one and the same. The INC has requested State Governments to create or establish separate examining bodies and have a separate registering authority. The Indian Nursing Council has not been able to regulate the quality of training in nursing schools and colleges because it lacks control over the State Medical Councils. Nursing personnel are not actively involved in policy formulation in India, even on matters that affect nursing practice, unlike in other countries. The enforcement of provisions of the Indian Nursing Council Act, 1947 is poor as there is lack of uniformity in many State Nursing Council Acts which were enacted prior to the Indian Nursing Council Act, 1947. Some powers prescribed in the Central Act are similar to those prescribed in some of
Production of doctors: The medical colleges

The Mudaliar Committee recommended establishing one medical college for a population of 50 lakh. This comes to 218 medical colleges according to the current estimated population. As of July 2004, there are 229 medical colleges in India, out of which 125 are in the government sector and remaining 104 in the private sector (Table 7). Of these 229 medical colleges, 67 have been permitted under Section 10A of the MCI Act. There is wide inter-state disparity in the number of medical colleges and the admission seats available every year. The admission capacity in these colleges is 25,500 students per year-7700 undergraduate seats in north India compared to 18,000 in the south. Viewed from the norm of one medical college for 50 lakh population, Andhra Pradesh, Karnataka, Maharashtra and Tamil Nadu have an excess of medical colleges while states such as Uttar Pradesh, West Bengal, Chhattisgarh, Madhya Pradesh, Orissa, Assam and Rajasthan have a shortfall (Table 8). These are also the states where the health indicators are relatively poor.
161
SECTION III
Table 6 Cumulative number of allopathic doctors registered with State Medical Councils till 30 September 2004
State Medical Council Number of registered doctors Projected population on 1 April 2004 No. of registered doctors per lakh population
With < 50 registered doctors /lakh population

Haryana Uttar Pradesh North-eastern states Bihar and Jharkhand Madhya Pradesh and Chhattisgarh Orissa Rajasthan 1285 44,927 15,723 35,110 29,003 14,712 22,506 21,000,000 186,293,000 49,389,000 107,362,000 86,681,000 37,520,000 57,463,000 6.1 24.1 31.8 32.7 33.5 39.2 39.2
With 50-100 registered doctors per lakh population

Andhra Pradesh West Bengal Gujarat Jammu andKashmir Maharashtra Kerala 48,402 52,274 36,521 7,993 90,855 32,412 78,892,000 83,079,000 514,24,000 10,716,000 94,839,000 33,444,000 61.4 62.9 71.0 74.6 95.8 96.9
With >100 registered doctors per lakh population

Tamil Nadu Karnataka Goa Punjab Delhi Total
Source for number of registered doctors: Medical Council of India
71,157 65,789 2,332 33,705 28,402 633,108
64,991,000 54,692,000 1,768,000 25,526,000 16,047,000 1,061,126,000
109.5 120.3 131.9 132.0 177.0 59.7
As per the norm of one medical college for 50 lakh population, 218 colleges are required. There are 229 colleges in India which 45% are in the private sector. There are wide inter-state variations in the number of colleges, with Karnataka, Maharashtra, Andhra Pradesh and Tamil Nadu having an excess, and Uttar Pradesh, West Bengal, Chhattisgarh, Bihar, Orissa and Madhya Pradesh having a shortage. Both private and government medical colleges have a shortage of teachers. Often governments resort to a mass reshuffle of teachers of different specialties from one medical college to another on a temporary basis at the time of inspection by the MCI. Keeping a fake roll of medical teachers and showing expenditure under the salary head is a common tactic adopted by managements of private medical colleges, which has an adverse impact on the quality of instruction. The problem of shortage of medical teachers is more acute in private medical colleges, especially in pre- and paraclinical specialties such as Anatomy, Physiology, Biochemistry, Pathology, Microbiology, Pharmacology, Forensic Medicine, Community Medicine.
Requirement of teaching faculty for imparting training in medical colleges

The MCI has laid down Minimum Standard Requirements for medical colleges for 50/100/150 Admissions Annually Regulations, 1999, which also contain the requirement of minimum number of teachers for imparting training in the colleges. Based on the MCI Regulations, the minimum number of teaching faculty required (total, and annually on an assumption of 30 years of service as a teacher) for different subjects is shown in Table 9. However, the current levels of production of postgraduates in some disciplines clearly falls short of meeting the present and future needs of the teaching faculty is Anatomy, Physiology, Biochemistry, Pharmacology, Forensic Medicine, Community Medicine, Radiotherapy, and Physical Medicine and Rehabilitation (Table 10).
Quality of training in medical colleges and factors associated with it

One challenge in medical education is to induct and retain competent teachers who can transfer their expertise to the students. However, under the current system of recruitment only the technical knowledge of the person is assessed and
162
SECTION III
not his/her aptitude for teaching. The medical curriculum is meant to help decide what knowledge needs to be given to the students. Due to very high emphasis on passing examinations, students tend to focus only on those areas that would be covered in the examination. Often teachers fail to determine what the students should know, what would be interesting to know and what the student need not know. Consequently, instead of focusing on community needs depending on local health problems, students have no option but to focus more on recent advances, which are of limited relevance to the needs of the population at large. The medical curriculum helps to decide what knowledge needs to be given to students, what skills they should acquire and what methodology needs to be adopted to impart these skills. Training of MBBS students should be skills-oriented for dealing with the needs of the community, keeping in view the demographic and epidemiological transition. Teachers need to differentiate between what the student should know, what would be interesting to know and what the student need not know. Training during internship should be oriented towards the acquisition of skills. Another problem contributing to the poor quality of training of doctors is the vacant posts among the teaching faculty. The process of filling up vacant posts is tedious and timeconsuming. It takes almost a year from the date of advertisement to the appointment of a specialist doctor as a teaching faculty by which time the selected person may have joined elsewhere. Vacancies adversely affect the quality of training of doctors and, in many colleges, the position is very seriousthere are some government colleges with just two to three faculty members in pre- and paraclinical departments. An Assistant Professor may have to work as Head of the Department; in such cases, often teaching faculty members from some other college are deputed as internal examiners for examination purposes. In addition to regular teaching faculty, many medical colleges have posts of Senior Residents who serve as a vital link for the training of medical students. However, about 30% of the posts of senior residents are lying vacant, which affects both the quality of training and patient care. The development and training of the faculty of medical colleges has been a greatly neglected area. There are no avenues or incentives for the teaching faculty to undertake research, or introduce innovative methods of training to effectively transfer skills to undergraduate and postgraduate medical students. Doctors getting an opportunity to attend conferences/workshops abroad without any financial liability on the government, whether Central or State, often undergo the unpleasant experience of obtaining clearance to leave the country. Only a few states have time-bound promotions for teaching doctors; there are doctors who have not been promoted for over 10-20 years. Because of the mushrooming of medical colleges in the private sector which require teaching
doctors and are willing to pay higher salaries with other benefits as well, many senior in government medical colleges opt for voluntary retirement and move to private medical colleges. As mentioned earlier, it often takes years to fill up these vacant posts. These have a demoralizing effect. Large vacancies exist of teaching doctors in pre- and paraclinical specialties because there are insufficient postgraduate seats in these specialties, and students do not find it worth their while to undergo postgraduate training in these subjects as they are non-practising branches. Consequently, often students join post-graduation in these subjects as a stop-gap arrangement till they get a seat in one of the clinical subjects. In States that have an excess of medical colleges as per norms prescribed by the Mudaliar Committee, almost two-thirds (63.4%) are private medical colleges. Their basic objective is to earn money. These private medical colleges charge exorbitant fees from students, for which students often raise money through loans. It is understandable that these students treat the expenses on their training as an investment, and would later want to recover their investment, and are likely to be
Table 7 State-wise number of medical colleges in India (as on 30 July 2004)

State/Union Territory Number of medical colleges Government Private Total Total number of seats
Andhra Pradesh Assam Bihar Chandigharh Chhatisgarh Delhi Goa Gujarat Haryana Himachal Pradesh Jammu and Kashmir Jharkhand Karnataka Kerala Madhya Pradesh Maharashtra Manipur Orissa Pondicherry Punjab Rajasthan Sikkim Tamil Nadu Uttar Pradesh Uttaranchal West Bengal Total
Source: Medical Council of India
10 3 6 1 2 5 1 8 1 2 3 3 4 6 5 19 1 3 1 3 6 1 13 9 9 125
17 27 3 2 8 1 2 5 1 5 13 2 3 2 1 4 3 27 31 8 14 2 7 19 38 1 3 4 5 3 6 2 8 1 7 20 3 12 2 2 9 104 229
3475 391 510 50 200 560 100 1625 250 115 350 190 3905 1600 820 4200 100 364 475 520 800 100 2315 1262 200 1105 25,682
163
SECTION III
disinclined towards primary health care or public health. The Re-orientation of Medical Education (ROME) Programme was ambitiously launched in 1977 to prepare doctors to contribute effectively to the improvement of community health. To support this programme each medical college was assigned three rural development blocks where buildings and hostels were constructed to provide teaching complexes. It was a very well conceived programme with political will but before its actual implementation it got lost in oblivion. Major hindrance to good quality training of undergraduate and postgraduate medical students is the private practice that most States permit. While the MCI and State/Central governments and most professional bodies have endorsed the need for teachers of medical colleges to be full-time and nonpractising, clandestine or officially sanctioned private practice is commonplace. The logic behind banning private practice was to ensure that medical teachers are available for improving and sustaining the quality of care in teaching hospitals. However, in several states this objective has been overwhelmed by market forces that make clinical practice more remunerate than teaching.
Box 1 Medical education should remain a charitable enterprise

Christian Medical College (CMC), Vellore is an institute of excellence for medical education that is at par with other national and international training institutes. It admits 60 students annually through a rigorous selection procedure that includes an assessment of the aptitude; suitability of the candidate and the tuition fee is also very low. There is considerable emphasis on community-based training through the Community Health and Development (CHAD) programme. Faculty members of the Medical College go to the field practice area under CHAD and provide services to people and training to students. In addition, there is a linkage between CHAD and medical college hospitals wherein a referral system is functional. This kind of a linkage does not exist in most medical college areas and community-based training is also poor. The training of students at CMC is such that two-thirds of the MB,BS alumni work in India and 80% of them in non-metropolitan areas of the country. The students and faculty members are required to stay within the campus. Although the payment structure for faculty members is not at par or competitive with the private sector, they are provided excellent housing facilities and a good working environment, thereby contributing to their motivation and job satisfaction. This is evidence that quality of training and non-financial incentives matter far more than financial incentives to provide better health services to the masses.
Skills mix
India is passing through demographic and epidemiological transition. Hence, an MB,BS doctor should be able to provide care for such communicable diseases as TB, malaria, respiratory infections, diarrhoea, etc. and non-communicable diseases such as accidents/injuries, hypertension, diabetes, psychiatric illness, other heart diseases. As a minimum, therefore, a doctor must have the competency to diagnose and provide basic emergency obstetric care for maternal complications and neonatal care. In addition, he/she should be able to provide services such as normal delivery, medical termination of pregnancy (MTP), cardiopulmonary resuscitation, etc. For a student to acquire these skills, he/she has to have adequate exposure to and interaction with patients. However, with many colleges admitting over 150 students annually, not all students obtain adequate exposure and acquire enough skills to be able to handle most problems independently in a primary care setting. Besides, about 45% of medical colleges are in the private sector and many of them do not have a sufficient caseload of different diseases for good quality training of medical students. For example, an MB,BS doctor should have acquired the skills to perform vasectomy which implies that he/she should be able to perform about two vasectomy operations under supervision. For over 25,000 admissions to medical colleges annually, this implies over 50,000 operations by MB,BS students under the supervision of a surgeon, while it is well known that only about 1.4 lakh vasectomy operations are conducted every year in India, a large number of these District Hospitals. The quality of training of postgraduates should also be given due importance. It has been observed that postgraduate doctors posted as specialists do not perform surgeries, e.g. an ophthalmologist not doing cataract surgeries, or an obste-
trician-gynaecologist not performing caesarean section operations. This is probably because private practice is allowed in some states; thus, private patients of some specialists who teach in medical colleges are admitted to hospitals in the government sector and postgraduate trainees do not get an opportunity to examine or operate upon them. Moreover, the training environment for postgraduate students in medical colleges is different from what they find in District Hospitals or in CHCs, which have a relatively poor infrastructure.
Duration of training: Is it adequate?

The four-and-a-half years of undergraduate medical education is followed by 12 months' internship wherein students are meant to enhance their clinical skills and understand health care delivery in a community/rural setting. Internship is currently implemented only on paper, particularly since various universities have started the Pre-PG (MD/MS) entrance examination, making internship redundant. Throughout his/her internship, a medical graduate prepares for the MD/MS postgraduate entrance examination.
Increasing trend for postgraduation

It is estimated that almost half the medical graduates opt for postgraduation and settle in urban areas. So serious is the problem that currently most states have failed to ensure the
164
SECTION III
Table 8 State-wise number of required and existing medical colleges

State/Union Territory Population as on 1 April 2004 Number of medical colleges required colleges required @ one per 50 lakh population Existing number of of medical colleges Difference between existing and required number
States/UTs with more than the required number of medical colleges

Karnataka Maharashtra Andhra Pradesh Kerala Tamil Nadu Pondicherry Gujarat Delhi Jammu and Kashmir Punjab Himachal Pradesh Chandigarh Goa Sikkim Uttaranchal Arunachal Pradesh Manipur Mizoram Nagaland Andaman and Nicobar Island Dadra and Nagar Haveli Daman and Diu Lakshadweep Haryana Tripura Meghalaya Chhattisgarh Jharkhand Rajasthan Assam Orissa Madhya Pradesh Bihar West Bengal Uttar Pradesh TOTAL 54,692,000 94,839,000 78,892,000 33,365,000 63,755,000 1,236,000 51,057,000 16,047,000 10,716,000 24,536,000 7,270,000 990,000 1,768,000 621,000 9,051,000 1,327,000 2,798,000 1,061,000 1,884,000 429,000 212,000 155,000 79,000 21,000,000 4,203,000 2,705,000 22,205,000 26,315,000 57,463,000 27,520,000 37,091,000 64,476,000 81,047,000 83,079,000 177,242,000 1,061,126,000 11 19 16 7 13 0 10 3 2 5 1 0 0 0 2 0 1 0 0 0 0 0 0 4 1 1 4 5 11 6 7 13 16 17 35 212 31 38 27 14 20 5 13 5 4 6 2 1 1 1 2 0 1 0 0 0 0 0 0 3 0 0 2 3 8 3 3 7 8 9 12 229 20 19 11 7 7 5 3 2 2 1 1 1 1 1 0 0 0 0 0 0 0 0 0 -1 -1 -1 -2 -2 -3 -3 -4 -6 -8 -8 -23 17
States/UTs with an adequate number of medical colleges
States/UTs with a fewer number of medical colleges
Source of information on existing number of medical colleges: Medical Council of India
availability of anaesthetists, obstetricians and surgeons for effective delivery of emergency obstetric care services including newborn care through the public health facilities. There is a shortfall of 10.1% in the number of sanctioned posts for doctors at PHCs and even out of the sanctioned posts, about 13.4% are lying vacant.
Induction training and in-service re-orientation training in the public sector

Doctors who join services in the public sector should be oriented and indoctrinated so that they can manage the health services effectively and efficiently. However, most states do not impart an indoctrination/induction training to fresh
165
SECTION III
Table 9 Number of teaching faculty for Undergraduate Medical education in 229 Medical colleges
Subject For U/G education Additional for P/G education Total Number of new faculty members required annually
Anatomy Physiology Biochemistry Pathology Microbiology Pharmacology Forensic Medicine Community Medicine General Medicine General Surgery Paediatrics TB and Chest Dis. Skin and VD Psychiatry Orthopedics ENT Ophthalmology Gynaecology/Obstetrics Radiodiagnosis Radiotherapy Anaesthesia Physical Medicine and Rehabilitation Total
2346 2346 1216 3476 1273 1931 1230 2618 2961 2961 1545 687 687 687 1545 687 687 2003 1688 744 2575 959 36852
315 315 315 315 315 315 315 315 0 0 0 0 0 0 0 0 0 0 315 315 315 0 3465
2661 2661 1531 3791 1588 2246 1565 2933 2961 2961 1545 687 687 687 1545 687 687 2003 2003 1059 2890 959 40317
89 89 51 126 53 75 52 98 99 99 52 23 23 23 52 23 23 67 67 35 96 32 1344
Note: Number of teachers estimated on basis of norms from Medical Council of India (1) This is the number of subject-wise faculty members required for imparting undergraduate training in the medical colleges where undergraduate medical education is being imparted (2) There are at least 105 medical colleges imparting undergraduate medical education that also impart postgraduate training. As per MCI's Postgraduate Medical Education Regulations, 2000 for conducting postgraduate courses, extra staff is required in the departments of Anatomy, Physiology, Biochemistry, Pharmacology, Pathology, Microbiology, Community Medicine, Radiodiagnosis, Radiotherapy, Anaesthesia and Forensic Medicine (four additional teaching faculty in each of these departments, in addition to those prescribed for undergraduate education). (3) There are also some institutions that impart only postgraduate training, and information on the faculty requirement of such institutions has not been incorporated in this table.
appointees and, therefore, they tend to concentrate more on curative services than on following a preventive and promotive approach. In view of the changing demographic and epidemiological situation, and developments in the field of medicine, doctors need to be re-oriented from time to time to keep them abreast of the latest changes in programme management. At present, individual vertical programmes impart in-service training to doctors in the public sector but these are fragmented. The training is disease-specific and there is no integrated retraining of doctors.
Regulation of training in medical colleges: The incompletely addressed agenda

The MCI is the apex body for ensuring maintenance of uniform standards of medical education, both graduate and postgraduate. In 1997, the Council revised Graduate Medical Education Regulations dealing with eligibility criteria for admission to the MB,BS course, a detailed curriculum for the MB,BS course as well as internship. The regulations also provide for
integrated teaching, objectives of the study in each subject and the skills that a student shall acquire at the end of study of a particular subject. The MCI has also stipulated Minimum Qualifications for Teachers in Medical Institutions Regulations, 1998, which prescribe the minimum qualifications required for a person to be appointed as an Assistant Professor, Associate Professor or Professor. However, the basic problem in the MCI structure is that since education is a state subject, it is the primary responsibility of the states and their universities. Effectively, the MCI can only recommend de-recognition of a particular college for the MB,BS course. However, in the prevailing sociopolitical environment no medical college has so far been de-recognized on account of the reported deficiencies. The MCI has recommended to the Government of India to make the necessary provisions in the MCI (Regulations) Act for renewal of registration of medical practitioners every five years, linking such renewal with attendance of compulsory continuing medical education (CME) programmes. Such CME programmes have been found to be beneficial in keeping med-
166
SECTION III
ical practitioners up to date with the latest developments in the field of medicine. The MCI has a limited role in prescribing regulations for undergraduate and postgraduate education and inspecting the sites to verify on a set checklist the resources required to start and continue running a medical college. In the context of the norms available at present, MCI inspectors tend to concentrate primarily on infrastructure and staff position rather than quality/methodology/orientation of medical education. Hence, even in colleges reputed to impart excellent education, the quality is declining because of shortage of teaching staff, inadequate laboratory facilities and budgets to update libraries, lack of research, inadequate clinical load and lack of first-hand experience in examining and managing patients, etc. To compound the problem the facilities at medical colleges are not upgraded on a regular basis. While shortage/non-availability of funds is one important factor for non-upgradation of facilities, the apathy of the teaching faculty due to their commitment to private practice is another factor contributing to the continuous decline in standards of medical education. Over the past 1-2 years, the MCI observed that a large
Table 10
number of doctors were claiming employment as medical teachers in more than one medical college at the same time, apparently to show to the inspection team of the Council that the colleges concerned fulfilled the minimum requirement for teaching staff for seeking permissions/renewals under Section 10A of the MCI Act. To curb this practice, the MCI introduced Declaration Forms to be signed by doctors claiming employment as medical teachers in any given medical college, and a provision for endorsement by the Dean/Principal of the medical college was also introduced. To seriously deal with the persisting problem, in 2004 the Council unanimously decided that the names of 65 erring doctors furnishing more than one declaration form and claiming teaching employment in more than one medical college at the same point of time be erased temporarily from the Council up to 31 July 2007. Of these 65 doctors 59 (90.8%) are from pre- or paraclinical specialties. Just as the MCI has not been able to fully discharge its responsibilities, so is the case with respective State Medical Councils. The scope of work of State Medical Councils should be increased and they should be brought under the purview of the MCI by amending the concerned Act(s) so that the State Medical Councils can act as extended arms of the MCI.
Teaching faculty requirement and production of fresh postgraduates annually for Medical Colleges in the country
Speciality No. of faculty required annually Average No. of Postgraduate Degrees awarded during 1999-2000 & 2000-2001* Shortfall in teaching faculty required annually
Medical and health universities

The professor Rais Ahmed Committee recommended setting up a University of Health Sciences to help medical colleges maintain high standards by upgrading facilities, instituting faculty development programmes, adopting a multidisciplinary approach to professional development relevant to the socioeconomic conditions of India and vigorous research activity. In an effort to make medical teachers appreciate the basic shift in medical education and draft an action plan to bring about needed corrections in the curricula at every level of medical education, a system of establishing separate medical universities in each state was started. Andhra Pradesh took the lead in this direction by setting up a medical university in 1986. Some other states have also recently set up health universities to improve the quality of training in medical colleges as well as other nursing and paramedical disciplines.
Anatomy Physiology Biochemistry Pathology Microbiology Pharmacology Forensic Med. Community Med. General Medicine General Surgery Pediatrics TB & Chest Dis. Skin & VD Psychiatry Orthopedics ENT Ophthalmology Gynae / Obs Radiodiagnosis Radiotherapy Anesthesia PMR
89 89 51 126 53 75 52 98 99 99 52 23 23 23 52 23 23 67 67 35 96 32
23 29 21 141 57 33 11 39 346 324 162 32 53 28 123 76 126 286 82 11 197 2
66 60 30 -15 -4 42 41 59 -247 -225 -111 -9 -30 -5 -71 -53 -103 -219 -15 25 -100 30
* Source: Health Information of India 2000 & 2001, Ministry of Health & FW, Government of India, 2003 Many Medical Colleges also impart M.Sc. courses in pre- and para-clinical specialities, but there was no information on these aspects and these have not been considered for the calculations After obtaining a postgraduate degree a person may join as a teaching faculty, provide health services in public or private sector, or migrate to another country. There could also be attrition due to non-practicing of medicine. These factors have not been factored. This interpretation is also based on the assumption that the doctors, after obtaining postgraduate degree, would opt first for a teaching post and only after such posts have been filled would doctors opt for joining the health care delivery system.
Accreditation
There is no system of accreditation of medical colleges to ensure that that training at these institutions meets acceptable levels of quality. The Srivastava Committee
167
SECTION III
(1975) noted that it was necessary to restructure the entire programme of medical education, as the existing system did not prepare the right type of personnel needed for a national health services programme. The Committee recommended that immediate steps be taken to set up a Medical and Health Education Commission, comprising the MCI, INC, DCI, PCI, representatives of Central and State Governments, and leading persons in the field of health services and medical education. Its role was suggested to be promotive. However, no action has been taken to establish such a Commission.
Specialist services in the public health care delivery system: adequacy and availability
As per norms, at the CHCs, there are four specialists-a general physician, a general surgeon, a paediatrician and an obstetrician-gynaecologist. There is a shortfall of 61.3% in the number of sanctioned posts of specialists (surgeons, obstetrician-gynaecologists, physicians and Paediatricians) at CHCs, and of these, 37.7% are lying vacant. It may be noted that very few CHCs have sanctioned posts for anesthetists. The number of postgraduate degrees or diplomas being awarded today in India is not adequate for effective delivery of specialist services (anaesthesiologists, public health specialists, paediatricians, gynaecologists and psychiatrists). The number of seats for admission to postgraduate degree and diploma courses should be determined according to the health needs of the people and this should be a dynamic process. Specialist services could be rendered by a postgraduate degree or diploma holder. The number of specialists required in some fields and those being produced annually are shown in Table 11. There is a severe shortfall of specialists in all dis-
ciplines, and this shortage is more worrying in the case of Community Medicine/Public Health, Paediatrics, Anaesthesia and Ophthalmology. If we are to achieve the Millennium Development Goals as well as the goals laid down in the National Health Policy 2002, in addition to focusing on primary health care, there is an urgent need to rationally create posts for specialists (Anaesthetists and ophthalmologists) at CHCs (and of paediatricians and gynaecologists if they have not already been created). Posts should also be created for public health specialists at all levels from the Centre to the districts, and all vacant posts filled up. Efforts should be made to keep the specialist workforce motivated, especially by way of non-financial incentives. This is essential because brain drain was estimated to have reached an alarming 30% of the annual output in 198687 (IAAME, 1992). The migration of doctors-both external (from India to other countries) and internal (from the public sector to the profit-oriented private sector) -has shown an increasing trend.
Integration with indigenous systems of medicine

Ayurveda, Unani, Siddha and Homeopathy are the four broadly recognized Indian systems of medicine (ISM). Successive committees set up by the Government have highlighted the competitive advantage of ISM doctors due to their easy accessibility to and acceptability by the masses, especially in rural areas. There is a need to 'integrate' ISM&H in the health care delivery system and national programmes, and ensure optimal use of the vast infrastructure of hospitals, dispensaries and physicians. There are almost 7 lakh registered ISM practitioners in India. They have widespread availability and better acceptability in communities.
Table 11 Requirement versus availability of specialists in selected disciplines for delivering health services in the public sector in India
Subject PGs available for health annually (1) Average number of diplomas 1999-2000 and 2000-01 (2) Available annually (3) = (1)+(2) Number of new specialists for the health care delivery system Net available annually after migration*(4)= (3) x 0.6 Total required(5) Required annually@ (due to attrition) (6)=(5)30 Shortfall(7) = (6)-(4)
delivery system awarded during
Community Medicine Paediatrics Skin and VD Psychiatry Ophthalmology Gynaecology/Obstetrics Anaesthesia
0 111 30 5 103 219 100
11 153 29 21 71 198 171
11 264 59 26 174 417 271
7 158 35 15 104 250 162
3750 7952 1200 1200 4296 7952 7952
125 265 40 40 143 265 265
118 107 5 25 39 15 103
* Considering migration of 10% of specialists to other countries and 30% of specialists to the private sector in the country, thereby implying availability of 60% manpower for providing service in the public sector @ Considering average length of service to be 30 years Number of specialists required has been based on the following assumptions: a)Community Medicine: One CMO at each district headquarter, assisted by four other public health specialists (5 per district x 600 districts= 3000; 35 States/ UTs- 20 per State / UT Hq = 700; 50 for national level) b)Paediatrics, Obstetrics/Gynaecology, Anaesthesia: 3 at District Headquarters and 2 at CHCs = 3 x 600 + 2 x 3076 = 7952) c)Skin/VD and Psychiatry: 2 at District Headquarters = 2 x 600 = 1200 d)Ophthalmology: 2 at District Headquarters and 1 at CHC = 2 x 600 + 1 x 3076 = 4296
168
SECTION III
There is enormous scope for integration between the allopathic system of medicine and ISM (AYUSH), which could be done in a phased manner. To begin with, there could be increasing coordination between the system of medicine allopathic and ISM by posting ISM doctors at the same hospital s/dispensaries as allopathic doctors. Later, there could be functional integration at all levels including at the level of training. There were 691,470 AYUSH (Ayurveda, Yoga, Unani, Siddha and Homeopathy) practitioners registered in India as on 1 April 2002 (Table 12). If these are added to the number of doctors registered under the allopathic system of medicine, the total number of doctors registered in India becomes 1,315,296, giving a doctor to population ratio of 1:827 population. However, there is no integration between allopathic doctors and those following ISM. Of late, some coordination between these two systems has been attempted by the Central Government, and clinics of Ayurveda, Unani and Homeopathy have been set up at some allopathic hospitals and dispensaries. Under the RCH Programme, seven Ayurvedic and five Unani medicines were included and as a pilot project they were distributed in nine states and four cities. However, in the absence of proper orientation of the health personnel, they are not being effectively utilized.
Recommendations
1. To improve the quality of laboratory procedures, the minimum course for Laboratory Technicians should be upgraded from a diploma to graduate degree course-Bachelor in Medical Laboratory Technology. This course should be offered only to students with a science background and not to students from an Arts/Humanities background. 2. The existing diploma programme should be continued till there are 80%-90% of required pharmacists in the country. Once this is achieved, the diploma in pharmacy education may be replaced by the upgraded pharmacy practice course. Existing diploma holders should be allowed to maintain their registration to practice only by participating in specified condensed courses so that their knowledge and competence are brought as close as possible to the new programme for registration of pharmacists. By 2015, there should be only one category of registered pharmacists. Thereafter, anyone not coming up to the mark can at best be categorized as a pharmacy assistant. 3. Every state should have a full-time Director (Nursing) for better management and development of this vital human resource for health. There is a need to formulate and implement a national strategic plan for nursing and midwifery development as has been done by countries such as Bangladesh, Thailand, Indonesia, Myanmar and Sri Lanka. To develop leadership skills among nurses, the Government should invest in multidisciplinary leadership and management development programmes for nurses and midwives as has been done in countries such as New Zealand, Bangladesh, Myanmar, Nepal, Sri Lanka and Thailand.
4. A quality assurance system for nursing should be introduced to ensure good quality care and nursing outcomes as expected by clients and according to professional standards. This would ensure commitment of the care provider towards providing the best care for consumers-to lessen patient suffering, shorten the length of hospital stay, reduce health care costs, and decrease infection, complications and death. For better care of patients, the posts of nursing staff at all levels need to be increased in hospitals as per the recommendations of earlier committees, and all vacancies need to be filled. 5. There should be only two levels of nurses, first, a professional nurse who undergoes a 4-year BSc (Nursing) programme offered at university level. Second, an auxiliary nurse who undergoes a 2-year Auxiliary Nursing (Certificate) programme offered at schools of nursing. A Bachelor of Nursing Sciences should be offered to nurse from diploma school to upgrade their qualifications. The auxiliary nurse should be able to pursue studies in Bachelor of Nursing Sciences degree. To expand the role of nurses in India, Advanced Nurse Practitioner (APN) programmes should be established. A Master of Science programme in nursing should focus on advance nursing practice. A plan for the production of APNs should be included in manpower planning and their scope of practice and roles and responsibilities clearly identified by the INC. Posts for Clinical Nurse Specialist (CNS) and Nurse Practitioner (NP) should be created first at big hospitals and later in peripheral health facilities. 6. The roles and responsibilities of nurses need to be redefined by suitable amendments in the INC Act so that they can administer some injectable drugs in case of emergencies. APNs should be able to deal with complex health problems and have clinical judgements. They will also be able to provide education and consultation, as well as help to improve patient care and health (Annexures XVI and XVII). 7. The MCI and INC should be strengthened by amendments in the respective Acts to enable them to discharge their role as regulators and the State Medical and Nursing Councils should be brought under the purview of the MCI and INC. The scope of work of State Medical and Nursing Councils also needs to be increased so that they can function as extended arms of the MCI and INC. The INC and nurses should be actively involved in health policy formulation, especially those that would have an effect on the nursing profession. 8. There should be integrated planning and development of human resources in health-doctors, nurses and other paramedical personnel. As has been recommended by earlier committees, a live register needs to be introduced for all categories of medical and para-medical personnel. Regularly updated information on the number of postgraduates available in different specialties in India and registered in different State Medical Councils should be maintained. 9. There is a need to regulate and address the interregional imbalances in medical colleges. The Government must also take the responsibility of opening additional medical colleges in the public sector in those states that have fewer medical colleges than required. 10. The standards of training in medical colleges, nursing
169
SECTION III
schools and colleges, and in those institutions that impart training to other categories of paramedical personnel have to be improved. There is an urgent need to establish a Commission for Excellence in Health Care and Human Resources for Health so that grants can be provided to medical and health universities and medical colleges to improve the standards of training. 11. A system of accreditation of institutions imparting training to various categories of human resources for health should be introduced. 12. Sufficient incentives-financial and non-financial-should
be given to attract medical teachers to join and continue in pre- and paraclinical specialties in medical colleges. As an alternative, non-MBBS postgraduate seats could be increased in these specialties and a record of the number of such people passing out should be kept for rational health manpower planning and production. Teachers in medical and nursing training institutions should be provided fellowships to undertake higher studies, and attend conferences and workshops. 13. There should be an increasing focus on rural and community orientation of MB,BS and nursing students. The compulsory three months' training of medical doctors in a community setting during internship should be implemented seriously so that students Table 12 acquire the skills to interact with the comState-wise number of registered doctors (allopathic and AYUSH) munity and can effectively and efficiently deliver services in a community setting. State/Union Number of AYUSH Number of allopathic Total number of Students in nursing schools and colleges Territory doctors registered doctors registered till registered doctors should also be posted in peripheral health till 30 September 2004 1 January 2002 facilities to help them develop an understanding of effective delivery of health Andhra Pradesh 29,238 48,402 77,640 services in a rural setting. A system of reArunachal Pradesh 0 0 0 registration of doctors and nurses once Assam 1284 15723 17007 every five years and linking re-registration Bihar 161,010 34,975 195,985 with a minimum number of hours of CME Chhattisgarh 0 186 186 should be introduced. Goa 0 2332 2332 14. The examination system for students Gujarat 22,425 36,521 58,946 needs to be revamped so that evaluation Haryana 26,047 1285 27,332 of skills is done on the basis of what the Himachal Pradesh 8466 0 8466 student should know for the delivery of Jammu and Kashmir 505 7993 8498 essential health interventions. The entrance Jharkhand 0 135 135 examination for postgraduation should Karnataka 18,792 65,789 84,581 be conducted before the start of internKerala 22,968 32,412 55,380 ship so that students can focus on learnMadhya Pradesh 56,009 28,817 84,826 ing clinical and housemanship skills durManipur 0 0 0 ing internship. Maharashtra 83,167 90,855 174,022 15. The number of seats in specialties Meghalaya 229 0 229 such as Anaesthesiology, Paediatrics, Mizoram 0 0 0 Obstetrics-Gynaecology, Psychiatry and Nagaland 1,997 0 1,997 Community Medicine should be increased. Orissa 8,781 14,712 23,493 Postgraduate students in these clinical Punjab 33,542 33,705 67,247 specialties should be posted in District Rajasthan 29,261 22,506 51,767 Hospitals for one month in the second year Sikkim 0 0 0 of their postgraduation and for two months Tamil Nadu 37,053 71,157 108,210 in the third year so that they can get adeTripura 0 0 0 quate hands-on experience in managing Uttaranchal 0 0 0 patients in District Hospitals/CHCs. For Uttar Pradesh 94,898 44,927 139,825 this, due amendments in the rules of West Bengal 45,280 52,274 97,554 respective universities should be made. A&N Island 0 0 0 16. For management of public health Chandigarh 297 0 297 programmes by public health specialists D&N Haveli 0 0 0 at all levels, institutes of excellence for Daman & Diu 0 0 0 training in public health need to be estabDelhi 10,221 28,402 38,623 lished, as has already been declared by the Lakshadweep 0 0 0 Central Government. There is also a need Pondicherry 0 0 0 to introduce an All-India Cadre of Public Total 691,470 633,108 1,324,578 Health on the lines of the All-India Civil Source: Department of AYUSH, Ministry of Health and Family Welfare, Government of India, New Delhi Services.
170
SECTION III
17. In an effort towards integration of allopathy with ISM, there could be increasing coordination between practitioners of the allopathic system of medicine with AYUSH practitioners by posting doctors of the two systems of medicine in the same facility as has been done in some CGHS dispensaries. This could be followed by functional integration between the two systems of medicine at all levels-training, placement and programme implementation. 18. Since doctors do not stay at PHCs and the community is deprived of health services. There has to be an increasing focus on paramedicalization of primary health care services in India. To make the delivery of health and family welfare services more effective through a primary health care set-up, there should be two ANMs at the subcentre. 19. A pilot should be tried wherein doctors are not posted to PHCs and are instead posted at CHCs, which could have six specialists (a general physician, a general surgeon, a paediatrician, an obstetrician-gynaecologist, an anaesthetist and an ophthalmologist) and four MB,BS medical officers. The
MB,BS doctors could be given transport to provide OPD services at PHCs and they could stay at CHC headquarters. Routine services and emergency care at PHCs could be provided by nurse practitioners, a separate cadre that would then need to be introduced. In such a case, there would be a need to upgrade the skills of nurse practitioners to provide medical care in emergency situations. 20. When one compares the number of undergraduate and postgraduate seats, it becomes obvious that a vast majority of doctors do not receive any training after MB,BS. The current pattern of medical training, often followed by an inadequate internship, does not provide adequate skills for the doctor to take up independent practice. When this is viewed in the context of India's need to have a large number of qualified general practitioners it is clear that adequate opportunities need to be created for MB,BS doctors to undergo postgraduate studies in Family Medicine (General Practice)a professional rather than academic degree.
171
SECTION III
References
Academy Press. The Trained Nurses' Association of India. Indian nursing yearbook 2000-2001. Noida: Academy Press; 2000. All India Institute of Medical Sciences. Inquiry driven strategies for innovations in medical education in India: Consortium of medical institution. New Delhi: All India Institute of Medical Sciences; 1991. Alwan A, Hornby P. The implications of health sector reforms for HRD. WHO Bulletin 2002;80:56-60. Anderson RD, Sweeney SD, William AT. An introduction to management science: Quantitative approach to decision making. New York: South-western College Publishing; 2000. Banerjee A, Deaton A, Duflo E. Health care delivery in rural Rajasthan. Economic and Political Weekly 2004;944-9. Bangalore: Community Health Cell. Research project on strategies for social relevance and committee orientation in medical education. Building on the Indian experience; Step by Step. Bangalore: Community Health Cell; 1992. Bateman ST, Snell AS. Management: Building competitive advantage. New York: Irwin Mc.Graw-Hill; 1999. Buchan J, Dal Poz MR. Skills mix in the health care: Reviewing the evidence. WHO Bulletin 80:2004;80:57580. Chandra S, Mathur GM. Health care through mobile camps. WH Forum 1985;6:153-4. Chaudhary N, Hammer J, Rogers H, Mulidharan K. Teacher and health care provider absence-a multi-country study at E:\PS\PS Website (2001-2004) Absenteeism 2004. Dhruv Mankad M (ed). Medical education reexamined. Mumbai: CED; 1991. Government of India. Report of the Expert Committee on Health Manpower Planning, Production and Management. New Delhi: Ministry of Health and Family Welfare; 1987. Government of India. Report of the XTH Plan Working on Development of Nursing Service. New Delhi: Ministry of Health and Family Welfare. Government of India. Report on Health Survey and Development Committee (Bhore Committee). New Delhi: Government of India; 1946. Government of India. Report on Health Planning and Development Committee (Mudaliar Committee). New Delhi: Government of India; 1961. Government of India. Report of the Medical Education Review Committee (Mehta Committee). New Delhi: Government of India; 1982. Government of India. Report on Job Responsibilities of Staff of Primary Health Center by Rural Health Division. New Delhi: Ministry of Health and Family Welfare, Government of India;1986. Government of India. Report of Expert Committee on Health Manpower Planning, Production and Management (J S Bajaj Committee) New Delhi: Ministry of Health and Family Welfare, Government of India;1987. Government of India. Report on Medical Education and Support Manpower (Srivastava Committee). New Delhi: Government of India; 1975. Government of India. Report on National Policy on ISM and Homeopathy-2002 and National Health Policy-2002. New Delhi: Ministry of Health and Family Welfare; 2002. Government of India. Report on mainstreaming of ISM&H and integration of different medical system. New Delhi: AYUSH, Ministry of Health and Family Welfare; 2004. Government of India. Websites of different Medical Institutions of India MHFW, GOI and MCI 2004. ICSSR and Indian Council of Medical Research. Health for All-an Alternative Strategy. (Report of a study group set joint by IIE, Pune); 1981. India Today. Best medical colleges of India. India Today 17 May 2004. Educational casualty. India Today 26 July 2004. Indian Nursing Council. Teaching material for quality assurance model: Nursing quality and commitment. Indian Nursing Council. Master of Nursing: Syllabus and regulation.
172
SECTION III
Indian Nursing Council. The Indian Nursing Council Act, 1947. New Delhi. (Amended in 1950 and 1957) Indian Nursing Council. Syllabi and regulations for the courses of studies for auxiliary nurse and midwife. New Delhi;1977. Indian Nursing Council. B.Sc. Nursing syllabus and regulations (basic programmed). 3rd ed. New Delhi; 1981. Indian Nursing Council. Golden jubilee celebration: Nursing in the new millennium. Daryaganj, New Delhi: Jaina Offset Printers; 2000. Indian Nursing Council. Regulations. 2001. Indian Nursing Council. Syllabus and regulations for Diploma in General Nursing & Midwifery New Delhi; 2001. Indian Nursing Council. Syllabus for post basic B.Sc. Nursing. New Delhi;2001. Kothari LK. Medical education. Rajasthan: SIHFW; unpublished. Lerberghe WV, Conceicao C, Damme WV, Ferrinho P. When staff is underpaid: Dealing with the individual coping strategies of health personnel. Bulletin of the World Health Organization 2002;80:581-4. Lele D, Pai MR. Doctors, Patients and Consumers Protection Act by Rotary Club of Bombay, Bombay, 1993. Lingam S. University of Health Science University News 1989;XXVII. Marchal B, Kegels G. Health workforce imbalances in times of globalization. National Health Policy-2002. New Delhi: Ministry of Health & Family Welfare. McIntosh N. Medical education for the 1990s and beyond: An approach to clinical training. Paper presented in the MCI convention on Need Based Medical Education at New Delhi, 25 August 1992. Medical Council of India. Annual Report of the Medical Council of India, 2002-03, New Delhi. Medical Council of India. Minimum Standard Requirements for the Medical College for 50 Admissions Annually Regulations, 1999. Medical Council of India. Minimum Standard Requirements for the Medical College for 100 Admissions Annually Regulations, 1999.
Medical Council of India. Minimum Standard Requirements for the Medical College for 150 Admissions Annually Regulations, 1999. Mishra R, Chatterjee R, Rao S. India Health Report. New Delhi: Oxford; 2003. Narsimhan V, et al. Responding to global human resource crisis. Lancet 2004;363:1469-72. Pai S. Review on Munnabhai MBBS. British Medical Journal 2004;328. Ramamurthy B. Unfolding standards of medical education. Paper presented at the International Symposium on Medical Education, New Delhi, 28 August 1992. Sood R (ed). Postgraduate training in medicine-key issues (Technical series). New Delhi: API; 2002. Voluntary Health Association of India. Independent Commission on Health in India-a report. New Delhi: Voluntary Health Association of India (VHAI); 1997. Voluntary Health Association of India. State of India's Health. New Delhi: VHAI; 1991. World Health Organization (WHO). Personnel for health care-case studies of educational program. Public Health Paper 71. Geneva: WHO; 1980. World Health Organization. Assessing health workers performance. Public Health Paper 72. Geneva: WHO; 1980. World Health Organization. Increasing the relevance of education for health professionals. Public Health Paper 838. Geneva: WHO; 1993. World Health Organization. Macroeconomics and health: Investing in health for economic development. Newton, MA: Digital Design Group; 2001. World Health Organization. Nursing and midwifery workforce management: Conceptual framework. New Delhi: Regional Office for South-East Asia; 2003. World Health Organization. Nursing and midwifery workforce management: Guidelines. New Delhi: Regional Office for South-East Asia; 2003. World Health Organization. Nursing and midwifery workforce management: Analysis of country assessment. New Delhi: Regional Office for South-East Asia; 2003. Yamey G. Interview of Ron La Porte in the News column. BMJ 2004;328:1158.
173
SECTION III
Annexure I Community-based service providers

Name of community based workerVillage health guide Started in 1977 throughout the country
Training requirements
Village community selected a volunteer to act as a link between the community and government health system. Three months' training was provided at PHCs.
Support provided
During training, the village health worker was paid a stipend of Rs 200 per month. He/she was given an honorarium of Rs 50 per month. A kit containing common articles of use, medicines and a manual was provided. Rs 1200 per year allocated to every MSS member for arranging its monthly meetings. Information, education and communication material was supplied. An ANM is the Member Secretary of the MSS. Licensed by the zila panchayat to practise in villages. If there was a Jan Swasthya Rakshak in a village, he / she was the nominated Member Secretary of the Village Health Committee. Mitanins recognized as depot holders for chloroquin in many areas, and have also been identified as DOTS providers in many areas.
Services to be provided
Provided health education and created awareness on maternal and child health and family welfare services. Kept a track of communicable diseases and managed minor ailments and provided first aid to patients.
Evaluation, if any
Was reviewed by a group of experts. The scheme was discontinued with effect from 1 April 2002.
Mahila Swasthya Sangh (MSS) Started in 1990-91 throughout the country
Formed in villages with 1000 population in plains and 500 population in hilly areas. The group comprise 5 grassrootslevel functionaries and 10 prominent women from village community. The members were given a short training. 6 months' training given to an educated person in the village under TRYSEM.
The MSS helps ANM in educating and motivating community, and obtains support from other women colleagues for programmes such as immunization, antenatal care and family planning.
In some states, the MSS is very active while in others, it has been formed as a formality. In many states, monthly meetings are not organized, and meetings held once in 6-8 months. In some states, the existence of fake MSS are reported.
Jan Swasthya Rakshak (Madhya Pradesh)Started in 1995, but taken up in right earnest in 2001 under the Rajiv Gandhi Mission
Provide curative services and deliver public health services in villages.
The rate of attrition is90% . Practising Jan Swasthya Rakshaks were found to be providing only curative services and injecting saline and other drugs.
Mitanin (Chhattisgarh)Started in December 2002
Women were identified from villages and given training on immuniza-tion, antenatal care, prevention and control of gastroenteritis, larval control, prevention and management of malnutrition.52,000 persons have been trained. Couples with the wife in the reproductive age identified from villages and given 10 days' training on methods of family welfare.
Create awareness and provide support for immunization, create awareness on control of mosquito breeding, opposing irrational practices by private practitioners and opposition to domestic violence.
An evaluation needs to be undertaken to examine the impact and sustainability of this scheme.
Jan Mangal Couple (Rajasthan)Started in 1995
Bimonthly meetings held at PHCs for which the couple is paid Rs 200 towards travel expenses and wage loss. Depot holders for oral contraceptives, condoms and oral rehydration solution (ORS). Dai Kit provided a Disposable Dai Delivery Kit. Rs 100 per day for 10 days of training, and cost of travel from residence to training site and back
Promotion of small family norm, and motivating people to use spacing or terminal methods of contraception.
In many areas, couple protection rate increased; however, due to erratic supply of funds, this scheme lay dormant for quite some time. The scheme was revived in the state last year.
Traditional birth attendant (TBA)Started in 2001 under the RCH Programme in about 180 districts with safe delivery rates less than 80%
Varied from state to state from four to several days. Under the RCH programme, imparted 10 days' residential training to practising TBAs at CHCs / PHCs in two phases of 6 and four days respectively at an interval of 4 to 12 weeks A woman from among dairy cooperative workers. 16,323 community-based workers trained
Antenatal careDelivery care - perform safe deliveriesEarly identification of complications in mother and childPostnatal careHealth education and counseling
None documented. However, TBAs did not stay at place of training for the period of training. Consequently, learnt lesser skills for safe delivery and early identification of complications
Community-Based Worker (CBW) (SIFPSA, UP)Started in 1994 in 6 districts, now extended to 40 districts
Given Rs 1000 per month as honorarium
To assist the ANM in her work for the delivery of services at the village level.To mobilize the community to promote spacing methods of family planning and maternal and child health services. To conduct safe deliveries, to provide resuscitation to the newborn , to manage the sick child and to inject gentamicin
Record -keeping by the ANM improved, while she started delegating most of her tasks to the CBW.
Community worker (Gadhchiroli)Started in 1995 in 39 villages (39,000 population)
A IV to X standard pass person from the community intensively trained for 42 days in six phases spread over one year for safe delivery and newborn care, including administration of Inj Gentamicin 350 women from 50 villages were trained
Intensive training, follow-up supervision and guidance with refresher training for two days every two months
Infant mortality rate (IMR) in the area reduced by 47% and neonatal mortality rate reduced by 62%
Bharatvaidya (Osmanabad)Started in 1994 in 50 villages
Trained women supported to diagnose and manage mild illness and minor ailments
Conduct health survey in village, registration of births and deaths, daily home visits, creating awareness on STDs/AIDS, treatment of mild illnesses
IMR reduced from 72 per 1000 in 1995 to 39 per1000 in 2004
174
SECTION III
Nursing for the delivery of essential health interventions
EALTH AND POVERTY ARE CLOSELY RELATED. IMPROVING HEALTH IS THE fundamental goal of economic development. Nurses and midwives play a major role in the health care system. The main functions of nurses are health promotion, prevention of diseases, nursing therapeutics and rehabilitation. Qualified nurses can contribute to achieving positive health outcomes such as reducing mortality, morbidity and disability, promoting healthy lifestyles, improving maternal and child health, and combating HIV/AIDS, malaria, tuberculosis (TB) and other diseases to achieve the Millennium Development Goals. To attain this, policies are required on an effective nursing workforce, appropriate distribution, deployment and utilization, and strong political leadership and commitment of the government, professional organizations and nurse leaders. This paper attempts to identify how nurses and midwives can contribute to health care. It reviews existing situation of nursing and midwifery in India regarding nursing services, nursing education, nursing management, evidence base, nursing research and regulation. In view of the experiences of other countries, future scenario of nursing and midwifery in India is suggested. Strategies to meet the challenges and recommendations are outlined for policy-makers and organizations.
Existing situation of nursing and midwifery in India

According to the National Health Policy (NHP) of India 2002, the major health problems are infectious diseases. These diseases can be prevented by mid-level health personnel such as nurses. However, the quality of nursing and midwifery services, education, research, management and regulation is inadequate and the workforce insufficient.
Nursing and Midwifery Services

The contribution of nurses and midwives to the quality and efficiency of health services is insufficient. The nurse to population/patient ratio is low compared to other countries. In 2004, the ratio was 1:2250 in India and 1:100-150 in Europe. This ratio in African countries, Sri Lanka and Thailand is 1:1400, 1:1100 and 1:850, respectively. Many States in India face a shortage of nurses and midwives. Most of the States have no system of re-registration of nurses. As on March 2003, 8,398,620 nurses were registered with the State Nursing Registering Councils. Only 40% of registered nurses are active because there is no system of live register in India, the said figure includes all the nurses who have been trained since 1947. Nursing positions are created due to financial constraints, poor working conditions, low pay-scales, emigration, retirement or death. Further, the optimum nurse to patient ratio recommended by the Staff Inspection Unit (Ministry of Finance) is implemented in only 7 Central Government hospitals due to economic constraints. This means that one nurse has to care for more patients than he/she should. Nursing and midwifery services do not receive high recognition from the public. The roles and responsibilities of nurses are not clearly defined. As a result, they spend most of their time in non-nursing tasks. In India, the nurse to doctor ratio is almost 1.5:1 while it is 3:1 in developed countries. Most nurses in the service hold a diploma and some hold a Bachelor's degree in nursing and midwifery. There are no specialist nurses in clinical practice. Nurses and midwives do not have much opportunity for
DILEEP KUMAR
PRESIDENT, INDIAN NURSING COUNCIL COMBINED COUNCILS BUILDING, KOTLA ROAD, AIWAN-I-GHALIB MARG, NEW DELHI E-MAIL: presidentinc@yahoo.com
175
SECTION III
continuing their education as no such system exists in most hospitals. In the interest of patient care, the NHP 2002 emphasizes the need for an improvement in the ratio of nurses vis-vis doctors/beds. It also emphasizes on improving the skill level of nurses and increasing the ratio of degree- holding nurses vis--vis diploma-holding nurses. It further recognizes the need for establishing training courses for superspecialty nurses required for tertiary care institutions. In 2004, the Indian Nursing Council (INC) conducted a workshop to develop a quality assurance (QA) model for the nursing services. It has not yet been implemented. The model focuses on the code of ethics and professional conduct of nurses, nursing standards, nursing process and nursing care plan, patient teaching, management techniques, continuing education and research, and the nurse's role during a disaster. However, the INC does not have control over the nursing services. The quality of service depends on the policy and administration of each hospital. Many private health care institutions provide on-the-job training to their health workers instead of hiring nurses. The INC cannot control such exercise as it is not under the Indian Nursing Council Act. Working conditions in many hospitals and communities in India are poor and unsafe. Medical equipment and supplies are inadequate. Incentives are limited. Washing and uniform allowances have, however, been revised and increased for nurses in Central Government hospitals. Limited opportunities are available for career advancement due to non-creation and non-existence of clinical specialty nurse and nurse practitioner positions. There is a shortage of nursing personnel owing to non-implementation of recommended nursing staffing norms. Promotions are limited to 2-3 times throughout the career and few supervision posts are offered. The Trained Nurses' Association of India was established in 1922 by amalgamating the Association of Nursing Superintendents and Trained Nurses' Association and many other associations and unions. It aims to uphold the dignity and honour of the profession, promote cooperation among nurses and provide service for its members.
deliveries; (ii) providing referral facilities close to the community; and (iii) ensuring that bulk of the remaining domiciliary deliveries were conducted by ANMs. However, even after having set up the required number of subcentres, the number of deliveries by ANMs is very low in most States. In the community setting, it is observed that ANMs/LHVs face problems related to transportation, accommodation, gender-based harassment, lack of security, incentives and career prospects, and inadequate provision for living with their families and educating their children.
Nursing and Midwifery Education

There are 635 nursing schools and 165 nursing colleges in India. Some are attached to medical colleges. The nursing and midwifery education programmes offered are given in Table 1. The INC has a equivalancy system in accepting and recognizing qualifications or certificates awarded by other universities or countries. The INC has set standards for all educational programmes by identifying the curriculum structure and syllabi, and has a procedure for the inspection of nursing education institutions every 3-5 years. Common problems identified are inadequate number of nurse teachers and nurse teacher specialists, non-adherence to the Council's norm for teacher to student ratio, inadequate infrastructure, facilities and budget, lack of commitment and accountability among educators for clinical supervision and guidance of students, and inadequate and improper clinical facilities for students. Between September 2004 and October 2004, 61.2% institutions were found unsuitable for teaching. When the Council withdrew the recognition of the institutions, they continued with the permission of the State Nursing Council. There is an overlap in the functioning of the State Nursing Registration Council and the Indian Nursing Council Act with regard to opening of educational institutions of nursing, which has resulted in the mushrooming of such institutions in selected States. The postgraduate curriculum in nursing is not adequate. Teachers with Master's and Doctoral degrees are few. Research and academic work is scarce. There is no national development plan for nursing and midwifery services to enhance the quality and quantity of nurse educators, students and staff nurses.
Community Health Nursing Services

At the community level, there are no positions for nurses. Health care is provided by auxiliary nurse-midwives (ANMs), lady health visitors (LHVs) and female health workers. There are 5,025,030 registered ANMs, and 40,536 registered health visitors and female health workers. Due to the heavy workload, nursing care or home health care cannot be properly provided. Antenatal and delivery care are mostly provided by traditional birth attendants (TBAs), which results in high maternal and infant mortality rates. The maternal mortality varies from 79-135 lakh in better-performing States to 498-707 lakh in low-performing States. The infant mortality rate varies from 14-52 per 1000 live-births in better-performing States to 63-97 per 1000 live-births in low-performing States. The popular norms for the three-tier rural health care infrastructure were evolved with the objective of comprehensive care. For delivery care, this meant (i) promoting institutional
Evidence and Nursing and Midwifery Research

The use of evidence and research to improve practice is inadequate. Data and evidence for research are not managed systematically. They are inaccurate and out-of-date. Nursing research as a subject is taught at the basic and graduate level. There are a number of Indian nursing journals. However, the number of nursing research studies and publications are not many. The reasons for nursing staff not doing research are inability to do research, no idea of research problems, heavy workload, no time, inadequate resources, no support from administrators, and being unaware of its
176
SECTION III
Table 1 Nursing and midwifery education programmes in India

Programme Eligibility Duration of training (years) Examination Registration
ANM
Matriculation
1.5-2
State Nursing Council/DME State Nursing Council/ State examination board/DME University University
State Nursing Council
GNM
10+2 (arts or science)
3-3.5
BSc (Basic) BSc (post-basic)
10+2 (science) 10+2,GNM 2 years' experience BSc MSc
4 2 (regular) 3 (distance) 2 1 (full-time) 2 (part-time) 3-5
MSc Mphil
University University
PhD
MSc/MPhil
University
ANM: auxiliary nurse-midwife; GNM: graduate nurse-midwife
importance. In addition, funds for research in nursing are insufficient and not provided by the national research fund.
Management of Nursing and Midwifery Services

Nurses and midwives are not well accepted or recognized as leaders or administrators. Nursing management skills, leadership, lobbying and negotiating skills are poor. There are an inadequate number of nurse and midwife leaders at the national and State levels for nursing practice, research, education, management, planning and policy development. Although the nurse is a member of the health team, she/he is never asked to represent the profession in planning and policy formulation for nursing services, education, etc. The nursing chief only looks after the nursing personnel and has no authority to make decisions on pay scales, number of posts, staff development or new interventions. In response to the demand of the Delhi Nurses' Union, the Government of India has sanctioned 5 nursing posts at the national level.
Regulation of Nursing and Midwifery Services

The INC was constituted by the Indian Nursing Council Act, 1947. The Act was amended in 1950 and 1957 to set a uniform standard of regulation and practice for nurses, midwives and health visitors by specifying the minimum requirements for courses in nursing education, institution inspection and accreditation for quality of education, and maintaining registration by compiling data from the State Nursing Councils. There are 22 State Nursing Councils whose functions are to
inspect and accredit schools of nursing in their State, conduct examinations, prescribe rules of conduct, take disciplinary action and maintain a register of nurses, midwives, ANMs and health visitors in the State. State Nursing Councils ensure that the prescribed syllabi are followed and standards maintained. Nurses and midwives are required to register with the State Nursing Registration Council after successful completion of nursing courses. The policy of renewal of registration every 3-5 years will be implemented soon and 1-2 continuing education programmes will be required for re-registration. Enforcement of the provisions of the Indian Nursing Council Act, 1947 is found to be weak as many of the State Nursing Council Acts which were enacted before the Indian Nursing Council Act lack uniformity. The State Nursing Councils are virtually governed within the jurisdiction of the State Act under the Constitution of India. The Indian Nursing Council Act, 1947 being a Central legislation, should have powers overriding the State Nursing Councils Acts, by following the Medical Service Act controlled by the Ministry of Health & Family Welfare. Some of the powers prescribed in the Central Act are similar to those prescribed in some of the State Nursing Council Acts. These are major hindrances to maintaining uniform standards by the INC. In some States, the examining body and the registering authority are the same. The INC has requested State Governments to create or establish separate examining bodies and registering authorities. To carry out its functions, the INC works in collaboration with the State Nurse Councils, schools and colleges of nursing and examination boards.
177
SECTION III
Major issues
According to the existing situation in India, major issues that need to be solved are as follows: Insufficient contribution of nurses and midwives to health care development due to few positions for nurses and midwives at the State and national levels inadequate nursing leadership and strategic management inappropriate nurse to population/patient ratio inadequate preparedness of nurses and midwives inadequate recognition of the nurse's status in the health care system limited active involvement of professional organizations. Poor quality of nursing and midwifery care due to
shortage of nurses and midwives due to
limited roles prescribed in the Indian Nursing Council Act,
1947
inconsistency in the Indian Nursing Council and State
Nursing Council Acts insufficient information systems in nursing and midwifery services shortage of staff at the INC and State Nursing Councils.
Future of nursing and midwifery in India

Nurses and midwives in India should play a major role in improving the health and quality-of-life of people. The Millennium Development Goals can be achieved with their active work in the community. Infectious diseases such as HIV/AIDS, TB and malaria can be prevented and health promotion strengthened for all ages, especially maternal and child health. There should be a sound research base, strong leadership, policy formulation and unity of professional organizations with the strong commitment and continuous support of policymakers and the government. Nurses and midwives should be deployed in the hospital and community as per the recommended staffing norms. Roles, responsibilities and competencies for each category of nurses and midwives are to be clearly defined and implemented. Nurses and midwives should actively provide care based on evidence or research and implement nursing practice standards. They will have to pursue continuing education for selflearning. There should be nurse specialists in various clinical areas. The working conditions for nurses and midwives should be good and their safety should be ensured. Nurses and midwives should be active members of the health care team and work in collaboration to provide holistic and comprehensive care for the patient and family. The gap between nursing and midwifery staff and nurse educators needs be bridged. They should work together to improve nursing services and provide a suitable setting for students to practise clinical nursing. Nursing education programmes must be strengthened. Nurses and midwives should have Bachelors' degrees and those with diplomas or certificates must be upgraded. Masters' programmes aimed at producing advanced practice nurses in various clinical areas need to be strengthened and expanded. Doctoral education in nursing must produce good researchers and leaders. Students can learn in an environment equipped with qualified teachers, adequate classrooms, libraries, information technology systems and nursing laboratories. Student-centred learning, self-learning and lifelong education must be emphasized. A quality assurance system for nursing services and education should be implemented. The INC should work with State Nursing Councils to regulate quality effectively. Nurses and midwives will be appreciated by society because of their good work. Nurses and midwives should be involved in policy formulation at the State and national levels. The nursing and midwifery development plan can be integrated in the health care development plan. In addition, the government should recognize the significant role of nurses and midwives in health care and be willing to support and invest heavily in nursing and midwifery.
(i) inadequate number of nursing positions as per the recommended staffing norms (ii) migration (iii) insufficient number of nurses with Bachelors' and Master's degrees and in clinical specialties. limited competency of nurses and midwives due to (i) too many categories of nurses and midwives with overlapping roles (ii) unclear roles and responsibilities of nurses and midwives (iii) ineffective clinical preparation and supervision during training (iv) inadequate continuing education system (v) limited utilization of evidence and research (vi) insufficient clinical nurse specialists and nurse practitioners (vii) inadequate facilities and opportunities for clinical nurse specialists (viii) non-creation of posts for clinical nurse specialists. inadequate standards and guidelines for nursing practice ineffective regulation of nursing and midwifery practice inadequate infrastructure for nursing and midwifery practice inadequate motivation to provide effective care. Poor quality of nursing education to produce qualified graduates for service due to an inadequate national nursing and midwifery education plan and development limited involvement of nurses and midwives at the policy level shortage of qualified nurse educators inadequate infrastructure for nursing education too many categories of nursing and midwifery personnel limited production of academic work and research. Limited role and authority of the INC in nursing development due to
178
SECTION III
Meeting the challenges

1. Strengthen involvement of nurses in health and nursing policy formulation and planning
Nurses need to study policy formulation and planning at all levels of education. Techniques for negotiation and lobbying should be taught. Networking within and outside the nursing profession should be built and strengthened. Data and information on nursing and health should be available, updated and accessible online, if possible. The INC should take the lead and actively participate in health policy formulation, especially policies that will affect and impact the nursing profession. More positions for nurses are needed at the policy-decision level. The Thailand Nursing Council and nurse leaders participated in the formulation of the National Health Act and National Universal Health Coverage Act. In Canada, the US, the UK, Australia, New Zealand, Thailand, Korea and Norway, nurses and midwives have been either elected or appointed to Parliament. The President and members of the Thailand Nursing Council sit on many national health committees to formulate policies on health care services and reimbursement, quality control of health care services, and health manpower development.
fied nurses, renewal of licence, and setting up a nursing service and nursing education accrediting system. If possible, a hospital QA system should have nursing as an integral part and involves nurses in a surveyor team. In Thailand, a hospital is accredited by an autonomous organization. The nursing component is included in the assessment criterion focusing on nursing activity, nurses' notes, participation of nurses in patient care teams, and nursing activity in infection control. Nurses are members of the hospital surveyor team. This accreditation activity stimulates quality improvement. The Thailand Nursing Council is also developing nursing service standards in addition to nursing care standards for each level of the health services. For example, at the primary care level, the structure standard indicates that there should be two professional nurses to care for 5000 people and at least one of them has to be a nurse practitioner. The outcomes standard gives indicators such as reduction of infection rate, increase of self-care capacity among patients with chronic illness, and enhanced patient satisfaction at the tertiary level as well as no bedsores, shorter length of hospital stay, reduction of urinary tract infections, etc. The process standard indicates that nurses should provide holistic health care using nursing processes by ensuring a professional code of ethics and the patient's sociocultural context.
2. Empower nurse leaders

There should be a nursing division led by a nursing director in hospitals. The nurse director has to develop leadership and management skills to enhance the quality of the nursing workforce and nursing care to improve the health of the people and achieve the United Nations' Millennium Development Goals.
4. Ensure nursing workforce management as an integral part of human resource planning and health system development
A well-managed nursing workforce requires an effective and efficient nursing workforce policy and planning. As a follow up to the Resolutions WHA 45.5 (1992) and WHA 49.1 (1996), Member States were urged to formulate and implement national strategic plans for development of nursing and midwifery services. Bangladesh [1994], Thailand [1994], Indonesia and Maldives [1997], Myanmar [1999], Sri Lanka and Nepal [2001] have developed national strategic plans for achieving this goal. An essential component of the nursing and midwifery development plan is manpower planning. Planning can prevent shortage of nurses and increase efficiency in deployment, utilization and development. It is important to include nursing workforce management in human resource and health system development.
3. Establish a quality assurance system for the nursing service

A quality assurance system comprises vision, mission, objectives, strategic and operational plans, nursing service activity, nursing manpower management, roles and responsibilities, nursing standards, nursing indicators, nursing research, nursing administration and management, resource allocation and financial support. The objective of this system is to ensure quality care and nursing outcomes as expected by clients (less suffering, shorter duration of hospital stay, and reduction of health care costs, infection, complications and mortality), and according to professional standards. It also indicates the commitment of the care provider towards providing the best care to consumers. Successful development and implementation of the system depends on the commitment of nursing leaders, hospital administrators, mutual goal-setting, participation of all personnel in the process, continuous quality improvement and good communication. The role of the INC in regulating nursing practice should be strengthened by amending the Nursing Act to include maintaining of registration of quali-
5. Enhance nursing autonomy in practice

The roles and responsibilities of nurses are identified by professional organizations, nursing education and nursing services, and they can be adapted and expanded to meet universal nursing standards. In India, there are a number of care activities that nurses can undertake because of their educational background but cannot carry out because doctors do not delegate responsibility to them. Having nurses take on some of the care that they are trained for independently will be cost-effective. Nurses with a Master's degree in advanced nursing practice can deal
179
SECTION III
with complex health problems, have a better clinical judgement and can select the proper option for the patient by using evidence-based practice. They can also provide education and consultation.
6. Enforce implementation of recommended norms on nurse to patient ratio

The quality of nursing care also depends on the number and categories of nurses who provide care. In hospitals and community settings, there should be a norm or standard for nurse to patient ratio. Norms recommended by the Health Manpower Planning, Production and Management Committee in 1986 and INC for different wards and outpatient departments should be reviewed. The INC must propose to the government the need for more posts and develop mechanisms to enforce the recommended norms for quality of care. The Thailand Nursing Council has developed norms for nurse to patient ratio; in medical, surgical, paediatric and gynaecological wards it is 1:4-1:8, in the delivery room or intensive care unit it is 1:2-1:1 and 1:150-1:100 in the outpatient department. The nurse to technical nurse or practical nurse ratio is 1:3-1:2. This is the standard for all hospitals at the three levels of the health care services.
has an equal right to receive quality and accessible care. Community health care workers must understand the community and work as a partner. To ensure quality of care, the subcentre, primary health centre, community health centre and district should have the infrastructure given in Table 2.
8. Produce advanced practice nurses

Advanced practice nurses (APNs) are prepared at the Master's level. An APN can be categorized into a clinical nurse specialist (CNS), nurse practitioner (NP), nurse anaesthetist and midwife. The roles of the APN are clinician expert, educator, researcher, consultant and manager. APNs have the competency of clinical judgement, leadership skills, are an agent of change, and help in collaboration and communication. In the United States, it has been found that the APN can make an early diagnosis so that the patient receives proper treatment in time, with a shorter length of hospital stay, reduced complications and high patient satisfaction. In Australia, NPs are required to work at the community level as case managers and may have an independent practice as well. In Australia and New Zealand, there are midwives whose education is at the Master's level. In Thailand 30 years ago, the Master's programme in Nursing aimed to produce nurse educators and nurse administrators, and later nurse specialists.
7. Create posts for professional nurses at the community level and strengthen the competency of the auxiliary nurse-midwife
In India, there is a doctor and nurse at the community health centre but at the primary health centre and subcentre, only the female health worker, ANM and LHV are there. One ANM has to take care of 5000 people, which prevents her from providing effective health promotion activities, maternal and child care; conducting home visits and preventing illness. Frequently, babies are delivered by TBAs who do not have formal training, which leads to a high rate of infant and maternal mortality. ANMs sometimes cannot provide comprehensive care or make proper judgement due to their limited training. To ensure quality of service at all community level, a public health nurse (PHN) should work with an ANM. The ANM should be qualified to provide effective maternal and child care to reduce maternal and infant mortality rates, and be able to replace the TBA. The ANM should be taught more about infectious diseases and their prevention such as HIV/AIDS, TB, malaria and be responsible for midwifery work in the community. A PHN, who is a graduate in nursing, should learn more about epidemiology, health promotion, disease prevention, primary medical care, alternative medicine, health and culture, and community nursing. Community nursing should include community assessment, family health care, school health, home health and long-term care. The capacity of PHNs and ANMs should be strengthened so that they can provide health information and education, which are important means to improve the health behaviour of individuals, family and the community. The community
Table 2 Infrastructure required at various levels of the health care services

Subcentre population 5000 To be manned by 2 ANMs with 2 years' training as per the revised syllabus male health worker and one PHN (PHNGraduate nurse/GNM + DPHN) Strengthening the infra structural facilities One PHN practitioner (with additional training) and one PHN supervisor to effectively supervise all MCH and FW services 4 staff nurses for 24 hours' service
PHC 30,000population
CHC 14 staff nurses 1,00,000 population 3 PHN supervisors 1 PHN practitioner 1 independent midwifery practitioner District level Strengthen the institution of the DPHN officer to supervise and monitor the nursing and midwifery system 2 PHN officers
ANM: auxiliary nurse-midwife; PHN: public health nurse; DPHN: district public health nurse ; MCH: maternal and child health; FW: family welfare; PHC: primary health centre; CHC: community health centre.
180
SECTION III
To expand the role of nurses in India, APN programmes should be established and should be included in manpower planning. The scope of practice should be clearly identified by the INC.
9. Ensure appropriate facilities and adequate medical equipment and supplies

Facilities, medical equipment and supply form the infrastructure required for providing health service. The health care facility should have a standard for rooms and space for outpatient departments and inpatient wards, and a standard for essential medical equipment and supplies. Good environmental sanitation and waste management can reduce the outbreak of infectious diseases such as hepatitis B and C, and reduce injuries and health risks such as needlestick injury. Adequate medical equipment and supplies provide the patient with proper treatment and care, reduce nursing time and the rate of infections. However, there should be a good maintenance and control system. The National Infection Control Committee of Thailand sets a minimum standard for health care facilities such as room, sink, isolation room, incinerator room to prevent infection.
cation is essential for self-development, knowledge-building and learning. In the US as well as in Thailand, nurses are required to have continuing education credits for renewal of their licences. Continuing education stimulates nurses to keep up with new knowledge and technology, to increase their skills and competency, and to be able to contribute to the health care team. The nursing service department or hospital should formulate a policy on staff development and set aside a budget to strengthen their competency in providing quality nursing care. This is an incentive for nurses. The existing continuing nursing education programmes should be strengthened or new units established. The appointment of responsible persons for continuing education activity is needed. Continuing education programmes should get approval from the INC.
12. Strengthen payment scales, incentive systems and working conditions

Emigration of nurses has been a critical issue in recent years in many countries including India, because developed countries faced with nursing shortages import skilled nurses from developing countries by offering higher salaries. In addition, there is evidence of nurses resigning from the workforce. Both events affect the number of nursing personnel, which is already inadequate. A shortage of nurses and understaffing have been linked to many negative consequences including increased incidence of cross-infection rates, accidents, injuries and poor delivery of services. Factors encouraging Caribbean nurses to emigrate were financial, poor working conditions, lack of opportunities for professional development, non-involvement in decision-making and lack of support from supervisors. Therefore, good payment and incentive systems, and better working conditions should be established. The payment scales of nurses in many countries are low compared to other health care workers. Payments should reflect education, type of work, roles and responsibilities, and workload. A nurse's job requires good knowledge, skills, hard work and commitment. In India, the payment scale was adjusted in 1996 but is still low compared to other professions; it should be increased. An incentive system for nurses could include allowances, uniform, housing, reimbursement for health care services, extra payment for working in the evening and night shifts or overtime, or working at remote or unsafe areas, and opportunities for continuing education. Transportation and safe housing for nurses who work in the community or remote areas should be provided for the convenience and safety of home care service. Recognition should be given to good workers at the institutional, local and national levels. Opportunities to obtain a higher degree, short-course training or to attend nursing conferences or workshops should be given to each nurse at least once a year for self-improvement and career advancement. Potential nurses should be encouraged to study for a higher
10. Promote evidence-based practice and nursing research

Establishment of policies on the use of evidence in practice is required. Nurses with a Master's degree should be encouraged to provide evidence, read nursing research and use evidence to improve or change nursing practices. An academic atmosphere should be created in the workplace. An information system and library should be provided. Multidisciplinary research should be encouraged. At the hospital, there should be a person who is responsible for nursing research activity including fund seeking for research and building of research network. Nurse educators should develop a short-course training on evidence-base and research or to supervise research activity. Resources such as journals and books can be shared. Joint research between nurse educators and clinical staff should be encouraged to strengthen the capacity of both groups and improve education and practice. The INC can be a part of nursing research development. The INC should set nursing research priorities in collaboration with nursing and non-nursing organizations to provide research funds and promote nursing activities for policy formulation. Establishment of a nursing research information system is encouraged to monitor research work, areas of research and researchers. Dissemination of nursing research and models for best practices should be established.
11. Establish a continuing nursing education system

Continuing education is an informal study or activity to gain knowledge and learn about new technology. Lifelong edu-
181
SECTION III
degree and take study leave with pay. Career ladders for promotion of nurses should be established at the national level. Therefore, each nursing service must have a human resource development plan and implement it effectively. Good working conditions including adequate and appropriate working facilities, cleanliness and safety can also facilitate productive work and the quality-of-life of nurses.
13. Ensure quality of nursing education by strengthening nursing programmes, increasing qualified nurse educators and allocating appropriate resources to maximize efficiency and effectiveness
Education is a key factor for human resource development. With good education, people can learn and earn money. Education programmes should be reviewed intensively and revised. There should be only 2 levels of nurses, first, a professional nurse who studies for four years in the Bachelor of Science programme in nursing offered at the university level. The admission criterion is 12 years' schooling. Second, an auxiliary nurse who studies for 2 years in the certificate Auxiliary Nursing programme offered at the school of nursing. There should be a continuation Bachelor of Nursing Science so that those from diploma school can upgrade their qualification. Auxiliary nurses should be able to continue their studies in the Bachelor of Nursing Science as well. The Master of Science programme in nursing should focus on advanced nursing practice. The INC has set standards and syllabi for all nursing programmes. However, the roles and responsibilities of nurses at each level should be clearly defined, and the curriculum structure and training experience may have to be revised. Inspections for nursing education institutions are being carried out by the INC. A workshop for inspectors should be held to discuss common issues in nursing education, review the inspection process and revise the inspection criteria and guidelines. The Thailand Nursing Council accredits nursing education every 1-4 years using criteria similar to the quality criteria used by the QA system of higher education. The quality components are (i) vision, mission, objective, strategic and operational plan; (ii) teaching, learning including educators, teaching learning activity, evaluation method; (iii) student development including financial support, student activities, alumni; (iv) research including a number of projects, grants, publications, utilization; (v) community service including projects, outcome; (vi) cultural and environmental reservation; (vii) finance and budgeting; (viii) administration including leadership, supportive staff, management system; and (ix) QA system. The quality of education depends on the quality of the educators. The teacher for the BSc programme in nursing should be at least a Master's degree holder and have teaching experience as prescribed by the INC. The teacher at the graduate level should do research and publish at least one article every two years. Educators should coordinate closely with the nursing staff in hospitals to achieve education that is relevant to the needs of the service. Educators should collabo-
rate with the nursing service in research and nursing service development. The teaching-learning activity should emphasize participatory learning and cultivation of lifelong education. Infrastructural needs such as a library, information technology system and nursing laboratory should be of good quality. In addition to learning activity in the classroom, students should participate in extracurricular activities such as sports, music, student clubs, social work and community development projects. The curriculum should be revised regularly, and alumni and stakeholders should be involved in the process so that the curriculum meets the demands of society. A nursing development plan should be developed at each nursing institution and at the national level. Effective nursing education management requires planning to develop a group of nursing education leaders with the involvement of policy-makers.
14. Expand the role and authority of the Indian Nursing Council on Nursing development by revision of the Act, Restructuring and Networking
To maintain control of the quality of practice, the Indian Nursing Council Act and regulation may be reviewed and revised. Standards of implementation should be enforced. Control over the State Nursing Councils should be considered and a clear line of command initiated. Networking with other nursing professionals is necessary and should be strengthened to create unity and power for nursing development. Strategies should be developed to work with the Ministry of Health & Family Welfare, Division of Nursing Service or other organizations both within and outside the country to improve the quality of nursing and of nurses themselves.
Recommendations
Policy level
1. Manpower planning and development for nursing must be an integral part of human resource planning of the health system and should involve nursing experts and stakeholders. 2. A study on nursing manpower should be carried out to support health manpower planning and development. This would include projections on need, production, deployment and utilization to respond to the required changes in health care reforms. 3. Adequate positions must be created for nurses working in hospitals and the community to facilitate population coverage, accessibility and quality care. 4. Budget allocation should be done for human resource development, research and infrastructure. 5. Pay scales, working conditions and incentive systems should be improved 6. A policy on QA system for health care should be established and implemented.
182
SECTION III
Nursing education
1. Planning and development for nursing education must be done at the national level. It should involve the nursing, professional organizations and stakeholders. 2. A collaborative research network should be established to strengthen research in nursing. 3. The nursing curriculum at all levels should be reviewed and revised based on nursing competencies. 4. Nursing education should be upgraded to BSc, MSc and PhD levels. 5. The qualifications of the teaching faculty should be strengthened, and facilities and equipment provided to facilitate quality nursing education. 6. Clinical practice and supervision should be strengthened. 7. Nurses should be empowered so that they can be involved in policy decisions by enhancing leadership, communication and public speaking skills. 8. There should be conformity of nursing education standards with QA systems. 9. APN programmes should be created to train nurse specialists. 10. Creative, critical thinking and innovation in education and practice must be encouraged.
continuing education, training or direct experience and a mentor system.
Community health nursing

The NHP 2002 states that public health delivery centres need to make a beginning by increasing the number of nursing personnel. Therefore, to overcome the shortages and as per NHP 2002 plans, there is a need to modify nursing staffing norms to provide essential health interventions to the community health nursing services at various levels.
Professional organizations and regulation

1. Establish a system for renewal of licenses. 2. Amend the Indian Nursing Council Act, 1947 for autonomy of functioning so that the INC can enforce the standards for regulating nursing education and service.. 3. Develop a comprehensive information system on nursing and midwifery in the areas of education, service, clinical practice and management of workforce. 4. Conduct an assessment of the nursing and midwifery manpower according to the need for education and service. 5. Develop an in-service education centre for nursing. 6. Ensure active participation of nurses/midwives in multidisciplinary teams to advocate health regulation. 7. Increase awareness among nurses and midwives on councils and regulations. 8. Strengthen the infrastructure and manpower in the INC, i.e. create positions of Joint Secretary and Deputy Secretary (Nursing). 9. Create positions of Nurse Registrar and Deputy Nurse Registrar in the State Nursing Councils.
Nursing service
1. Identify clearly the roles and responsibilities of nurses at each level. 2. Establish a QA system for the nursing service and ensure implementation of care standards and norms. 3. Create the post of Advanced Nurse Practitioner. 4. Establish networking among nursing directors and educators to develop and implement nursing service planning and development. 5. Strengthen independent nurses' role in the health care service. 6. Demonstrate to the public the quality of nursing services at all levels. 7. Enhance continuing education for nurses to improve quality care. 8. Plan to budget for appropriate equipment and facilities. 9. Utilize research findings and evidence-based nursing practice. 10. Establish training programmes for independent nurse practitioners. 11. Create positions for independent nurse practitioners. 12. Submit a proposal to request for the improvement of working conditions, pay scales and incentives. 13. Improve leadership and management skills of nurses by
Conclusion
Nurses and midwives can make major contributions to health care development and achieve the Millennium Development Goals only if there is strong support at the policy level to ensure policy implementation. Strong commitment and close collaboration between professional organizations, nursing service institutes and educational institutes are needed in planning, implementation and evaluation of nursing workforce management. Maximal use of resources within the country is essential. Best practices from each State need to be shared, learned and recognized. In addition, nurses and midwives should commit themselves to continuously improve the quality of nursing services by strengthening their competencies.
183
SECTION III
Bibliography
Anderson RD, Sweeney SD, William AT. An introduction to management science: Quantitative approach to decision making. New York: South-western College Publishing; 2000. Bateman ST, Snell AS. Management: Building competitive advantage. New York: Irwin Mc Graw-Hill; 1999. Indian Nursing Council. Teaching material for quality assurance model: Nursing quality and commitment. Indian Nursing Council. Master of nursing: Syllabus and regulation. Indian Nursing Council. The Indian Nursing Council Act, 1947. New Delhi. (Amended in 1950 and 1957). Indian Nursing Council. Syllabi and regulations for the courses of studies for auxiliary nurse and midwife. New Delhi; 1977. Indian Nursing Council. BSc Nursing syllabus and regulations (basic programme). 3rd ed. New Delhi; 1981. Indian Nursing Council. Golden jubilee celebration: Nursing in the new millennium. Daryaganj, New Delhi: Jaina Offset Printers; 2000. Indian Nursing Council. Regulations; 2001. Indian Nursing Council. Syllabus and regulations for Diploma in General Nursing and Midwifery, New Delhi; 2001. World Health Organization. Nursing and midwifery workforce management: Analysis of country assessment. New Delhi: Regional Office for South-East Asia; 2003. Indian Nursing Council. Syllabus for post basic BSc Nursing. New Delhi; 2001. National Health Policy 2002. New Delhi: Ministry of Health & Family Welfare; 2002. Government of India. Report of expert committee on health manpower planning, production and management. New Delhi: Ministry of Health and Family Welfare; 1987. Government of India. Report of the Xth Plan working on development of nursing service. New Delhi: Ministry of Health and Family Welfare. The Trained Nurses' Association of India. Indian nursing year book 2000-2001. Noida: Academy Press; 2000. World Health Organization. Macroeconomics and health: Investing in health for economic development. Newton, MA: Digital Design Group; 2001. World Health Organization. Nursing and midwifery workforce management: Conceptual framework. New Delhi: Regional Office for South-East Asia; 2003. World Health Organization. Nursing and midwifery workforce management: Guidelines. New Delhi: Regional Office for South-East Asia; 2003.
184
SECTION III
Access to Essential Drugs and Medicine
N DEVELOPING COUNTRIES, HEALTH CARE HAS BEEN A NEGLECTED ISSUE IN the overall policy framework. With low public budgets, providing a universal social security cover to the population is difficult. On the other hand, households spend a sizeable portion of their income on food items, leaving little for health care. The single most vital component of health care is drugs, as they account for a substantial part of household health expenditures. The market for drugs, particularly the allopathic category, has been growing rapidly in India-in terms of production, trade, investment and employment. However, the industry is characterized by supplierinduced demand (and therefore loss of consumer sovereignty), uncertain demand for the patients, oligopoly elements, monopoly profit, etc. This has far-reaching implications on the health care of the masses, whose essential problem lies in lack of purchasing power, lack of access and knowledge of modern medicine. In view of the above, the specific objectives analysed in this chapter are the following: 1) Provide an estimate of drug expenditure patterns of both the government and households across States; 2) Examine the pattern of drug production; 3) Investigate whether the Indian drug industry is highly concentrated across therapeutic classes; 4) Assess the price change in drugs over the years, under different policy regimes; 5) Examine issues relating to drug regulation; 6) Critically evaluate public procurement of essential drugs, needed for the public health system; 7) Examine the impact of the patent regime since the 1970s and the likely consequences of Trade-Related Intellectual Property Rights (TRIPS) in the post-2005 period;
Analysis of state-wise drug expenditure in India

Share of household expenditure on drugs and medicines
Drugs and medicines form a substantial portion of out-of-pocket spending on health among households in India. Estimates from the National Sample Survey (NSS) for the year 1999-2000 suggests that over 5% of the total consumption expenditure of households went into health spending. However, there are significant variations among different categories of the population. For instance, Table 1 shows that the share of health in the total expenditure of households in rural areas is little over 6%, while that for urban India is little less than 5%. An analysis across States indicates that Kerala, which is one of the highly advanced State in terms of health indicators, spends a relatively larger share of household expenditure on health, both in rural and urban areas. Bihar and Assam, which are the poorest States in the country, spend relatively less. Table 2 depicts annual out-of-pocket health expenditure across states in India during 19992000.
S. SAKTHIVEL
INSTITUTE OF ECONOMIC GROWTH, UNIVERSITY OF DELHI ENCLAVE, NORTH CAMPUS, DELHI 110007 E-MAIL: sakthivel327@hotmail.com
Household drug expenditure in India

Estimates from the 55th consumption expenditure survey reveal that three-fourths of the total out-of-pocket (OOP) health expenditure is spent on drugs, in rural and urban areas. Tables 3 and 4 show that drug spending is high in lesser-developed States
185
SECTION III
Table 1 Share of health to total household expenditure

State Rural Urban (in %) Aggregate
Andhra Pradesh Assam Bihar Delhi Goa Gujarat Haryana Himachal Pradesh Jammu and Kashmir Karnataka Kerala Madhya Pradesh Maharashtra Orissa Punjab Rajasthan Tamil Nadu Uttar Pradesh West Bengal All India
6.56 2.47 4.40 4.57 4.28 5.03 6.99 5.25 2.90 4.58 7.79 6.05 7.50 5.46 7.66 4.79 5.80 8.20 4.64 6.05
4.13 4.04 2.96 3.34 5.16 4.22 6.56 3.91 3.61 4.17 7.15 5.25 5.98 4.51 5.60 4.70 4.45 5.64 4.84 4.91
5.60 2.83 4.15 3.40 4.76 4.63 6.84 5.04 3.12 4.37 7.59 5.74 6.59 5.24 6.87 4.76 5.02 7.45 4.73 5.57
(except Himachal Pradesh) such as Orissa (90.56%), Bihar (88.26%), Rajasthan (87.67%), Jammu and Kashmir (87.09%) and Himachal Pradesh (87.14%). Economically advanced States such as Maharashtra, Gujarat, Tamil Nadu and Karnataka reportedly spend less. Estimates further show that out of per capita expenditure of households amounting to Rs 577 spent annually on health in urban India, Rs 400 goes into buying drugs, accounting for around 70%. In rural India, however, the share was 77%, while the spending pattern has been Rs 380 and Rs 295, respectively. Kerala, Haryana and Goa-all small States, appear to be spending over Rs 600 per annum per capita in urban areas while households in poor States such as Bihar and Orissa spend relatively less. Tables 5 and 6 further indicate that out of Rs 400
Box 1 Household health spending in India: Key statistics

In 1999-2000, 5% of total household consumption expenditure went into health spending; drugs accounting for the bulk of 4%. Rural households spend over 6% and urban households 5%. Kerala (4% of GSDP) is at the top of the spenders' list while Bihar and Assam incur relatively less (<1% of GSDP).
Source: Extracted from the Unit-level Records of Consumer Expenditure Survey, 55th Round of NSS, 19992000
Box 2 Household drug expenditure

Household drug spending is high in less developed States- - Orissa (90.64%); Bihar (89.14%); Rajasthan (89.43%); Jammu and Kashmir (90.39); while Advanced States spent less on drugs-Maharashtra (68.75%); Gujarat (63.90%), Karnataka (68.75%) and Tamil Nadu (61.41%). Urban India spends around 70% of OOP expenditure on drugs and 77% in rural India. In India, the share of drugs in total outpatient treatment is 83% in rural and 77% in urban areas. In India, the share of drugs in total inpatient treatment is 56% in rural and 47% in urban areas.
Table 2 Annual household out-of-pocket health expenditure (19992000)

State Rural (Rs in crore)
Urban Aggregate GSDP/GDP % to GDP/GSDP
Andhra Pradesh 2105 872 2976 125236 Assam 301 143 444 29263 Bihar 1609 234 1842 72083 Delhi 51 706 757 52914 Goa 35 51 86 6749 Gujarat 1065 882 1948 106427 Haryana 901 452 1353 48270 Himachal Pradesh 262 37 299 11983 Jammu and Kashmir 182 104 286 13961 Karnataka 1008 907 1915 96179 Kerala 1834 763 2597 62514 Madhya Pradesh 1418 790 2207 99322 Maharashtra 2639 3154 5793 241410 Orissa 781 198 979 36283 Punjab 1135 522 1656 62361 Rajasthan 1413 654 2067 80019 Tamil Nadu 1343 1430 2773 126500 Uttar Pradesh 6323 1804 8127 187641 West Bengal 1516 1222 2738 127933 All India 27280 15576 42856 1761838
2.38 1.52 2.56 1.43 1.28 1.83 2.80 2.50 2.05 1.99 4.15 2.22 2.40 2.70 2.66 2.58 2.19 4.33 2.14 2.43
Box 3 Government budget expenditure on drugs

Approximately Rs 2000 crore incurred by both the Central and State Governments during 2001-02. The share of drugs in the health budget in the Central Government is around 12%. Southern States incurred the highest expenditure, with Kerala and Tamil Nadu spending around 15% each. Assam, Bihar, Uttar Pradesh and Orissa spent about 5% or less on drugs and medicines.
Note: Applied per capita figures to mid-year survey population of Census Source: Extracted from the Unit-level Records of Consumer Expenditure Survey, 55th Round of NSS, 19992000
186
SECTION III
Table 3 Share of drugs in inpatient and outpatient expenditure of rural households

State Inpatients Drugs to Inpatients Inp to OOP Outpatients Drugs to outpatients Out to OOP (in %) Share of Drugs in OOP
Table 4 Share of drugs in inpatient and outpatient expenditure of urban households

State Inpatients Drugs to Inpatients Inp to OOP Outpatients Drugs to outpatients Out to OOP (in %) Share of Drugs in OOP
52.66 46.31 68.46 67.02 58.72 44.11 41.45 68.94 77.78 50.18 49.42 67.53 52.92 78.88 53.62 71.99 40.45 70.25 58.72 56.04
24.06 28.66 11.48 15.17 24.26 35.29 25.90 25.37 11.76 26.31 34.21 20.52 26.70 15.88 23.96 20.75 25.67 13.09 17.55 21.52
78.68 80.42 91.83 60.90 85.75 74.69 89.16 95.76 92.07 75.39 83.48 84.83 74.52 92.86 87.62 93.99 68.65 89.25 75.91 83.17
75.94 71.34 88.52 84.83 75.74 64.71 74.10 74.63 88.24 73.69 65.79 79.48 73.30 84.12 76.04 79.25 74.33 86.91 82.45 78.48
72.42 70.65 89.14 61.83 79.19 63.90 76.80 88.96 90.39 68.75 71.83 81.28 68.75 90.64 79.47 89.43 61.41 86.76 72.89 77.33
40.20 45.13 64.16 57.72 58.07 50.34 46.99 80.49 62.07 39.27 40.97 60.63 48.21 83.26 31.89 61.27 36.63 59.61 39.18 47.34
19.67 27.35 11.80 37.09 34.29 30.02 34.27 24.04 23.55 32.70 37.21 22.03 30.18 21.05 26.22 18.05 31.63 17.00 22.72 26.64
78.99 77.29 84.57 81.52 82.12 77.80 91.55 72.46 87.26 64.07 77.73 83.18 63.78 92.13 88.83 88.86 72.92 85.95 76.21 77.10
80.33 72.65 88.20 62.91 65.71 69.98 65.73 75.96 76.45 67.30 62.79 77.97 69.82 78.95 73.78 81.95 68.37 83.00 77.28 73.36
71.36 68.49 82.16 72.69 73.87 69.56 76.28 74.39 81.33 55.96 64.05 78.21 59.08 90.26 73.90 83.88 61.44 81.47 67.80 69.18
spent on drugs in urban India, a substantial part of it is by way of outpatient payments totalling around Rs 325 and the remainder (about Rs 75) is on account of inpatient payments. A similar pattern can be observed in rural India. Out of the total of Rs 295 spent on drugs annually, inpatient expenses accounted for Rs 45, and outpatient expenses were about Rs 250. Results from the 55th National Sample Survey (NSS) consumption expenditure survey also reveal that the share of drugs in total outpatient treatment is extremely high. In rural India, the share of drugs is observed to be the highest, accounting for nearly 83%, while in urban India, this worked out to 77%, as depicted in Tables 2 and 3. In fact, in a few states, the share of drugs is more than 90% (Bihar, Himachal Pradesh, Jammu and Kashmir, Orissa and Rajasthan in rural areas and Haryana and Orissa in urban areas). On the other hand, the share of drugs in inpatient treatment is not as high as in the outpatient category. The respective share of drugs (inpatient) in rural and urban India was roughly 56% and 47%. It is interesting to note that in rural India, if both inpatient and outpatient expenses are taken together, the share of drugs to total household expenditure accounts for roughly around 5% while in urban India, it is around 3.5%. Overall, as indicated earlier, it appears that little over 4% of the OOP spending of households goes into buying drugs.
Government expenditure on drugs

The magnitude of expenditure incurred on drugs by households does not show a similar pattern in public expenditure. The component of drugs and medicines in the overall budget of both the Central and State Governments is only a minor share, as salaries account for the bulk of the health sector expenditure in India. The analysis involves 16 major Indian States, which accounts for roughly 85% of the total health budget in the country. The expenditure pattern on drugs of the State Government, as depicted in Table 7 shows that there are wide-ranging differences across States, from as little as less than 2% in Punjab to as much as 17% in Kerala during 2001-02. The southern States such as Kerala and Tamil Nadu spend over 15% of their health budget on drugs. Many backward States, both in economic and health indicator terms, incurred the lowest expenditure on drugs. States such as Assam, Bihar, U.P. and Orissa spent about 5% or less of their health budget on drugs and medicines. It appears from the analysis that approximately Rs 2000 crore was spent in India by the State and Central Governments together on procuring drugs and medicines during 2001-02. The Central Governments share of drugs in its total health budget is around 12%. In all, roughly 10% of the health budget goes into procuring drugs in India.
187
SECTION III
Indian pharmaceutical sector: An overview

The pharmaceutical industry has witnessed tremendous transformation since the 1950s. The size of the Indian pharmaceutical industry, both bulk drugs and formulations is estimated at Rs 35,471 crore in 2003-04 (IDMA 2004), which is just over 1% of the global market (ICRA 1999). This is against the value of the production of pharmaceuticals of a mere Rs 10 crore in 1950 (Narayana 1984). At present, there are about 6,000 units operating in this sector (if only bulk drugs, formulations and large parenterals are taken into account) (Mashelkar Committee 2003). Investment in the industry has steadily grown over the years from a mere Rs 23.64 crore in 1950 to a moderate Rs 500 crore in 1980 and went up considerably to reach around Rs 4000 crore in 2003. Propelled by the booming demand, the production of pharmaceuticals has registered a tremendous increase over the years. The growth rate of bulk drugs recorded in the 1970s and 1990s is almost double-around 20%-that of the production registered for the 1980s is evident from the Table 8. The output of formulations has seen a phenomenal increase during the period under consideration but is less than 4% as against bulk drugs, in both the 1970s and 1990s. The 1980s is the only period in which formulation growth had outperformed the growth of bulk drugs by a marginal 1%. The massive growth of the pharmaceutical industry could be
Fig 1 Trends in the production of bulk drugs and formulations in India since the 1970s
attributed to a few domestic and international developments that took place particularly since the 1950s. At the global level, the industry in general was then experiencing a major overhaul by vertically integrating operations such as production, marketing and research. The protection given to the pharmaceutical industry through patents and brand names saw many top companies switch over to the production of specialty medicines.
Table 5 Percapita annual drugs and other medical expenditure (rural)

(Rs.) State Drugs Inpatient Total Drugs Outpatient Total Total drugs Aggregate OOP
49.47 17.70 17.00 53.44 73.50 56.17 67.06 86.09 23.15 39.11 134.07 43.14 68.84 32.16 92.89 51.21 39.57 45.70 27.92 45.91
93.94 38.23 24.83 79.74 125.16 127.34 161.81 124.87 29.76 77.95 271.31 63.88 130.09 40.77 173.23 71.14 97.84 65.05 47.55 81.93
233.34 76.54 175.86 271.55 335.05 174.42 412.81 351.80 205.54 164.62 435.63 209.94 266.10 200.56 481.69 255.45 194.50 385.26 169.56 248.53
296.57 95.18 191.52 445.92 390.74 233.52 463.01 367.39 223.25 218.37 521.82 247.49 357.07 215.99 549.77 271.78 283.31 431.68 223.38 298.83
282.81 94.25 192.86 324.99 408.55 230.59 479.88 437.89 228.69 203.73 569.70 253.07 334.94 232.72 574.58 306.66 234.07 430.95 197.48 294.44
390.51 133.41 216.35 525.66 515.90 360.86 624.83 492.26 253.01 296.31 793.13 311.37 487.16 256.76 723.01 342.92 381.15 496.72 270.94 380.76
Sources: Extracted from the Unit-level Records of Consumer Expenditure Survey, 55th Round of NSS, 19992000
188
SECTION III
Table 6 Percapita annual drugs and other medical expenditure (urban)

State Drugs Inpatient Total Drugs Outpatient Total Total drugs Aggregate OOP
(Rs.)
34.14 55.39 19.31 127.95 162.98 76.87 129.17 128.62 64.54 68.36 142.83 66.59 118.66 66.51 56.31 57.80 65.29 55.27 50.21 72.76
84.91 122.74 30.10 221.68 280.65 152.70 274.87 159.80 103.98 174.08 348.62 109.83 246.14 79.89 176.58 94.34 178.24 92.73 128.13 153.68
273.94 251.96 190.20 306.51 441.53 276.92 482.63 365.97 294.49 229.55 457.30 323.41 363.23 276.13 441.35 380.54 280.93 389.11 332.13 326.27
346.81 325.99 224.90 375.97 537.69 355.92 527.21 505.06 337.48 358.29 588.28 388.81 569.46 299.70 496.85 428.23 385.28 452.72 435.78 423.16
308.08 307.35 209.51 434.46 604.51 353.80 611.80 494.60 359.04 297.91 600.13 390.00 481.89 342.64 497.66 438.34 346.22 444.38 382.34 399.02
431.72 448.73 255.00 597.65 818.34 508.62 802.08 664.86 441.46 532.37 936.90 498.64 815.60 379.60 673.43 522.57 563.52 545.45 563.91 576.83
Sources: Extracted from the Unit-level Records of Consumer Expenditure Survey, 55th Round of NSS, 19992000
Table 7 State-wise government drug expenditure in India (200102)

State Andhra Pradesh Assam Bihar Chhattisgarh Gujarat Haryana Karnataka Kerala Maharastra Madhya Pradesh Orissa Punjab Rajasthan Tamil Nadu Uttar Pradesh West Bengal Central Govt.* All India* Drugs 7923.09 0.00 1996.90 1822.47 1253.76 N.A. 6927.17 N.A. 10.00 3965.86 1768.98 N.A. 3952.80 16428.68 5938.25 5005.25 56993.21 Materials and supplies 4781.45 1530.10 206.29 680.22 1440.06 3096.12 856.82 12420.68 20295.91 3956.04 361.30 916.32 5092.25 1668.57 1166.04 793.23 72649.23 131910.63 Total 12704.54 1530.10 2203.19 2502.69 2693.82 3096.12 7783.99 12420.68 20305.91 7921.90 2130.28 916.32 9045.05 18097.25 7104.29 5798.48 72649.23 188903.84 (Rs in lakh) Health Expenditure (Rev.) 131424.08 32690.82 71348.49 22587.10 71547.95 31470.98 98633.19 72931.59 178379.51 66689.30 42135.78 61826.45 97311.61 118432.85 135578.81 131948.35 597700.00 1962636.86 Drug Expenditure as % of Health Expenditure 9.67 4.68 3.09 11.08 3.77 9.84 7.89 17.03 11.38 11.88 5.06 1.48 9.29 15.28 5.24 4.39 12.15 9.63
Note: Many states report drug expenditure under the category of Materials and supplies. Materials and supplies include hospital accessories, beding cloth, materials supply, laboratory cahrges. charges, Others and X-ray materials. Here we have included materials supply only. * Includes only 16 states total reported in the table, which account for around 85%. ** The drug budget for the Central Government includes expenditure incurred on four National ProgrammesBlindness Control Programme, TB Programme, Leprosy Programme and Vector Borne Disease Control Programme. Source: Budget documents, respective State and Central Government
189
SECTION III
Table 8 Growth rate of bulk drugs and formulations production in India since the 1970s
Growth of production 1970s 1980s 1990s 200003 19702003
Bulk drugs Formulations Total production
20.28 16.88 17.39
10.08 11.07 10.91
19.49 16.42 16.95
19.76 14.68 15.72
12.38 11.05 11.17
Note: All values are percentages. Growth rates refer to compound growth rates based on current prices. Source: Computed on the basis of IDMA, various issues.
This resulted in the invention and introduction of products ensuring high growth and monopoly profits. Radical invention and introduction of new drug technologies stimulated the industry to transform massively. Unprecedented attention to wonder drugs through the magic-bullet technology shifted the focus from treating the symptoms to healing the disease itself. On the domestic front, the government intervened by establishing a few public-owned life-saving and essential drugs-producing companies. The Indian Patents Act, 1970, the FERA (Foreign Exchange Regulation Act), 1973, acted as a boost to the domestic growth of this all-important industry.
have flooded the market. Take for instance, the top twentyfive formulations sold in the Indian market in 1999. Of the top 10 products, two belong to the category of irrational vitamin combination and cough syrup while the other drug is a useless liver drug. Ten of the top 25 products sold in India in 1999 belonged to either one of these categories: blood tonic, cough expectorant, non-drug, analgesics, nutrients, liver drug, etc. which are either hazardous, non-essential or irrational. These ten inessential and irrational drugs together accounted for nearly 10% of the total value of 300 products. Table 10 clearly sums in part the changing pattern of drug requirements according to the shifting disease profile of India. Lifestyle drug categories such as cardiovascular drugs, hormones, nutraceuticals are growing in magnitude with every passing year. These categories together accounted for over one-fifth of the total pharmaceutical sales in 2002.
Fig 10 Market share of drugs by therapeutic segment (2002)

( Rs crores) Therapeutic Segments Total retail sales Percent to total market
Skewed production priorities

Even though drug production witnessed a phenomenal upsurge during the past three decades of the century, the production and sale of products also reflected market potential. [However, the market potential for drugs is largely induced-through marketing, advertising and distributional network.] Thus, drug production mainly catered to those who have enough purchasing power. Table 9 above exposes skewed production priorities by the drug industry. Irrational, non-essential and hazardous drugs
Table 9 Pattern of pharmaceutical sales in India, 1999

Product rank Products Sales Market Product (In crore) share(%) description
Alimentary system Cardiovascular system Central Nervous system Musculoskeletal disorders Hormones Genitourinary system Infections and infestations Nutraceuticals Respiratory system Eye Allergic disorders Skin Metabolism Total market
2137 2182 1157 1367 2362 738 3726 463 1734 273 31 911 21 17,102
12.50 12.76 6.76 7.99 13.81 4.31 21.79 2.71 10.14 1.60 0.18 5.32 0.12 100.00
Source: Calculated from ORG-Retail Sales Audit, June 1999 Note: The above analysis is based on the leading 300 products reported in the source. The value of these leading products accounts for 46.81% of the total market value.
1 3 9 11 12 17 20 21 22 25
Becosules Corex Liv-52 Dexorgange Digene Combiflam Polybion Glucon-D Evion Revital
79.42 61.27 62.17 47.40 46.69 43.05 40.76 39.66 39.19 38.98
1.39 1.07 1.09 0.83 0.82 0.75 0.71 0.69 0.69 0.68
Irrational vitamin combination Irrational cough mixture Useless liver drug Blood tonic Needless antacid Irrational analgesic combination Irrational vitamin combination Useless nutrients Irrational vitamin combination Oral ginseng tonic
Source: Compiled from ORG-Retail Sales Audit, June 1999
Expensive pre-digested nutritious supplements such as protein foods, malt tonics, various vitamins, calcium, haemoglobin, iron, etc. form a notable share of the total market sales of drugs. Similarly, dehydration caused by diarrhoea, particularly in children, also accounts for innumerable prescriptions, although dehydration can be easily treated with the combination of simple household items such as water, salt and sugar or oral rehydration solution (ORS). Similarly, the share of the antituberculosis drug market accounts for a meagre 2%, while a significant burden of disease and death in India is caused by tuberculosis. In this context, it is interesting to note the contribution made by different players in the market. Table 11 provides an
190
SECTION III
Table 11 Contribution of essential and inessential drugs by domestic and multinationals firms
(Rs in crore) 1985-86 Formulations Top 50 companies MNCs Domestic companies Top 50 companies 1998-99 MNCs Domestic companies
Total turnover Turnover of few inessential, irrational drugs Antacid, antiflatuent Vitamins Anti-anaemia preparation Rubs and balms Cough and cold preparations Turnover of few essential drugs Antibiotics Anti-tuberculosis drugs Vaccines Anti-diabetic drugs Antiseptic, anti-infective Antiparasitic (amoebicides)
1008.7 573.2(56.82) 276.2 195.1(70.7 ) 30.0 19.2(64.0) 91.2 76.1(83.4) 42.0 18.5(44.1) 12.3 12.3(100) 48.7 37.5(77%) 327.8 128.8 (39.3) 225.4 81.1(36) 20.6 4.0(19.4) 1.5 0.5(33.3) 10.4 9.4(90.4) 21.5 10.8(50.2) 16.7 5.8(34.7)
435.5(43.17) 81.1(29.3) 10.8(36.0) 15.1(16.6) 23.5(55.9) _ 11.2(23) 199(60.7) 144.3(64) 16.6(80.6) 1.0(66.7) 1.0(9.6) 10.7(49.8) 10.9(65.3)
2113.29 1146.16(54.24) 591.68 511.97(86.53) 157.50 77.99 (49.45) 260.35 260.35 (100) 47.40 47.40(100) 32.09 32.09(100) 94.14 94.14(100) 690.68 198.34(28.72) 522.51 129.08 (24.70) 38.88 _ 32.59 32.59(100) 36.67 36.67(100) 29.14 _ 30.89 _
967.12(45.76) 79.71(13.47) 79.71 (50.55) _ _ _ _ 492.34(71.28) 393.43 (75.30) 38.88(100) _ _ 29.14(100) 30.89(100)
Note: Figures in parentheses denote the percentage share between domestic and multinational corporations. The above figures relate only to the top 50 products analysed among 300 products listed in the source. Source: Computed from ORG-Retail Sales Audit, March 1986 and June 1999
analysis based on the top 50 products for the period 198586 and 1998-99. While examining the leading 50 formulations, only products that fall under the categories mentioned in Table 10 are taken up for consideration here while the rest are excluded from the analysis. The dominant proportion of transnational corporations in the production and sale of inessential and irrational combinations of drugs is apparent from Table 11 wherein vitamins, rubs and balms, and cold and cough preparations account for 100% in the production in 1998-99, mirroring almost similar trends witnessed in 1985-86. The strength of domestic Indian companies lies in categories such as antibiotics, anti-tuberculosis and antiparasitic, anti-infective and antiseptic preparations. The analysis clearly demonstrates that while multinationals concentrate on high-value, low-volume products, the domestic industry, on the other hand, concentrates on high-volume, low-value products. The former promotes brand names while the latter by generic brands.
maceutical industry to a marginal Rs 4 lakh during the period. But the strength of the Indian pharmaceutical trade lies in the formulations market due to its cost advantage. Compared to bulk drugs, the export of formulations steadily increased from nearly Rs 35 crore in 1980-81 to around Rs 413 crore in 1990-91 to a staggering Rs 3038.5 crore in 1998-99. The export of Indian pharmaceutical products witnessed a quantum jump in the 1990s; the growth rate was 32.85% (Table 12). The export of bulk drug items, however, has been on a very small scale for a long period in the history of the drug
Table 12 Growth of trade in pharmaceuticals in India in the 1980s and 1990s

Growth of pharmaceutical trade 1980s (%) 1990s (%)
Trends in the pharmaceutical trade

The role of the drug trade assumes importance, as India was historically dependent on drug imports when the domestic industry was in a nascent stage. The import of bulk drugs has been considerable and increased during the 1950s and 1960s, as formulations were prepared with basic bulk drugs, mainly by transnational companies. For instance, the import of bulk drugs and formulations registered an upward trend from Rs 13.17 crore in 1963-64 to Rs 37.54 crore in 197374. Nevertheless, export marked a sharp and significant increase during this period from a meagre Rs 2 crore to Rs 37.54 crore, thereby helping to reduce the trade gap in the phar-
Bulk drugs export Formulations export Total pharmaceutical export Bulk drugs import Formulations import Total pharmaceutical import
Note: Growth rate indicates the compound growth rate Source: Worked out from IDMA and OPPI Annual Reports
50.58 19.54 30.31 18.13 34.54 20.17
24.20 30.84 32.85 27.71 27.72 32.97
trade. However, trends in the late 1990s indicate a reversal of this trend. The amount of exports during 1998-99 was estimated at about Rs 2400 crore. Indian domestic pharmaceutical companies have made major inroads into the highly competitive generic segments of the world market. It is this market which is fetching a high value for Indian companies and
191
SECTION III
steadily building an excellent infrastructure network around the world. Bulk drug import has been a significant item in the basket of total imports in the 1950s and 1960s. Although the extent of its significance has undergone tremendous change, bulk drug still accounts for one-half to one-third of the total bulk drug consumption. In 1980-81, against the total bulk drug production of Rs 240 crore, import amounted to nearly Rs 90 crore. Further, trade figures reveal that during 1998-99, the total import of bulk drugs was roughly Rs 2000 crore as opposed to the domestic production of around Rs 3200 crore. From Table 11 it is apparent that the total import growth calculated for the 1990s point to a growth rate of around 33%.
Concentration in Indian Drug Industry

A casual observer might assume from Table 13 that the Indian pharmaceutical market is extremely competitive as even the top most firm could not garner more than 7% of the total market share, while the share of the top 10 companies is only around 30% (ICRA 1999). A comparative analysis of Table 13 reveals some interesting insights. Over a span of two decades, the contribution made by the top 10 players have come down from around 40% in 1976 to 30% in 1998. The other point that needs to be noted is that a majority of the leading companies in 1976 (7 out of 10) were multinational drug corporations. A complete reversal of the trend was seen in 1998, wherein 7 out of 10 top companies were domestic ones. However, a simple analysis of the above pattern is misleading because the market for drugs is not a homogeneous, single-product category but a multiproduct one. Thus, the market for pharmaceuticals can be subgrouped into a large number of independent submarkets (characterized by low
Table 13 Retail market share of top 10 pharmaceutical companies in India

Company Market share 1976 (%) Company Market share 1998(%)
Sarabhai Glaxo Pfizer Alembic Hoechst Lederle Parke Davis Abbott Ciba-Geigy Sandoz Total for above companies
7.1 6.2 5.9 4.2 3.6 2.5 2.3 2.3 2.3 2.2 39.6
Glaxo-Wellcome Cipla Ranbaxy Hoechst-Marrion-Roussel Torrent pharmaceuticals Alembic Wockhardt-Merind Lupin Labs Knoll pharmaceuticals Pfizer Total for above companies
6.7 4.2 3.5 3.2 2.4 2.4 2.3 2.3 2.3 2.3 31.6
cross-elasticities of demand). This is because the medicines prescribed for cardiovascular disease cannot be administered to a patient suffering from cancer. Consequently, one cannot observe drug manufacturers competing on an industrywide basis. The following paragraphs would give a fair idea of the concentration in the Indian pharmaceutical market, which is measured by (i) the dominant market share held by a handful of companies, in terms of sales, and (ii) dominance of a small number of products within each therapeutic class. Table 14 provides information on the market share enjoyed by leading drugs under each therapeutic class. It also shows the market share of drug companies manufacturing top products. Table 14 displays an extreme concentration persisting across therapeutic groups in the Indian drug industry. A detailed analysis by various therapeutic segments of drugs demolishes the claim of the industry lobby that competition prevails in the industry. The number of drugs covered in the ORG-MARG database under all the therapeutic segments for 1998-99 is the top 300 products, which is close to half of the total retail market in India. This is against an estimated 20, 000 drugs in the Indian drug market. The table apparently establishes the supremacy of a few companies and correspondingly a handful of its drugs in each therapeutic category. Out of 32 therapeutic classes considered in the analysis, in 19 markets, four and less than four companies retain dominant shares. Their respective market shares range from 30% to more than 90% in a few cases. For instance, the market for streptomycin points to an extreme concentration, wherein just one company commands the entire market (93.27%) while the class of vaccines, rubs and other inhalants, antiseptics and other penicillin markets are held by 3-4 companies, respectively. The share of top products also follows a similar pattern in various therapeutic groups. Closely followed by these patterns, another 13 therapeutic segments show less extreme concentration. Included in this category are 5-8 companies whose market share in each therapeutic market is in the range of 30%-70%. Another noteworthy pattern that emerges from Table 14 is that the element of oligopoly cuts across the entire spectrum of the therapeutic class, whether it is the case of essential drugs like antibiotics, anti-tuberculosis drugs or inessential drugs such as vitamins, cough and cold preparations, tonics, etc. The drug industry is extremely concentrated, debunking the theory that the drug market in India is competitive.
Price, procurement and regulation of essential drugs

Drug price control in other countries
Worldwide, drug prices are subject to controls and regulations. A host of policy instruments are exercised to rein in drug prices from increasing to unreasonable levels. Such controls take the following forms, either singly or in combination with more than one instrument: cap on mark-ups, fixed margins to wholesalers/pharmacists, price freezes, reimbursements, reference pricing, contributions to insurance pre-
Source: Figures for 1976 are adapted from Singh (1985), while those for 1998 have been compiled from PROWESS Database, Centre for Monitoring Indian Economy (CMIE), Mumbai, 2000
192
SECTION III
Table 14 Concentration in the Indian drug industry

Therapeutic group Leading products Amount of sales (Rs in crore) % Number of products
Chloramphenicol* Streptomycin* Mineral supplements Tonics Laxatives Anticoagulants All other antibiotics* Vaccines Trimethoprim combinations* Systemic corticosteroids Antiepileptics Rubs, other inhalants Antidiabetic drugs Antispasmodics, etc. Antiseptics, disinfectants Other penicillins* Sex hormones Topical corticosteroids Cough preparations Tuberculostatics Tetracyclines and combinations* Hypotensives Antiasthmatics Vitamins Cardiac therapy Cold preparations Antacid, antiflatulents, etc. Anti-inflammatory, anti-rheumatics Analgesics Ampicillin/amoxycillin* Cephalosporins* Macrolides and similar preparations*
20.84 14.06 49.15 25.76 22.81 21.61 39.22 66.96 65.74 122.77 64.05 100.19 108.78 51.18 71.47 52.97 73.71 115.22 73.05 153.58 79.69 92.56 119.61 296.91 135.96 197.30 158.05 255.04 150.17 250.37 297.41 179.10
43.41 93.27 39.00 31.00 32.00 32.43 35.62 74.65 65.31 71.00 44.27 83.41 38.34 41.30 78.63 86.53 31.23 42.00 41.23 44.69 57.73 36.84 45.88 38.87 31.64 41.65 55.71 41.09 51.28 38.92 43.3 69.37
1 1 2 2 2 2 2 3 3 4 4 4 4 4 4 4 5 6 6 6 6 7 7 8 8 8 8 8 8 8 8 8
* broad antibiotics category Note: Percentages indicate the value of leading products to total sales in each therapeutic group. Source: Calculated from ORG Retail Sales Audit, June 1999
mium, patient copayments, generic substitution, ceiling on promotional expenditure, differential value added tax on drugs, etc. Governments in various countries undertake cost studies to determine drug prices. While criteria vary from one country to another, countries largely follow comparable methods of pricing between a new product and that of an existing product in a similar therapeutic class. For new breakthrough products, prices are worked out based on therapeutic merit. France, Canada, Egypt, Mexico follow this pattern of price fixation. Most countries have some form of reimbursement mechanism to purchase drugs for the benefit of patients. France regulates margins allowed for wholesalers and pharmacists of
reimbursable drugs. In fact, in France, 91% of medicines sold by retail drug stores are on the reimbursable list. All reimbursable drugs in Italy are price controlled. Reimbursed generic prices Number are allowed to be sold 20% below the original of companies price. In Italy the prices of prescription drug that are reimbursed cannot exceed the European aver1 age price. In case the reimbursed price of a drug 1 exceeds the European average, the product under 2 question will automatically be removed from the 2 reimbursement list. However, non-reimbursable 2 drug prices can be changed once a year. 1 Reimbursement drug prices are controlled by a 2 reference pricing system in Germany, although 3 prescription drug prices are allowed to be changed 3 freely. Reference pricing is one in which drugs that 4 are therapeutically equivalent fall into one class 4 and are reimbursed at similar levels. The differ4 ence that arises between the reference and mar4 ket price is to be paid by patients. All prescription 4 drugs in Japan are effectively on the reimburse4 ment list. Reimbursement drug prices are subject 3 to price control based on the average level of prices 4 of similar categories. Britain allows for reim5 bursement on all drugs unless they are on the neg6 ative list. Although newly introduced prescription 4 drugs ones are not controlled at the launch of 6 the new product, generic prices are subject to con7 trols. 4 Interestingly, Spain controls even the launch 6 price of prescription drugs. The criteria for price 8 control are based on cost of production, profit 8 allowance and anticipated volume of sales. One 5 of the highlights of the health care scheme in Spain 8 and Switzerland is that a strict monitoring of the 8 doctors prescribing behaviour is undertaken and 8 those who are found to indulge in high prescrib5 ing are warned by the government. Switzerland 6 allows manufacturers to freely fix prescription drug prices, which are not on the reimbursable list. On the other hand, the reimbursable generic drug price is set at below 25% of the original. Over-the-counter (OTC) prices are generally free of price control in most countries. Although drug prices are largely control-free in the US, the government fixes a specified discount on the market prices of those drugs that are sold to Medicaid programmes. Reimbursement of drug prices in the US varies, since pharmacies have different agreements with various insurance companies. Fixing margins on the profit of pharmaceutical companies also forms part of drug price control/management. Egypt sets a maximum limit of 20% and 12% on profit to manufacturers of locally produced drugs and imported drugs, respectively. The respective profit margins for wholesalers and retailers are 7% and 20%. While manufacturers are allowed a margin of 7%10% in Mexico, wholesalers can retain up to 18.5%. Pharmacists in Mexico are provided two options of marginsFinancing and Delivery of Health Care Services in India
193
SECTION III
Box 4 Drug Price: Headed northwards?

Analyses show that 11 out of 15 antibiotic drugs witnessed a price rise in the range of 1%-15% annually during 1994-2004. Anti-TB drugs-Eight out of 10 drugs had shown price increases ranging between 2% and 13% per annum. General rise in the prices of drugs across all sub-therapeutic categories of cardiovascular diseases: cardiac disorders-2%-16%; anti-anginals-5%-6%; peripheral vasodilators and antihypertensives-1%-7% annually. Vaccines and antitoxins registered a meagre price rise during the period. Antimalarial drugs registered mixed price trends and similar conditions prevailed among cancer drugs. A price declining trend was observed among HIV/AIDS drugs
6% while OTC drugs are imposed a tax of 17.5% in the Netherlands. Switzerland imposes only 2% VAT on drugs. In Indonesia, VAT is fixed at 10% of manufacturers selling price except on drugs that are on the essential drugs list (GOI 1999).
The drug price control system in India

In the post-independence period, statutory control on drugs was first introduced in 1962. However, owing to criticism from the industry, the Government made changes in the Drug Price Control Order. Subsequently, the Government identified a list of 18 essential drugs and referred them to the Tariff Commission. The Tariff Commission was asked to go into the various aspects of the cost structure of these essential lifesaving drugs and asked it to recommend reasonable prices. Realizing the importance of checking the prices of drugs from escalating to phenomenal heights, the Drug Price Control Order, the first of its kind with a thorough analysis, was introduced by the Government of India in 1970. The Price Control Order was meant to keep the prices of drugs at affordable limits to the consumers and at the same time ensure that producers received reasonable returns. The Order captured 347 bulk drugs under its net, which were placed in various categories. The minimum percentage of profit margin was granted to different categories and producers were allowed to charge a maximum amount of post-manufacturing expenses. The other vital feature of this Order relates to the stipulation of minimum ratio of bulk drugs to formulations.
one for drugs launched before 1975 and those post-1975. Margin ceilings also exist in Spain with wholesalers availing 12.4% and pharmacies getting 43.5% of manufacturers retail price. Switzerland allows wholesalers to avail of a margin in the range of 11.1% to 17%, depending on the drug price. Margins allowed for pharmacies are in the range of 26% to 70%. The margin allowed for wholesalers by the manufacturer is only around 2% and for pharmacies it is about 15% in the US. Drug prices in many countries are linked to a ceiling on promotional spending by pharmaceutical firms. France levies an impost ranging between 9%and 20% on the proportion of promotional expenditure to sales of drug companies. Britain restricts promotional spending to a percentage of drug sales to the National Health Service. Spain places a ceiling on the promotional expenditure for a drug at 12%-14% of the producers sale price. The health care system in developed market economies has evolved in such a way that patients need not bear the entire amount OOP. The system is funded mainly by employee/employer contributions towards insurance payments and patient copayments. France has a system in which even copayments are borne either by private or by non-profit insurance plans. In Italy, however, patients are required to pay 50% of the price, depending on the category of drug. While half of Spains universal coverage of health care essentially comes from taxation, the rest is contributed by social security schemes, copayments and other OOP payments. Price control/management becomes cumbersome and impossible when the number of formulations is very large. Egypt successfully manages this impossible scenario by approving only 4000 medicines. In fact, the number of manufacturers producing a drug is restricted to four or five. For a new entrant beyond this level, the new firm is permitted to sell the drug 30% below the average price of the drug sold in the market. Similarly, to ensure the rational use of drugs, Mexico encourages single-ingredient formulations. Prescription drugs attract a value added tax (VAT) of only
Objectives of drug price control

A broad-based drug policy was formulated based on Hathi Committee Report of 1975. Based on Hathi Committees recommendations, the Government announced Drug (Price Control) Policy, 1979. Some of the key objectives of the Policy were: to ensure adequate availability of drugs to provide drugs at affordable prices to ensure the quality of drugs and check medicines from being adulterated to achieve self-sufficiency in production and self-reliance in drug technology.
Rationale for drug price control

The drug market is unique. Besides market failure, the overall health condition in India has made it all the more necessary to have a stricter price control regime. Detailed discussion of these follows: The demand for medicines is uncertain and consequently becomes inevitable. A patient with a potential disease cannot afford to ignore taking medicines as the disease may turn out to be fatal or result in permanent disability, In view of the above, it becomes pertinent on the part of the patients to buy medicines as advised by their doctors irrespective of the price. Demand inelasticity of consumers thus provides added advantage to drug firms to charge a rent-seeking price and, moreover, the pre-condition of con-
194
SECTION III
Fig 2 Criteria for inclusion of drugs under price control, DPCO, 1995
INCLUDE
All drugs with turnover of Rs 4 crore or above
All drugs with a turnover above Rs 1crore but below Rs 4 crore and single formulator having 90% or more of the market share
EXCLUDE
All drugs which satisfy existence of sufficient market competition with the conditions of: a) minimum 5 bulk drugs producers b) minimum 10 formulators c) market share of single formulator not more than 10%
Note:(i) Bulk drug turnover includes local production and import values. (ii) A formulator is a manufacturer of a single-ingredient formulation containing the subject bulk drug. (iii) Market share of a single formulator of a single-ingredient formulation of the subject bulk drug marketed in the country (as per Operations Research Group (ORG) (iv) Reference for the market share determination is data reported in ORG, March 1990 Source: Department of Chemicals and Petrochemicals (1999), Report of the Drug Price Control Review Committee, Government of India, October, New Delhi
sumer sovereignty for perfect competition cannot be ensured. The market for formulations, particularly if one goes by various therapeutic categories, is either monopoly or oligopoly. Price competition does not exist. The above market imperfections apart, with nearly onethird of the Indian population below poverty line, health conditions make it pertinent to allow for price control. Over the years, however, the controls are being dismantled gradually and the number of bulk drugs that were under price control has been brought down gradually to a minimum level. In 1979, 347 bulk drugs were under the Price Control Order, which came down to 142 in 1987. Drastically pruning the list further, the Drug Price Control Order of 1995 sought to limit the control list to just 76 drugs. Along with gradual reduction in the number of drugs under price control, certain procedures were greatly simplified and coverage of pricecontrolled drugs underwent enormous changes over the years. Table 15 shows that the number of categories of bulk drugs was pruned from three in 1979 to just one in 1995. With a reduction in the number of categories, the percentage of maximum allowable post-manufacturing expenses (MAPE) were unified to 100% in 1995 against 40%, 55% and 100% in 1979. As the process of globalization and liberalization are intensifying in India, controls and regulations on a lifeline industry such as the pharmaceutical industry is being lifted. The Drug Price Control Order of 1995 does away with many controls and regulations. The purview of price control was limited to just 76 drugs in 1995. The DPCO delineates certain benchmarks on which price control will be based. These are (i) sales turnover, (ii) market monopoly, and (iii) market competition. Across the board, the price control order fixed 100%
maximum allowable post-manufacturing expenses (MAPE) to all drugs. MAPE refers to the mark-up on the ex-factory costs provided to cover all selling and distribution costs, including the retail and wholesale trade margins.
Impact of price control on drugs and pharmaceuticals

In India, prices of drugs were once considered to among the highest in the world. This trend of high prices has tended to reverse since the 1970s in the wake of a series of policy measures. The Drug Price Control Order of 1970 brought all drug formulations in two categories: essential and non-essential. While those in the essential category were allowed a mark-up of only 75% in view of their importance, the latter category was allowed 150%. Later, in the 1978 Drug Policy, a slight modification was made by classifying formulations into four categories. The four categories and their respective markups are as follows: (i) Category I attracted only 40% markup; (ii) Category II was allowed 55% mark-up; (iii) Category III was permitted to charge 100% mark-up; and (iv) Category IV was totally exempt from price control. One of the earliest price analyses at the disaggregated level was done by Rane (1990) while attempting to assess the impact of the DPCO, 1987. The study was essentially carried out a simple relative comparison of drug prices between two periods-1986 representing the pre-policy period and 1990 denoting the post-price policy regime-as lack of adequate data on weights forced him to settle for this method. The price increase between these periods as a result of price policy categorywise show that most of them were in double digits in the
195
SECTION III
period following price decontrol. The highest rise was registered in the case of skin preparations, accounting for nearly 50% followed by respiratory system preparations recording a 32% price increase. The lowest increase was registered in the case of musculoskeletal preparations with 9.96%. Therefore, such a disaggregated assessment of increase in drug prices uncovers many facts which are not captured in the wholesale drug price index.
Box 5 Exorbitant profit and trade margins: Is monopoly purchase the answer?
The initial drug price is set high. The distribution network is extremely complex. Trade margins range from 100% to a whopping 5600%. The highest and lowest price differences between market and tender prices in the case of cancer drugs were about 275% and 1166%, respectively. A huge price difference is observed among maternal and mental health conditions. Maternal health drugs are 117% and 4028% for the lowest- and highest-priced drug category. Mental health drugs are 329% and 5102% for lowest- and highest-priced drug category. Monopsony can save up huge cost. Fixing trade margins is another solution.
Analysis of the impact of the DPCO, 1995

This section comprehensively probes into the impact of the DPCO, 1995 on drug prices. A comparative analysis of preand post-DPCO, 1995 price trends of major essential drugs is considered. The analysis basically involves examining price trends of essential drugs that are part of the DPCO, 1995 and those that are outside price control. Methodology for the analysis The basis for delineating drugs under price control and decontrol are derived from the government list of drugs under control (GOI 1995). Further, from this list only essential drugs were considered. Price data for the analysis are basically culled out from various December (except 2004, where the issue of August has been obtained) issues of the Monthly Index of Medical Specialities (MIMS), India spanning 11 years from 1994 to 2004. The number of essential drugs considered for analysis is as follows: (i) a total of 152 formulation packs relating to 14 disease conditions from the Essential Drugs List were taken into account. These 152 formulation packs were of similar strength and numbers (dosages). Out of 152 medicines, a total 115 medicines, constituting around 75% of the total medicines were decontrolled drugs while the rest 37 are price-controlled
drugs, accounting for 25%. Subsequently, the retail price of the formulations of each of these drugs was obtained from MIMS India. Although the number of products considered initially was 600 plus, after elimination the number came down to 152 products. Then we arrived at the annual price change (in percentage terms) of formulations under each drug. Elimination of such a large number of products became imperative due to the following: (i) for consistency-different dosages and strengths were ignored and only packs containing similar units, dosage forms and strengths were included; (ii) products that are not listed in MIMS India continuously for 11 years are also ignored from the analysis. The price change during the period from 1994 to 2004 is captured by working out the year-on-year percentage change and cumulative 11-year price change. The observed price change-annual percentage price change-is Table 15 given in Tables 16 and 17. The observed price change among 12 formulaComparative chart summarizing various drug price tions packs accounting for around 8% of total control orders formulation witnessed more than 10% price rise Items under DPCOs DPCO 1979 DPCO 1987 DPCO 1995 annually during 1994-2004. Another 38 medicines, accounting for 25% of the total formulaNumber of drugs under price control 347 142 76 tions considered, showed price increases in the Number of categories under which the above 3 2 1 range of 5%-9% during the same period. A moddrugs were categorized erate price rise of less than 5% was registered MAPE allowed on normative/national ex-factory costs to meet post-manufacturing among 56 formulations packs, which accounts expenses and to provide for margin to the manufacturers (5) for 37% of the total packs considered. Among Category I 40 75 100 them, 19 formulations are under the DPCO, 1995. Category II 55 100 NA Virtually no price change was recorded among Category III 100 NA NA another 19 formulations during this period, conCategory IV 60 NA NA stituting around 12% of the packs. Such price rises Percentage of total domestic pharmaceutical sales 90 70 50 was observed across all therapeutic categories. covered under price control (approximately) A general trend that emerges from Tables 15 and 16 clearly point out that over one-fifth of the MAPE: maximum allowable post-manufacturing expenses 36 price-controlled drugs under consideration Source: Indian Drug Manufacturers Association (IDMA) Bulletin 1998;XXIX:202 have tended to be either stable or have shown a
196
SECTION III
Table 16 Price change in drugs used for the treatment of cardiovascular diseases
Drug name Formulation Therapeutic group % Price change (1994 to 2004)
Bisoprolol Carboprost Atenolol Metoprolol Hydrochlorothiazide Ramipril Dopamine Nifedipine Enalapril maleate Indapamide Enalapril maleate Nifedipine Indapamide Hydrochlorothiazide Atenolol Digoxin Digoxin
Source : Calculated from MIMS India, 1994 to 2004
Concor Prostodin Tenofed Selopres Arkamin-H Cardace Dopinga Cardules Plus En.Ace. Natrilix SR Enace-D Depin Lorvas Adelphane-Esidrex Tenolol Cardioxin Lanoxin
Anti-anginals\ Haemostatics Peripheral Vasodilators Anti-Hypertensives Anti-Hypertensives Cardiac Disorders Cardiac Disorders Anti-anginals Anti-Hypertensives Anti-hypotensives Anti-Hypertensives Anti-anginals Anti-hypotensives Anti-Hypertensives Peripheral Vasodilators Cardiac Disorders Cardiac Disorders
-5.55 -2.74 -0.35 0.00 2.05 2.13 3.49 3.74 4.52 5.87 5.87 6.24 6.78 7.00 7.51 11.28 16.64
Price change in drugs used for the treatment of central nervous system disorders
Drug name Formulation Therapeutic group % Price change (1994 to 2004)
Carbamazepine Lithium Carbonate Lorazepam Paracetamol Trihexyphenidyl Clozapine Haloperidol Phenobarbitone Phenytoin Sodium Clozapine Imipramine Diazepam Metoclopromide Diazepam Diphenyl Hydantoin Fluoxetine Lorazepam Sodium Valproate Trihexyphenidyl Phenobarbitone Paracetamol Phenytoin Sodium Fluphenazine
Mazetol Licab/XL Larpose Disprin Paracetamol Trinicalm Forte/Plus Sizopin Serenace Gardenal Dilantin Lozapin Antidep Paxum Reglan Elcion CR Epsolin Fludac Ativan Epilex Pacitane Garoin Zimalgin Eptoin Anatensol Inj.
Analgesics and Antipyretics Antidepressants Sedatives and Tranquillisers Analgesics and Antipyretics Sedatives and Tranquillisers Sedatives and Tranquillisers Sedatives and Tranquillisers Anticonvulsants Anticonvulsants Sedatives and Tranquillisers Antidepressants Sedatives and Tranquillisers Antiemetics and Antinauseants Sedatives and Tranquillisers Anticonvulsants Antidepressants Sedatives and Tranquillisers Anticonvulsants Neurodegenerative Disease Anticonvulsants Hypnotics Anticonvulsants Sedatives and Tranquillisers
-1.36 -1.04 -0.42 -0.26 -0.13 0.00 0.00 0.00 0.00 1.76 2.12 2.40 3.57 4.13 4.21 5.35 5.66 5.83 6.70 7.00 7.48 8.32 12.99
197
SECTION III
Table 16 Price Change in Drugs of Infections and infestations

Drug name Formulation Price control Therapeutic group % Price change (1994 to 2004)
Ceftriaxone Ceftriaxone Penicillin Chloramphenicol Ciprofloxacin Penicillin Cefotaxime Metronidazole Erythromycin Ciprofloxacin Cloxacillin Metronidazole Chloramphenicol Amoxicillin Erythromycin Amoxicillin Ampicillin Cloxacillin Ampicillin Zidovudine Zidovudine Chloroquine Isoniazid Pyrazinamide Primaquine Mebendazole Tetanus Toxoid Tetanus Toxoid Mebendazole Streptomycin Rifampicin Rifampicin Chloroquine Amphotericin B Pyrazinamide Clofazimine Streptomycin Ethambutol Isoniazid Ethambutol
Monocef I.V. Oframax Pentids Reclor Ciprowin Pencom Claforan Flagyl Eltocin Ciplox Supremox Inj. Aristogyl Chloromycetin Novaclox Erythocin Novomox Campicilin Amplus Ampipen Retrovir Zidovir Melubrin Myconex 600 P-Zide PMQ-INGA Mebex Dual Antigen Tripvac Wormin Strepto-Erbazide Rimactane Rifacilin Emquin Fungizone Intravenous PZA-Ciba Hansepran Ambistryn-S Inabutol Forte Rimpazid 450 Combunex
DPCO 95 DPCO 95 DPCO 95 Decontrolled Drugs DPCO 95 DPCO 95 DPCO 95 DPCO 95 DPCO 95 DPCO 95 DPCO 95 DPCO 95 Decontrolled Drugs Decontrolled Drugs DPCO 95 Decontrolled Drugs Decontrolled Drugs DPCO 95 Decontrolled Drugs Decontrolled Drugs Decontrolled Drugs DPCO 95 Decontrolled Drugs Decontrolled Drugs Decontrolled Drugs Decontrolled Drugs Decontrolled Drugs Decontrolled Drugs Decontrolled Drugs DPCO 95 DPCO 95 DPCO 95 DPCO 95 Decontrolled Drugs Decontrolled Drugs Decontrolled Drugs DPCO 95 Decontrolled Drugs Decontrolled Drugs Decontrolled Drugs
Antibiotics Antibiotics Antibiotics Antibiotics Sulphonamides and other Bact. Antibiotics Antibiotics Anti-amoebics, anti-giardiasis Antibiotics Sulphonamides and other Bact. Antibiotics Anti-amoebics, anti-giardiasis Antibiotics Antibiotics Antibiotics Antibiotics Antibiotics Antibiotics Antibiotics Antivirals Antivirals Antimalarials Antituberculosis Antituberculosis Antimalarials Anthelmintics and other anti-infestive drugs Vaccines and anti-toxins Vaccines and anti-toxins Antivirals Anti-T.B. Anti-T.B. Anti-T.B. Anti-malarials Antifungals Anti-T.B. Antileprotics Anti-T.B. Anti-T.B. Anti-T.B. Anti-T.B.
-2.58 -0.17 -0.05 0.00 0.00 0.58 2.22 2.52 3.03 3.38 3.89 4.74 5.73 7.19 8.57 8.67 10.90 12.29 15.22 -8.40 -7.67 -2.58 0.00 0.00 0.00 0.00 0.30 0.72 1.62 2.41 2.90 4.97 7.29 7.50 7.56 7.85 8.96 9.96 12.39 13.45
198
SECTION III
Table 17 Price change in drugs of Alimentary, Musculo-Skeletal Disorders, Hormones and Genito-Urinary System
Prednisolone Frusemide Nalidixic Acid Nalidixic Acid Oxytocin Insulin NPH Magnesium Sulphate Insulin NPH Spironolactone Ibuprofen Glibenclamide Ibuprofen Frusemide Glibenclamide Magnesium Sulphate Beclomethasone
Wysolone Frusenex Gramoneg Negadix Pitocin Lentrad Pepticaine Actrapid Aldactone Brufen Daonil Combiflam Frumil Euglucon Solacid Anovate
DPCO 95 DPCO 95 DPCO 95 DPCO 95 Decontrolled Drugs DPCO 95 Decontrolled Drugs DPCO 95 DPCO 95 DPCO 95 Decontrolled Drugs DPCO 95 DPCO 95 Decontrolled Drugs Decontrolled Drugs Decontrolled Drugs
Corticosteroids and related drugs Diuretics and Antidiuretics Antidiarrhoeals Urinary anti-infectives Drugs acting on uterus Throid and antithroid drugs Gastro-intestinal sedatives and Ulcer drugs Hyper and hypoglycaemics Diuretics and Antidiuretics Non-Steroid anti-inlm. Drugs Hyper and hypoglycaemics Non-Steroid anti-inlm. Drugs Diuretics and Antidiuretics Hyper and hypoglycaemics Gastro-intestinal sedatives and Ulcer drugs Drugs Acting on th ecolon and Rectum
-5.10 -1.03 -0.41 -0.38 0.77 1.24 3.44 3.99 4.34 4.66 6.81 7.16 8.24 8.35 10.60 16.37
Price change in drugs of Nutrition and Respiratory System

Iron (Salts/complex) Salmeterol Theophylline Chlorpheniramine Dextrose Dextrose Vitamin A Folic Acid Salbutamol Calcium Carbonate Salmeterol Salbutamol Iron (Salts/complex) Terbutaline Chlorpheniramine Vitamin A Terbutaline Theophylline Budesonide Calcium Carbonate
Ferradol Salmeter Asmapax Depot Corex Electrobion Leclyte Rovigon Astymin Forte Salbetol Filibon Serobid Salmaplon Imferon Grilinctus-BM Piriton Expectorant Ossivite Bro-Zedex Alergin Pulmicort Anemidox
Decontrolled Drugs Decontrolled Drugs DPCO 95 Decontrolled Drugs Decontrolled Drugs Decontrolled Drugs DPCO 95 Decontrolled Drugs DPCO 95 Decontrolled Drugs Decontrolled Drugs DPCO 95 Decontrolled Drugs Decontrolled Drugs Decontrolled Drugs DPCO 95 Decontrolled Drugs DPCO 95 Decontrolled Drugs Decontrolled Drugs
Tonics; appetite stimulants Bronchospasm Bronchospasm Expectorants, cough suppressants, mucolytics and decongestants Mineral and parenteral nutritional suppl. Mineral and parenteral nutritional suppl. Vitamins Tonics; appetite stimulants Bronchospasm relaxants Mineral and parenteral nutritional suppl. Bronchospasm Bronchospasm relaxants Anaemia; Neutropenia Expectorants, Cough Suppr., Decongestants Expectorants, cough suppressants, mucolytics and decongestants Mineral and parenteral nutritional suppl. Expectorants, Cough Suppr., Decongestants Bronchospasm Bronchospasm relaxants Anaemia; Neutropenia
-2.06 -0.67 -0.34 -0.24 -0.13 0.00 0.44 2.05 2.57 2.90 3.40 4.06 4.36 4.67 5.42 5.62 5.69 7.36 8.93 14.18
Source: Calculated from MIMS India, 1994 to 2004
199
SECTION III
Table 17 Price change in drugs on ENT, Skin and Surgicals

Ketamine Hydrocortisone Ketamine Hydrocortisone Atropine Sulphate Fluticasone Lignociane Beclomethasone Atropine Sulphate Bupivaccine HCl Gentamycin Silver Sulphadiazine Gentamycin Adrenaline Lignociane Bupivaccine HCl Heparin Adrenaline Meglumine Antimonate
Ketmin Inj. Furacin-S Ketalar Crotorax-HC Bellpino-Atrin Zoflut Kemicetine Otological Beclate/N/C Atrisolon Marcaine Genticyn SSZ Aplicaps Andregen Xylocaine C Adrenaline Otek-AC Sensorcaine Beparine/Beparine Cream Gesicain C Adrenaline Urografin
Decontrolled Drugs Decontrolled Drugs Decontrolled Drugs Decontrolled Drugs Decontrolled Drugs Decontrolled Drugs Decontrolled Drugs Decontrolled Drugs Decontrolled Drugs Decontrolled Drugs DPCO 95 Decontrolled Drugs DPCO 95 Decontrolled Drugs Decontrolled Drugs Decontrolled Drugs Decontrolled Drugs Decontrolled Drugs Decontrolled Drugs
Anaesthetics Topical steroid pre.. Anaesthetics Topical steroid pre.. Mydriatics and Cycloplegics Topical steroid Preps Anti-infective prep. Local reactants on the nose Anti-inflammatory and anti-allergic prep. Surgical antibacterials Anti-infective prep. Anti-infective prep. Anti-infective prep. Anaesthetics Anti-infective prep. Surgical antibacterials Misc. skin prep. Anaesthetics Diagnostic Agents
-3.58 0.00 0.00 0.18 1.08 1.22 1.27 1.97 2.31 2.38 3.26 3.56 4.19 6.44 7.88 8.44 8.45 9.99 18.19
Price change in drugs of Cancer and Other Related treatments

Paclitaxel Etopside Fluorouracil Methotrexate Tamoxifen Vincristine Heparin Cisplatin Fluorouracil Bleomycin Methotrexate Vincristine Doxorubicin Tamoxifen Cyclophsphamide Allopurinol Chlorambucil
Intaxel Etosid Fivefluro Neotrexate Mamofen Neocristin Thrombophob Kemoplat Fluracil Bleocin Biotrexate Cytocristin Doxorubicin-Meiji Nolvadex Endoxan-N Zyloric Leukeran
Decontrolled Drugs Decontrolled Drugs Decontrolled Drugs Decontrolled Drugs Decontrolled Drugs Decontrolled Drugs Decontrolled Drugs Decontrolled Drugs Decontrolled Drugs Decontrolled Drugs Decontrolled Drugs Decontrolled Drugs Decontrolled Drugs Decontrolled Drugs Decontrolled Drugs Decontrolled Drugs Decontrolled Drugs
Carcino-chemotherapeutic Drugs Carcino-chemotherapeutic Drugs Carcino-chemotherapeutic Drugs Carcino-chemotherapeutic Drugs Carcino-chemotherapeutic Drugs Carcino-chemotherapeutic Drugs Carcino-chemo-therapeutic drugs Carcino-chemotherapeutic Drugs Carcino-chemotherapeutic Drugs Carcino-chemotherapeutic Drugs Carcino-chemotherapeutic Drugs Carcino-chemotherapeutic Drugs Carcino-chemotherapeutic Drugs Carcino-chemotherapeutic Drugs Carcino-chemotherapeutic Drugs Gout Carcino-chemotherapeutic Drugs
-5.06 -0.91 0.00 0.00 0.00 0.00 0.00 0.32 0.71 0.85 1.04 1.08 1.10 2.84 3.64 4.80 8.87
200
SECTION III
downward movement. These drugs are mostly in the category of antibiotics and belong mainly to the class of infections and infestations. Nearly 50% of controlled drugs have shown a moderate 1%-4% increase in price over the past 11 years (19942004), across all therapeutic categories. Eight out of 36 pricecontrolled drugs have witnessed over 5% rise in price; these belong to the category of infections and infestations, bronchospasm, diuretics, etc. The change in prices has been uneven across therapeutic categories. For instance, out of the 15 packs in the antibiotics category, 11 of them registered a price increase of between 1% and 15% per annum, while the other four medicines witnessed either stable prices or a decline over the 11-year period. In the case of anti-TB drugs, eight out of ten drugs had shown a price rise ranging between 2% to 13% annually during 19942004. Among antimalarial drugs, one of them had shown an increase while the remaining two had registered either a stable price or a price decline. In the class of central nervous system drugs, anticonvulsants recorded an increase in price of 0%-8% annually, while antidepressants witnessed no price change in a few categories and others showed a price increase of around 5% per annum. Anti-HIV drugs show a general price decline. Vaccines and antitoxins registered a meagre rise of less than 1% during this period. One can generally expect this trend as the vaccines market is mostly controlled by the Government and many vaccines have been introduced a long time ago and hence their price cannot be high. Anti-diarrhoeals have also tended to show a price decline during this period. In the class of cardiovascular drugs, there appears to have been a general rise across all therapeutic segments. Drugs for cardiac disorders registered a price rise in the range of 2%16% annually. The observed price rise per annum in the category of anti-anginals was in the range of 5%-6%. Peripheral vasodilators and antihypertensives witnessed a price rise in the range of 1%-7%. Anti-cancer drugs have remained stable, except one drug under consideration, which had shown more than 8% price increase annually.
Drug price and retail margins

The analysis above reveals that the drug price rise has displayed an enormous upswing during the past decade despite price controls. The year-on-year annual average price increase for certain categories of drugs has been more than 10%. However, it must be noted that the initial price per se is fixed with enormous margins. Trade margins are among the highest in the pharmaceutical industry. The extra sales taxes are levied by respective State Governments, as drugs come under the State-level taxes. Local sales taxes differ from one State to another. Recently, efforts are under way by the Department of Chemicals and Fertilizers to bring the sales taxes of different States under a uniform rate (4%). The exorbitant trade margins in the pharmaceutical market have become evident recently from the tender purchase
of drugs by Tamil Nadu Medical Services Corporation (TNMSC). Monopoly price can be challenged by monopsony purchase. Tender purchase of drugs by the TNMSC has revealed postmanufacturing margins running into four digits in retail purchase while in tender purchase the prices were at rock bottom (Srinivasan 1999). The following analysis gives an idea of the exorbitant trade margins and sky-high profits in the drug industry. A simple comparison of formulation packs of comparable size and strength between the market price and tender price is considered here. Therapeutic drugs are basically drawn from the Essential Drugs List. The market price of formulations has been obtained from August issue of MIMS India, 2004 while the tender price has been downloaded from the TNMSC website, applicable during the year 2004. Although nearly 80 formulation packs were considered earlier, only 30 of them have been analysed here. Therefore, two formulations for each of 15 disease conditions (from the Essential Drugs List) were taken into account - the one with the highest price difference and the one with the lowest. As can be seen from Table 18, price differences ranged from around 100% to 5600%. No systematic pattern in price difference could be deciphered across various health conditions. The highest and lowest price difference in the case of cancer drugs were 275% and 1166%, respectively. The observed price difference of drugs on maternal health are 117% and 4028%, respectively, for the lowest and highest drug category. In the case of mental health, the respective difference is 329% and 5102%. What has been noted above is only the price difference observed among one single pack (tablets/capsules). If one were to convert this price difference and apply it to the entire retail drug market sales, the resulting trade margins/profit would be mind-boggling. Drug companies are not welfare societies and hence one can assume that a normal profit margin has been included in the quoted tender price. The present-day drug industry is characterized by a complex distribution chain. Therefore, a multipronged strategy needs to be devised to smash this network. A ceiling on trade margins is the need of the hour. The monopoly power of drug companies can also be challenged by monopsony purchase, as the TNMSC procurement as shown. For the Essential Health Intervention package, the necessary drugs could be procured directly from the drug companies by a tender purchase for the entire country. Involving 33 States and Union Territories would only weaken the monopsony power. States could be persuaded to adopt a centralized procurement mode.
Public procurement of essential drugs*

The Central and State Governments spent approximately Rs 2000 crore during 2001-02 on procuring drugs. Apart from this, a few international organizations provide funds (or in kind) for drugs either through the central Government or directly to the States for specific programmes such as leprosy
* This section is largely derived from Parameshwar, (2004). Drug procurement systems in India, paper submitted to National Commission on Macroeconomics and Health, December.
201
SECTION III
Table 18 Drug price difference between retail market and tender purchase
Disease conditions Therapeutic drug Formulation Strength and No. Retail Price (Rs.) TNMSC price (Rs.) Price difference (%)
Cancer Cancer Child and infectious disease Child health COPD and Asthma COPD and asthma CVD CVD Diabetics Diabetics Injuries Injuries Japanese encephalitis Lymphatic Filariasis Malaria Maternal health Maternal health Mental health Mental health Tuberculosis Tuberculosis Others Others Others Others Others Others Others Others
Cyclophsphamide Fluorouracil Chloramphenicol Phenytoin Sodium Betamethasone Salbutamol Verapamil Atenolol Insulin NPH Glibenclamide Bupivaccine HCl Ketamine Ceftriaxone Diethylcarbamazine Chloroquine Carboprost Ferros Sulphate Chlorpromazine Alprazolam Rifampicin Pyrazinamide Rantidine Dopamine Ciprofloxacin Paracetamol Diclofenac Sodium Diazepam Dexamethosone Sodium Phosphate Cetrizine
Endoxan-N Fluracil Chloromycetin Dilantin Walacort Asthalin Veramil Aten Actrapid Daonil Sensorcaine Ketalar Lyceft Banocide Melubrin Prostodin Ferrochelate-Z Chlorpromazine-NP Alprocontin Rifacilin PZA-Ciba Consec Dopinga Ciplox Calpol Diclonac Calmpose Decdan Alerid
50mg;10 5ml 250mg;10 100mg;10 0.5mg; 10 4mg;10 40mg;10 50mg;14 10ml 5mg;10 0.5%;20ml 50mg;10ml vial 1g;vial 50mg;10 250mg;10 1amp 150mg;10 25mg;10 0.5mg;10 150mg;100 500mg;10 150mg; 10 5ml 200mg;100ml 500mg;10 50mg;10 5mg;10 2ml 10mg;10
36.35 11.67 30.76 131.55 3.55 5.21 5.02 25.75 129.28 6.60 34.34 89.50 90.00 3.88 4.36 80.13 19.94 5.95 22.55 99.68 42.46 7.51 25.00 27.00 8.78 11.03 13.70 10.36 31.50
13.218 1.001 4.4 9.75 1.043 0.522 4.392 1.2 86.85 0.454 15.5 15.15 16.11 0.707 2.233 68.5 0.495 1.81 0.442 66.6 5.188 2.205 6.05 6.41 1.24 0.686 0.4 0.222 0.561
275 1166 699 1349 340 998 114 2146 149 1454 222 591 559 549 195 117 4028 329 5102 150 818 341 413 421 708 1608 3425 4667 5615
Source:For Retail PriceMonthly Index of Medical Specialities, India, August, 2004 For TNMSC PriceTamil Nadu Medical Services Corporation (TNMSC). Available from URL: http:/www.tnmsc.com/system.html
control, etc. The current funding of drugs in the Central and State Governments is reported to be grossly inadequate. For instance, in Orissa, the current level of spending per public institution is found to be extremely low: ranging from Rs 16,000 annually in PHCs to Rs 50,000 in CHCs (6-15 beds). At the secondary care level, with more than 30 beds, the amount spent for outpatient care works out to roughly Rs 0.50 per patient per day and Rs 9.50 per patient per day for inpatient care (Table 19). It is clear that public institutions spend grossly inadequate amounts on drugs. Scaling up funds to increase spending on drugs is extremely important. At the same time, optimum utilization should be made of available resources. Efficient procurement policies have a significant bearing on ensuring the right medicines in sufficient quantities procured at lowest price to secure the maximum therapeutic value to the largest number of beneficiaries with the available resources.
An efficient procurement policy would have an integrated approach starting from (i) preparation of an essential drugs list, (ii) assessment of the quantity of drugs needed, (iii) quality assurance from suppliers, (iv) procurement process, (v) supply chain management, and (vi) prompt payment to suppliers. In India, Central and State Government institutions follow one or more of these arrangements for public procurement: (i) Central Rate Contract System, (ii) Pooled Procurement either by the government or through an autonomous corporation, (iii) decentralized procurement, and (iv) local purchase. The Tamil Nadu Medical Service Corporation (TNMSC) set up in 1994, is a pioneer in the current drug procurement and distribution system. The success of the TNMSC lies in its centralized drug procurement and distribution system supported by a computerized system of drug management. The TNMSC has set up warehouses at all district headquarters from where supplies are provided to hospitals and other health
202
SECTION III
Table 19 Inadequate public spending on drugs in Orissa

Level of public institutions Amount (in Rs)
Public institutions (>30 beds) OPD (per patient per day) IPD (per patient per day) Block-level CHC hospital (16-30 beds) PHC (6-15 beds) Block-level PHC Subcentres
OPD: outpatient department; IPD: inpatient department; PHC: public health centre
0.50 9.50
110,000 50,000 30,000 16,000
Table 20 Centralized procurement price: A comparison

Drug Strength and pack Tamil Nadu 1996 2003 1.18 10.71 28.80 3.50 0.80 1.36 1.12 5.13 20.90 2.37 0.52 1.04 1996 1.24 7.98 29.20 3.94 0.72 1.45
Paracetamol Norfloxacin Rifampicin Chloroquine Gilbenclamide Atenolol
500 mg10 tablets 400 mg10 tablets 450 mg 10 tablets 250 mg10 tablets 5 mg 10 tablets 50 mg10 tablets
facilities. A passbook system has been introduced where the entitlement of each facility is given in monetary terms. The institution can obtain any drug in the approved list if funds are available in the passbook. The TNMSC has also developed a unique Drug Distribution Management System (DDMS) which is put to use in effective monitoring of procurement and distribution of drugs and supplies. Under this system, each district warehouse is linked by computer to the central computer in the Head Office. Receipt and issues of drugs have been computerized resulting in instantaneous adjustments to the stock position. This has facilitated movement of drugs from one warehouse to another based on needs, thus avoiding shortages. Usually States adopt a two-envelope system (technical bid and price bid being sent in separate envelopes). This system ensures a speedy and transparent mechanism in procurement of drugs. Contracts are awarded to only those manufacturing units, which have a Good Manufacturing Practices (GMP) certificate of the WHO and should ideally have a minimum ceiling of annual turnover. Karnataka and Rajasthan, however, follow a decentralized system. In the former, a major part of drug procurement, accounting for 60%, is sourced by zila panchayats at the district level while the remaining 40% is sourced by government medical stores. In Rajasthan, in the order of priority, drugs are procured from public sector units (Rajasthan Drugs and Pharmaceuticals Ltd.). Tenders are invited only for those drugs not supplied by
Public Sector Undertakings and Small Scale Industries. The direct benefits flowing from the TNMSC model seem to support lower prices contributed by competitive bidding and bargaining power. Table 20 illustrates the phenomenon of stable or declining prices due to centralized tender procurement of drugs. A simple comparison of drug price is carried out here, involving the procurement system in Delhi and Tamil Nadu, of drugs from different therapeutic categories with similar strengths and pack sizes. The analysis reveals that drug prices have tended to decline gradually or even steeply in some cases, during the period 1996-2003. The tender prices are not only declining but the analysis in the earlier section shows that even the initial price quoted is well below the market price, indicating a wide drug price difference. Further, the IT-driven logistics management system facilitates monitoring of procurement, distribution and issue of medicines. Quality control is achieved through building in quality requirements in the procurement process and drawing samples from each batch and testing them. Delhi However, these developments and the success 2003 achieved by the States does not appear to have any impact on the Central Government, which 1.17 continues to have multiple agencies for procur6.48 ing and distributing drugs to its various health 20.90 schemes/programme. This is depicted graphically 2.75 in (Annexure I). While the Medical Stores Depot 0.64 under the Ministry of Health and Family Welfare 1.55 has seen gradual reduction in its handling of procurement and storage of drugs meant for a few States and paramilitary forces, drugs required under the Central Government Health Scheme (CGHS) are procured through the Hospital Services Consultancy Corporation (HSCC). Under the CGHS, orders for both generic drugs and proprietary drugs are placed through the HSCC. As expected, the price difference between generic and proprietary drugs is extremely high. It is a matter of concern that the Government of India, which brings out as Essential Drugs List covering only generic drugs is actually procuring and dispensing proprietary drugs for its employees under the CGHS scheme. The total value of proprietary drugs is many times the value of drugs purchased by generic name. Second, the price difference not only results in sub-optimal utilization of resources but is also a major drain on Central Government resources. This discrepancy should be resolved and only generic drugs should be procured and distributed. There is no quality check on proprietary drugs whereas generic drugs are procured from prequalified bidders whose products are also subjected to sample testing. Under different National Health Programmes (NHPs), the Central Government either provides financial aid or supplies drugs to States through centrally procured arrangements. Each of the six NHPs has its own procurement procedures resulting in duplication of effort with no attendant benefits of lower prices that a bulk purchase would entail. Currently, the NHPs are (i) Revised National Tuberculosis Programme, (ii) National Leprosy Elimination Programme, (iii) Reproductive and Child
203
SECTION III
Health (RCH), (iv) National Malaria Control Programme, (v) National AIDS Control Programme, and (vi) National Blindness Control Programme. The amount spent on drug procurement under the first three programmes worked out to Rs 480 crore during 2002-03. However, the procedures adopted in the procurement of drugs in this case appear to be a lengthy one with significant time over-runs. For instance, the action plan for procurement and supply of PHC kits under the RCH programme for the period 2003-04 had envisaged a timeframe of only nine months. However, procedural delays resulted in the entire process being completed in exactly double the time. Similar delays have also been observed in the procurement process involving another agency RITES, which deals with the Malaria Control Programme. Much of these delays can be attributed to the absence of a system of pre-qualification of bidders. As a consequence, lower bids get rejected on the ground that the bidder does not have the required qualification or ability. In the absence of clear and well-defined criteria, the chances of an element of subjectivity in making decisions cannot be ruled out. The solution lies in introducing a two-envelope system, one on technical and another on price bids. Once technically unqualified bidders are rejected, the selection of the lowest bid becomes automatic.
Drug Regulation in India

In India, the drug regulatory system has been poor and neglected over the years, although much has been written and recommended by various committees. Poor enforcement mechanisms and multiple interpretations of the Drugs and Cosmetics Act 1940 have made regulation in this sector an unviable proposition (GOI 2003). In some States such as West Bengal, Rajasthan and Punjab, there is no testing laboratory. Assuming a norm of one inspector for every 50 manufacturing units and one inspector for 200 sales units, the gap between the required norm and the actual number of available drug inspectors is woefully inadequate. Given the currently available figure of 935 drug inspectors, one inspector serves around 320 wholesale and retail units instead of a norm of 200. This could be the reason why the number of spurious and substandard drugs detected was relatively less. With adequate manpower and infrastructure, inspection of manufacturing and sales premises alongside a strong surveillance mechanism relating to the movement of spurious/counterfeit drugs could be carried out and unearthed more rigorously. As far as the manufacturing units are concerned, the Government of India noted that roughly around 5900 units require intense surveillance/inspection and not all the 20,000 units (Mashelkar Committee Report 2003). Further, the Committee noted that the 1333 bulk drug units, 4354 formulation units, 134 large volume parenterals (LVP) and vaccine manufacturing units-accounting for 5877 units-are the ones that require intense inspection. The other major categories are cosmetics, loan licences, blood banks, etc. According to the
Mashelkar Committee, around 120 drugs inspectors are needed to monitor about 5877 units and another 100 inspectors are required for the remaining categories. Other observations and recommendations made by the Committee are as follows: Strengthen the infrastructure and manpower relating to the monitoring/surveillance/inspection mechanism, both at Central and State level. Information received by the Mashelkar Committee reveals that only 17 out of 31 States has a drug-testing facility; of 17 only 7 appear to be reasonably equipped/staffed. Measures are needed to tone up the Drugs and Cosmetics Act 1940, providing it with more powers (penalties) against manufacturers and distributors. The Mashelkar Committee proposes a Central Drug Administration (CDA) to be set up under the Ministry of Health and Family Welfare with autonomous status. The Committee recommended the setting-up of the CDA by the end of 2004 and State-level regulatory systems be strengthened accordingly. Review C & C1 licenses under the Drugs and Cosmetics Rules issued against manufacturing and distribution (wholesalers & retailers) to keep abreast with recent developments in the drugs sector. Review the Schedule H drugs which provides a list of prescription drugs. Comprehensively review Schedule K of OTC (over-the-counter) drugs. Curb inter-State movement of spurious drugs, tone up the existing communication network and freely exchange information between States. As far as the health food/therapeutic foods/dietary supplements are concerned, regulation relating to their quality and safety is needed as the demand for such products is increasing, and producers/sellers are indulging in exaggerated claims. These products should be brought under the purview of the relevant food law. However, any product that has distinct medicinal claims would be qualified as a drug and not food products. There is a growing market for Indian Systems of Medicine (ISM), herbal products and drugs of natural origin. Concern has been voiced over their efficacy and quality. Efforts need to be made to update the requirements for licensing such products. Since for many such products long-term safety data are not available on their usage, additional safety data need to be obtained. The other area of concern is uncontrolled growth of medical devices and equipment. The standards and quality of many newly emerging equipment are questionable in nature. Clinical research is an area of concern as human lives in developing countries has become an experimental theatre for pharmaceutical firms. The Committee is of the firm opinion that responsibility must be shared between all concernedinvestigators, sponsors, ethics committees and regulators. According to the Committee, It is absolutely essential to institutionalize Good Clinical Practices to achieve credibility for the data generated in India (GOI 2003).
204
SECTION III
Regarding Phase I clinical trials, the Committee accepted the revised Schedule Y, which stipulates that data generated from such trials in foreign countries need to be furnished to the Indian licensing authority and permission granted to repeat Phase I studies. As far Phase II and Phase III trials are concerned, the Committee observes that since the trials undergo rigorous review by the International Conference on Harmonization (ICH) signatory countries, approvals could be accorded and expedited by the regulatory authorities simply based on the technical documents submitted in ICH countries.
Patent protection under WTO, 1995

The Patents Act, 1970 has been instrumental in encouraging and developing the indigenous drug industry and indirectly containing medicine prices, but is currently under threat with the conclusion of the last Uruguay Round of General Agreement on Tariffs and Trade (GATT) negotiations in 1993 and the establishment of World Trade Organization (WTO) on 1 January 1995. In fact, extension of pharmaceutical product patents to all member countries was the key and controversial issue and also the last issue to be hammered out prior to tabling of the Draft Agreement at the end of 1991. A gist of the patents system, 1970 and the change-over envisaged under TRIPS is given in Table 21. The erstwhile GATT (since 1995, WTO) sought to radically transform the patent Act in many countries. The specific article dealing with patents-Trade-Related Intellectual Property Rights (TRIPS)requires that the signatories to GATT must necessarily amend their Constitution in accordance with this Article. The Article on TRIPS requires member countries to change their Act in such a way that they grant product patent to the pharmaceutical, chemical, food and agricultural sectors as well. The period of patent rights is to be changed in the Indian case from seven to twenty years. A proper amendment needs to be made to the Constitution of respective member countries amending the present rules. For developing countries, 1 January 2000 was fixed as the deadline for amending the Constitution. Developing countries like India have, however, been granted a five-year transition period till 2005. Until then, exclusive marketing rights (EMRs) would have to be granted to those companies introducing newly invented products. Domestic production of the patent-protected products is not mandatory wherein import is to be considered as a working of the patent. Even the Paris Convention specifically nails non-working or import of patent-protected products as an abuse of exclusive rights. The other retrograde step in the direction of TRIPS is the restrictions imposed on the free use of compulsory licensing provisions, which were hitherto
Drug patents in India

The Indian Patent Act, 1970
The Indian Patents Act, 1970 (effective since 1972) sought to provide only process patents for chemical substances including pharmaceuticals, agrochemicals and food products, and it granted product patents for non-chemical substances. The duration of process patents was fixed at seven years from the date of filling of the patent, or five years from the sealing of the patent, whichever is earlier. Considering the importance of sectors such as pharmaceuticals, the Indian Patents Act, 1970 added a few provisions, which sought to significantly restrict the scope of protection. (i) Under the license rights, a process patent owner is obliged to sell the license to any third party fetching a maximum royalty of 4% in turn. (ii) The Government retained the right to issue compulsory licenses (after 3 years from the date of sealing of a patent) if the product under question was above reasonable prices or if it did not satisfy public interests. (iii) Import of patent protected products is not considered to be working of patent and therefore the patentee must necessarily produce the same in the country within three years from the date of sealing of a patent.
Table 21 A synoptic comparison of Indian Patents Act, 1970 and TRIPS, 1995
Different provisions of patent acts Indian Patents Act, 1970 TRIPS-WTO, 1995
Type of patents Effective duration of patent Compulsory licensing
Working of patent Burden of proof
Only process patent allowed in the case of pharmaceuticals, chemicals and food Seven years from the date of filing or five years from the date of sealing, whichever is earlier Compulsory licensing allowed after three years of sealing of patent if the price of product under question is above a reasonable level or if it did not satisfy public interests Domestic production alone is considered as 'working of patent' In case of patent infringement, the burden of proof lies with the complainant of the patents
Product patent allowed in all sectors Twenty years from the date of filing of patent application Restrictive use of the provision of compulsory licensing - allowed only when there is national emergency/public non-commercial use/government use Whether products are manufactured locally or imported, it would amount to 'working of patent' The burden of proof would fall on the alleged defendant of patent infringement
205
SECTION III
available in the present India Patents Act of 1970. The provision of compulsory licensing (under the new dispensation) can be harnessed only when there is a clear case of national disaster or calamity.
TRIPS and its likely impact

Several issues need attention in the wake of a change from process to product patent. These issues include price rise, market structure, foreign investment inflows, technology transfer, royalty and hence foreign exchange outflow, importdependence, etc. A sensitive and a highly controversial issue with regard to TRIPS is the concern about the high price of medicines. India was at the forefront in raising this issue backed by strong evidence. It is natural that many recent findings on this matter focused on likely price trends in India in the event of amending the present patents Act. Lanjouw (1998) found drug prices in India, particularly in the post-patent 1970 period, among the lowest in world. As a sequel to a transition to the product patents regime, drug prices in India are expected to considerably escalate to a high level. Simultaneously, however, he and a few others (Vohra 1999) argue that given the current market conditions, it is estimated that only 10%-20% of the pharmaceutical products are under patent, and hence there is no need to focus on negative trends on the drug price front. It needs to be noted that once patented products start proliferating in the market, the composition of patented products in the total pharmaceutical market would undergo a drastic change in favour of the former. This would have a far-reaching influence on price. Recent studies, mostly of simulation exercises carried out by Challu (1991), Nogues (1993), Fink (2000) and Watal (2000) all clearly show the extent of price increase that would be
likely in the near future with a changeover from the present system to a patent monopoly era. Table 22 provides a synopsis of each of these studies. The study by Fink (2000) suggests a surge in pharmaceutical prices in the range of 9%76% if product patent rights are introduced. However, as far as the impact on various therapeutic categories is concerned, the upsurge in price would depend on the demand for new patented products or on the available alternative treatments, whichever dominates the market. Interestingly, Fink suggests that rapid acceleration in drug prices could be countered by various price control measures available with the local government, a provision allowed in the TRIPS agreement. Compulsory licensing is another tool to counter the adverse implications of conferring patent protection. Price ceilings, if put into effective practice, by allowing firms to charge normal profits in addition to production costs, would reduce or eliminate an inventors patent-induced market power, argue Braga et. al. (2000). They further assert that when normal profits are granted the potential disincentive to invest would wither away resulting in recouping of R&D investment. In any case, the price of patented products is bound to be high. This could be because of several reasons: (i) formulation activity would be costly as multinationals would normally set high prices for the bulk drugs imported in view of global reference pricing; (ii) issuing compulsory licensing to any company in India would amount to enormous royalty fees, in return. This would naturally be reflected in the base price of the patented products; (iii) any effort to locally produce the patented medicine is nothing but monopoly production and consequently monopoly pricing, which will always be higher than the competitive price. However, a point worth noting in this context is that one must actually analyse the entire gamut of issues related to the pharmaceutical market and one cannot merely take such
Table 22 Summary of Studies: Simulation exercise on pharmaceutical product patents and their impact
Studies Price Capital Transfer Welfare Loss/Gain
Challu (1991) (Argentina) Nogues (1993) (developing countries) Fink (2000) (India)
Estimated price increase for the market segment subject to patents: 273.2%. Given different demand and substitution elasticities, the lower priced among quinolones such as ciprofloxacin, the price could range from 233.5% to 276.7%, while for the highest, such as ofloxacin, it could range from 318.6% to 370.5%. The price rise could be as high as 242% with a constant elasticity-type demand function.
Money transfers abroad: US$ 367 million per year -
Consumer welfare loss: US$ 309 million per annum The losses from consumer misallocation could be as high as US$ 7.7 billion Taking the case of quinolones, welfare losses range from US$ 28.7 million to US$ 69.9 million per annum, assuming certain elasticity
Watal (2000) (India)
Moving from current market structures to patent monopoly could yield a loss of US$ 140 million annually
206
SECTION III
provision as given. An appreciation of the overall structural adjustment in economies such as India show that over the years, particularly since the early 1990s, pharmaceutical prices have been decontrolled to a substantial degree and in fact presently only a few drugs (75 essential drugs in 1998) are actually controlled. Many more of these are likely to witness lifting of controls in the immediate future, as made evident in the intentions of government policy pronouncements (GOI 2001).
Compulsory licensing and parallel imports

Merely ten years after the establishment of WTO, the darker side of TRIPS is unfolding before the world. There has been perceptible damage to public health particularly in many developing countries. The controversy over accessibility of cheap drugs to combat HIV/AIDS in Africa and other developing countries has put the TRIPS agreement under vigilance. In 2000, Brazil was taken to the Dispute Settlement Mechanism of WTO for the alleged violation of TRIPS by the US and the next year saw 39 drug MNCs aligning against the South African Government for not conforming to the new global patent regime. The AIDS epidemic has assumed serious proportions triggering off national emergencies, particularly in the African countries, in which 2.5 crore people are infected with HIV. The South African government pressed into service the provision of compulsory licensing and parallel import under its new patent regime. Accordingly, this provision enabled the South African Government to either direct domestic companies to manufacture or import cheap branded generic drugs from developing countries such as India. Western pharmaceutical companies have been marketing a cocktail of antiretroviral drug therapy at unaffordable prices in many developing countries. With the arrival of Indian generic substitutes, drug multinationals raised a protest. For instance, Cipla was the first to enter the African market with its triple drug cocktail (Hindu, 16 May 2001) of antiretrovirals-lamivudine, stavudine and nevirapine-offering it at a price of US$ 350 per patient per year, a small fraction of the US$ 10,000 that a western patient pays. An unimaginably high price in developed countries and an on-going competition fostered by generic varieties in South Africa has brought prices tumbling down. Intriguingly, despite an offer of knock-off versions at rock-bottom prices, market prices are far beyond the purchasing power of an average African. Drug MNCs initiated action against the South African Government. A spirited national and international challenge was mounted on these companies, which forced them to make a tactical retreat. Another controversial case involves Brazil. Its newly amended patent policy allows for local production by providing license to domestic companies if the foreign-patented products are not produced locally. The Brazilian patent law requires a foreign patentee to manufacture a product locally within the stipulated three years of the grant of patent. Importing such patentprotected products is not considered to be working of patents in Brazilian law. Under this provision, Brazil recently allowed domestic production of generic anti-HIV/AIDS drugs, which has been contested by the US. With the heat of international pressure mounting heavily on the US, it withdrew the case reg-
istered at WTO against Brazil (The Hindu, 26 June 2001). Stung by increasing criticism and battered image, drug multinationals have subsequently joined the race to slash anti-AIDS drug prices. This move is, however, seen as a ploy to retain their market share, which is threatened by inexpensive generic competition. Merck and other five multinationals (The Hindu, 16 May 2001) have since come forward to sell antiretroviral drugs at lower prices to developing nations. It is argued that TRIPS allows for certain flexibility in its clauses to protect public health. The monopoly abuse of the patent system that emanates from exclusive rights conferred on the patentee could be controlled or restricted by means of resorting to granting compulsory license or through parallel imports. The principles articulated under Article 8 and Article 31 of the TRIPS agreement (www.wto.org/english/ tratop_e/trips_e/trips_e.html) appear to enable member countries to adopt measures that would safeguard them in the event of public health emergencies. The specific instances under which compulsory licensing could be conferred are: (i) insufficient or non-working of patents; (ii) failure to produce locally and therefore continuously import the product even after the issue of patent for 3-4 years; (iii) in the event of charging an unreasonably high monopoly price. Unfortunately, it took years for many developing countries to realize and challenge the lethal provisions of the TRIPS agreement. The toll and suffering that the AIDS epidemic inflicts on impoverished Third World nations triggered the latest patent battle. It needs to be reiterated here that apart from the devastating AIDS pandemic, there are other killer diseases in countries such as India (diarrhoea, malaria, TB, etc.) which require immediate attention and pose a continued threat to health security.
Data exclusivity
Article 39.3 of the TRIPS agreement requires that member countries safeguard the interest of inventing companies from unfair commercial use of products arising out of disclosure of data submitted by the companies. However, TRIPS allows for exception to this rule. Member countries can waive this article to protect public health exigencies and thereby grant generic manufacturers the opportunity to produce drugs thereby limiting evergreening of patents. In pharmaceutical industry parlance, data exclusivity is one in which the originator company registers with a regulatory authority of a country by submitting data demonstrating the safety, quality and efficacy of the innovative drugs. However, the generic manufacturer need not get such an approval, as, while applying for approval of their drug, they refer to bioequivalence data already established by the originator. If data exclusivity are granted for a specific time period, it would deny the generic manufacturer from availing the reference data of the originator. The period of data exclusivity ranges from five years in the US, six to ten years in the EU, etc.
207
SECTION III
Indian Patents (Amendment) Bill, March 2005: Significance and implications

India has moved into a product patent regime in 2005 complying with the TRIPS provisions of WTO. In a series of amendments to the Indian Patents Act, 1970, the latest and the crucial amendment to the Act was made in March 2005. The new Indian patent amendment suffers from ambiguity, technical loopholes and fails to incorporate some of the flexibilities incorporated in the TRIPS regime. This has serious implications for access to drugs and medicines in India, and the developing world in general. The issues that still need to be addressed in the newly amended patent acts are: (i) issues relating to the scope of patentability; (ii) cap on royalty payments; (iii) plugging all ambiguities and technical loopholes in the amendment to avoid unnecessary and expensive litigation in the future; (iv) vesting discretionary powers in the patent office in terms of timelines of rules, making them vulnerable to vested interests. Let us discuss each of them in detail.
Cap on Royalty Payments

Another related issue with compulsory licensing is the cap on remuneration to the patent holders. The amended Act leaves open this issue and assures reasonable royalty to the patent monopoly. In many countries, there is a cap on royalty payments made to the patent holders, say 4% of the total turnover of the medicine. Patents monopolies can simply refuse to issue compulsory licensing by demanding excessive royalty payments. What constitutes reasonable is only to be decided in the court, multiplying litigations.
Mailbox Products
Product patent regime, all over the world, thrives on frivolous claims for me-too drugs of similar chemical entities. This is clearly in evidence before the advent of product patent regime in India. Under the mailbox provisions (India is accepting applications for product patents in the areas of pharmaceuticals and agrochemicals since 1999, although not granted any patents since the amendment was made only in March 2005), there were reportedly 4792 applications for product patents although during 19952004, only 297 new chemical entities have been bestowed with product patent status in the world. It is therefore clear that the rest of the applications for patents are only frivolous in nature. Moreover, according to the amended Act, any generic producer who were manufacturing these mailbox products, before January 1, 2005 can continue to produce such medicines but are required to pay reasonable royalty. Accordingly, the generic manufacturers were required to show that they made significant investment in their venture. This ambiguity is likely to throw up infringement suits and more litigation.
Definition and scope of patentability

The new amendment does not clearly state what is patentable. In the amended Act, pharmaceutical substances are described as any new entity involving one or more inventive steps. This could mean anything involving formulations, pharmaceuticals, isomers, polymorphs and their combinations. Ideally, and for practical purposes, it should have been new chemical entity. While the Indian Patent Act, 1970 clearly defined the terms invention, patents, inventive step, and industrial application, the new amendment suffers from ambiguity and leaves several loopholes in defining these terms. The other criterion for patentability in the new Act, namely inventive step, unnecessarily broadens the scope. Accordingly, the patentee is either required to display that the invention incorporates a technical advance or has economic significance, or both. Thus, by simply showing economic significance of an inventive step over technical advance, patent holders get the benefit of this broad and ambiguous definition.
Pre-grant and post-grant opposition

Through the present amended Act, initiation of opposition proceedings against a grant of patents is allowed only by way of representation and not in the form of notice. Further ambiguity is reinforced as it is unclear whether access to documents of patent holder is possible. If not, how will any opposition proceedings be carried forward or to start the process?
Compulsory Licensing
One of the central themes of the Doha Declaration is the issue of compulsory licensing. Patent monopoly abuse is sought to be restricted by issuing a compulsory licence to a generic producer in the pharmaceutical market. The Doha Declaration reaffirmed the members the right to protect public health and extolled the members to interpret and implement TRIPS, which would help them promote access to medicines for all. According to the newly amended patent Act, a compulsory licence can be issued only during a national emergency, extreme emergency or public non-commercial use, and it will be issued only after three years from the date of grant of the patent. By leaving out the grounds on which a compulsory licence can be issued, the bill barters away the flexibility brought in during the Doha Declaration.
Policy suggestions
Out-of-pocket spending on drugs by households in India is extremely high. Given the low purchasing power of the population, virtually no health insurance in place and with an inadequate and malfunctioning public health institutions, the need of the hour is to focus on the following: The present list of drugs under the DPCO, 1995 needs to be expanded by including all the drugs under the Essential Drugs List. The criterion of essentiality must form a vital part in deciding whether a drug is to be included in the controlled category or not. The Indian pharmaceutical market is flooded with irrational drugs (particularly combination products). Estimates sug-
208
SECTION III
gest that there are about 20,000 products in the market. With barely 300-plus drugs recommended by the WHO and 268 in the Essential Drugs List as proposed by the Government of India, 2003, it would be entirely possible to tackle all disease conditions in the country. In one stroke, all combination products could be wiped out by a government edict. Towards this goal, a committee of physicians, pharmacologists, microbiologists, etc. should be appointed to review irrational combination drugs. It is absolutely essential to encourage only generic drugs. Formulation of standard treatment guidelines is imperative in the backdrop of an essential drugs list. Moreover, a National Formulary updated on a two-yearly basis also needs to be put in place. We also recommend that a National Antibiotics Committee be set up and continuous surveillance ensured on the availability of antibiotics. Manufacturers and retail pharmacy stores may be provided with a variety of incentives to produce and sell essential drugs. Fiscal incentives such as lower duty, subsidy, etc. could be provided. For instance, in the post-April 2005 period, a VAT of 4% is being proposed on drugs and medicines. Although 4% could be levied on other inessential drugs, essential drugs could attract a minimum of 1% VAT. Since trade margins are exorbitantly high in the drug industry, fixing ceilings on trade margins is necessary. Currently, price-controlled drugs under the DPCO, 1995 attract 8% on wholesale and 16% on retail. This margin can continue at this rate and uncontrolled drugs should also be brought within the purview of margin ceilings. As suggested by the interim report of the Sandhu Committee (Government of India 2004) we propose wholesale and retail margins on branded drugs to be 10% and 20%, respectively. On generic-generic drugs, the respective margin could be 15% and 35%. This should be inclusive of various trade discounts offered to dealers. Monopsony purchase can check exorbitant profit and trade margins of drug corporates. Centralized public procurement is definitely a way to save the public exchequer. i) As a first step, procurement of drugs meant for all Central Government health programmes and health facilities of different ministries and other autonomous bodies need to be centralized. Health institutions under the Central Ministry must strictly dispense only generic drugs and do away with proprietary drugs. ii) The success of the TNMSC in drug procurement could be replicated in other States as well. During 200304, with a total budget of around Rs 120 crore, the TNMSC served nearly 11,000 medical institutions (from medical college hospitals to subcentres, including autonomous institutions and other departments in the State). The administrative cost involved in running TNMSC is in the range of 0.5%-1%. By simply procuring in bulk and streamlining the procurement system, the present budget spending on drugs could save huge resources. This would enable States to procure more and make them available to the needy. Drug regulation has become a complex and neglected
issue over the years in India. Strengthening the drug regulatory authority, as prescribed by the Mashelkar Committee, is the need of the hour. The drug regulatory authority could be provided with an autonomous status to ensure transparency and effective functioning. To this end, we suggest that the government set up a National Drug Authority (NDA) with an autonomous status to take up the functions of drug pricing, quality, clinical trials, etc. Consequently, the present National Pharmaceutical Pricing Authority (NPPA) could be merged with the proposed NDA. For strengthening the drug regulatory system as suggested by the Mashelkar Committee, the Central Government needs to allocate Rs 1.6 crore annually for the additional posts (mostly inspectors) that would be created and another Rs 50 lakh as contingencies for the creation of additional offices. While the Essential Drugs List is prepared by the Ministry of Health and Family Welfare, the Ministry of Chemicals and Fertilizers is involved in formulating and exercising price controls on drugs. For an integrated approach, it would be a better idea to transfer the functions of both the ministries relating to drugs to the proposed NDA. The NDA must make efforts to collect, tabulate and disseminate data on drug production, therapeutic-wise sales, company level information on drugs, etc. It is absolutely essential for the government to collect such data. To ensure a transparent mechanism for new drug approvals, a Public Hearing could be organized involving physicians, pharmacologists and specialists in that specific therapeutic group by the drug controller. The drug company could be requested to furnish data indicating with which of the existing drugs the new drug has been compared in clinical trials and provide justification for the introduction of the new drug. This would ensure that the new drug in question is not only safe but also less expensive than the existing ones. The new Indian Patent (Amendment) Bill, March 2005 which was passed by Indian parliament suffers from ambiguity, technical loopholes and still fails to incorporate some of the flexibilities incorporated in the TRIPS, WTO regime. This has serious implications for the access to drugs and medicine in the country in specific and to the developing world in general. The amendments need to clearly spell out the scope of subject matter on patentability. The question of reasonable royalty to be paid on the issuance of compulsory licensing should be stated upfront and specific by indicating a cap on royalties to be paid to the patentee, say 4% of the total turnover in a year. For the mailbox drugs introduced post-1995, the Government need to specify what constitutes significant investment for the Indian companies manufacturing these drugs otherwise it may lead to unnecessary litigation. Government should consider incorporating in the immediate future, mechanism for automatic compulsory licensing. Given the fact that the Indian Patent Offices suffer from lack of adequate manpower and infrastructure, the discretionary powers vested on the patent office in terms of timelines of Rules could make the patent office vulnerable to vested interests. This is because Rules can be amended as and when
209
SECTION III
the patent office deems it fit leading to excessive discretionary powers. We recommend therefore that the rules be made more
transparent and at the same time strengthen the Patent Office in order to carry out its duties more efficiently.
210
SECTION III
References
Braga CAP Fink C, Sepulveda CP. Intellectual property rights and economic development. World Bank Discussion Paper No. 412. Washington: The World Bank; 2000. Challu, Pabblo. The consequences of pharmaceutical product patenting. World Competition 1991;15:65-126. Carsten F. How stronger patent protection in India might affect the behaviour of transnational pharmaceutical industries. Policy Research Working Paper No. 2352. Washington: The World Bank, Development Research Group; 2000:1-45. Government of India. Interim report of the Committee to Examine the Span of Price Control (including the trade margin) for Medicines. New Delhi: Department of Chemicals and Petrochemicals; 2004. Government of India. Mashelkar Committee Report, 2003. Government of India. Budget speech of the Finance Minister for the year 2001. Government of India. Demand for Grants, Ministry of Health and Family Welfare, 2001-02. Government of India. National Essential Drugs List, Ministry of Health and Family Welfare, April 1996. Government of India. Report of the Drug Price Control Review Committee. New Delhi: Department of Chemicals and Petrochemicals; 1999: 23-46. Government of India. The Drugs (Prices Control) Order, 1995. http:/www.tnmsc.com/system.html Government of India. National Sample Survey Organization. Household Consumer Expenditure Survey, 55th Round (1999-2000). Government of India. Interim Report of the Committee to Examine the Span of Price Control (Including the Trade Margin for Medicines. New Delhi: Department of Chemicals and Petrochemicals, Government of India; November 2004. Indian Credit Rating Agency (ICRA). The Indian Pharmaceutical Industry. Industry Watch Series, New Delhi: ICRA; July 1999: 37. Indian Drug Manufacturers Association (IDMA). Annual Reports, various issues. Lanjouw J.O. The introduction of pharmaceutical product patents in India: Heartless exploitation of the poor and suffering. National Bureau of Economic Research, Working Paper 6366;1998:1-53. Monthly Index of Medical Specialities (MIMS), New Delhi, various issues. Narayana PL. Indian pharmaceutical industry: Problems and prospects. New Delhi: National Council of Applied Economic Research; 1984. Nogues J.J. Social costs and benefits of introducing patent protection for pharmaceutical drugs in developing countries. The Developing Economies 1993;XXX: 24-53. Operations Research Group (ORG), Pharmacy Retail Sales Audit, Baroda, various issues. Organisation of Pharmaceuticals Producers of India, Annual Reports, various issues. Parmeswar, M. Drug Procurement Systems in India, paper submitted to National Commission on Macroeconomic and Health, December, 2004. Rane W. Rise in drug prices since 1987-an analysis. Economic and Political Weekly June 30, 1999:1375-9. Shiva M. Medicines, medical care and drug policy. New Delhi: Voluntary Health Association of India; 2000. Singh S. Multinational corporations and Indian drug industry. New Delhi: Criterion Publications; 1985. Srinivasan S. How Many aspirins to the rupee? Runaway drug prices. Economic and Political Weekly 1999; 34. The Hindu 16 May 2001, p. BS 1. The Hindu 26 June 2001, p. 13. Vohra U. TRIPS and the health sector in the South East Asia Region. New Delhi: WHO/SEARO; 1998: 1-59. Watal J. Pharmaceutical patents, prices and welfare losses: Policy options for India under the WTO Rules Agreement. World Competition: Review of Law and Economics 2000; 24:733-52. World Bank. World Development Report, various issues. World Trade Organization (WTO). Websites gateway to TRIPS. Available from URL: www.wto.org/english/ tratop_e/trips_e/trips_e.html. World Health Organization (WHO). World Health Report, various issues.
211
SECTION III
Annexure 1
Drug procurement and distribution system (Central Government) Ministry of Health and Family Welfare
Govt. medical stores depot
CGHS
Tertiary hospitals
National Health Programme
Supplies to paramilitary Few States forces
HLL
HSCC
RITES
HSCC (Delhi bulk)
Medical stores depot (Rest of India bulk)
Local purchase (Reimbursement)
States/regional warehouses
District warehouses
CHCs
PHCs
Subcentres
212
SECTION III
Appropriate Policies for Medical Device Technology: The Case of India
T
AJAY MAHAL
SRINIVAS TAMAN
BIOMEDICAL ENGINEER ANDHRA PRADESH VAIDYA VIDHANA PARISHAD HYDERABAD, ANDHRA PRADESH INDIA E-MAIL: anilvarshney@hcsintl.com
ANIL VARSHNEY
HEALTHCARE MANAGEMENT CONSULTANT, HEALTHCARE CONSULTANCY SERVICES E-MAIL: anilvarshney@hcsintl.com
HE TERM MEDICAL TECHNOLOGY IS GENERALLY TAKEN TO ENCOMPASS THE entire set of attributes associated with inputs that go into the provision of medical services. These include pharmaceuticals, medical devices, medical procedures and the organization of health services themselves (Mohr et al. 2001).1 A change in medical technology is usually taken to imply a change in one, or more, of the above attributes. Thus, the development of new drugs to treat people with HIV, the emergence of angioplasty and coronary-stents for coronary artery disease, and the development of magnetic resonance imaging (MRI) and Positron Emission Tomography (PET) for diagnostic purposes are all examples of changes in medical technology under this definition. In India, policy and research concern with the introduction and spread of medical technology been limited, thus far. The exceptions are pharmaceutical drugs and the regulation of diagnostics for sex determination of the fetus (Balakrishnan 1994; Govindaraj and Chellaraj 2002; Mudur 1999). Discussions on medical devices, when they have occurred, have focused on corruption and other problems in public procurement (Johnson 2003; Sudarshan 2003). In contrast, in developed countries, the subject of medical technology has attracted research and policy attention over a considerably wider area. A particularly fruitful line of inquiry has been the impact of medical innovations on health expenditures, and the pathways through which these expenditure increases occur. An influential strand of this literature argues that technological change accounted for more than 20 percent of the multi-fold increases in health spending that occurred in the United States during the period from 1980 to 2000, mainly due to increased volume of utilization and higher prices (Mohr et al. 2001; Newhouse 1992). Following from this, research in developed countries has tended to follow two directions: first, to analyze factors leading to the development and subsequent increased use of advances in medical technology; and second, to inquire whether the added expenditures yield gains in health that outweigh the costs. Examples of the former include examining the role of provider payment mechanisms, the system of medical education, learning processes among practicing doctors, education levels among potential consumers of care, defensive medicine in response to malpractice law and government regulations on the spread of newly developed technology; and on factors that influence the development of malarial drugs (Baker and Wheeler 1998; Bikhchandani et al. 2001; Bryce and Cline 1998; Danzon and Pauly 2001; Finkelstein 2003; Jonsson and Banta 1999; Lleras-Muney and Lichtenberg 2002; Kremer and Sachs 1999; Ramsey and Pauly 1997; Rosenthal et al. 2001, Weisbrod 1991). As to the question of whether added expenditures on medical innovations yield sufficiently large health gains, the central conclusion of the existing literature is that increases in expenditures associated with medical technology are not a social bad. Thus Cutler and McClellan (2001) conclude that improved heart attack treatments (such as angioplasty with stents) and new methods for neonatal care and depression have yielded life-expectancy gains that, when valued in monetary terms, are at least six times their increased cost. Lichtenberg (2004) argues that the launch of new chemical entities (drugs) accounts for almost 40 percent of the increase in life expectancy
1. Each of these terms can, in turn, be more precisely defined. For instance, the Global Harmonization Task Force (GHTF) defines a medical device as Any instrument, apparatus, implement, machine, appliance, implant, in vitro reagent or calibrator, software, material, or other, similar article, intended by the manufacturer for human beings for diagnosis, investigationsupporting or sustaining life (GHTF 2003, p.5)
213
SECTION III
in a sample of 52 countries during the period 1986 to 2000. Cutler and Meara (2001) examined the declines in mortality in the United States during the 20th century, and found that most of the declines are associated with technological advances - initially the emergence of antibiotics and later, better procedures for addressing cardiovascular disease and neonatal mortality. A recent survey of 225 U.S. primary care physicians identified magnetic resonance imaging (MRI) and computed tomography (CT), along with angioplasty as having contributed significantly to the length and quality of life of patients (Fuchs and Sox Jr., 2001); although the value of such diagnostic devices is contentious because populations of other developed countries such as Canada, continue to have excellent health systems and with much less reliance on MRI and CT-scan technology. Presumably for the reasons above, Deaton (2004) suggests that the rapid transfer of knowledge and skills made possible by closer global links has the potential of leading to great improvements developing country populations' health and consequently, of reducing inequalities in global health status. It is also not surprising thus, that Cutler and McClellan (2001, p.12) conclude, medical spending as a whole is clearly worth the cost (Italics ours).
Relevance of Medical Technology Discussions to Indian Policy Makers

The above discussion ought naturally to be of to concern Indian policymakers, and for several reasons. First, there are likely to be continued pressures on the demand side towards adoption of medical innovations. An increasingly open trade environment in India and heightened global interlinkages will likely increase the awareness of newer medical technologies in India and rising incomes, along with the spread of voluntary insurance will make such technologies more affordable to the average Indian. These tendencies towards increased demand will be accentuated by an ageing Indian population. Indeed as its population ages, many of the innovations in developed countries that have significantly greater numbers of elderly populations will become increasingly relevant to India's population. These tendencies are likely to be further exacerbated by medical tourism that is currently being promoted by the private sector and some government officials in India (Fernandes 2003). Second, there will be supply side pressures, as medical institutions seek to adopt the latest innovations in a bid to attract not only customers, but also leading medical professionals who might otherwise choose to practice elsewhere, or to migrate abroad (for example, Baru 1998). This will likely have a cascading effect on the nature of training provided in medical institutions - more diagnostic intensive, with presumably less focus on clinical skills. To this one can add increased efforts of suppliers of medical devices and other products to sell their products in rapidly growing markets such as India. It is, therefore, easy to project that with demand- and supply side- effects, the volume of new medical products in
India will expand. Prices may rise as well, as suggested by some analyses of the impact of India's drug patent regime moving from process patents to product patents. The limited public resources currently available to spend on health means that governments at the center and the state levels in India may need to set priorities regarding the use and adoption of medical innovations, and their diffusion, at the very least, in public facilities. One might suspect that, by their very nature, public sector budgetary limits force new medical technology adoption in public facilities to progress at a slower rate than in private institutions. However, success in this endeavor is not guaranteed, if there are incentives to obtaining new equipment and adopting newer procedures, including greater prestige, and the need to prevent poaching of medical personnel by private sector institutions. The existence of corruption in procurement procedures may also positively influence technology adoption. A paradoxical situation may arise where health care costs could nonetheless be increasing at a fast rate in the public sector without any corresponding gains in health, if the public sector functions inefficiently. Effort may be needed to shepherd developments in the private sector as well. The large amounts currently spent out of pocket by Indian households on health care do not eliminate the need for public policy on medical innovation, given a setting where doctors and suppliers of new technology are in a position to decide health services consumption patterns. Thus, public intervention may be needed, or safeguards introduced to ensure that the innovations used yield the highest health benefits relative to expenditures; and intervention may also be needed to address any inequalities in access that might result on account of differential physical and financial access to innovations. In thinking about these issues in the Indian context, a major handicap is the lack of good information on medical technology flows and the factors driving these flows, and the resulting impact on outcomes of interest - such as the cost of care, inequality in access to care, and ultimately, health outcomes. This paper is a first step in the direction of filling this gap, by bringing together existing data and new information on the way medical technology is diffusing in India, its use patterns, and in its potential implications. To keep the discussion manageable, we focus on technology embodied in medical devices. Four main research and policy questions are addressed: What do we know about the spread of new medical device technology in India and what are the main factors underlying this tendency? How effectively is available medical device technology in India being used in terms of its impacts on the costs of providing health care and on inequalities in access to health care? What is the state of regulations in India with regard to medical device technology? What is the appropriate strategy (including public/private partnerships) towards medical innovations and the avail-
214
SECTION III
able basket of medical technology and what can we learn from the experiences of developed countries in this regard? In practice, addressing these questions in research has proved to be difficult even in the United States, a country with rich data sources. In India, where data are sparse it is difficult to meaningfully address these questions beyond a small-subset of issues and categories of medical innovations. Thus, in analyzing how modern medical technological innovations embedded in devices are spreading and are being used in India, we focused primarily, but not exclusively, on diagnostic equipment such as MRI and CT-scans. An absence of domestic production for such diagnostic devices means that reasonably accurate estimates of the flows of such devices in India can be constructed from foreign trade data. Our analysis of import flows of modern diagnostic medical devices is supplemented, in the paper, with a discussion about the efficiency with which available medical devices, diagnostic and non-diagnostic, are currently being used in India. This analysis is valuable because it has the potential of highlighting the cost and effectiveness implications of the introduction of new medical devices. Inefficient use of existing medical devices has implications in that in a regime of changing technology, it may be a pointer to rising health expenditures without corresponding improvements in health outcomes of interest. We used data from several sources for our analysis. These include import statistics (on both quantity, and unit prices) from official foreign trade data of the Ministry of Commerce, utilization and expenditure information from household consumer expenditure surveys and health care utilization and expenditure surveys of the National Sample Survey Organization (NSSO). These data were supplemented by selected case-studies of the utilization of imaging equipment in the public and private sectors, recently undertaken by one of the authors on behalf of the National Commission on Macroeconomics and Health (NCMH); and by a detailed analysis of the functional status of medical equipment in public sector hospitals operated by the Andhra Pradesh Vaidya Vidhana Parishad (APVVP).
Flows of Medical Devices into India

Tables 1 and 2 present data on the volume, and the value, of imports of a selected set of diagnostic medical devices into India, such as CT-scanners, MRI systems, the linear ultra-sound scanner; angiograph, endoscopes and electrocardiograph (ECG).2 These devices all have the characteristic that they are predominantly manufactured outside India, so that import flows offer a reasonably accurate picture about their pace of diffusion into India. The data in these tables were obtained from commoditylevel foreign trade statistics compiled by the Ministry of Com-
merce, and careful readers will note a number of obvious shortcomings in the information presented. First, the categorizations used are potentially overlapping - for instance whole body scanners as recorded by Indian customs can be both the X-ray (CT-scans), or of the magnetic resonance imagining (MRI) variety; unfortunately, the official trade statistics do not make a clear distinction between the two. Moreover, the distinction between a CT apparatus and a CT-scanner is not obvious, since these terms are used interchangeably in the profession.3 Second, it seems that the volume units used for MRI Apparatus in the trade statistics data are not identical to a full MRI system since a quick calculation using the information the two tables reveals that doing so would lead to unrealistically low unit cost estimates for, say the most recent years 2000-3. Now some of this could be the result of imports of older MRI models and/or imports of used equipment. It is also possible that the term MRI apparatus refers to individual components of the MRI system, including major replacement parts, so that several such components make up a fully functional system. Thus these statistics cannot allow us to immediately infer how many completed MRI systems have been imported into India. Presumably, a similar concern holds for items under the term CT Apparatus as well. Despite these obvious data issues, the information in Tables 1 and 2 is still quite illuminating. Note that with perhaps a slight blip during the period 1994-7 both the volume, however measured as well as the real value of imports of medical devices have experienced sharp increases in the 1990s. For items that serve essentially as consumables, or have well defined units, such as catheters and endoscopes, there is a clear increase in utilization. For devices such as MRI's and CT-scans, the increase could also be due to the increased rate of imports of spare parts, as the cumulative number of devices present (installed) in the country increases over time, or new equipment. Both factors are likely to be associated with increased utilization. The data in Table 1 on trends in CT-scan imports and the extremely sharp rates of increase in CT Apparatus units is not inconsistent with this claim. Although we do not provide the calculations here, it can also be easily checked from the numbers in tables 1 and 2 that per unit cost (value/volume) for almost all of the devices examined here has either remained stable, or declined during the period under consideration. There are three possible scenarios consistent with this: (a) lowered prices of older models and their spare parts with medical innovation in developed countries; (b) newer models becoming available at prices that are essentially similar to the past prices for what now have become older models; and (c) changing composition of the Apparatus category for MRI and CT scans. Since (c) applies only to the case of categories CT Apparatus and MRI Apparatus, we conclude that innovation in the medical device sector is accompanying price declines in medical
2. A cardiac catheter is used as a diagnostic device. But unlike other devices discussed in Tables 1 and 2, it is a consumable (thrown away after use). 3. Nor can one simply guess that a CT-scanner (non-whole body) and a whole body scanner is a subset of CT apparatus because it can be readily checked that in some years the sum of the value of the two types of scanners, exceeds the value of the CT apparatus category.
215
SECTION III
Table 1 Import of selected medical devices to India by volume, 1991-2003

Device type 1991-94 Three year totals 1994-97 1997-2000 2000-03
Table 2 Import of selected medical devices to India by value, 1991-2003 (Rupees in millions)
Device type 1991-94 1994-97 1997-2000 2000-03
CT apparatus NA >73 CT scanner (NW) 113 167 MRI apparatus NA 78 Scanner (whole body) 68 61 Cardiac catheters (000s) 1092.54 1000.35 Electrocardiogram 171 231 Linear ultrasound scanner 742 1135 Endoscopes 1862 2114 Fibroscopes NA 627 Angiogram NA NA
206 181 113 49 1171.03 3713 1737 2526 1049 72
1810 176 807 116 1774.93 9347 4733 9590 2691 176
CT apparatus CT scanner (NW) MRI apparatus Scanner (whole body) Cardiac catheters (000s) Electrocardiograph Linear ultrasound scanner Endoscopes Fibroscopes Angiograph
NA 357.08 NA 422.94 542.32 102.12 388.63 97.00 NA NA
>53.81 187.41 557.75 213.04 473.47 109.60 689.66 125.33 47.55 NA
544.01 234.58 713.67 312.33 1621.18 289.03 816.16 108.65 71.53 567.05
1647.47 464.46 2687.96 436.45 2364.04 226.43 2477.50 399.02 90.42 804.11
Note: NW = CT scanner other than for the whole body; measurement units of CT and MRI apparatus are based on Indian Customs definitions.Source: Foreign Trade Statistics of India
Note: NW = CT scanner other than for the whole body; measurement units of CT and MRI apparatus are based on Indian Customs definitions; GDP deflator used to convert Rupee prices into 1993-94 prices.Source: Foreign Trade Statistics of India
devices, or quality improvements, or some combination of both. Notice that our results would be even stronger if the prices were expressed in US$ terms, since the Rupee depreciated against the US$ during this period at a rate much greater than the rate of inflation (Government of India 2004).
The Demand Side

So far we have looked at the supply-side picture and inferred trends in the spread of medical diagnostic technology in India - both in terms of units, as well as in terms of actual utilization of the equipment from import data. Corroborating evidence is available, even if not sufficiently device-specific, from household survey data on the use of diagnostic services. Tables 3 and 4 present information from household surveys on health care utilization and consumer expenditures in India. Data from two large household health care and utilization surveys suggest (in Table 3) that the likelihood of undergoing a diagnostic test, by an average inpatient, or by an average outpatient, increased during the period from 1987 to 1996, the two points in time at which the two surveys were respectively conducted. To be sure, the different categories of diagnostic tests (ECG versus ESG versus CT-scans, say) were not distinguished by the household survey questionnaire; but the thrust of the data seems clear enough. Similarly, table 4 shows that diagnostic expenditures by households nearly doubled during the period from 1993-94 to 1999-2000, whether taken as a proportion of aggregate household spending, or as a proportion of aggregate health care spending by households. Even more remarkably, diagnostic expenditures accounted for one-fourth (25 percent) of the increase in the share of health care spending by households that occurred during this period. The evidence in table 3 on the increased per patient usage of diagnostic services suggests that at least some of the increase in expenditures would have been accounted for by increased use of diagnos-
tic services. Taken together these bits of information in tables 3 and 4 suggest that: (a) Diagnostics use is increasing over time in India; (b) that people are paying more often for diagnostic services; and the net result of these tendencies is that the overall share of diagnostic care spending (which is the result of some mix of increased use and increased payment) in total household budgets is also increasing over time. Why did this happen? There are a number of candidate reasons. For a start, the spread of new diagnostics can be expected to be the natural outcome of scientific progress. This process would also likely have been facilitated by the liberalization in the foreign trade regime in India, a process that took root in the early 1990s. Unfortunately, it is not straightforward to test this latter hypothesis since the definitions of various commodities in foreign trade records were not spe-
Table 3 Proportion of patients getting an X-ray/ECG/ESG scan in India, 1986-87 and 1995-96
Care type and residence X-Ray/ECG/ESG/ scan 1986-87 1995-96
Inpatient Rural Urban Total Outpatient Rural Urban Total
33.63 45.16 36.82 2.90 5.47 3.57
43.06 52.07 46.39 3.61 6.34 4.41
ECG: electrocardiograph; ESG: electrosonogramSource: 42nd and 52nd rounds of the National Survey Sample Organization's household surveys.
216
SECTION III
Table 4 Diagnostic, health and total expenditure of Indian households, 1993-94 and 1999-2000
Expenditure categories Rural 1993-94 Urban Rural + Urban Rural 1999-2000 Urban Rural + Urban
Inpatient Diagnostic Exp /Total HH Exp (%) Diagnostic Exp /Total IP Exp (%) Total IP Exp/Total HH Exp (%) Outpatient Diagnostic Exp /Total HH Exp (%) Diagnostic Exp /Total OP Exp (%) Total IP Exp/Total HH Exp (%) Inpatient + Outpatient Diagnostic Exp /Total HH Exp (%) Diagnostic Exp /Total OP+IP Exp (%) Total IP+OP Exp/Total HH Exp (%)
0.05 5.47 0.89 0.06 1.23 4.55 0.10 1.92 5.44
0.05 3.99 1.19 0.09 2.52 3.42 0.13 2.90 4.60
0.05 4.85 1.00 0.07 1.60 4.15 0.11 2.23 5.15
0.09 6.82 1.37 0.15 3.08 4.72 0.24 3.92 6.09
0.10 7.16 1.44 0.15 4.21 3.62 0.26 5.05 5.06
0.10 6.95 1.40 0.15 3.43 4.31 0.25 4.29 5.71
HH = Household; IP = Inpatient; OP = Outpatient Source: Consumer Expenditure Surveys of the National Survey Sample Organization, 1993-94 and 1999-2000
cific enough to identify imports of specific items such as MRIand/or CT-scanners in the period prior to 1991. Other factors are likely to have played a role as well. Given India's health system, where the bulk of health care spending is by households, technology innovation will also be driven by consumer demand expressed in terms of purchasing power. The period since the early 1980s has been characterized by rapid increases in incomes in India, which may very well have contributed to the rising demand for better quality care, including better diagnostic services. Since this period has also been a time of severely constrained government budgets, one might naturally expect to see any evidence of such a tendency in a growing private sector. Thus table 5 which presents survey data on whether households who obtained diagnostic services paid for them, or not, indicates that the share of free diagnostic services has declined over time. This is entirely consistent with the evidence in table 4 which shows increased proportions of household spending directed to diagnostics.
Into the above mix, one can add the role of medical practitioners and diagnostic service suppliers themselves in promoting the use of diagnostic services. It is well-known, for instance, that many medical practitioners in both the public and private sectors have informal contracts with private providers of diagnostic services and pharmacies that yield them a commission on each referral made to the concerned pharmacy or diagnostic service provider. Financially large investments in diagnostic equipment put extra pressure on diagnostic service providers to offer incentives to individuals (qualified and unqualified practitioners) who may be in a position to offer such referrals. Baru (1998, pp.112-4) cites evidence from Hyderabad that this commission could be as much as 10-15 percent of the cost of a diagnostic test. Varshney (2004) found that an average of 10 percent of total expenditures of diagnostic service providers consist of business development payments to doctors; and the share may be as high as 30 percent for high-end diagnostics such as MRI and CT scans.
Table 5 Patients getting an X-Ray/ECG/ESG, by payment mechanism, All India, 1986-87 and 1995-96
Care type and residence Free 1986-87 (%) Part-free Payment Free 1995-96 (%) Part-free Payment
Outpatient Rural Urban Total Inpatient Rural Urban Total

ECG: electrocardiogram; ESG: electrosonogram Source: NSSO household surveys of 1986-87 and 1995-96
21.58 29.16 24.63 39.69 46.22 41.91
5.28 5.49 5.37 3.12 3.70 3.32
73.14 65.35 70.01 57.19 50.08 54.78
9.14 11.16 9.69 35.75 41.94 38.01
0.35 1.09 0.55 10.57 13.69 11.71
90.51 87.75 89.76 53.68 44.37 50.28
217
SECTION III
Not all centers give incentives, of course, and commissions are especially common among unqualified medical practitioners. Overuse may also result on account of internal referrals in corporate/private hospitals where there may be performance targets for consultants.
Cross-country evidence
To supplement the discussion on the role of different factors in influencing the spread of medical technology, we carried out a regression of a measure of MRI imports on a set of supply- and demand-side explanatory variables, for a set of non-MRI manufacturing, primarily developing, countries. Using data from the World Bank's World Development Indicator's database and the United Nations, we inquired whether inflows of MRI equipment into countries were systematically related to countries' levels of per capita income (a proxy for effective demand), doctor-to-population ratios (a catch-all for supplier driven factors) and the role of foreign-aid (a demand side factor). We used country-reported import data on MRI equipment flows in a sample of 49 MRI equipmentimporting countries (with negligible capacity to produce MRI equipment on their own). The main findings are reported in table 6. While it is difficult, at this point, to ascribe causality to the relationships for obvious econometric reasons, the fairly strong relationships (in the expected direction) between imports of MRI equipment per capita, per capita real GDP and the doctor-to-population ratios are worthy of note.
Are these flows achieving the desired objectives?

The obvious question is: Did the spread of diagnostic devices in India improve outcomes valued by policymakers, relative to the expenditures incurred? Answering this question is not
straightforward, because although one can claim on the basis of the household expenditure data that medical diagnostic devices import inflows reflect increased demand and contributed to increased diagnostic services utilization and spending in India, the impact on outcomes such as access and equity is less clear; even less so for health improvements One rough method to check for efficiency in resource use is whether the supply of equipment such as magnetic resonance imaging systems in India was excessive, relative to some pre-agreed norm. Alternatively, one could look to whether the equipment is underutilized, relative to some notion of full capacity. A study for the state of Pennsylvania in the United States suggests a norm that ranges between 3,000 and 3,500 scans per MRI per year, as appropriate (Bryce and Cline 1998). Alternatively, one can try examining the number of MRI sites per capita in other countries and take that as the norm. Baker and Wheeler provide an estimate of about 1.45 MRI sites per 100,000 people in the United States in the mid-1990s. The use of United States data to develop a norm is, needless to say, troublesome given that there are quite legitimate concerns about excessive medical technology and medical expenditures in that country, relative to health outcomes achieved. Thus, the situation in other countries that may have managed their health resources somewhat more efficiently ought also to be considered. Rublee (1994) provides estimates of 0.11 MRI per 100,000 people in Canada. This range of MRI per-population estimates - between Canada and the United States - can serve as a norm for our purposes. The Radiology Association of India (RAI) website estimates that roughly 50 MRI's and 350 CT-scanning facilities currently exist in India, whereas a recent estimate based on discussions with wholesalers of diagnostic equipment assesses the number of MRI's to be of the order of 70-100, and CT scans to be about 300 (Varshney 2004). However, these appear
Table 6 Correlating MRI imports to potential explanatory variables

Explanatory variable Dependent variable: Average MRI imports (2001-2003) per capita Model I Model II Model III Except Africa Model IV Except Africa
Constant Per capita GDP (1995 US$) Doctors per 1000 population Foreign aid per capita (US$) Average MRI imports 1998-2000 per capita N R2
-0.395 (0.155) 0.083** (0.016) 0.092** (0.039) 0.021 (0.020)
49 0.508
-0.369 (0.154) 0.070** (0.018) 0.084** (0.039) 0.020 (0.020) 0.300 (0.202) 49 0.531
-0.377 (0.184) 0.084* (0.045) 0.086** (0.018) 0.011 (0.027)
40 0.488
-0.365 (0.183) 0.076** (0.020) 0.080* (0.045) 0.011 (0.027) 0.253 (0.226) 40 0.506
Note: Regressions are based on data from the United Nations and World Bank. All major exporters of MRI products were excluded from the sample.** Statistically significant at the 5-percent level of significance.
218
SECTION III
to be serious underestimates of the number of CT scans, since Table 1 suggests at least 931 CT-scans in India (we have included only whole-body scanners in our list of CTscans), even if we ignore items listed under CT-apparatus.4 That would suggest that the actual number of CT-scans exceeds RAI estimates by nearly 166 percent. If we assume similar rate of RAI underestimation for MRI's, the estimated number for magnetic resonance imaging sites in India is 133, which translates into 0.0133 per 100,000 people. If we combine our estimates of the number of CT-scan and MRI facilities, that still comes to only about 0.11 CT/MRI units per 100,000 people. Overall, therefore, the number of CT-scan and MRI diagnostic facilities in India does not seem to be excessive, even in comparison to Canada. There may be an issue about distribution of diagnostic equipment sites though, since high-end diagnostic facilities such as these are typically located in urban areas, particularly major metropolitan areas. Taken as a proportion of India's total urban population only, the estimated number of MRI/CTscans in India constitute about 0.39 per 100,000 people. Even this is substantially lower than just the number of MRI sites per capita in the United States, and is almost certainly likely to be lower than the combined MRI/CT per capita numbers for Canada. Another way to try to infer excessive supply (or otherwise) of diagnostic equipment is to examine utilization rates in relation to some standards. For instance, if utilization rates are too low, one may judge that there are too many medical devices in the market.5 A recent study obtained information on two Delhi hospitals, one public and one private, and one stand-alone private diagnostic facility in Delhi, for this purpose (Varshney 2004). The findings of the Varshney case studies are rather stark. In the private sector, the MRI unit conducted 7,500 scans per year while being operational for a total of 360 days a year. In contrast, the public sector MRI facility was used for only 740 scans, and the facility was operational only 300 days per year. Clearly the public MRI unit appears to be seriously underutilized. Whether this indicates excess capacity, relative to need, is unclear since the poor may forgo diagnostic services altogether if there are problems of access. The functioning of this unit at below capacity, if symptomatic of a broader problem with public sector facilities, would suggest that poorer groups have unequal access to new technology, even when subsidized by the public sector. There are good reasons to believe, however, that there is geographic inequity in the location of diagnostic sites, and that may indicate spatial inequity in access as well. Data for 70 MRI sites identified in Varshney (2004) suggest a lopsided distribution: 63 percent (44) of the sample MRIs were located in 5 major cities (Bangalore, Chennai, Delhi, Hyderabad and Mumbai) with a combined population of no more than 45 million (or 4.5 percent of India's population), and composed of the most well off individuals in India. Thus, one adverse outcome of the introduction of state of the art diagnostic
services, at least at the present time, is inequity in access to high technology health care, whether valuable or not for health outcomes. The cross-country evidence in table 6 suggests similarly that modern diagnostic technology is likely to be directed towards richer countries/areas with high doctor-topopulation ratios. Another area of concern is misuse of technology. Policymakers in India have been particularly concerned about the use of diagnostic services such as ultrasound for sex determination, and implications for female feticide. While the practice has been banned in India, it is commonly understood that it still continues illegally, given that both the user (demander) and the supplier of diagnostic services gain from it, and monitoring is potentially costly.
Efficacy of Medical Equipment Use in the Public and Private Sectors

The previous section focused on advanced medical diagnostics and suggests that in the aggregate there may not be an excess of diagnostic devices such as CT-scans and MRI sites in India. It presented some evidence of regional inequity in location; and it briefly pointed to a problem with an existing mechanism (public sector provision) to partially address inequities related to financial access. That is, government facilities that are often the sole affordable source of advanced technological devices to the poor do not keep their equipment functional, or are otherwise unable to preferentially provide services to the poor. How policymakers handle the entry of new technology obviously has important health policy outcomes. The above discussion also suggests that thinking on policy approaches to address medical devices and the technology they embody needs to go beyond the effective harnessing of the new technologies. In particular, an examination of the effectiveness with which health facilities in the public and private sectors currently use their equipment is potentially very valuable. This would help focus attention on health system features that might lead to wastage of resources if left unattended at a time of technological change, and refocuses attention on the challenge of efficiently providing public sector health services to the less well off.
Equipment Use in the Public and Private Sectors

We use two sources of information on the public sector utilization of medical devices, mainly durable equipment: for the state of Andhra Pradesh from the Andhra Pradesh Vaidya Vidhana Parishad (APVVP); and from a study undertaken for the National Commission on Macroeconomics and Health by Varshney (2004). Information on the private sector is primarily from Anil Varshney (2004). The data from APVVP covered 74 community health centers, 55 area hospitals and 21 district hospitals run in Andhra
4. We assume that all CT-scans purchased prior to 1991 are no longer in use. 5. Of course, assessing utilization rates of equipment may be tricky in assessing optimal capacities if there is supplier induced demand.
219
SECTION III
Pradesh. Unfortunately this data cannot always be separately broken down by diagnostic and non-diagnostic equipment. A priori, however, there is no reason to believe that findings for the two sets of equipment ought to be different. This data highlights several areas of concern to policymakers with respect to equipment in government health facilities. In particular, (a) government facilities face an acute shortage of basic equipment; (b) the equipment on the premises is not always functional; (c) and there are potentially serious problems with regard to time taken for installation and repairs. Similar findings for a selected set of developing countries are presented in Mavlankar et al. (2004). Consider availability. If even the most basic equipment is unavailable, the introduction of newer technologies, if it were to occur, would lead to inefficient use of resources, especially if cheaper investigations were substituted by more expensive ones. As of 2004 the value of medical equipment at community health centers, district and area hospitals under APVVP ranged between 70-85 percent of that required under norms focused on acquisition of basic, not the most advanced technology. Notice that this superior situation occurred after a long period of stewardship and World Bank support, and unlikely to be representative of other, more backward states in India. Their situation would be more akin to APVVP hospitals in 1993 - when available equipment ranged from 25 percent to 75 percent in value relative to norms set by the government. Even when equipment was available, it was not fully functional. This possibility raises questions about whether the new technologies, if introduced, can effectively be used at all. In 2002, between 45-51 percent of major medical equipment at area hospitals and community health centers, was classified as either non-usable, idle, or with low utilization rates. Only at the high-level district hospitals was the situation better, with an average of 15 percent for the three categories; in 1993, the situation was, of course, much worse with 28 percent of the major equipment, even in district hospitals, being either underutilized, or not functional. There are other kinds of wastage as well. For instance, it took an average of between 2-4 months to install X-ray and ultrasound equipment at the from the time it was received at a APVVP run district hospital during the period 2000-2, with the lag being substantially greater for lower-level area hospitals and community health centers. Even these lag times were substantial improvements over previous periods. The findings for APVVP hospitals are reflected in the casestudies undertaken by Varshney (2004) of diagnostic devices at public hospitals in New Delhi. For instance the time from ordering to actual commissioning of MRI, CT-scan and Ultrasound equipment at the public hospital was four times that of comparable private facilities. Delays occurred at every-stage in the ordering and delivery process at the public hospital deciding upon the type of equipment needed, clearance of payments to the supplier of the equipment, incomplete electrical and other pre-installation preparatory work at the time of receipt of the equipment. This does not include the time taken for needs assessment a process that could poten-
tially take years at a public hospital. In addition, utilization rates following installation were not always up to the mark, as indicated by the number of cases scanned by the MRI unit at the public hospital. The latter may reflect more than just a breakdown of equipment - as we discuss in some more detail later. Varshney (2004) also undertook an analysis of public and private diagnostic facilities in one district in Rajasthan. The findings are similar to those from Andhra Pradesh and New Delhi: that relative to private facilities, the down-time in public hospital equipment was greater, reflecting fewer operational hours as well as the poor functional status of equipment. The obvious implication of the inefficiencies outlined above is that the cost of production of diagnostic services (and indeed for all other types of equipment) and the overall quality of service is likely to be different under public and private sector managements and operation. Table 7 reports unit cost findings that have been derived from data presented in Varshney (2004, Table 4.2.1). Even ignoring the costs of delays and consequence foregone benefits in improved health in the public sector, the evidence suggests that private sector investi-
Table 7 Unit cost calculations of diagnostic investigations in New Delhi

Institution Estimated average cost of investigation Ultrasound CT Scan MRI scan Others
Public hospital Private hospital Stand alone diagnostics centre
589 350 503
2700 3333 1999
50250 NA 4285
29 45 26
All figures in INR (total cost includes fixed consumable cost) Note: Estimates are based on calculations and numbers reported by Varshney 2004.
gation costs are somewhat lower than in public facilities, with the case of MRI being especially stark in this regard. Unit cost calculations based on data from Rajasthan are similar in spirit to the results from New Delhi, and in some ways highlight the unique problems faced by lower-level public health facilities in India. The ultrasound equipment at the facility was non-functional, so that even though technicians were being paid and space occupied, no diagnostic investigations were done. But these unit cost estimates form only a part of the picture, since there are significant quality differentials in service provision. For instance, an outpatient visitor scheduled for an ultrasound had a typical waiting time of 2 months, and a month or more for a CT-scan in the public hospital in New Delhi. The wait list for an inpatient ranged from 3-10 days for diagnostic services in the public hospital. Moreover, following completion of an examination, the report was available typically after a delay ranging from 3 to 5 days, and hard copies of the report were not usually accessible to the patient. This is to be contrasted with private services where the services and the report were typically available on a
220
SECTION III
same day basis. In addition, the mode for reserving slots for undertaking the tests, making payments and completing other administrative activities appear to be considerably more complicated in public facilities (Varshney 2004).
Why are there such differences in quality and cost in the public and private sectors?
The proximate causes are obvious: non- or partly-utilized equipment, resulting in fewer investigations, with personnel and other costs either similar to, or higher than in the private sector. But what are the underlying reasons for this state of affairs? The causes of the poor functioning of equipment in the public sector, relative to the private sector range cover a wide range - from the unavailability of personnel needed to operate it (the absence of a radiologist in the district hospital in the Varshney study explains the lack of utilization of the ultrasound machine), poor co-ordination of procurement and installation processes, poorly trained staff and a general lack of accountability. Many of these same factors, together with financial constraints explain why when equipment in public facilities runs into a shortage of spare parts or otherwise experience technical problems, it takes a long time to get running again. For instance, suppliers of medical equipment point out that, public sector facilities take a long time to pay outstanding dues and there are problems with corruption. Moreover, personnel in these facilities tend to delay the reporting of problems with equipment. Poor follow-up and/or financial shortages mean that government agencies sometimes do not insure equipment once the warranty period has expired - and that may render equipment non-functional without any financial redress as soon as it runs into a technical hitch. These problems are particularly severe in public facilities that lie outside the major metropolitan areas, since their financial and human resource constraints are even greater. In contrast, Varshney (2004) points out the obvious advantages that arise on account of a clear line of accountability and financial risk bearing in the private sector. He compellingly argues that the direct consequence of financial accountability are that response time to potential problems is much faster, getting better trained staff and careful handling of equipment gets high priority and maintenance and insurance contracts that minimize financial risk are common, particularly for major pieces of equipment.
Problems with the medical device supply and maintenance industry in India
Of course, the private sector has its own problems, as reflected previously in the discussion on the possible overuse and misuse of diagnostics and other medical devices in India. These concerns often lie at the root of policymakers' efforts to regulate private providers. Then there are problems further down the supply line. First, there is effectively no quality regulation on the sale of high-tech medical devices, with existing ISI (Indian Bureau
of Standards mark) standards limited to a small subset of lowcost medical equipment. This is in contrast to strict quality controls on what can be sold/imported in the countries of the European Union and the United States (see below), and even China. Imports of second-hand medical devices in some categories of up to 10 years old are also allowed into India (Harper 2003) with the consequence that a lot of substandard second-hand medical devices are currently flowing into and around the country. The only regulation that currently exists relates to protections relating to radiation. But there is little or no control on what the equipment does relative to its claimed effects, its technical specifications and the like. In addition, however, both private and public health facilities and diagnostics providers face problems related to the continued operation of medical equipment in India, so that costs of medical device operation are higher than they would otherwise be. Availability of good quality spare parts is a serious problem faced by both the public and private health service providers in India. While especially acute for older equipment spare parts for which are no longer made by the original manufacturer, the absence of any sort of oversight in the medical device market means that there are a lot of equipment suppliers who simply do not deliver follow-up services, making it costly search exercise for purchasers to sort through alternative providers. This is an important issue, because the expenses on spare parts of diagnostic equipment typically tend to exceed by several times, the original cost of the equipment over its lifetime; and because of the rapidly changing imaging technology which makes new models obsolete almost as soon as they begin operation. A related challenge is a severe shortage of technical experts for repair work when needed, on medical equipment. Varshney (2004) notes that companies selling the equipment have probably the best engineers, but they often engage third parties, whose personnel are not as skilled, to help with the execution of maintenance contracts. The shortage of company engineers means that only the very persistent clients are able to get hold of them for maintenance and repair needs. In general, and for reasons mentioned above, the private sector is able to manage this process better than the public sector. Public sector facilities located in areas outside major cities are the most severely hampered, thereby contributing to long idleness times for equipment and a resulting wastage of resources. The option of engaging company engineers is not even available to those who obtain second-hand equipment. One might reasonably argue that the market will do the sorting, with more reliable suppliers pushing out the less-reliable ones. Over time this may well turn out to be the case. But the adjustment process may well be long and costly, as appears to be the case in India. And it is unclear how and whether a resource-constrained public sector, and its facilities in remote areas, will be able to adequately respond to these adjustments as they occur. With rapidly changing technologies, the process may be even more arduous as purchasers are asked to sort through increasingly complex technical specifications.
221
SECTION III
What needs to be done?

The set of recommendations proposed here are intended to serve only as a guide for more detailed policy responses, and mainly reflect the concerns outlined in the preceding sections. We divide our discussion on policy recommendations relating to medical devices into two: (a) regulatory recommendations on the new and second-hand medical devices market; and (b) recommendations on health systems aspects of the medical device use, including the potential for public-private partnerships. As noted above, in India there is essentially very little regulation of the medical device industry; even less by way of quality-, or benefit-cost assessment. In thinking about the appropriate policy steps to take, note that countries in the European Union, the United States, and Canada have, at the minimum, regulations that require devices perform as claimed by their manufacturers, or sellers, before any product can be marketed. In the United States this regulatory responsibility is executed by the Food and Drug Administration (FDA). In the European Union, this function is essentially that of an autonomous implementing agency, known by different names in various countries (e.g., Medical Device Agency in the United Kingdom). Typically the process involves suppliers being required to produce documentation on performance, and it may also involve verification, such as by independent (privately run) notified bodies that undertake this for the EU in consideration for a fee. The assessment also typically includes meeting the requirement that any harmful effects of the device (adverse health outcomes) are an acceptable risk. Next, there are requirements that ensure that any harmful effects that come to light after approval of market entry are also covered by regulation, including possible withdrawal of the permission to enter the market. Typically, this process involves some form of record-keeping in the form of a history of adverse incidents, and associated steps and sanctions. It may also involve voluntary reporting by patients and users of the equipment, or statutory reporting by manufacturers and diagnostics service providers. The regulatory authority is also responsible for putting out safety notices for information to the general public. These two requirements appear sensible. However, it is arguable whether an India-based regulatory authority and/or autonomous entities are capable of undertaking the quality checks required at this point in time. We understand that a committee of the Indian Council for Medical Research (ICMR) recently proposed the setting up of an Indian Medical Devices Regulatory Authority (IMDRA) along these lines. This recommendation needs to be acted upon, but as an independent authority, and NOT under the Director General of Health Services (DGHS), as proposed by the government. When formed, the IMDRA may find it worthwhile to piggyback on publicly available information on licensing status and medical device performance from either the European Union, or the FDA, or both. There are, however, areas where the proposed Indian Medical Device Regulatory Authority can potentially be extremely
useful. This is in the area of ensuring some order in the medical device market - to distinguish fly-by-night operators from more reliable sellers of devices, to ensure that sellers of equipment provide adequate levels of spare parts and technical training, to maintaining price lists and the like. Presumably, the effectiveness of this effort may require working in collaboration with the buyers of such equipment and its sellers. In particular information on the different sellers and their terms and conditions ought to be available at this regulatory agency. This could be linked to some compulsory registration mechanism, again developed in consultation with the sellers of equipment and purchasers. Once we are past basic quality requirements and the requirements of clinical efficacy, issues of cost-effectiveness become pertinent - that is, are the outcomes achieved by the medical device worth the cost? This raises questions about whether there need to put limits on the number of medical devices overall, across regions and the like. It also raises questions as to how to rank different medical devices by health sector priority; whether the public sector ought to purchase them; and ultimately what to do once priorities have been defined. The technical method of addressing such questions comes under the rubric of technology assessment, and several countries do departments undertaking medical technology assessment. To set up such offices means having personnel with a collective range of skills in bio-engineering, law, medicine and social sciences, and they are often politically extremely sensitive because of the potential impact their recommendations may have on medical device markets. For partly these reasons, the office of technology assessment in the United States was a casualty in the early 1980s, having been set up just a few years earlier. The role of technology assessment is obviously valuable, however. In a resource constrained setting such as India's, relying solely on the market to guide the growth of medical technologies may lead to a lot of wastage. Nor is it easy to focus on the public sector alone, if doing so leads to manifest inequities in access; or, a loss of highquality personnel to the private sector. Several countries have experimented with (and many continue to do so even today) on various additional restrictions on the number of medical devices. These include certificate of need (CON) requirements, which require establishing a need for a facility in an area, prior to getting a license for it. Some provinces in the United States, such as Pennsylvania, as well as countries such as Australia, Canada and Netherlands do have CON requirements, although the effectiveness of such measures in curtailing the spread of technology has been questioned (Cline and Bryce 1998). It might also be argued that in India, it is much too early to be thinking about CON requirements given that many major diagnostic medical devices such as MRI's and CT-scans are barely making their entry into the market. Another issue of concern is that CON requirements might restore the license-permit raj with its concomitant implications for corruption. Thus attention has been directed to other, market-centered, mechanisms by which the objectives of cost-effectiveness can be met, fully or partially. One option is to create incentives
222
SECTION III
for the integration of the role of health care providers (including diagnostics service providers) with that of health insurers, in conjunction with a prospective payment (capitation fee) system. This combination ought to reduce incentives for over-consumption of health care generally, including diagnostic care. HMOs are an obvious example of this phenomenon, and there is some evidence that HMO concentration has curtailed the pace of MRI diffusion in the United States (Baker and Wheeler 1998). Another method may be to educate physicians better about medical technologies, including not only their benefits also their economic and potentially harmful side-effects; similar efforts could be directed at students in medical schools in India. There is some evidence that physicians do respond to information of this type in a way as to reduce the use of harmful expensive technologies, although relying on this option is probably not the preferred option.
Recommendations on health system features

In addition to a better regulatory approach to the medical device market, there are other health policy-related activities the government could do, to address several of the inefficiencies discussed earlier in the paper. In thinking about appropriate strategies, we are guided by the consideration that excessive reliance on regulation and outright bans is, in light of the Indian experience historically, unlikely to work effectively. As our first example, consider the challenge of the misuse of diagnostic medical devices. The use of ultrasound scanning equipment for sex determination continues till today despite a government ban in India. With both the demander (consumer) of services and supplier of services in a mutuallybeneficially arrangement, the government is unlikely to be effective with purely punitive measures. Plus, overly strict regulation tends to be abused by authorities to harass service providers and doctors. Government policies may thus need to take the form of small steps that ensure that only trained radiologists operate such machines; that there are education campaigns against this practice (currently ongoing); and perhaps ensuring that better information is available on the spread of ultrasound equipment, so that policy efforts can be more effectively directed to geographic areas and communities deemed to be most at risk. Now consider the issue of overuse. It is generally difficult to pinpoint overuse as defensive medicine is likely to become increasingly prevalent in the future, patients are typically in favor of more technology intensive interventions, and there are no set norms for optimal use. As the Indian market
increasingly opens up to insurance companies, however, it is entirely possible that instead of the expected increases, diagnostic use may be limited by insurance audits. For instance, Ramsey and Pauly (1997) found that even fee-for-service insurance plans led to curtailing of excessive diagnostic equipment use. To this one may add a policy of extending the medical code of ethics to establishments that employ doctors (even if not owned by doctors), and grievance cells that involves professional associations and the medical device regulatory authority. One may also want to think about developing model guidelines for doctors to follow in assessing patients, although this again relies on self-regulation that has not worked well in India thus far. A final issue revolves around ensuring that resources invested in medical equipment in public hospitals are not wasted owing to non-use, particularly in smaller towns and cities. These mean first that procurement and installation processes have to improve. The example of APVVP cited above suggests that this can be done, by hiring technically proficient staff and by empowering them. Decentralized financing authority to hospital committees would also help. Finally, at least in smaller towns, the need for better trained staff to operate and repair equipment is critical. Perhaps large scale contracts with sole suppliers, in return for extensive skills training and maintenance support may be the way to go. Varshney (2004) suggests that one could explore the training of local districtlevel staff who could serve as franchisees to the supplier. This could help avoid the costs that result from delayed response to repair requests from the government hospitals. In connection with training, there is obviously also a great need to train clinical engineers through courses offered at, say the Indian Institutes of Technology. Such curricula are readily available at institutions in the United States and elsewhere.6 In the concern for more effective usage of medical devices in India, one could consider an alternative scenario whereby the public sector could hand over some of its responsibilities to private providers. Varshney (2004) gives examples of three case studies (in Delhi, Meerut and Kolkata) of privatepublic partnerships in the provision of diagnostic services along these lines - with the private partner operating the equipment in space made available in the premises of the public hospital. The chief gain to the private provider was in terms of a ready clientele, whereas the public sector hospital benefited in terms of proportion of patients getting free services and a functioning facility. The experience has tended to be mixed, owing mainly to a shortage of patients going to the facility - a combination of doctors referring patients to outside facilities in return for a consideration, patients seeking second opinions prior to get the diagnostic done, and the like.
6. We are grateful to Dr. Valiathan for this point.
223
SECTION III
References
Baker L, Wheeler S. Managed care and technology diffusion: The case of MRI. Health Affairs 1998;17:195-207. Balakrishnan R. The social context of sex selection and the politics of abortion in India. In: Power and decision: The social control of reproduction. Cambridge, MA: Harvard School of Public Health; 1994:267-86. Baru R. Private health care in India: Social characteristics and trends. New Delhi: Sage Publications; 1998. Bikhchandani S, Chandra A, Goldman D, Welch I. The economics of iatroepidemics and quackeries: Physician learning, informational cascades and geographic variation in medical practice. (Draft). Los Angeles, CA: University of California, Los Angeles, Anderson School of Business; 2001. Bryce C, Cline K. The supply and use of selected medical technologies. Health Affairs 1998;17:213-24. Cutler D, McClellan M. Is technological change worth it? Health Affairs 2001;20:11-29. Cutler D, Meara E. Changes in the age distribution of mortality over the 20th century. NBER working paper no. 8556. Cambridge, MA: National Bureau of Economic Research; 2001. Danzon P, Pauly M. Insurance and new technology: From hospital to drugstore. Health Affairs 2001;20:86-100. Deaton A. Health in an age of globalization. NBER Working Paper No. 10669. Cambridge, MA: National Bureau of Economic Research; 2004. Fernandes C. Promoting health and medical tourism in India. Express Travel & Tourism, September 2003. Finkelstein A. Health policy and technological change: Evidence from the vaccine industry. NBER working paper no. 9460. Cambridge, MA: National Bureau of Economic Research; 2003. Fuchs V. Health care for the elderly: How much? Who will pay for it? NBER working paper no. 6755. Cambridge, MA: National Bureau of Economic Research; 1998. Fuchs V, Sox H Jr. Physicians views of the relative importance of thirty medical innovations. Health Affairs 2001;20:30-42. Global Harmonization Task Force (GSTF). 2003. Information document concerning the definition of the term medical device. Document no. SG1/N029R13. http://www.ghtf.org/ index.html (accessed on 2 February 2005). Government of India. Economic Survey of India 2003-4. New Delhi: Government of India, Ministry of Finance; 2004. Govindaraj R, Chellaraj G. The Indian pharmaceutical sector: Issues and options for health sector reform. World Bank Discussion Paper No. 437. Washington, DC: The World Bank; 2002. Harper S. Global import regulations for pre-owned (used and re-furbished) medical devices. Washington, DC: United States Department of Commerce, International Trade Administration; 2003. Johnson T. Hospitals junk hi-tech equipment. Times of India, Bangalore, 4 March 2003. http://timesofindia.indiatimes.com/ cms.dll/html/uncomp/articleshow?artid=39207874 (accessed on 30 December 2004). Jonsson E, Banta D. Management of health technologies: An international review. British Medical Journal 1999; 319:1293. Kremer M, Sachs J. A cure for indifference. Financial Times, 5 May 1999. Lleras-Muney A, Lichtenberg F. The effect of education on medical technology adoption: Are the more educated more likely to use drugs? NBER Working Paper No. 9185. Cambridge, MA: National Bureau of Economic Research; 2002. Lichtenberg F. The impact of new drug launches on longevity: Evidence from longitudinal disease-level data from 52 countries 1982-2001. NBER Working Paper No. 9754. Cambridge, MA: National Bureau of Economic Research; 2004. Mavalankar D, Raman P, Dwivedi H, Jain M. Managing equipment for emergency obstetric care in rural hospitals. Working Paper No. 2004-03-08. Ahmedabad, India: Indian Institute of Management; 2004. Mohr P, Mueller C, Neumann P, Franco S, Milet M, Silver L, et al. The impact of medical technology on future health care costs. Bethesda, MD: Project HOPE, Center for Health Affairs; 2001.
224
SECTION III
Mudur G. Indian medical authorities act on antenatal sex selection. British Medical Journal 1999;319:401. Newhouse J. Medical care costs: How much welfare loss? Journal of Economic Perspectives 1992;6:3-21. Ramsey S, Pauly M. Structural incentives and adoption of medical technologies in HMO and fee-for-service health insurance plans. Inquiry 1997;34:228-36. Rosenthal M, Landon B, Huskamp H. Managed care and market power: Physician organization in four markets. Health Affairs 2001; 20:187-93. Rublee D. Medical technology in Canada, Germany and the United States: An update. Health Affairs 1994; 13:113-17.
Sudarshan H. The epidemic of corruption in health services. Presentation at the Institute for Health Systems, Hyderabad, India, 18 August 2003. Varshney A. Medical equipment use patterns in the private and public sector in India: Implications for policy. New Delhi: National Commission on Macroeconomics and Health; 2004. Weisbrod B. The health care quadrilemma: An essay on technological change, insurance, quality of care and cost containment. Journal of Economic Literature 1991;29:523-52.
225
SECTION III
Annexure 1: Medical equipment use pattern in the public and private sectors in India: Policy implications
M G
edical equipment: Present status

MEDICAL EQUIPMENT CONSTITUTES A MAJOR PART OF THE INVESTMENT IN the health care sector. It adds up to more than 60% of the capital cost. Dependence on medical equipment for diagnosis and management is increasing day by day. With rapidly advancing technology, digitalization, increasing computing powers, technological devices, equipment and non-equipment technology is becoming a necessity for early diagnosis, intervention, and prolonging and improving the quality of life. Studies conducted by WHO showed that 25%-50% of all health equipment in developing countries cannot be used for one reason or another, seriously impending efforts to improve heath services to the people. The main reasons for this are (i) difficulty in acquiring consumables and spare parts, (ii) lack of trained operators and service technicians, (iii) inadequate infrastructure for installation and operations, (iv) excessive amount of sophisticated equipment and insufficient basic equipment, and (v) obsolete and unsafe equipment. The above reasons are mainly due to inadequate management, which is result of lack/deficiency of policies and procedures for comprehensive technology management in the health system, both in the public and the private sectors, involving planning, acquisition, operations, maintenance and retirement. Decision-makers are seldom trained or have the awareness/knowledge of modern technology. Equipment are typically specified by department or doctors more for prestige, craze for the latest and best. These persons may not be responsible for its eventual operation and maintenance. There is a lack of coordination among agencies involved in various processes, from demand generation to procurement, finance and maintenance. Investment and recurring costs are non-sustainable. Manufacturers are only keen to sell the latest and the greatest. Equipment selection is not done as per the morbidity pattern or skills available to make the best use of the equipment. The total cost concept termed TCO (total cost of ownership) is not evaluated in both the public and the private sectors; only the purchase price is taken into account while making financial projections. Thus, the cost of installation, operations, maintenance, human resources, training, spares, support furnishings, calibration instruments, end-user cots, patient costs (charges + other expenses) and return on investment (ROI) are rarely given consideration, while these indirect or invisible costs constitute more than the purchase price, up to 80%-90% of the TCO.
Utilization rates
Both quantitatively and qualitatively, medical equipment is better utilized in private sector diagnostic centres as compared to the public sector and private hospitals. Utilization per machine is high in the private sector as to compared public sector hospitals; this is in spite of the fact that the latter have more manpower and a higher patient load. Qualitatively, in terms of early investigation, early reports and minimum patient visits, the private sector shows better utilization. Also, private sector doctors and technicians do more number of investigations per machine in a year than in the public sector. At the district level, utilization in both the public and the private sector is less, due to fewer requests, and a lower morbidity pattern as compared to Delhi hospitals
ANIL VARSHNEY
HEALTHCARE MANAGEMENT CONSULTANT, HEALTHCARE CONSULTANCY SERVICES E-MAIL: anilvarshney@hcsintl.com
226
Medical equipment use pattern in the public and private sectors in India: Policy implications
SECTION III
which are referral hospitals. Utilization of ultrasound in the public sector (district studied) is nil due to non-availability of a radiologist in the hospital. Utilization of other equipment in public hospitals, especially surgical departments, is better than in the private sector.
Misuse and overuse

Despite the existence of the Pre-natal Sex Determination Test Act, ultrasound machines are being used clandestinely for sex determination, driven by the demand from seekers and greed for money. Most insured patients are over-investigated and a similar trend is seen in large corporate hospitals. This trend is comparatively less in general practice.
Medical equipment market

The sale of medical equipment in India is worth approximately Rs 1500 crore annually. Fifty per cent of this is purchased from Indian manufacturers; 70% of the imported equipment is electronic medical equipment. Most high-end medical equipment are installed in cities with tertiary care hospitals. Concentration of equipment in a particular area has resulted in price competitiveness and the practice of incentives. Most of the equipment costing above Rs 10 lakh are installed in the metropolitan and large cities. Medium-level equipment is there at the district level for secondary care in both the public and the private sector. At the subdistrict level public sector equipment has poor installation and poor functioning as compared to the private sector. Servicing of and spares for medical equipment are major issues in terms of poor services, non-availability of trained technicians at district and subdistrict levels, planned redundancy by companies, non-availability of an annual maintenance contract (AMC) after 5 years by the companies, nonavailability of spares and their high cost, and monopoly of companies with respect to spares.
Procurement, maintenance, down-time and consumables

Procurement of medical equipment is faster and more efficient in the private sector, averaging 3 months from demand to commissioning; in the public sector it takes 18 months. Breakdown of medical equipment is very low in the private sector, averaging 3 days in 5 years per equipment, due to proper handling and better maintenance practices. The number of unusable equipment is also low in the private sector. Consumables are managed at minimum inventory carrying cost. Consumable supplies are faster (within 24 hours) due to better vendor compliance, as they are paid in time and regularly.
sound is barely at break-even point in private diagnostic centres, lower in private hospitals and high in public hospitals. The unit cost of ultrasound in a private hospital is 60% that of a public hospital and 70% that of private diagnostic centres. The price of as MRI is extremely high in public hospitals as compared to private ones (twelve times high). Unit costs for biochemistry are low in private diagnostic centres as compared to both private and public hospitals. Costs in the private sector are high due to interest, rentals and other establishment costs, as well as return on investment. Costs in the public sector are high due to human resources. As the unit cost is high, more investigations are required to achieve the break-even number in both sectors. For biochemistry services the unit costs are low in both the sectors as compared to the patient price, and the volumes of investigations are high. Public hospitals provide a subsidy (cost to hospital less patient price) of Rs 439 on ultrasound, Rs 1203 on CT scan, and Rs 46,750 on MRI to the public, and 100% subsidy in biochemistry tests (calculated for blood sugar). In private sector diagnostic centres there is no subsidy thus they break even at a high volume load. In private hospitals, there is a marginal element of subsidy of Rs 333 for CT scan. Patient charges are more flexible in private diagnostic centres as compared to private hospitals. Private diagnostic centres with a low patient load do not recover their costs at the present patient price; centres continue to provide these services, hoping to become viable in 4-5 years. Their income is augmented by laboratory medicine services. The percentage utilization in relation to break-even numbers (number of investigations required to recover fixed costs at the current patient price) are as follows: for ultrasound, it is 90%-120% in the private sector, 20% in the public sector for CT scan, it is 53% in public hospitals, 90% in private hospitals, 190% in private diagnostic centres for MRI, it is 7% in public and 120 % in private diagnostic centres for auto-analyser (calculated for sugar-biochemistry test) it is 360% in a private laboratory, 150% in a private hospital, since it is free in the public sector, a break-even point is never reached. (Breakeven number is the number of investigations required to recover the yearly fixed cost of establishment at the existing price to the patient.) At the district level, break-even numbers are not reached in both sectors in biochemistry. Ultrasound is cost-effective in the private sector.
Financing operation cost, tax benefits and incentives

Approximately 25%-30% of public hospital patients get their investigations done from private diagnostic centres. The investigation load in private diagnostic centres is 20% from public hospitals, 10% from private hospitals, and 70% from private practitioners. The price in the private sector is driven by what the market can bear as well as public sector prices. Cost calculations in the private sector are on an ad hoc basis. Operation costs in the private sector are borne by collections
Unit cost of diagnostic services

The unit cost (cost to establishment) of the investigation of CT scan and MRI in a private sector diagnostic centre is lower than the cost in public and private hospitals. The cost of ultra-
227
SECTION III
from patients, loans, depreciation funds, and instalment amounts (rescheduling bank payments). The only tax benefit available to the private sector is low custom duties on imported medical and life-saving equipment. The other sops for private players (large hospitals) are land at subsidized rates (so a lower rental value). Private diagnostic centres pay commercial rentals, property tax and commercial charges for electricity and water. Incentives play a key part in referrals to diagnostic centres initially. The expenditure on these incentives accounts for 10% of the expenses of diagnostic centres, which is lower than the electricity bills, rental and depreciation. Incentives are received by 40% of private sector and public sector doctors alike. However, only 30% of diagnostic centres (owned by doctors) are forced into this incentive-based referral network by competition from businessmen-owned centres where this practice is 100%. Old and established centres do not indulge in this practice. The code of medical ethics on splitting fees applies to doctors and not to business houses. The private sector, said to be highly priced relative to the public sector, is patronized by the upper- and middle-income groups, and insurance patients. This leaves the poor socioeconomic group to patronize public health facilities that are already overburdened resulting in poor quality and long waiting times.
should be developed in collaboration with professional bodies. The WHO publication medical device overview and guiding principles provides suggestions for standards. Norms for high-end medical equipment may be developed, based on the population or morbidity load. This will reduce excessive installation, which results in blocked investments both in the public and private sectors (e.g. 1 MRI 3 CT scan units in a population of 10 lakh).
Public-private cooperation
There exists an opportunity to develop public-private partnerships that will benefit all sections of society. There are different kinds of partnerships, from investment to maintenance and management. However, outsourcing all public sector investigations to the private sector seems to be the best option. This will reduce the burden on public hospitals, and enable the Government to reallocate funds for drugs, improve the quality of existing health services and strengthen delivery. Patients will also benefit from the quality of private sector services, and investigation time will be reduced with early diagnosis and treatment. Currently, public-private experiments are in a nascent stage and private service providers do not seem to benefit much from this relationship.
Policy steps needed

Misuse
The demand for pre-natal sex determination needs to be curbed through change of mindset, incentives for the girl child, counselling centres for parental education and socialcultural-behavioural changes. This should be managed by womens organizations and heavy penalties should be levied on persons seeking foeticide. It is also recommended that adequate records be maintained at the diagnostic centre and strict surveillance done of centres with Medical termination of pregnancy facilities.
Regulating the equipment market

Setting up a Medical Equipment Devices Regulatory Authority, on the lines suggested by WHO should be done so as to ensure quality equipment with adequate spare parts, and prioritize installations in undeserved areas as per needs. Procurement in Government hospitals needs to improve and various options should be evaluated to minimize the time. WHOs Essential healthcare technology package (EHTP software) should be utilized. Hospital and biomedical engineering should be promoted further as a specialty, to ensure the availability of trained technical experts for good quality repairs and maintenance of equipment. Long-term equipment management programmes should be initiated in all healthcare institutions for proper functioning of the equipment. There is an urgent need to make sure that equipment must be able to serve at least 15 years and companies should provide spares and AMC for the period. Third-party training of engineers who can handle various equipment needs to be encouraged and sourcing of spares from original subvendors should be initiated for equipment that are not attended to by the manufacturing companies. Local production of more sophisticated medical equipment needs to be given a priority, by way of limiting imports and imposing a manufacture in India clause in medical device regulation.
Overuse
This should be controlled through medical audit by the insurance companies as well as internal audit by the hospitals, and a grievance redressal cell for patients should be set up in the State medical councils.
Regulation of diagnostic facilities

Just like the National Accreditation Body for Laboratories (NABL) (voluntary) for registration and accreditation of Pathology and Biochemistry services, a similar body should be created for Radiology and Imaging, by the IRIA/IMA /PRSF and other professional bodies. Standards should be implemented by States (health being a State subject) for diagnostic centres, after categorization by size, type and range of services provided. The Central Government should provide the standard guidelines and this
Equipment invoicing
Manufacturers should regularly update themselves on the latest developments in medical equipment technology, and provide pricing lists in trade journals.
228
SECTION III
Table 1 Summary of utilization and costs in Delhi

Particulars US private US public US CT CT CT private private public 1 private hospital hospital MRI private MRI public Auto analyser private Auto analyser Public Auto analyser private hospital 2 40 300,000 150,000 2 150000 198,816 151 29 16 45 60 44 nil Same day same day
No. of machines 1 3 Cost of equipment(in lakh) 30 62 Number of investigations/ year 9000 26,784 Number of Investigations/ machine/year 9000 8928 Number of doctors or professionals/ modality 2 8 Number of investigations/ 4500 3348 professional/ year Number of investigations to break 9070 13,3674 even fixed costs Percentage utilization compared to 99 20 break even point Fixed cost per investigation (in Rs) 403 549 Consumable costs/ investigation(in Rs) 100 40 Cost per investigation (fixed + consumable) 503 589 Patient charges(in Rs) 500 150 Collection received to offset fixed costs/ 397 110 investigation(in Rs) Subsidy in real terms (in Rs) 3 439 Booking time Same 2 month day Reports delivered Same Same day day
2 1 15 180 18,438 10,000 9219 10,000 5(pt) 2 3688 5000 15,486 119 265 85 350 400 315 Nil Same day Same day
1 213 7392 7392 7 1056
1 145 4000 4000 4(pt) 1000 4518 89 2903 430 3333 3000 2070 333 Same day Same day
1 750 12,000 12000 3 4000
1 810 742 742 3 247
3 25 500,000 166,667 2 250000 136,682 366 9 17 26 50 33 Nil Same day Same day
5 80 800,000 160,000 3 266667 Infinity -61 18 11 29 0 0 29 3 days 3 days
5080 14,030 197 1549 450 1999 3500 3050 53 2543 160 2703 1500 1340
10,083 11,286 119 7
3739 50040 550 210 4289 50250 5000 3500 4450 3290 nil 50,250 Same 3 days day Same 2 days day
Nil 1203 Same 1 month day Same 3 days day
Conclusions
Medical equipment is essential for optimal healthcare at all levels of the health services. Their judicious installation and maintenance is the key to ideal healthcare delivery. Medical equipment are better utilized in the private sector as compared to the public sector in terms of number of investigations per machine, number of investigations per doctor and staff, unit cost per investigation, which are lower than that in the public sector despite the high interest rate and ROI. This is achieved by employing an optimal number of staff, more working hours and better quality in terms of early and timely delivery of reports. Since medical equipment constitutes a major part of the investment in any hospital, it would be advisable to outsource the high-end diagnostic services to the private sector managed by professionals, at charges that are lower but financially viable to the establishment. The capital money thus saved in the public sector could be utilized for improving the quality of services, and supplies of medicines and injectables. Installation of costly medical equipment in the private sector should also be regulated by encouraging group practice
(professionally owned), lower interest rates on the lines of housing loans, and encouragement to practise in underserved areas. The medical equipment market needs regulation on the lines suggested by WHO (Medical device regulation to safeguard professional and patient interest). Overuse of diagnostics needs to be curtailed through medical audit in large corporate hospitals and through internal and external audit by insurance companies/professional agencies. Research and analysis on essential health technology should be given priority and encouraged so that limited resources can be better utilized, by way of appropriate technology for the level and type of health care services being rendered at any institution. Equipment selection must be done in a scientific manner by looking at the total cost of ownership (TCO) rather than the purchase price alone. To get the best out of equipment, a management programme must be initiated in all institutions in both the public and the private sectors. This will increase the life and performance of the equipment and have less down-time.
229
SECTION III
Fig 1 Investigation per machine per year in each category of institution studied
Fig 2 Investigation per doctor per year in each category of institution studied
230
SECTION III
Fig 3 Numbers required to break even at present utilization and patient price
Fig 4 Percentage utilization at present in relation to break-even numbers
231
SECTION III
Fig 5 Fixed and consumable costs per investigation (unitized) and comparison with patient price (in Rs)
232
SECTION III
233
SECTION III
Fig 6 Financial cost components per unit investigation in INR (ultrasound investigation)
Fig 7 Investigation costs per unit
Fig 8 MRI investigation per unit component financial costs
Fig 9 Biochemistry financial cost component
Inner Circle Public Hospital
Middle Circle Private Hospital
Outer Circle Private Diagnostic
234
SECTION III
References
Varshney A. Concept paper on Technology in Health [to NCMH 2004] Wang B. Acquisition strategies for medical technology. World Bank Health Technology Forum. 2003. WHO. Medical device regulatory guidelines. 2003-4.
235
SECTION IV Financing of Health Care in India
SECTION IV
Financing of Health in India
F
K. SUJATHA RAO
SECRETARY NATIONAL COMMISSION ON MACROECONOMICS AND HEALTH, GOVERNMENT OF INDIA NEW DELHI E-MAIL: ksujatharao@hotmail.com
S. SELVARAJU SOMIL NAGPAL

somilnagpal@yahoo.com
S. SAKTHIVEL
INANCING IS THE MOST CRITICAL OF ALL DETERMINANTS OF A HEALTH SYSTEM. The nature of financing defines the structure, the behaviour of different stakeholders and quality of outcomes. It is closely and indivisibly linked to the provisioning of services and helps define the outer boundaries of the systems capability to achieve its stated goals. Health financing is by a number of sources: (i) the tax-based public sector that comprises local, State and Central Governments, in addition to numerous autonomous public sector bodies; (ii) the private sector including the not-for-profit sector, organizing and financing, directly or through insurance, the health care of their employees and target populations; (iii) households through out-of-pocket expenditures, including user fees paid in public facilities; (iv) other insurance-social and community-based; and (v) external financing (through grants and loans). While taxation is considered the most equitable system of financing, as tax is a means of mobilizing resources from the richer sections to finance the health needs of the poor, out-of-pocket expenditures by households is considered the most inequitable. Under a system dominated by out-of-pocket expenditures, the poor, who have the greater probability of falling ill due to poor nutrition, unhealthy living conditions, etc. pay disproportionately more on health than the rich and access to health care is dependent on ability to pay. Assessing how pro-poor a system of financing is again depends on how the different types of financing interact with each other. For example, a country may have a social health insurance policy but may not cover public hospitals as they are in theory expected to provide free care. In such a situation there may be greater incentives for patients to go to private hospitals as expenses are covered by insurance resulting in no incentives for the public hospitals to function well. In that case, the poor who have no immediate access to insurance or private hospitals may stand to lose with poor quality public care. In India, as in most countries, there is a clear urban-rural, rich-poor divide. Affluent sections, urban populations and those working in the organized sector covered under some form of social security such as the ESIS or CGHS, have unlimited access to medical services. The rural population and those working in the unorganized sector have only the tax-based public facilities to depend on for free or subsidized care, and private facilities depending on their ability to pay. The impact on equity then gets determined on whether the tax-based public facilities are able to provide a similar quality of care as provided under the Social Health Insurance Scheme. Because, if funding is low and the quality of care falls below expectation, is inaccessible, entails informal payments, etc. then the benefit of free care at the public facility gets neutralized with the second option of paying out-of-pocket to a relatively hasslefree private provider available close by, making the system of financing inequitable as well as inefficient. How and why this is so will be discussed in this section, as an understanding of the current structure of financing is important to identify future options for a better system.
Health Spending in India

Health spending in India is estimated to be in the range of 4.5%-6%. These estimates are based on a weak methodological background. Therefore, an exercise was undertaken to construct estimates of health spending based on a National Health
239
SECTION IV
Account (NHA) framework. Such an approach enables a better and more reliable understanding of the size and structure of health financing in India. Results from the NHA show that the estimated health expenditure in India for the year 2001-02 was approximately Rs 108,732 crore, accounting for 4.8% of the GDP at current market price (Fig. 1). while health expenditure as a percentage of the GDP measured at factor cost works out to 5.2%. Out of this, Central, State and local Governments together spend one-fourth of the total health expenditure. The share of other central ministries, which include railways, defence, posts and telegraphs, other civil ministries, etc. is estimated to be about 2.42% of total health spending in the country. The estimate is based on direct spending by the ministries as well as reimbursements provided to its employees. Local governments resources for health are through transfers from State Governments and their own resources. An estimated 2.2 % of total health spending comes from the local government. The estimate involves only spending by municipalities and not Panchayati Raj institutions. It is to be noted that municipalities (in metros and particularly Mumbai Municipal Corporation) are major contributors among local governments while the share of Panchayati Raj institutions are a miniscule part of the health budget, since a substantial part of the panchayats are mostly composed of either Central or State transfers.
Although the emergence and growth of NGOs have received much attention in India in recent years, their contribution to the health sector is a negligible 0.3% of the total health expenditure. As financial intermediaries, social insurance accounts for around 2.36% of the entire health budget in the country, with a significant contribution by the ESIS. While community insurance is a non-starter in the country, the share of private voluntary insurance schemes has a share of less than 1% of the total health budget.
Household Out-of-Pocket Expenditure on Health

The dominant role of the private sector in Indian health care system is well known, both in health provision and financing. India is one among the developing countries where households spend a disproportionate share of their consumption expenditure on health care, with the Governments contribution being minimal. Household consumer expenditure data of various rounds of the National Sample Survey Organization (NSSO) suggest that households spend about 5%-6% of their total consumption expenditure on health and nearly 11% of all non-food consumption expenditure. The analysis here shows the estimate of household expenditure on health for the year 2001-02, using the NHA framework. The estimate is based on the utilization pattern of health facilities and the expenditure involved by different sources of care and services provided. However, the mean expenditure and utilization pattern of morbidity for the year 1995-96 has been extrapolated (assuming a similar pattern of expenditure by different providers) and anchored to the 2001 Population Census and applying current growth rates worked out from the 50th and 55th rounds of Consumer Expenditure Surveys (CES) by both ruralurban and inpatient-outpatient populations. This growth rate takes into account both the price factor and growth of services during the period under consideration. Results from the survey suggests that for the year 200102, households out-of-pocket health expenditure is estimated to be Rs 72,759 crore which accounts for 3.2% of the GDP at current market price. Since 1995-96, household expenditure on health has been growing at the current rate of approximately 14% overall. In 1995-96, households in India spent an estimated Rs 33,253 crore at nominal prices which is then estimated to have increased to Rs 72,759 crore in 2001-02. With an overall growth rate of 14%, household spending is likely to be close to Rs 100,000 crore in nominal terms during 2003-04. Except the category of childbirth/delivery, all other categories registered a current growth rate in double digits. The growth in inpatient expenditure has been highest, in the range of 16%-18% during 1995-96 to 2003-04. (Table 1) In per capita terms, household expenditure measured in nominal prices has almost tripled from Rs 364 in 1995-96 to Rs 905 in 2003-04, while real per capita household expenditure is expected to only marginally increase from Rs 265 to
Fig 1 Sources of finance in the health sector in India during 2001-02
Regarding private spending on health, the NHA matrix reveals that 71% of the health budget is contributed by private sector, of which households alone spend 69%. As a percentage of the GDP at current market prices, households spend an estimated 3.3%. Spending by private firms is in various ways: either through their own health facilities, or by providing a lumpsum amount to the employee for health, or reimbursing a part of the health expenditure incurred or by contributions to insurance schemes such as ESIS or voluntary private insurance schemes. External aid to the health sector, either to the Government or NGOs, taken together forms 2% of the total health budget.
240
SECTION IV
Table 1 Household health expenditure by different source of care in India, 1995-96 to 2003-04
Type of service 1995-96 2001-02
(Rs in crore)
2003-04 Growth rate
UP met almost 90% of their health care needs by out of pocket means. In other States, which have strong public health systems such as Tamil Nadu, Delhi and Karnataka, households spending is relatively lower than the public expenditure, accounting for about two-thirds of the total health expenditure in these States. (Table 2)
Outpatient-rural 16,692.96 34,290.99 43,590.87 Outpatient-urban 7251.45 16,904.82 22,415.01 Inpatient-rural 3030.04 8536.86 12,057.25 Inpatient-urban 2092.90 5150.72 6954.10 Childbirth 1654.22 2258.14 2504.97 Antenatal care (ANC) 1053.90 2383.27 3128.22 Postnatal care (PNC) 390.85 1028.10 1419.21 Immunization 241.02 535.61 698.95 Contraceptives 207.14 422.74 536.22 Self-care 638.83 1247.47 1559.23 Total 33,253.31 72,758.71 94,457.19
12.75 15.15 18.84 16.19 5.32 14.57 17.49 14.23 12.62 11.80 13.94
Public Financing of Health

Even though public sector spending accounts for less than a quarter of the total health spending in India, it has a major role in terms of planning, regulating and shaping the delivery of health services. Such public provisioning is considered essential to achieve equity and to address the large positive externalities associated with health. As a result, a vast and widespread public health system grew over time across the country; there were 137,311 subcentres, 22,842 PHCs, 3043 CHCs, 4048 hospitals and a workforce of 345,514 in 2001-02. The way in which the sector is financed determines the effectiveness of service delivery and requires an understanding of the financing mechanisms in this sector. Health being a State subject, the sector is financed primarily by the State Governments. The per capita total health spending was estimated to be around US$23 during 19972000 (World Bank 2003). As compared to the levels of spending by countries such as Sri Lanka (US$31) and Thailand (US$71), the spending in India is substantially low. A breakdown of health expenditure reveals that expenditure by the public sector in these countries is twice that of India. Substantially higher levels of health outcomes in these countries as compared to India clearly indicate that there is a strong case to markedly increase public sector spending on health, as stated in the National Health Policy 2002 and the National Common Minimum Programme (CMP) 2004. The primary source of public financing is the general tax and non-tax revenues. These include grants and loans received from both internal and external agencies, which face competing demands from various ministries and departments. This pool of resources is used to finance the Centres and States own programmes. The Central Government plays a catalytic role in aligning the States health programmes to meet certain national health goals through various policy guidelines as well as financing certain critical components of centrally sponsored programmes implemented by the State Governments. In addition to tax revenues, a meagre amount is also raised through user charges, fees and fines from the sector, and further supplemented through grants and loans received from external sources. In the case of local governments, the respective State Governments largely finance their health programmes. Local governments do raise resources through user charges and certain fees though the quantum varies widely from States to States. Overall, the sector is underfunded, not without consequences. An issue that is often raised in the context of inadequacy of resources to the sector is the efficiency of the resources allocated. The current level of funding to the sector is grossly inadequate as brought out by various studies over the past
Source: Estimated from the 52nd Round of the NSS, using 2001 Population Census and applying growth rates worked out from the 50th and 55th rounds of the NSS
Rs 347, respectively. State-wise analysis (of only major 15 States in India) reveals that Kerala, which is a leading State in terms of health indicators, also accounts for the highest household spending in India, with a little over Rs 1700 per annum (Fig. 2). This is followed by Haryana and Punjab whose households spend an estimated Rs 1000 annually. It is interesting to note that these States comparatively have higher levels of public spending on health. Although Tamil Nadus public expenditure on health is high, household spending is among the lowest. In States such as UP, MP and Orissa, both public expenditure
Fig 2 Household spending in Indian States
and household expenditures are relatively low. It is disquieting to note that nearly 70% of the total health expenditure in India comes from households, while around 25% is financed by the Central, State and local Governments. If we consider only out-of-pocket and State and Central Government spending, then households in Bihar and
241
SECTION IV
Table 2 Household, public and total health expenditure in India (200405)

States Household Exp. (Rs. Crores) Govt. Exp. (Rs. Crores) Other Exp. (Rs. Crores) Aggregate Exp. (Rs. Crores) PC HH Exp. (Rs.) PC G. Exp. (Rs.) PC Other Exp. (Rs.) PC Exp. (Rs.) HH as % of THE (%) PE as % of THE (%) OE as % of THE (%)
Central Govt. 0 A. P . 6441 Arun. Pradesh 430 Assam 3054 Bihar 11854 Delhi 1004 Goa 524 Gujarat 4893 Haryana 3385 H.P . 2126 J&K 1759 Karnataka 3847 Kerala 8373 M.P . 6432 Maharastra 11703 Manipur 420 Meghalaya 58 Mizoram 38 Nagaland 1024 Orissa 2999 Punjab 3493 Rajasthan 3399 Sikkim 72 T.N. 3624 Tripura 253 U.P . 17158 W.B. 7782 U.Ts. 3160 State Totals 109308 GT [GOI+State] 109308
14819 1696 67 672 1091 721 116 996 421 306 471 1267 1048 1051 3527 89 94 58 84 684 827 1190 55 1590 100 2650 1715 325 17965 32784
730 640 0 52 202 55 22 424 175 40 47 353 281 228 726 8 8 0 7 111 273 267 0 760 13 550 433 227 5906 6636
15549 8777 497 3778 13147 1780 662 6313 3981 2472 2277 5467 9702 7711 15957 517 160 96 1116 3795 4593 4855 127 5974 366 20359 9929 3712 133178 148727
0 820 3776 1089 1021 664 3613 920 1518 3377 1609 702 2548 746 1156 1680 242 405 4897 786 1379 565 1274 566 760 924 931 11168 1012 1012
137 216 589 239 124 476 798 187 189 486 431 231 319 164 348 356 388 623 404 179 326 198 965 248 301 150 205 52 167 304
7 82 0 19 23 37 153 80 79 64 43 64 86 35 72 32 34 0 37 29 108 44 0 119 40 31 52 37 54 61
144 1118 4365 1347 1497 1177 4564 1187 1786 3927 2082 997 2952 1200 1576 2068 664 1027 5338 995 1813 808 2240 933 1101 1152 1188 598 1233 1377
0 73.38 86.51 80.84 90.17 56.41 79.17 77.51 85.03 85.99 77.26 70.36 86.3 83.41 73.34 81.24 36.45 39.39 91.74 79.04 76.05 70 56.89 60.67 68.99 84.28 78.38 85.13 73.5
95.3 19.39 13.49 17.78 8.3 40.48 17.48 15.78 10.56 12.38 20.69 23.18 10.8 13.63 22.1 17.2 58.37 60.61 7.57 18.02 18 24.5 43.11 26.61 27.35 13.02 17.27 8.74 22
4.7 7.29 0 1.38 1.53 3.11 3.35 6.71 4.4 1.63 2.05 6.46 2.9 2.96 4.55 1.56 5.18 0 0.7 2.93 5.95 5.5 0 12.72 3.66 2.7 4.36 6.12 4.46
Source : Based on National Health Accounts (NHA), 2001-02 Notes : i) Household Expenditure Based on NHA for the year 2001-02 and extrapolated for 2004-05 ii) Central Govt. expenditure includes transfer to states, other central ministries and central PSUs; and data obtained from Demand for Grants (Provisional), Govt. of India. iii) Govt. Expenditure includes Central, States, Local Govt., and PSUs; data obtained from States Finances (Provisional), RBI, Various issues iv) Others include foreign agencies, private firms and NGOs; data relates to 2001-02, which is subsequently extrapolated for 04-05. v) PC HH Exp. Per Capita Household Expenditure; PC G Exp. Per Capita Govt. Expenditure; PC Other Exp. Per Capita Other Expenditure; HH as % of THE Household as % of Total Health Expenditure; PE as % THE Public Expenditure as % of Total Health Expenditure; OE as % of THE Other Expenditure as % of Total Health Expenditure; C. Govt. Central Govt.; U.Ts Union Territories.
decade or so. A concern that is equally voiced is how judiciously the funds allocated currently are utilized. Countries such as Bangladesh and Indonesia spend about US$14 and US$19, respectively, per capita on health; relatively less than the per capita spending by India (US$23). But the health outcomes in terms of child mortality are considerably better in these countries-74 for Bangladesh and 45 for Indonesia compared to 93 for India (World Bank 2003). This clearly reveals that the current level of spending has the potential to improve the outcomes if properly allocated. In the following sections an attempt is made to understand the present trends and structure of public spending on health to critically evaluate the above issues in detail.
Trends in public spending on health in India

Public spending on health in India gradually accelerated from 0.22% in 1950-51 to 1.05% during the mid-1980s, and stagnated at around 0.9% of the GDP during the later years (ie. spending by only Central and State health departments) (Table 3). Of this, recurring expenditures such as salaries and wages, drugs, consumables, etc. account for more than 90% and is on the rise in recent years. In terms of per capita expenditure, it increased significantly from less than Re 1 in 195051 to about Rs 215 in 2003-04. However, in real terms, for 2003-2004 this is around Rs 120. Estimates, irrespective of the definition, reveal that the per capita spending by the
242
SECTION IV
Table 3 Trends in health expenditure in India (GDP is at market price, with base year 1993-94)
Per capita public Health expenditure as % of the GDP Year Revenue Capital Total expenditure on health (Rs)
Table 4 Utilization of government hospital services for inpatient treatment

Fractile groups 1986-87 (42nd Round) Rural Urban 1995-96 (52nd Round) Rural Urban
1950-51 1955-56 1960-61 1965-66 1970-71 1975-76 1980-81 1985-86 1990-91 1995-96 2000-01 2001-02 2002-03 2003-04
0.22 0.49 0.63 0.61 0.74 0.73 0.83 0.96 0.89 0.82 0.86 0.79 0.82 0.86
NA NA NA NA NA 0.08 0.09 0.09 0.06 0.06 0.04 0.04 0.04 0.06
0.22 0.49 0.63 0.61 0.74 0.81 0.91 1.05 0.96 0.88 0.90 0.83 0.86 0.91
0.61 1.36 2.48 3.47 6.22 11.15 19.37 38.63 64.83 112.21 184.56 183.56 202.22 214.62
0-10 10-20 20-40 40-60 60-80 80-90 90-100 All groups
12.94 10.59 22.94 18.69 19.73 8.82 6.30 100.00
11.59 11.42 25.66 23.91 17.28 3.63 6.52 100.00
3.13 6.30 16.70 17.20 24.30 14.07 18.30 100.00
12.87 7.40 20.90 18.17 19.77 10.23 10.67 100.00
All values are in percentages Note: Government hospitals refer to public hospitals, PHCs and public dispensaries. Sources: 1. NSSO. Morbidity and Utilisation of Medical Services. Report No.364, Department of Statistics, CSO, Government of India, September 1989, pp. A-8-13. 2. NSSO. Morbidity and treatment of ailments. Report No.441, Department of Statistics, CSO, Government of India, November 1998, p. A-65 and p. A-170.
Sources: Report on Currency and Finance, RBI, various issues; Statistical Abstract of India, Government of India, various issues; Handbook of Statistics of India, RBI, various issues
Government is far below the international aspiration of US$12 recommended for an essential health package by the World Development Report 1993 (World Bank) and, again by the Commission on Macroeconomics and Health (World Health Organisation 2002) CMH (WHO) for low-income countries. As a result of stagnant budgetary allocations, the quality of care suffered substantially and adversely impacted on the utilization of government services by households. Besides, health services that were earlier being provided free were in some cases charged, forcing patients to seek private health care.
the poor are, to some extent centrally funded vertical programmes such as immunization, ANC, TB, Malaria, Leprosy, etc. The inequity in the access to and distribution of public health services has been a concern because of the extent of impoverishment households face on account of ill health, and catastrophic illnesses in particular.
Health Expenditure by the Central Government

Major policy initiatives and reforms relating to health emanate from the Ministry of Health and Family Welfare (MOHFW), which plays a crucial role in financing this sector. The Union Ministry of Health and Family Welfare consists of three departments. The department-wise break-up of the Health Ministrys budget suggests that over one-third of the budget is spent by the Department of Health, while roughly two-thirds goes to the Department of Family Welfare. The Indian Systems of Medicine and Homeopathy (ISM&H) (AYUSH) Department receives a paltry 2%-3% of the total budget of the Ministry. There are 5 important aspects to the nature of central spending in recent years: 1. The gradual reduction in the proportion of funds released to States at a time when the States were themselves under fiscal stress; 2. The sharp reduction in capital investment in public hospitals at a time of technological innovation, shifts in the epidemiology and health needs and expectations of the people, besides the sheer increase in disease burden in absolute terms; 3. Increased subsidy for own employees; 4. Low priority to preventive and promotive health; and 5. Allocative inefficiencies under the National Health Programmes
Impact on equity due to low public spending

The results of the NSS of 1986-87 and 1995-96 showed a considerable decline in the utilization of public health services by the poor, especially the rural poor (Table 4). Besides, the study also showed that the rich consumed public services three times more often than the poor. The ratio of access to admission between the lowest 10% quintile and the richest 10% was reported to be 6.1 and 2.2 between the below poverty line (BPL) and the above poverty line (APL) populations. The 52nd Round of the NSS provided insights into the quintile-wise health-seeking behaviour. As per this data, of the poor who availed of services, 61% used public facilities compared to 33% among the rich. The poorest, however, benefit relatively more from spending on primary care only (Mahal 2001). This is primarily on account of the poor quality and irregular supply of these services which dissuade the rich from accessing them. Further, many of the services that benefit
243
SECTION IV
Centralization of funds and inadequate capital expenditures

The Ministry implements certain schemes such as the Central Government Health Scheme (CGHS), national disease control programmes, etc. by itself, and other schemes through the State Governments. A large part of the Ministrys budget is passed on as grants-in-aid to States for implementing various national health programmes. Such transfers accounted for about 43% of the total budget of the Ministry in 200304. Even though the size of the Central health budget has grown considerably from Rs 1670 crore in 1991-92 to Rs 7851 crore in 2003-04, transfers to States as a proportion of the total budget of the Ministry declined sharply from nearly 57% to 44%. This in effect reveals the increasing role that the Central Government has been assuming in health service delivery. As a result, roles such as stewardship and governance that the Central Government is expected to play are undermined. The share of the Central Government expenditure on health, including grant-in-aid to States, constitutes over a third of the combined expenditure by the States and Centre. Figure 3 shows that during the period 1991-2003, the overall increase in central allocations was 4%-6% annually. Despite this, there was a sharp decline in capital expenditure, which fell from about one-fourth of the Ministrys expenditure to less than 6.7% of the net MoHFW expenditure (excluding grant-in-aid to States and UTs but including capital expenditure incurred by the Ministry of Urban Development on hospitals such as RML and LHMC). During the same time, allo-
cations for materials and supplies for central sector public hospitals also fell from 22% to 15% to accommodate the increase in salaries from 56% to 63% on account of the Fifth Pay Commission. This has had an adverse impact on the declining level of quality in these once premier hospitals which are expected to act as a benchmark in the quality of care.
CGHS-a mandatory social health insurance scheme for the Central Government Employees
Six per cent of the combined budget of the department or 18% of the budget of the Department of Health was spent on 44 lakh beneficiaries or 0.5% of the countrys population under the Central Government Health Scheme (CGHS). Since the introduction of contracting of private hospitals for providing health services and permitting beneficiary members to purchase drugs at pharmacy shops in 2000, there has been an escalation in expenditure under this programme. Over and above the Rs 503 crore incurred on the CGHS by the Department of Health, an additional Rs 200 crore was spent by the various administrative departments on medical reimbursements of their serving employees during 2001-02. All taken together, the outpatient expense under the CGHS per card is estimated to be about Rs 3478 per year and the inpatient expense per card issued to retired civil servants and dependents is Rs 6692 per year.
Low priority for preventive health care

An important public health function that governments are expected to perform is expanding access to public goods by focusing on preventive and promotive education. Preventive and promotive education does not mean only disseminating disease-specific messages to raise awareness among people for behaviour change but also includes a range of other aspects such as laws for the use of helmets for preventing road accidents, or providing nutritional information to consumers regarding food products, on risky behaviours and exhorting people to adopt healthy lifestyles such as non-consumption of addictive substances such as tobacco, daily exercise, healthy diets, etc. In India, such an interventionist role of the State is negligible with some information, education and communication (IEC) activities undertaken under the National Health Programmes. This is a serious omission, given the huge treatment costs that will be required to cope with the emerging epidemic of non-communicable diseases. Under the NHP, the amount spent on preventive care aimed at prevention and behaviour change during the financial year is an estimated 21% as given in the Table 5; of this a large amount was for vaccines under the universal immunization programme (UIP). In terms of use of mass media and interpersonal communication, the expenditure under this head in the National Programmes is a mere 2% of the overall budget. For TB, the amount spent during 2001-2002 on preventive care is very low, as most of the expenditure was on drugs, equipment and staff. As people are unaware of the free services under the National Health Programmes, a large number
Fig 3 Trends in grant-in-aid allocations by MoHFW to States and declining capital expenditures
Note: 1. Figures in parentheses denote percentage share of central spending and grants-in-aid to states as a percentage of total MoHFW (GOI) expenditure. 2. Grants-in-aid has been calculated as the sum of expenditure under major heads 3601, 3602 and 3606. Source: Demand for Grants, Ministry of Health and Family Welfare, respective years
244
SECTION IV
Table 5 Expenditure on preventive and promotive activities under NHPs during 2001-02 (Rs in Lakhs)
Name of the programme Activity Malaria Leprosy TB FW* HIV /AIDS Blindness Total % TE
Distribution of IEC materials Immunization Supply of condoms Supply of bednets Supply of insecticides Total TE % TE
846.09 0 0 239.05 4301.36 5386.5 21978 24.5
1089.92 0 0 0 0 1089.92 6105.07 17.9
205.35 0 0 0 0 205.35 10,058.06 2.0
8542 54,722 11,821 0 0 75,085 391,663 19.2
15000
15,000 22,500 66.7
958.26 0 0 0 0 958.26 11,802.23 8.1
26,641.62 54,722.00 11,821.00 239.05 4301.36 97,725.03 464,106.36 21.1
6.0 12.4 2.7 0.1 1.0 21.1
* 2002- 03; Total expenditure IEC: information, education and communication; TE: total expenditure; TB: tuberculosis; FW: Family Welfare; Source: Ministry of Health and Family Welfare, GOI
of them continue to go to the private sector for treatment.. Expenditures under the Reproductive and Child Health (RCH) Programme pertain to the immunization programme which helps avoid vaccine-preventable diseases but is of limited value in achieving the goal of bringing down IMR and under-5 child mortality rate for which commensurate investments have to be made on propagating a range of behavioural practices. These include breastfeeding, use of oral rehydration solution (ORS), healthy practices such as consumption of boiled water or washing the hands with soap, prevention of acute respiratory infections (ARI), etc. These interventions are as important for child health as the use of vaccines for vaccine-preventable diseases. Since preventive and promotive expenditures are an investment on the demand side, it is necessary to not only increase the level of expenditure under this component but also implement these activities more rigorously to reduce the disease load and public expenditures on curative care.
in 1990-91 to Rs 513.26 crore in 2002-03, constituting almost 20% of the Departments expenditure during the year, as shown in Table 6.
Table 6 External funding of National Health Programmes (2002-03) (Rs in crore)

NHPs Total allocation Share of external funding Share of external funding (%)
Malaria 206.6 (29.3) TB 96.8 (13.7) Leprosy 75.0 (10.6) AIDS 241.4 (34.3) Blindness 84.6 (12) Total 704.3 (100)
97.96 95.10 67.99 239.96 12.25 513.26
47.4 98.2 90.7 99.4 14.5 72.9
Centrally sponsored schemes-National Health Programmes (1991-2003)

Of the total combined central budget 70% is spent on National Health Programmes related to the disease control programmes and family welfare. The allocation of funds for the 5 National Communicable disease control programmes (Leprosy, Malaria, TB, Blindness and HIV/AIDS) went up from 18.6% of the budget during 1991-92 to 26.8% of the budget in 2002-03, accounting for Rs 704.3 crore. Due to limited expansion of the budget, malaria got gradually crowded out giving way to HIV/AIDS. In 1991-92, malaria accounted for over 66% of the total outlay under disease control programmes of the Department of Health, shrinking to just 29.3% in 2002-03. During this period there was a corresponding increase in the HIV/AIDS programme: from 5% to 34.3%. In gross terms, the disease control programmes got a higher allocation as they were all funded under World Bank projects. The quantum of external funding received by the Department of Health on the communicable disease control programmes went up from a negligible amount
Figures in parentheses are the proportion of the total allocated for these 5 programmes Source: Demand for Grants, Ministry of Health and Family Welfare, respective years
Another major national programme that is centrally funded in substantial measure is the Family Welfare Programme. Under this programme, recurring expenditures of subcentres, the RCH Programme and free supply and social marketing of contraceptives are the main activities receiving 40%, 20% and 12%, respectively, of the budget allocations.
Health expenditure by State Governments

At the State level, public heath is also financed through general tax and non-tax revenue resources as the cost recovery from the services delivered has been negligible, at less than 2% (Selvaraju 2001). As a result, resource allocation to this sector is influenced by the general fiscal situation of the respective State Governments. For instance, the implementation of the recommendations of the Fifth Pay Commission during the late 1990s resulted in an increase in the fiscal deficit and a general resource crunch. Evidence
245
SECTION IV
from other countries also suggests that whenever there is a fiscal consolidation and stress, social sectors like health and education are targeted for pruning expenditures and reducing budget allocations (Tanzi and Schuknecht 2000). The figures presented in Table 7 confirm the above findings. The budgetary allocations to the health sector during the year 1999-2000 declined to the extent of about 2 percentage points as compared to 1985-96. Despite reduction in the health budget from 7.02% in 1985-86 to 4.97% in 2003-04 the fiscal deficit as a percentage of the GSDP recorded an increase, implying that allocation to health does not necessarily accentuate fiscal deficits. Public spending on the health sector in the States increased to about 0.9% of the GDP as per the estimates for 2003-04, from 0.8% in 1975-76 as seen in Fig 4. During the decade 1975-85, it registered a substantial increase and reached a high of 1.05%. Thereafter, it deteriorated steadily due to the general fiscal stress during the late 1980s followed by the reform measures initiated in the 1990s. The severity of the fiscal strain during the late 1980s forced the State Governments to introduce austerity measures and the soft sectors such as health were targeted for expenditure compressions. Similarly, when reform measures were initiated at the Centre during the early 1990s, fiscal transfers to States were compressed leading to reductions in health sector allocation at the State level. The recommendations of the Fifth Pay Commission in 1997 forced the governments to increase the budget to meet the increased salary cost of public sector personnel. However, these improved allocations could not be sustained beyond 1999-2000 when deceleration set in again. By the year 2001-02, the relative allocation to the sector reached levels closer to those prevailing in 1975-76. (Figure 4)
Fig 4 Trends in public health spending

Sources: State demand for grants for various years
Budgeting allocation and outcomes

The manner of resource allocation to and planning for the health sector shows a wide disparity in spending and outcomes across States, indicating the absence of appropriate norms for allocation and monitoring of health programmes. Table 8 gives the budget allocation function-wise (See Appendix 1). Although the table does not attempt to establish any correlations between such functional spending and key outcomes such as IMR or safe deliveries, it may be a proxy for assessing the functioning of the health system. Yet, the data are juxtaposed only to highlight the point that it is the low-performing (high IMR and low safe delivery) States which spend relatively higher amounts on pri-
Table 7 Share of health in revenue budget of major States (in %)

States Andhra Pradesh Assam Bihar Gujarat Haryana Karnataka Kerala Maharashtra Madhya Pradesh Orissa Punjab Rajasthan Tamil Nadu Uttar Pradesh West Bengal All States 1985-86 6.41 6.75 5.68 7.45 6.24 6.55 7.69 6.05 6.63 7.38 7.19 8.10 7.47 7.67 8.90 7.02 1991-92 5.77 6.61 5.65 5.42 4.19 5.94 6.92 5.25 5.66 5.94 4.32 6.85 4.82 6.00 7.31 5.72 1995-96 5.70 6.08 7.80 5.34 2.99 5.85 6.81 5.18 5.07 5.42 4.56 6.18 6.40 5.73 7.16 5.70 1999-2000 6.09 5.25 6.30 5.21 4.08 5.70 5.95 4.59 5.18 5.03 5.34 6.39 5.51 4.42 6.30 5.48 2003-04 (B.E.) 5.21 4.39 4.84 3.68 3.63 4.85 5.42 4.39 4.89 4.47 4.27 5.75 5.26 5.13 5.23 4.97
246
SECTION IV
Table 8 Sectoral allocation of health expenditure by States: 2001-02 (Rs in lakhs)

Primary Secondary Tertiary Social health insurance Administrative Research and IMR/1000* Safe delivery 2002** (% Rounded off) training live-births 2002
Well-performing States Andhra Pradesh Karnataka Kerala Tamil Nadu Medium performing States Punjab Gujarat Haryana West Bengal Maharashtra Poor Performing States Assam Bihar Chhattisgarh Madhya Pradesh Orissa Rajasthan Uttar Pradesh Total
63,241 (47.53) 51,334 (47.28) 19,389 (25.88) 52,700 (43.92) 26,078 (42.17) 30,336 (41.61) 16,217 (50.38) 46,184 (34.79) 102,106 (55.70) 21,002 (58.98) 46,349 (64.96) 17,166 (74.02) 41,650 (54.14) 20,370 (45.33) 57,831 (58.50) 142,193 (61.18) 754,143 (50.18)
22,844 (17.17) 23,883 (22.00) 26,460 (35.32) 18,120 (15.10) 10,078 (16.30) 4986 (6.84) 5060 (15.72) 35376 (26.65) 27722 (15.12) 6003 (16.86) 6047 (8.48) 2348 (10.12) 10,791 (14.03) 11,837 (26.34) 7556 (7.64) 50,257 (21.62) 269,369 (17.92)
27,625 (20.76) 23,626 (21.76) 21,198 (28.30) 34,114 (28.43) 9419 (15.23) 20,430 (28.02) 5507 (17.11) 30,153 (22.71) 36,292 (19.80) 6109 (17.16) 11,728 (16.44) 1541 (6.64) 14,420 (18.74) 6590 (14.66) 24,598 (24.88) 18,138 (7.80) 291,486 (19.40)
5419 (4.07) 4719 (4.35) 3502 (4.67) 8011 (6.68) 3131 (5.06) 6623 (9.09) 2436 (7.57) 6737 (5.07) 11120 (6.07) 0 (0.00) 768 (1.08) 328 (1.41) 2049 (2.66) 1054 (2.34) 2275 (2.30) 6680 (2.87) 64,850 (4.32)
11,592 (8.71) 4164 (3.83) 1979 (2.64) 5266 (4.39) 12140 (19.63) 8968 (12.30) 2518 (7.82) 12457 (9.38) 4645 (2.53) 2182 (6.13) 4765 (6.68) 1157 (4.99) 4915 (6.39) 4407 (9.81) 5159 (5.22) 12034 (5.18) 98,346 (6.54)
2326 (1.75) 844 (0.78) 2385 (3.18) 1772 (1.48) 995 (1.61) 1558 (2.14) 412 (1.28) 1839 (1.39) 1380 (0.75) 314 (0.88) 1692 (2.37) 394 (1.70) 1771 (2.30) 645 (1.43) 1419 (1.44) 621 (0.27) 20,366 (1.36)
62 55 10 44
68 62 97 80
51 60 62 49 45
61 59 44 43 61
70 61
20 18
85 87 78 80
32 37 38 26
Figures in Paranthesis are percentages to total spendings by states RE figures for 2001-02 have been used for Bihar, all others are actuals Source: Demand for grants for respective States, 2003-04 (2002-03 for Bihar) * SRS, 2004 ** Source: MICS 2000
mary care as compared to other States and yet continue to have such poor outcomes, raising the question as to whether there is any correlation of public spending to programme outcomes. However for such analysis longitudinal data are needed. Table 8, however, seems to suggest that an equitable spread of resources among all the three sectors-primary, secondary and tertiary-may be necessary. As can be seen in the case of UP-the skewed spending on primary and negligible amount on tertiary, which deals with medical colleges and training-
can have long-term effects in two ways: on the quality of people trained or in creating a shortage of skilled personnel. Whether poorly trained or low in numbers, the impact on access to primary care services will be adversely affected as the care provided will depend, in the ultimate analysis, on the availability of human resources.
Structure of health sector spending

Analyses of the structure of spending on health by State
247
SECTION IV
Governments shows that spending on salaries and wages account for more than 70% of health budgets. The huge network of services developed over time covering the length and breadth of the country is manned by doctors, nurses and paramedical staff and no doubt needs a large budgetary allocation. Of the remaining budget, nearly 12% is allocated for drugs, medicines, supplies and consumables; purchase of machinery and equipment account for 8%, and nearly 5% is allocated for maintenance of equipment, buildings, electricity, rent, taxes, etc. The remaining 5% is spent on other routine expenditures. The large proportion of the budget allocation for salaries is often criticized as unproductive. It is true that the rising share of salaries has squeezed out other components causing severe imbalances. With the less-than-proportionate increase in the total budget to the sector and political compulsions to not cut the salary head, the non-salary component used for fuel, drugs and medicines, maintenance and repair of equipment and buildings, etc. declined sharply.
Fig 5 Fig. 5. Burden of fiscal deficit

Source: RBI, prices are in real terms
Does fiscal deficit impact on per capita public health spending?

To analyse the spending behaviour of States and its connection if any with fiscal deficits, an analysis of five States was taken uptwo wellperforming and thee poor-performing States. The analysis shows that the burden of fiscal deficit is much higher in Orissa, Rajasthan and Uttar Pradesh compared to Tamil Nadu during the past decade as seen in Fig 5 below. This might have seriously hampered resource allocations to the health sector in these States affecting their ability to perform. As can be seen, with the increase in the overall fiscal deficit as a percentage of the GDP, there is a decline in per capita public health spending. The fall has been sharpest in UP. Public spending on health plays an important role in the imperfect health market. It ensures minimum service delivery under the difficult circumstances that prevail in backward States such as Orissa, UP, Bihar and, at the same time, acts as a corrective force for market failures where a number of players deliver services. Studies on health financing emphasize that even though the aggregate spending level in India is comparable to a few developing countries, the levels of per capita public spending on health needs to be stepped up (Prabhu 1993). This gains further importance as a large share of out-of-pocket expenditure by users of public hospitals goes to pay for drugs and diagnostic tests from private providers. This expenditure actually substitutes the governments expenditure. As seen in Table 8, States that allocate larger resources per capita are also the States with better health outcomes as also seen in Fig. 5. Therefore, in States such as Orissa and Uttar Pradesh, per capita public health spending needs to be increased more than proportionately because of low levels of out-of-pocket spending due to low incomes and poor purchasing power. In fact, out-of-pocket spending as a share of the household expenditure is among the highest in UP-the State where per capita public expenditure is also low, calling
Fig 5A
for an increase in public spending. It is, however, true that no correlation can be established between per capita public spending and household expenditure as the actual access to services depends on other factors such as the efficiency with which the system is functioning. In other words, if the health system is inefficient or poorly managed, mere increase of financial resources may have little impact. Suggestions to increase budgetary allocatios for health are often questioned because of the widely prevalent opinion that the budget allocated is seldom utilized. An analysis of the budget allocated and utilized at the end of the year for five States showed a mixed trend (Fig. 6). For instance, Kerala has been underutilizing about 7% of its budget allocated to the health sector whereas, in Tamil Nadu, expenditure exceeded allocation by about 6%. The evidence does not seem to fit a pattern. At periods of higher fiscal deficit, percentage utilization should be low, but in UP during the four-year period 1990-1994 there was consistently excess spending. While reasons for this will need a closer analysis, intuitively, it could be inferred that at times of fiscal stress, budget allocations are reduced to the bare minimum such as for salaries, which get
248
SECTION IV
Fig 6 Extent of underutilization of the Health budget

Source: Reserve Bank of India
larger than the caseload in those States, while the position was reversed in Karnataka, Orissa and West Bengal. Another important case of misplaced emphasis is the Pulse Polio Initiative, introduced in 1996. Implemented as a vertically driven scheme, an estimated total of Rs 3592 crores has been spent so far. This amount does not include the extrabudgetary expenditures incurred by the WHO on the appointment of over 1000 consultants in the country to monitor the programme and the amounts being incurred by the UNICEF on IEC. It is estimated that one drop of polio vaccine is almost 30 times more than the drop given in routine UIP. Moreover, almost 13% of the departments budget during 2003-2004 was spent on this single activity, which has limited impact on reducing the IMR, a principal national and Millennium Development Goal.
Box
utilized without effort quite automatically. Conversely, the lower utilization of funds at better times could perhaps have more to do with the budgeting process than in the States ability or capacity to absorb the funds, calling for a restructuring of the way in which health is financed.
Typical case of the National Malaria Programme

The State Malaria Officer (SMO) plans and places the demand for funds. When the budget is approved, it is about 50% of the demand placed by the SMO. The SMO still carries on with the 50% budget by rationalizing the funds. The SMO selects areas/blocks that have a high incidence and goes on selecting the second highest, third-highest blocks until the budget is exhausted. When the next budget allocation is made, the SMO continues the implementation process. By the time the SMO controls the incidence in about 50% of the blocks by the third or fourth year, the blocks where the programme was implemented in the first year again show up on the high incidence list. This cyclical process continues and eradication of the disease is further complicated. Increase in the drug resistance for every reoccurrence makes eradication a herculean task.
Issues of Concern
Financing of National Programmes-not as per need
Financing of disease control programmes are effected through societies created for the specific programmes at State and district levels. The mechanism for allocating funds directly to district societies was found to be effective as it enabled quicker absorption of funds. However, there has been a measure of scepticism. For instance, it was envisaged that such decentralization of funds to district societies would enable needbased, bottom-up programme planning and budgeting. However, this seldom happened. Most programmes are designed at the Centre and funds are released with strict guidelines and well-defined budget line items, not very different from for regular health programmes except that the unspent budget does not lapse at the end of the fiscal year. In addition, these programmes have little flexibility in issues such as contracting selected services or procurement of critical supplies. More importantly, analysis showed that in a number of instances budget allocations are not need-based and in consonance with the extent of the disease burden. For example, while the disease burden and caseload under leprosy in Bihar was 21.3% of total cases, the State received only 9.4% of the funds, while West Bengal having a caseload of 7.5% got over 10% of the allocation. Likewise, UP and MP together accounted for 37% of the total caseload under child morbidity but received only 24% of the total budget for RCH. Figures 7a and b show such mismatch between funds and need explicitly. For instance, out of 20 major States, the extent of funds allocated to States such as AP, Bihar, Madhya Pradesh and Maharashtra for the Malaria Programme was substantially
Gross underfunding of National Health Programmes: A mismatch between policy and practice
Policy governing the National Health Programmes is that services being provided under them are free for all. Theoretically, therefore, regardless of income class, all citizens of the country are eligible for availing of services free of cost under the NHP that cover vector-borne diseases, TB, leprosy, Family Welfare, cataract blindness and HIV/AIDS. Our calculations show that such a policy would need a minimum of Rs 12,000 crore against which the total amount that is spent by the Centre and States on these programmes is about Rs 5000. The suboptimal functioning of the delivery system due to gross underfunding explains the huge out-of-pocket expenditures being incurred by individual households in seeking services guaranteed to them under the NHP. A survey of households conducted by the IIHMR, Jaipur (IIHMR 2000) showed that a married woman in the age group of 15-49 years of age spent an average of Rs 400 for RCH servFinancing and Delivery of Health Care Services in India
249
SECTION IV
Fig 7a National Malaria Programme: A case of mismatch in funds and incidence-I

Source: Government of India
Fig 7b National Malaria Programme: A case of mismatch in funds and incidence-II

Source: Government of India
ices, with urban households spending Rs 604 and rural households about Rs 292. Of this, Rs 835 was spent for delivery, Rs 440 for RTI treatment and Rs 160 for child care. Similar studies show that the reluctance of women for institutional deliveries and the persistently high proportion of domiciliary deliveries are driven by cost factors. A delivery in a public hospital is reported to cost an average of Rs 601 while in the private sector it costs about Rs 3593, while at home it costs only Rs 93. The major item of expenditure was also found to be drugs, which constitute 62%. Such findings are not surprising as government spending on RCH is very low. Of the Centres total FW budget during the period 1997-98 to 200304, the amount for activities directly impacting on maternal health was Rs 2531 crores accounting for 9.7% of the total budget and Rs 17 per capita per annum for women in the age group of 15-49 years of age. Thus, it is clear that if we are to achieve the National Goals of IMR and MMR, there is a need to step up public spending and also develop social health insurance schemes to address the financial barriers that hinder women from seeking good quality care.
Lack of stability in budgetary processes

State Governments normally pass the budget between April and June every year. Once the budget is passed, treasuries located at various districts are intimated of the budgets allocated to various sectors, followed by a budget authorization. The amounts authorized vary widely depending on the financial situation of the State, and the current priorities and reasons could range from political compulsions to debt repayment. Several times during such a bad fiscal situation, budget authorizations are released but instructions are issued informally to treasury officers not to release money, disrupting ongoing activities and processes, such as finalizing a contract for procurement of drugs or equipment. The department does not only lose the unutilized funds at the end of the fiscal year but these are also shown as surrender of funds and the next years allocations accordingly pegged onto the funds actually spent. Secondly, expenditure items are also fixed and no discretion is given at any level to reallocate available funds for meeting a need or an emergency. For any such deviation the approval of the State Finance Department (and if a centrally sponsored scheme then the Central Government) is required which normally takes a few months at the minimum. Thirdly, utilization of funds also does not take place as the first instalment could be inadequate for any meaningful activity necessitating the release of subsequent instalments. Finally, in the month of December, the expenditure levels are reviewed and revised estimates for the department fixed. At times of acute fiscal stress, budget cuts are arbitrarily imposed across the department. All these factors are mainly responsible for the lumping of releases, non-timeliness of the availability of drugs or other inputs for any meaningful utilization, the lack of synchronization of the mix of inputs, etc. There are another two worrying aspects. One, at times, to not let funds lapse the amounts are spent on inessential items; and two, across the board budget
Weak absorption capacity in the Government

Even while there is mounting evidence to justify a quantum jump in public budgets for health, the Central Ministry routinely surrenders budgets allocated to it. Under World Bank projects also, there have been frequent expressions of concern at the slow pace of expenditure and poor drawals. What is the reason for this apparent disconnect between a shortage of funds and an inability to spend? Why does money not get translated into an outcome, particularly in the poor-performing States where the people are so desperate for subsidized health care? The reasons for the slow pace of expenditure are both systemic and institutional as well as poor designing and sequencing of expenditure items.
250
SECTION IV
cuts could often also mean interruptions in the supply of essential drugs carrying unintended consequences of drug resistance or burdening the patient to buy drugs.
Analysis of annual accounts of district societies for 2002-03

As has been argued earlier, various societies have been created at the district level for the implementation of the National Health Programmes, including the District RCH Society, District Blindness Control Society, District Tuberculosis Control Society, District AIDS Control Society, District Leprosy Control Society, etc. These societies are registered under the Societies Act, and the governing body of these societies usually include representatives of District administration, District Health officer (or his equivalent), and the respective programme Officers, often along with elected representatives from the district and some representation from civil society. The purpose of creating these societies is to provide autonomy for programme implementation, decentralize the planning, implementation and monitoring of the programme, and also as a funding mechanism wherein funds do not lapse at the end of the financial year and can be carried over to the next year. Funds are provided by the Central, and/or the State Government in the form of grant-in-aid to society, which is to be spent for the purpose of programme implementation. However, funds often reach the societies late, and sometimes the last instalment for the year is not even received at the end of the financial year (and is accounted for as Funds in Transit). The first instalment for a financial year can take
anywhere between 1 and 4 months or more to be released, and even more time to finally reach the District Society. At the society level, often this fund is not adequately utilized during the financial year, resulting in a high closing balance with the society at the end of the financial year. To some extent, this closing balance is also necessary, as it allows for a buffer to meet the expenditure for the new financial year before a new instalment of funds is received. However, in some cases, huge balances lie unutilized with the district societies. In a study of 17 such societies from five districts, for which the annual accounts and balances were available at NCMH, one society had a reserve balance at the end of 2002-03 which was adequate for 5 years of its current annual expenditure, while 5 other societies had reserve balances at the end of 2002-03 which were adequate for more than 1 years requirement. Table 9 Notably, 4 societies (out of 17) had negative balances, i.e. they had spent more than what had been released from the Centre/State, and this exemplifies delayed releases where funds did not reach even by the end of the financial year. A break-up of the items of expenditure showed that the predominant item of expenditure for Blindness societies was grantin-aid to NGOs, constituting between 41% and over 80% of the total expenditure incurred by these societies in 2002-03, but was insignificant for other societies. Similarly, barring one district, the remaining 4 Tuberculosis societies had salaries of the contractual staff as the largest component, varying from 55% to 83% of their expenditure during the year. If all the resource is spent on staff, what does the district society do to raise awareness? Further analysis showed no uniformity in focus.
Table 9 Break-up of expenditure of district societies in different States for various programmes in 2002-2003 (Rs in lakhs)
District Society Closing balance on 31-3-03 Expenditure during year 2002-03 Funds available at end of year (in months of requirement)
Vaishali Nadia Kozhikode Vaishali Nadia Nadia Vaishali Vaishali Kozhikode Kozhikode Jalna Nadia Pune Pune Pune Pune Pune
AIDS Leprosy Leprosy Leprosy Blindness Tuberculosis Tuberculosis Blindness Blindness Tuberculosis Tuberculosis RCH Tuberculosis Blindness Malaria Leprosy AIDS
150,314 318,502 416,071 916,235 -683,361 1,873,881 1,421,618 493,284 -660,950 2,762,057 1,131,043 263,555 0 -1,298,941 872,747 -137903 1,266,104
117,686 1,931,458 185,273 1,003,017 688,734 3,211,496 1,125,594 84726 1,019,019 851114 1,034,195 5,875,035 1,445,881 1,814,016 1,127,253 346,704 1,309,104
15.3 2.0 26.9 11.0 -11.9 7.0 15.2 69.9 -7.8 38.9 13.1 0.5 0.0 -8.6 9.3 -4.8 11.6
Source: Accounts of the various district societies made available to NCMH
251
SECTION IV
Dysfunctional system of financing

Departmental budgets are made in a five-year cycle, categorized into various heads and subheads. The broader divisions are revenue and capital, plan and non-plan. Revenue budgets finance current consumption such as salaries and wages of staff, purchase of drugs and medicines, and repair and maintenance of machinery, equipment and buildings including purchase of minor equipment, machinery, etc. Capital budgets are a one-time investment for purchase of land, building construction, equipment, machinery, etc. The budget of the health directorate is further categorized into rural, urban, allopathic, other systems of medicine, medical education and public health, and again into activities such as training, urban and rural family welfare services, contractual services, transport, and so on. All these budget heads are further allocated to numerous minute budget heads thereby making the allocations very specific. The budget process so developed over decades has resulted in fragmentation of the health sector budget into more than 4000 small heads. The funds allocated under those numerous budget heads are non-transferable and are surrendered to the States general pool of funds if they remain unutilized at the end of the fiscal year. This is strictly followed to ensure that the funds budgeted for specific activities at the beginning of the year should be spent on those activities to fulfil the intended objective. Such systems of budgeting are extremely useful for audit and accounting purposes as the key objective is expenditure control. Such procedures also help insulate the budgets from arbitrary diversions, misuse of funds and deviation from stated objectives. However, the system, from the perspective of achieving health system goals, is archaic and needs to be changed. Firstly, fixing budgetary allocation on five-year and annual plan cycles is not based on any meaningful programme audit. There are neither baselines, nor endlines, evaluations nor reviews taken into account or made available to serve as the basis for resource allocation. The exercise is routine with incremental shifts and some programmatic targets that move from year to year. In fact, targets have little to do with the professed goals that in turn have little to do with financial allocations. Therefore, since the physical targets have no bearing on the financial allocation, the focus shifts to budget utilization to protect future allocations. And since financial expenditure is the key indicator for achievement, the major proportion of the cumulative energy of the department go towards obtaining utilizations certificates and releasing funds to States and district societies, rather than focusing on the promotional activities that impact on health outcomes. Secondly, health sector needs are different, requiring a measure of flexibility as, barring some broad heads of expenditure where advance planning can be done, under operational costs, the level of unpredictability could be high. The type, nature and intensity of diseases change with seasonal variations, demographic shifts and the macroeconomic environment. Health managers cannot therefore be tied down to a five-year plan of activities nor can they foresee their needs five years in advance, as
Box
In Ontario province of Canada, all hospitals are required to furnish detailed financial returns to the department once a quarter based on which budgets are released. The returns run into over 2000 budget lines provided department-wise and indicating not only utilization of the budget but also utilization of the services. Such concurrent utilization, financial and physical line item-wise, is what gives the hospital manager an understanding of the kind of services for which the demand is growing, where there is an excess of drug budgets or the workload of staff allocated can be calculated. This then helps them to re-deploy staff to needy areas by training, wherever required; reconfigure resources, shutting down departments where there is inadequate demand; bring in control on prescription of drugs or tests wherever they are found to have crossed reasonable limits, etc.
a SARS epidemic can upset the whole budget allocation and priorities. Similarly, at the local level, also, hospital managers have to be taking multiple decisions all the time requiring flexibility and some autonomy in financial decision-making. Besides, for a policy-maker, the structure of budgeting makes it impossible to identify the cost centres, where expenditure control needs to be exercised, the type of skills mix needed, which departments should be closed down and which expanded in keeping with the changing disease burden, etc. Such lack of flexibility is the reason for the low occupancy of beds in public facilities. Since hospital budgets are not global and are factored based on bed strength, which determines the staff and drug support, etc. there could be situations where one department has funds though few patients, while another may have restricted funding but have two patients on one bed.
Complex design
Funds also do not get spent if the design of the scheme or intervention is very complex and process-oriented. Participatory systems that involve all stakeholders do provide, in the long run, greater sustainability to the programme. However, such approaches are time-intensive as different constituents of stakeholders have different and varied ideas, expectations and needs. Harmonizing them takes time, as community responses are not always uniform. Therefore, when any activity has to be implemented within a strict time-frame, then such processes get short-circuited and data are fudged or money not spent. Second, and more so in donor-funded projects, the emphasis is on spending. The release of funds is in equated instalments spread over all the project components. In such a system, delay in the completion of one activity upsets the implementation of others. For example, training may get held up due to delays in the preparation of training modules or training of trainers or the procurement of equipment may get stalled due to delay in the construction activity.
Inadequate allocation of funds under externally funded projects

A frequently heard issue in relation to externally funded
252
SECTION IV
projects is the slow pace of expenditure. This happens for three reasons: First, while the Government approves several standalone projects and agrees to a yearly funding plan, in practice, funds made available under the EAP (Externally Assisted Projects) component are normally short of the amounts agreed to. This is because of the system of capping the proportion of the EAP for each ministry based on the total resource position of the country calculated on the basis of total domestic and external revenues. Second, the procedures for implementing activities are cumbersome and require multiple clearances at several stages. Construction activities and procurement of equipment take, on an average, eighteen months to two years for starting the activity, or obtaining the equipment. Third, complex procedures are involved in the recruitment of staff, and the process of selection is highly time-consuming, taking over a year. States are also often reluctant to create posts for which they will have to pay after the cessation of the project in five years, adding to their non-plan budget. Besides, due to low salaries, often posts do not even get filled up. To circumvent this problem, increasingly projects are recruiting persons on contract. While this enables quicker placement of people, it affects the human resource issue of the department in the long term, since contractual appointees are neither provided training nor given any financial delegation of powers and responsibilities as they are seen as temporary workhands. A serious matter related to external funding is that such funding is not provided as an additionality. In such a system, instead of the health system being strengthened by external funding, priorities get skewed and distortions created, as nonfunded programmes, which could be equally if not more important, get lower funding priority. Besides, since external funding is not an additionality, there is little incentive for the department to mobilize donor aid. At times of acute fiscal stress, again, two things happen: either the externally aided components of the budget are protected to the exclusion of all else as seen in AP during the late 1990s, distorting departmental priorities once again, or the crises may end in curtailing the availability of funds to the externally aided projects also, affecting the spending and credibility of the State for raising future funding.
is aware of and responsible only for budgets released to hospitals having more than 100 beds, whereas budgets related to facilities having less than 100 beds are administered and monitored by the respective Zila Parishads with funds released by the Department of Finance. Moreover, as already observed, the budget lines are useful only for accounting purposes and not for policy planning. We tried, for example, to calculate how much government departments spend on the health care of its serving employees. For the Government of India, this information is spread over 8000 drawing officers, 700 autonomous bodies, 38 departments and 220 PSUs. Each DDO again has to scrutinize the salary bills to disaggregate the amount spent on medical care! In State Governments, obtaining this information was impossible.
Weak financial capability

At almost every level-central, State or district, administrative directorates or hospital units-the staff dedicated for financial oversight functions are few and their capacity weak. In most cases, the staff consists of one or two officers and a few clerks. None is trained on either financial management or on health needs. Several times their knowledge of financial rules is superficial. While the Central Government has an internal audit system, at State and district level, such concurrent audit systems do not exist. Computerization is poor and so is the capacity for planning and budgeting. Weak systems give room for discretion and scope for fraud and, more importantly, for delays due largely to raising meaningless and frivolous queries. This therefore calls for greater professionalism of the finance set-up and sharing of responsibility, making them equally responsible for poor expenditure. Changing their mindset from account-keeping to being facilitators for achieving certain goals should be the key for the future.
Conclusion
Health sector in India suffers from gross inadequacy of public finance and therefore an immediate and significant scaling-up of resources is an imperative. The undue burden on households for spending on health cannot be wished away. Further, it is also clear that there is an urgent need to restructure the budgeting system to make it more functional, amenable to review of resource use to take corrective measures in time and be flexible enough to have the capacity to respond to an emergency or local need. Rules and procedures for actual release of funds, appointment of persons, labour laws, procurement systems all need a thorough review. Greater decentralization of funds, aligned with functional needs and responsibilities, is necessary. However, any decentralization and financial delegation needs to be carefully calibrated and sequenced. In other words, decentralization can only be done after developing the requisite financial capability and laying down rules and procedures for accounting systems. Unless such restructuring takes place, greater absorption of funds will continue be difficult.
Budgeting not functional

In other countries, budgets have two heads-capital and operational. A budgeting system based on an artificial classification of plan and non-plan makes it impossible to know what money is going where. Since the annual planning process only considers the plan or new activities, the maintenance of assets never gets the required attention under the nonplan budget. So while under the plan budget, buildings are constructed, the cost of maintenance is not factored in from year to year. Second, the aggregation of budget heads keeps changing making any trend analysis difficult. Third, there is no uniformity in budget lines in the country. For example, in Karnataka and Maharashtra, the Director of Health Services
253
SECTION IV
Appendix 1
Primary care: For the purpose of this paper, primary care has been taken to include all facilities which provide outpatient care, which may be preventive, promotive or curative, and also those facilities which provide outpatient as well as limited inpatient care, wherein the admissions are primarily for stabilization or observation, etc. Thus, in this paper, primary care includes subcentres, PHCs and CHCs, as also ISM dispensaries. Secondary care: This category includes institutions with inpatient facilities, above the level of CHCs, but not providing superspecialty care. Therefore, this category includes taluk and district hospitals. While these institutions also provide primary care, and for some of these, this could be a major activity, the entire expenditure of such institutions is classified as secondary. Tertiary care: Teaching hospitals and medical colleges have been included in this category, which includes allopathic as well as Indian systems of medicine. While these institutions also provide primary and secondary care, and for some, these could be major components of their activities, the entire expenditure of such institutions is classified as tertiary. Research and training: Expenditure incurred on institutions mainly engaged in health-related research or health-related training activities have been classified in this category. Teaching hospitals and medical colleges have not been included here, for example, as patient care is the major activity undertaken by them. Social health insurance: Expenditure incurred on subsidizing the running of schemes such as CGHS, ESIS and other such funding mechanisms for the health care of specific groups of people, have been included in this category. International cooperation: This includes expenditure incurred on contributions to UN agencies and other international agencies such as Red Cross. Administrative: Although each of the above categories will have administrative components therein, this category only lists those expenditures that are exclusively administrative, for example, the secretariat of the ministry, or a regulatory agency. As will appear later in this chapter, a similar methodology for these categories has also been followed in the case of State budgets, except that small amounts of other expenditure that could not be allocated in the above categories are also grouped under the category of Other Expenditure. Since there was no expenditure by the status on International Cooperation, the category does not figure in the table on expenditures of the State governments. The budgetary data were categorized according to subheads and detailed heads in rows, and object heads in columns, and were entered into spreadsheets. Each of these detailed row items was then categorized as belonging to one of the above categories, and then the sum totals of expenditure under each category were obtained and tabulated. For the Central Government, actual expenditures for 5 different financial years, from 1991-92 till 2002-03, were similarly categorized and analysed. As the heads of accounts used in the 1991-92 budget were different from the ones in use now, these have been adapted to the ones in use at present. For States, this analysis was similarly done for 16 major States, for the year 2001-02, based on the actual figures published in the Detailed Demand for Grants of the respective health ministries. The only exception here was Bihar, where the 2003-04 demand did not contain all the actual expenditures, and the 2001-02 Revised Estimates (RE) figures for Bihar have been used throughout this study.
254
SECTION IV
References
Government of India. Statistical abstract of India. Various years. Mahal A, Singh J, Afridi F, Lamba V, Gumber A. Who benefits from public health spending in India. New Delhi: National Council of Applied Economic Research; 2001. National Sample Survey Organisation (NSSO). Morbidity and treatment of ailments. Report No. 441. New Delhi: Department of Statistics, Central Statistics Organisation, Government of India; 1998:A-170. National Sample Survey Organisation (NSSO). Morbidity and utilization of medical services. Report No. 364. New Delhi: Department of Statistics, Central Statistics Organization, Government of India; 1989:A-13. Prabhu KS. Social sector expenditures and human development: A study of Indian states. Bombay: Development Research Group, Reserve Bank of India; 1993. Reserve Bank of India (RBI). Handbook of statistics. Various years. Reserve Bank of India (RBI). Report on currency and finance. Various years. Selvaraju V. Budgetary subsidies to health sector among selected States in India. Journal of Health Management 2001;3. Selvaraju V. Health care expenditure in rural India. Working Paper No. 93. New Delhi: National Council of Applied Economic Research; 2003. Tanzi V, Schuknecht L. Public spending in the 20th century: Global perspective. Cambridge: Cambridge University Press; 2000. World Bank. World Development Report 2004: Making services work for poor people. World Bank; 2003:256-7.
255
SECTION IV
Annexure 1 National Health Accounts for India
O G
AJAY MAHAL
UR ESTIMATES OF HEALTH SPENDING IN INDIA ARE BASED ON THE EXHAUSTIVE study commissioned by the National Commission on Macroeconomics and Health (NCMH). These estimates are constructed using a National Health Accounts (NHA) approach which is explained below. The NHA methodology views financial flows as occurring across primarily three sets of agents and/or categories-ultimate sources of funds, financial intermediaries, and the uses to which funds can be put. These uses can be classified in a number of ways-by type of provider (e.g. government, private, non-profit); or by functional classification (e.g. inpatient care, outpatient care, collective goods, direction and administration, training and research). As our goal is to estimate aggregate spending and its major components, and the way such expenditures are financed, while avoiding any double counting, we limit ourselves to describing financial flows of only three types: (i) from sources of funds to financial intermediaries; (ii) from financial intermediaries to providers of care; and (iii) from financial intermediaries to a functional classification for the purpose of care. Given that there are several sources of funds, multiple financial intermediaries and different providers/functions in a country, these flows are best presented in the form of three matrices (here we provide only the first two matrices in Tables 1 and 2), each corresponding to a different part of financial flows related to health. The three sets of matrices relate to the following financial flows: From ultimate sources of funds to financial intermediaries From financial intermediaries to functional categories From financial intermediaries to care providers
The Meaning of Health Expenditure

A key step in trying to estimate financial flows linked to health is to specify the meaning of expenditure on health. For the purposes of this note, health expenditure is defined to include spending on care and treatment associated with illnesses, on administrative expenses associated with such treatment, spending on public health programmes (such as tuberculosis, malaria, blindness and HIV/AIDS), on medical research and training, rehabilitation, immunization programmes and selected components of programmes associated with maternal and child health. Both recurrent and capital expenditures are included. We have followed the convention of the literature on national health accounts and have not included in our analysis expenditures for nutrition, education, clean water and sanitation programmes, referred to as health-related services in George and Pattnaik (2004). It can be argued that the omitted categories of expenditure have implications for health, and some studies of health spending have, in fact, included such expenditure flows. This note presents information on health expenditure for 2001-02, the latest fiscal year for which data were available under many NHA categories.
S. SAKTHIVEL
The entities that spend money on health in India
SOMIL NAGPAL
somilnagpal@yahoo.com
Given the above working definition of health expenditure, who are the players (or agents) that spend money on health? From the perspective of the ultimate sources of funds, this group includes primarily the government, households (their out-of-
256
National Health Accounts for India
SECTION IV
pocket spending and contributions to insurance premiums, whether in the public or private sectors), non-governmental cum non-profit entities (NGOs), firms (whether in the public or private sectors), and international institutions. This set of ultimate sources of funds may also include banks that finance health sector investments, although the George and Pattanaik (2004) study does not consider this possibility. This study considered the following categories as the ultimate sources of funds: The Government (State, Central, local) Households Firms (public and private enterprises) Quasi-government organizations other than public sector enterprises Non-governmental organizations International agencies/rest of the world Others (such as surpluses of certain organizations that fall outside the above categories) Besides being ultimate sources of funds, many of the above agencies also serve as financial intermediaries as funds move from ultimate sources to ultimate uses. In particular, the following are the major financial intermediaries in the Indian setting: Department of Health, Medical and Family Welfare (DOHMFW) Other State government departments that spend money on health Central government ministries that spend money at the State level Local governments Societies/autonomous bodies Public and private enterprises (especially in their role as payers of health services for their employees) Social insurance [Employees State Insurance Scheme (ESIS), Central Government Health Scheme (CGHS)] Voluntary insurance (individual and group) Households (when they directly pay for services received by them)-sometimes they may be reimbursed for such expenditures. Financial intermediaries allocate funds to the ultimate providers of health and health-related services. Of the several categories, two types of ultimate uses were considered for this report: provider-based classification and functional classification. These are discussed further below. What are the uses and/or functions on which health expenditure is incurred? The two major classifications are provider-based and function-based.
Public providers of health services

Public providers include (i) hospitals of the State government (separately, if needed for Indian systems of medicine and non-Indian systems of medicine); (ii) dispensaries of the State government; (iii) sub-centres; (iv) rural and urban family welfare centres of the State government; (v) facilities of various Central ministries (such as Defence, Railways and Posts and Telegraphs); (vi) facilities of public enterprises; (vii) facilities and services of local governments; (viii) facilities of CGHS; (ix) facilities of ESIS; (x) facilities of autonomous institutions and societies (xi) facilities of other State government providers not captured above; (xii) collective health services (of DOHMFW and other government entities); (xiii) administrators (DOHMFW); (xiv) administrators (ESIS, CGHS and other social insurance); (xv) providers of training, education and research in the public sector, such as State Institute of Health and Family Welfare (SIHFW), medical colleges, OSM colleges, nursing colleges, auxiliary nurse midwife (ANM) training colleges, etc.; (xvi) providers of training, education and research in the private sector. It should be noted that Collective health services include expenditures on prevention of disease; family welfare and prevention of food adulteration. The categories include collection of statistics and statistical analyses, information/advocacy efforts in health, testing of water and food quality, family planning, antenatal care, etc.
Private Providers of Health Services

These include (i) private hospitals; (ii) private doctors; (iii) facilities of private firms/enterprises (iv) traditional health providers; (v) traditional birth attendants; (vi) ancillary care providers; (vii) administrators for private insurance; and (viii) medical education/research and training in the private sector. It should be noted that ancillary services include expenditure on drug purchases, clinical laboratories, diagnostic imaging, and ambulance services. This classification is difficult to undertake in the public or the private sector, although some estimates can be made based on the NSS data.
Non-profit institutions
(i) NGOs (charitable hospitals and dispensaries) and others that provide clinical services; (ii) NGOs that provide disease control and health promotion services; and (iii) medical education/training/research provided by non-profit institutions;
Provider-based classification
Public providers Non-profit providers Private providers Other providers Rest of the world
Other providers
(i) Rest of the world-this may include international NGOs and health services obtained abroad; (ii) self-care. Typically, data on (i) is difficult to find.
257
SECTION IV
Function based classification

Personal health services Collective health services Direction and Administration Health-related services
Personal health services

These include (i) outpatient care; (ii) inpatient care; (iii) selfcare; and (iv) treatment by unqualified practitioners. One can, if needed and data were available, consider a further sub-classification into public and private providers.
Collective health services

These consist of (i) disease prevention (expenditures on government programmes for control of communicable diseases and non-communicable diseases, surveillance of diseases, surveys and statistics, vaccinations other than primary vaccinations for children); (ii) health promotion-(a) family planning and welfare: expenditures on family welfare programmes undertaken by the government and all expenditure pertaining to childbirth, abortion (except spontaneous abortion, medical attention for which is considered a curative service), antenatal care, postnatal care, family planning and primary immunization to children; (b) Control of food adulteration and drugs-Includes expenditure on prevention of food adulteration and drug control administration
Direction and administration

In general, this information is not readily available for private providers. Thus, the standard approach has been to take account of all Direction and Administration expenditures in DOHMFW, in CGHS and ESIS, in private insurance and, if possible, in other government health services expenditures.
Health-related services
Here again, we have medical education, training and research and ICDS spending by public, private, or non-profit providers.
Methodology and Sources of Data for Estimating Health Expenditure in India

Matrix 1 summarizes information on the major ultimate sources of finances for health expenditure for which data were collected for India, the major recipients of such funds and the sources from which data were obtained on the magnitude of the various financial flows. Matrix 1 shows that the Central Government contributes to State health expenditure in several ways by supporting (i) State health departments, other State departments and societies by means of centrally sponsored schemes. Many (but by no means all) of the Centrally sponsored schemes in question are funded, at least partially, by international agencies;
(ii) health expenditure of its current and retired employees (and their dependants) based in different States: via the Central Government Health Scheme (CGHS), dispensaries of the Department of Posts and Telegraphs and Department of Telecommunications, and the Central Services (Medical Assistance) scheme; (iii) expenditure by Ministries such Railways and Defence on their current and retired employees (and their dependants); and (iv) grants to non-governmental organizations. Similarly, State governments contribute to health spending by supporting (i) the State department of health and family welfare, known as the Department of Health, Medical and Family Welfare (DOHMFW) and the various directorates that come under its responsibility; (ii) contributions to the social insurance scheme known as the Employees State Insurance Scheme (ESIS) established for employees earning less than a pre-specified amount in firms, public, or private organized sector. (iii) supporting health expenditures by their current and retired employees (and their dependents); (iv) supporting hospital societies in the form of stoppage charges; (v) contributions in the form of expenditures incurred by other State departments, such as Tribal Welfare, Governor and the Council of Ministers and the Department of Women and Child Welfare and Disabled Welfare; and (g) grants given to local governments for specific purposes. Matrix 1 also highlights the role of local governments in financing health expenditures. Local governments belong to two categories, depending on whether they relate to urban settings (Municipal Council, or Corporation), or rural areas (Panchayati Raj institutions [PRIs]). In principle, both sets of governments can raise funds on their own, in addition to benefiting from transfers from the State Government. Expenditures financed by own resources highlight the role of local governments as an ultimate source of funds. Most of the health-related activities of municipalities and municipal corporations are confined to that of public health (registration of births and deaths, antimalaria programmes, etc.), sometimes in conjunction with the operation of a small set of primary care centres. In this study, only urban municipalities/councils were considered as resources flowing into PRIs on health sector are insignificant. Households are a major ultimate source of health spending. These include contributions to insurance schemes (CGHS, ESIS, Armed Forces Group Insurance, Mediclaim) and user fees paid for health care at both public and private health care facilities. As some health care expenses of households are reimbursed by insurers, the government and private employers, the net out-of-pocket expenses incurred by households are less than that suggested by household surveys. Our study took this problem into account and the health expenditure estimates reflect adjustments for reimbursements. How did we arrive at households spending on health? Using data from the 52nd Round of NSS, we estimated the mean expenditure for 1995-96, while the number of treated inpatient and outpatient cases were anchored to the 2001-02 population. The mean expenditure obtained for 1995-96 was then projected forward to 2001-02 by adjusting it for both
258
Matrix 1
From Financial Sources to Financial Intermediaries Government Centre State Local Enterprises Agencies NGOs Private Firms Households PubSecBanks Other Public Foreign Private Sector TOTAL % Share (Rs in crore)
National Health Accounts for India (200102)
Financial Intermediaries
Central Government 1898.7 1299.1 113.4 1603.2 2339.0 2629.8 1149.8 807.4 315.4 603.3 1264.4 41.1 340.4 6.0 1929* 15048.6 1770.9 1898.7 1299.1 113.4 18422.7 2339.0 2629.8 1155.8 807.4 356.5 2208.1 0.0 99.9 439.0 793.5 7859.6 7.2 14.4 2.2 1.9 15651.9 2339.0 2043.2 2209.9 2.0 366.0 0.3 366.0 1186.0 3257.8 3.0 656.8 963.0 72758.7 74760.1 68.8 238.8 245.0 0.2 0.0 0.0 0.2 756.9 1768.0 74977.0 108732.5 100 1.75 1.19 0.10 16.94 2.15 2.42 0.00 1.06 0.74 0.00 0.33 2.03 0.00 0.00 0.70 1.63 68.96 100
Medical and Public Health
Family Welfare
Indian systems of medicine
State Government
Local Government
Other Central Ministries
Firms
Public
Private
Social Insurance
CGHS
ESIS
Private Insurance
Community Insurance
Other Voluntary Insurance
NGOs
Households
TOTAL
Percent Share (%)
DoHMFW: Department of Health, Medical & Family Welfare, CGHS: Central Government Health Scheme; ESIS: Employees State Insurance Scheme Note: ii) Household expenditure figures based on Health Round figures of 1995-96, extrapolated to 2001-02 in the ratio of the growth in Consumer Expenditure, which incorporates price change, growth in demand for health services, etc.. iii) PSU reimbursements and PSU Medical Allowance have been treated as transfers to households iv) Figures for flows from Foreign Agencies to State Govts are from CAAA data, and this also accounts for negative figures, and funding of NHPs through Central Government Budget has also been added. v) Central Department figures exclude CGHS, transfers to states, external funding vi) Foreign Agencies to States includes the sum of external funding of NHPs given to Central Government vii) Breakup of ESIS Contribution assumed in the ratio of 4.75:1.75, which slightly overestimates households because people earning low wages are exempted from their share viii) *Disbursements by Banks in 2001-02 has been shown in the table but not included in the calculations for the flows. The figure also includes interest accrued on the amount outstanding (6127 crores) on 31-3-01 for the year 2001-02 at an assumed rate of 10% ix) Figures for Local Government spending based on AP-NHA, actual data from Maharashtra, Statistical Abstract of India 2003, NIPFP study on Municipal Sector for XII FC. Does not include PRI. Average of actual data from major states applied to states for which sample data from corporations/municipalities was not available. x) PSU figures only include projections based on data pertaining to 255 Central PSUs, and do not include state PSUs xi) Defence expenses on employees are from IHS APNHA and pertain to 2002-03 xii) Figures for Private firm reimbursements projected from AP-NHA xiii) Foreign Agencies to NGOs based on 2000-01 Annual report of FCRA, MHA, indicating total funds received for health and FW activities (quoted in AP-NHA)
SECTION IV
259
260
Ministry of Health and Family Welfare Health FW ISM Government Government Ministries Public Private CGHS ESIS Insurance holds State Local Other Central Firms Social Insurance Private NGOs HouseTotal
Matrix 2
(Rs in crore) Share (%)
SECTION IV
From Financial Intermediaries to Providers
Providers
0.4 1.6 1095.7 31.6 3.8 20.5 6.7 1260.1
3733.8 3450.5 9660.3
11.13
60.6
4619.1 3467.7 11414.8 31.6 3.8 1329.3 19.8 10.0 0.8 29.8
4.25 3.19 10.50 0.03 0.00 1.22 0.02 0.01 0.00 0.03

1155.84 1942.30 439.06 200.00 27.30 21.16 65.8 165.4 23.26 545.0 3916.49 8101.11 41083.78 3274.06 807.40 1768.00 488.76 125.3 97.1 68.0 260.9 114.28 83.7 1299.1 1.19 3.4 113.4 0.10 57.1 18422.7 16.94 2339.0 2339.0 2.15 0.0 2629.8 2.42 0.0 1155.8 1.06 0.0 807.4 0.74 0.0 356.5 0.33 1685.0 2208.1 2.03 151.4 756.9 0.70 0.0 1768.0 1.63 1247 17239.8 74977.0 68.96
1155.8 1942.3 439.1 200.0 27.3 21.2 0.0 4550.5 8101.1 41249.2 3274.06 807.4 1768.0 603.0 125.3 674.2 1247.5 21619.9 108732.48 100
1.06 1.79 0.40 0.18 0.03 0.02 0.00 4.19 7.45 37.94 3.01 0.74 1.63 0.55 0.12 0.62 1.15 19.88 100
Governmen (non-ISM) Teaching hospital, etc (tertiary) 813.2 District hospital (secondary) 15.6 CHC+PHC+HSC+Dispensary+ NHPs (primary) 658.8 Government (ISM) Teaching hospital Other hospitals + clinics Other D & A Secretariat , direction and administration 42.0 Public health laboratories 19.8 Food/drug Control 10.0 Statistics/others 0.8 Other Administrative expenses 29.8 Other Public Providers Public firms Defence Railways Other Civil Ministries (Reimbursements) P&T Urban Development (Capital expenditure for hospitals) Others Private Providers Private hospitals Private doctors Drugs provider Diagnostics provider Other Private Providers Private firms NGOs Social Insurance Providers ESIS facilities CGHS facilities Train/Research Providers Training and research 248.2 Other Providers Traditional Others* 60.5 TOTAL 1898.7 Share (%) 1.75
Note: *Items under Others mainly include lodging, transport, etc. which works out to Rs 13,984.64 crore Note: Reimbursement data from NSS ignored in our calculations. DOHMFW=Department of Health, Medical and Family Welfare; CGHS=Central Government Health Scheme; ESIS=Employees State Insurance Scheme Figures in Column B, C and D from Detailed Demand for Grants, 2003-04, MoHFW and the breakup into tertiary, secondary, primary and Admn based on the category wise classification done at NCMH, excluding GIA to states. State Govt expenditure allocated into categories based on the ratio derived from the category wise classification of 16 major states at NCMH, and from this value, the ISM expenses have been removed and classified separately. Insurance allocations based on 80% claim ratio and an assumed 10:90 ratio of utilization of public and private hospitals by Mediclaim beneficiaries State govt. allocations to Public Health Laboratories, Food and Drug Control , Statistics, etc.. are not provided seperately above in the matrix but are included in total.
SECTION IV
Table 1 Health care spending in India, by source of funds, 2001-02

Ultimate source Total resources (Rupees in crore) Share in total expenses (%) As proportion to GDP (%)
Government Central State Local Households External Funding Firms Public enterprises Public sector banks Privateenterprises Others/NGOs Total
25850.5 7859.6 15651.9 2339.0 74760 2209.9 5301.0 2043.2 245.0 3257.8 366.0 108732.5
23.8 7.2 14.4 2.2 68.8 2.0 5.1 1.9 0.2 3.0 0.3 100.00
1.14 0.35 0.69 0.10 3.30 0.10 0.23 0.09 0.01 0.14 0.02 4.79
Source: Based on Matrix 1: Financial sources to financial intermediaries
Table 2 Health care spending in India, by financial intermediary, 2001-02

Ultimate source Total expenditures (Rupees in crore) Share in total expenses (%)
current employees and their dependants and, in some cases, retired employees (and their dependants). These contributions mainly take the form of (i) direct provision of health services by some firms; (ii) contributions by firms to insurance schemes such as ESIS and Group Mediclaim; (iii) Reimbursements of health expenditures; and (iv) lump sum allowances as part of salary. Whether the last category ought to be included in health expenditure estimates is debatable as, it need not correspond to any health spending. International agencies also support Statelevel health spending in a number of ways: (i) By providing grants/loans to the Ministry of Health and Family Welfare and other Ministries in the Government of India that, in turn, support centrally sponsored schemes at the State level; (ii) support to State Governments (health and other departments), and funding State-level societies; (iii) direct support to NGOs in the State for care, training and research. The funding agencies include bilateral and multilateral institutions, international NGOs, and individuals. Data on direct NGO support by individuals are particularly difficult to obtain.
Government Central Other Central Ministries State: DOHMFW Local Firms Public enterprises Private enterprises Social insurance Private insurance NGOs Households Reimbursements Total
Source: NHA Matrix 2, from financial source to financial intermediaries
26702.7 3311.2 2629.8 18422.7 2339.0 1963.2 1155.8 807.4 2564.6 756.9 1768.0 74977.0 2218.3 108732.5
price changes and growth in real terms, based on information from the 50th (1993-94) and 55th (1999-2000) rounds of consumer expenditure surveys. Calculations were also undertaken separately for inpatient spending and outpatient spending, for reproductive and child health, and expenditures incurred on self-care for the rural and urban populations in each state and union territory of India. Finally, households estimates for the year 2001-02 was obtained for different categories: rural-urban, outpatient-inpatient, reproductive and child health, self-care and by all States and union territories. A third set of major contributors for health expenses are firms, in the public and private sectors who support their
What is the total spending on health and as a share of GDP in India? India spent approximately Rs 108,732 crore on health and health-related expenditures during the fiscal year 2001-02. This amounted to about 4.8 percent of the estimated Gross Domestic Product (GDP) at market prices in 2001-02. National health expenditures, when taken as a proportion of GDP at factor cost, were 5.2 percent. As a proportion of GDP, our estimates of national health expenditures are on the lower end of previous estimates for India. [5.2 percent (Peters et al. 2001), 6.0 percent (Peter Berman, as cited in World Bank 1995)]. Who are the major financiers of health care spending in India? Table 1 shows that households ultimately financed about 69% of all health spending in India, with the different branches of the government (Central, State and local) contributing about 24%. If one were to add the contribution of public sector enterprises and quasi-government institutions, the governments share increases to a little more than 26% of all health spending. International funds support about 2 percent of national health spending, with private-for-profit enterprises contributing another 3%. Although not large as a proportion
24.55 3.04 2.42 16.94 2.15 1.80 1.06 0.74 2.36 0.70 1.63 68.96 2.04 100.00
Findings
261
SECTION IV
of total health spending, international support could potentially be quite substantial for state governments' health spending, amounting to roughly 10% of the latter. How important is Insurance as a source of financing? Note first that insurance is a form of financial intermediation, whose ultimate contributors could be households, the government, firms, and other groups. Insurance could both be social (such as ESIS and CGHS) as well as voluntary (in the form of group and/or individual Mediclaim policies, AGIF and other packages offered by insurance companies in India). Viewed in this narrow sense, about 3.1% of health expenditures in India were insurance-supported spending (Table 2). Of course, government health facilities that offer subsidized services can also be considered a form of health insurance, albeit financed by the state. Irrespective of the method used for arriving at health expenditures covered by insurance, it is clear that a substantial
amount of health expenditures (presumably curative care) in India is not covered by insurance schemes, and thus have the potential of leaving people who incur such expenditures worse off. Moreover, it may be the relatively economically worseoff households who bear the brunt of these expenditures for at least two reasons. First, most private and social insurance schemes do not cover them. Second, they may not be as able to access subsidized public health facilities as the better-off groups corner them the most, as a study using NSS data suggests (Mahal et. al.). In addition to the financial risk borne by people on account of the relative lack of access to health insurance, excessive reliance on out-of-pocket spending is economically inefficient because individuals are much less effective in bargaining for better prices and services than groups. Indeed, mechanisms such as reimbursement for health expenses incurred by individual households are also likely to be inefficient for the same reason.
262
SECTION IV
References
George C., Pattanaik G. Andhra Pradesh State Health Accounts 2001-2. Hyderabad: Institute for Health Systems; 2004. Mahal A et al. Who benefits from public health spending in India? New Delhi: HNP, The World Bank; 2001. D. Peters, A. Yazbeck, G. Ramana, R. Sharma, L. Pritchett, A. Wagstaff. 2001. Raising the Sights: Better Health Systems for India's Poor. Washington, D.C.: The World Bank. The World Bank. 1995. India: Policy and Finance Strategies for Strengthening Primary Health Care Services. Report #13042IN. Washington, D.C.: Population and Human Resources Division, South Asia Country Department II.
263
SECTION IV
User charges in Indias health sector: An assessment
Background
AJAY MAHAL
N. VEERABHRAIAH
ASSISTANT COMMISSIONER ANDHRA PRADESH VAIDYA VIDHANA PARISHAD HYDERABAD, ANDHRA PRADESH
SER CHARGES CAN BE DEFINED AS 'CONTRIBUTIONS TO COSTS BY INDIVIDUAL users in the form of a charge per unit of service consumed, typically in the form of cash' (Reddy and Vandermoortele 1996). Thus, user charges are explicitly distinguished from insurance arrangements that require payment into a pool without reference to a specific service received. They are also to be distinguished from health care financed through general revenues supported by taxation. In the present context, user charges are referred to as being for health services provided in the public sector. The case for user charges in the health sector has typically been made on three grounds all of which are central to India's health policy objectives: that they have the potential for adding to scarce public resources (and presumably, therefore, quality and coverage), enhancing efficiency and promoting equity. User fees can raise resources for health by charging for services provided in government-run facilities. The success of such a step will depend on (i) the magnitude of user charges per unit of service provided; and (ii) the responsiveness of service utilization to user charges, often referred to in technical jargon as the price elasticity of utilization. If the utilization of services in public facilities falls sharply in response to the imposition of user charges (i.e. it has a high price elasticity of health service utilization), the move may not be very successful in raising revenues. Moreover, if the decline in utilization is somehow not made up by a rise in utilization of needed health care services elsewhere, e.g. in the private sector, then the goal of raising resources may have the counterproductive impact of reducing health care consumption and potentially worsening health outcomes. To this one can add a third critical concern. Publicly provided subsidized health care, especially inpatient care in hospitals serves as a device for insuring the population against the financial risk from catastrophic illnesses. Thus, raising finances by means of user charges, even if feasible, may end up sacrificing another key goal of health policy, i.e. protection against the financial risk from illness. In theory, user charges can contribute to improved efficiency in several ways. For instance, if health care services in the public sector are being overutilized because they are free, then the imposition of user charges may help in curtailing some of this excess usage. In developing countries such as India, the most obvious example of such excess usage is the high use of outpatient departments in high-end secondary and tertiary hospitals, and the under use of primary health care centres (PHCs) for minor illnesses. By imposing user charges in public hospitals, especially for individuals who visit without a referral from a lower-level facility such as a PHC, policy-makers may be able to achieve more rational use of health facilities. This rationalization can be strengthened if revenues from user charges can help enhance the quality of care at lower-level facilities. Efficiency can also be promoted if user charges help to provide essential complementary items (such as drugs and consumables) to health facilities. A health facility with personnel but no consumables is one example where additional resources to fund consumables would also help improve the yield from the available medical personnel. If user charges not set at too high a level, they might lead administrators to make efforts to produce the associated service at a lower cost, which would enhance the efficiency of the health services.
265
SECTION IV
Finally, user charges can also be used to divert governments from subsidizing care that is less effective in achieving health improvements towards public health interventions such as immunization, clean water, nutrition and sanitation, with substantially greater health benefits per rupee spent. In this connection, the objective of resource enhancement, which requires low price elasticity, ends up conflicting with the objective of resource diversion towards public health interventions, which requires that people respond to user charges by curtailing consumption (Gertler and Hammer 1997). However, these efficiency gains are not guaranteed. For instance, if the funds from user charges get transferred to the treasury with no guarantee of the health department getting access to these additional resources, these efficiency gains will not occur. Also, user charges at hospitals seeking to divert individuals to lower-level units will not work in the absence of quality improvements at lower-level facilities, which led to excess hospital visits in the first place. All that may happen is a decline in overall utilization of care. To this one may add another caveat-the goal of efficiency may sometimes conflict with that of financial risk protection; for example, when user charges are imposed on hospital-based inpatient care to divert public subsidies to public health interventions such as immunizations, nutrition, health promotion activities, etc. User charges may also influence equity. If revenues from user charges are used to improve the quality of care at lowerlevel facilities, which typically are close to where the poor live, utilization of services by the poor as well as their health might improve. Alternatively, if user charges somehow dissuade richer sections from using public facilities, they may help the poor to obtain better access to subsidized public facilities. If these steps can be combined with fee exemption schemes for the poor, then there is a real potential for the poor to benefit from improved quality care facilitated by revenues from user fees. Equity gains will depend on the extent to which funds are retained at the facility and health department level, are used for improvement in quality, and on how effectively the poor can be identified and exempted from user charges. If exemption is difficult to administer or if quality improvements do not happen, how the health care utilization by the poor responds to user charges becomes critical from the standpoint of a policy focused on equity in access. Supporters of user fees have also pointed out other potential gains. Retention of user fees by facilities will help empower community-level managements that oversee their expenditures, thereby lending support for efforts at decentralization. Finally, user charges may also lead to increasing political acceptance of insurance mechanisms that help protect individuals against out-of-pocket expenses due to illness (Shaw 1996).
First, user fees, even if designed optimally, do not simultaneously achieve all the goals that are of interest to policymakers. Some goals such as efficiency and resource enhancement may conflict with others such as financial risk protection (Gertler and Hammer 1997). Second, the impact of user fees in achieving health goals depends substantially on the way the fees are implemented and on the responsiveness of health service utilization to user charges. Thus, both the appropriateness of user charges and their optimal design are dependent on information that is best obtained by empirical analyses.
User Charges in Public Health Facilities: The international literature

There is now an extensive literature on user charges in health, much of it focused on sub-Saharan Africa; but also from Asian countries such as China, Indonesia and Viet Nam. This section summarizes the main findings of this literature and the resulting implications for the perceived benefits of user charges with regard to resource raising, efficiency and equity.
Resources from user fees

There is considerable debate about the effectiveness of user charges in raising revenues. In sub-Saharan Africa, user fees have tended not to exceed 5% of total government spending on health (Creese 1997). At the micro level, there are large variations across countries and facilities, even within sub-Saharan Africa. Some lower-level health facilities reported a rise in their revenues by 40%-90% of their total expenditure. Hospitals were able to raise their revenues through user charge by 15%-45% of their non-salary expenditure (Shaw 1996). On the other hand, in China, revenues from user charges accounted for roughly 36% of all government spending, and the number seems to be high for countries such as Viet Nam as well. At the facility level, the share of revenues from user charges in total expenditure is substantially higher (Creese 1997; Gertler and Hammer 1997; Shaw 1996). Have there been any negative effects of user fees on health care utilization? There is substantial evidence that utilization of public health services fell in response to user fees in sub-Saharan Africa, Indonesia and China (Gertler and Hammer 1997; Shaw 1996, Hsiao and Liu 1996). Several early multivariate analyses to measure the responsiveness of utilization of services to their prices were plagued by a variety of methodological problems and yielded estimates that varied widelyfrom -0.002 to -3.6 (Reddy and Vandermoortele 1996; Shaw 1996). However, methodologically more careful analyses for Indonesia and elsewhere do suggest that health care utilization can be quite responsive to user charges; that the poor were more responsive to price increases than the rich; that urban residents were more price-responsive than rural; that inpatient care was less price-responsive than outpatient care; and that the care for children was more price-responsive than that for adults (Gertler and Hammer 1997). The mere fact that utilization falls in response to user fees
An empirical question
While considering the appropriateness or otherwise of user fees and their design, two sets of issues must be distinguished.
266
SECTION IV
need not necessarily be worrisome, if people also care about quality. Findings from a study in the Cameroon suggest that utilization can improve when quality improvements accompany user charges (Gertler and Hammer 1997). This suggests that the problem may not be with user charges per se, but with the way they are implemented. Moreover, if declining use of public services were accompanied by simultaneous increase in the use of private services, it may lessen the adverse health impacts of user charges. This last argument will not apply to the poor who are unlikely to be able to afford reasonable quality private sector care. In China, the evidence seems to suggest that utilization and health levels fell considerably on account of the increasing reliance on user charges to fund services of public providers (Creese 1997; Hsiao and Liu 1996).
Efficiency and user fees

A second issue is whether the introduction of user charges leads to an improvement in efficiency. While direct evidence is not available, there are good empirical reasons to believe that these gains are unlikely to have been fully realized in many cases. In some countries, the necessary 'price signals' for efficient use of different tiers of health care were undercut by exemptions at higher-end facilities for special groups-government employees, doctors and their dependants, and others. This is an example where there might have been a trade-off between the objective of risk protection and the objective of efficiency. Inefficiencies have not been eliminated in some countries owing to a referral system that does not provide any special benefits to people being referred to a higher-level facility from lower-level facilities as compared with those who directly access the higher-level facility. In other countries, revenues raised from user fees were not used to provide additional supplementary resources, or were not retained at the facility level, but were directly transferred to the treasury. Thus, efficiency gains from complementarities between existing facility resources and additional revenues could not be exploited. However, the picture is not bleak everywhere. In Zaire (now the Democratic Republic of Congo), the implementation of user fees was found to have reduced the utilization of district hospitals as the place for a first visit, and increased utilization of PHCs. This may point to the efficiency gains referred to above. However, it is important to keep in mind that if suitable exemptions are not in place, the objectives of efficiency may conflict with those of equity. Moreover, child health may disproportionately be affected, because their utilization of health care facilities is more price-responsive as compared with that of other age groups.
may harm them. The higher the responsiveness of the poor to fees charged for services, the higher may be the potential harm to their health; and increased inequity may result, if the poor respond by reducing utilization more than the rich do. The early literature on the subject did not provide a conclusive answer to the question: Are the poor more responsive to prices (user charges) for health services than the rich? These early studies were plagued by a variety of methodological problems. In some country studies, price elasticities of demand for health care were found to be statistically indistinguishable from zero (Shaw 1996). However, later studies have been able to empirically establish that the poor are indeed more responsive to health service prices than the non-poor (Gertler and Hammer 1997). As shown by the Cameroon study, user charges may not negatively affect utilization by the poor if the imposition of user charges was simultaneously accompanied by quality improvements. Indeed, the poor were likely to respond more strongly to quality improvements than the non-poor. Equity may, however, be considered along other dimensions: child versus adult health care utilization; rural versus urban population; and across gender. As recent international studies suggest, the price-responsiveness of health care utilization does vary across these different groups. Addressing these inequities requires a somewhat more nuanced approach to the designing of user fees to take into account different elasticities of health services utilization. For instance, one may propose lower user fees for children. At the same time, one cannot immediately conclude that user fees for adults ought to be higher, especially for inpatient care, which could be very expensive. In this case, lower user fees serve the objective of risk protection, especially for the poor.
User Charges: The Indian experience and policy implications

There are hardly any studies that provide insights into the potential impact of user charges in an Indian setting. The only pertinent study in India is a demand analysis undertaken by Gupta and Dasgupta (2000) who used data from a nationally representative survey carried out by the National Council for Applied Economic Research (NCAER). The study found that, across the economic spectrum, the price elasticity of demand for outpatient health care was statistically indistinguishable from zero. If so, one might expect user charges to be not so harmful for the objective of raising revenue (since demand will not be affected much by price increases). The low price elasticity also suggests that price incentives for bringing about efficiency improvements are unlikely to work well because people will not change their utilization much in response to price changes. Finally, to the extent that the price elasticity of demand for outpatient care was close to zero at all income levels, the study's findings suggest that user fees may not have an adverse effect on equity of utilization. The above study suffers a number of methodological problems, perhaps the most significant being the absence of a quality indicator (Shaw 1996; Gupta and Dasgupta 2000).
Equity and user fees

User charges are considered especially problematic from the standpoint of equity. To the extent that most African countries do not have exemption policies for the poor, user fees
267
SECTION IV
Moreover, its use of the income variable as the indicator of economic status may be problematic to the extent that it may not properly reflect the household's earning potential (permanent income) and is likely to be under-reported, more so at higher income levels. One can learn about the impact of user fees on the various policy objectives from the actual experience of public sector health facilities that have imposed user fees in India. There are now a large number of facilities in India-mainly secondary and tertiary hospitals in the public sector-that have imposed user charges as part of the World Bank's health system development projects and other reforms instituted by states. Another feature of these projects is the extensive record-keeping of utilization of services at the various facilities where user charges were imposed. One can examine the utilization of services at a baseline date and compare it to a later date to assess the impact of user fees in the interim period. To address our questions about user charges, we now examine some of the information made available to us from the states of Andhra Pradesh and Maharashtra. It should nonetheless be emphasized that the 'before-after' studies like the ones we resort to now, which take little account of confounding elements in the price-utilization relationship must, in the language of Gertler and Hammer (1997), be treated with 'extreme caution'. To take account of at least one confounding element, we supplement our findings on utilization of services with supporting evidence on the impact of user fees on the quality of care.
medical staff is also available. In addition, APVVP hospitals also collect information on revenues from user charges and various stoppage charges (charges for rent of parking spaces, shopping spaces and the like, not required for health service provision), and the way these revenues were used by hospital societies. We used APVVP data to ask three questions: What is the magnitude of user charges in APVVP hospitals? Is there any association between user charges and quality of services provided? Is there any association between user charges and health care service utilization by the poor and the non-poor? Sixty hospitals were purposively selected from the World Bank-supported 159 hospitals in Andhra Pradesh, after stratifying by geographic region (23 from Andhra, 12 from Rayalseema and 25 from Telengana) and were considered representative of rural and district-level hospitals (Table 1). The data on user charges and stoppages, the annual budget registers and the departmental audited financial reports of the selected hospitals formed the basis for the financial data used in the study. The data are for three consecutive financial years from 2001-2002 to 2003-2004. Table 2 gives region-wise data on the magnitude of user charges, in absolute terms and as a proportion of total hospital expenditure for the sample of health studies. The evidence presented clearly points to the rising importance of user charges that have grown considerably in importance in absolute terms and as a percentage of total non-salary expenditure even over the short time period for which we were able to obtain data. Andhra Pradesh allows APVVP hospitals that levy user charges to retain the revenues. This raises two questions: Did the greater potential for user charge revenues influence budgetary allocations from the government? And did this translate into improvements in the quality of care provided at APVVP hospitals? The first question can be examined in two parts: The (overall) amounts allocated by the State Government to the APVVP as a share of its health budget; and the amount allocated to individual APVVP hospitals from the overall APVVP budget. Data for the latter were not readily accessible. As to the former, it is quite apparent from the data that the share of the
User fees in Andhra Pradesh Vaidya Vidhana Parishad (APVVP) hospitals

The Andhra Pradesh Vaidya Vidhana Parishad (APVVP) manages 228 public sector hospitals, categorized as district hospitals, area hospitals, community health centres, specialty hospitals and dispensaries. The APVVP regularly collects data on utilization of services from 159 hospitals supported by the World Bank, on inpatients and inpatient days, outpatient visits, diagnostics, surgeries and deliveries. This information is available separately for individuals living above or below the poverty line (BPL), based on their possession of cards identifying them as such, and by gender. Facility-level information on total bed capacity in each department, the number of doctors by specialty, nursing personnel and other para-
Table 1 Sample of hospitals selected from the various regions of Andhra Pradesh
Region District hospitals No. Sample Area hospitals No. Sample Community health centres No. Sample Specialty hospitals No. Sample No. Total Sample
Andhra Rayalseema Telengana Total
8 3 9 20
3 2 3 8
19 9 27 55
6 4 10 20
33 15 29 77
12 5 11 28
4 1 2 7
2 1 1 4
64 28 67 159
23 12 25 60
268
SECTION IV
APVVP in the total expenditure/budget of the State Government has declined in recent years-from 16.7% in 2001-02, to about 10% in 2003-4 (Mahal et al. 2003, authors' estimates using APVVP data]. Thus, user charges have become important for APVVP hospitals as a response to declining
Table 2 User fees in the samples of APVVP hospitals, by region, 2001-04

Region and expenditure 2001-02 2002-03 2003-04
Andhra User fee revenues (Rs in lakh) User fees/total expenditure (%) User fees/total non-salary expenditure (%) Rayalseema User fee revenues (Rs in lakh) User fees/total expenditure (%) User fees/total non-salary Expenditure (%) Telangana User fee revenues (Rs in lakh) User fees/total expenditure (%) User fees/total non-salary expenditure (%)
APVVP: Andhra Pradesh Vaidya Vidhana Parishad Source: Authors' estimates, using APVVP data.
36.52 2.10 15.50 11.50 1.08 9.85 43.85 2.22 18.11
62.00 3.15 21.56 35.72 3.31 26.18 86.71 3.79 26.33
82.13 4.18 35.36 44.21 3.39 37.75 106.20 4.47 38.16
revenue sources and not as an independent additional source of revenue. Despite these findings, it might still be useful to inquire whether the retention of revenues from user charges by the hospital societies in APVVP hospitals led to increased investments in the quality of services provided, owing to the increased flexibility with which such funds could be used. Table 3 presents findings on the aggregate utilization of funds generated from user charges and their trends over the years. It is immediately clear from the data that the utilization of APVVP funds has been extremely tardy, although it has been improving over time - user fee utilization rates were barely 43% in
Table 3 Proportion of user fee revenues utilized by the APVVP (trends, 2001-04)
State/region 2001-02 2002-03 2003-04
All Andhra Pradesh Andhra Rayalseema Telengana
42.5 82.8 27.9 12.8
53.3 90.5 52.9 26.9
72.7 93.5 80.1 53.5
APVVP: Andhra Pradesh Vaidya Vidhana Parishad Source: Authors' calculations, using APVVP data. We have assumed that the utilization rate for user charges is the same as the utilization rate for the total of user charges and stoppages since both are in the same bank account and under the control of the hospital committees attached to the hospitals.
2001-2, rising to about 74% in 2003-4. Moreover, the rates differ across regions - ranging from 53% in Telangana in 2003-4 to more than 90% in Andhra. Taken together, these data suggest not only a potential inefficiency in resource use, but also a geographical inequity in the way revenues from user fees were utilized. The precise reasons for these inter-regional and inter-temporal differences in utilization rates of revenues from user fees are unclear. Potential explanations could lie in indivisibilities in priority needs-equipment, large maintenance costs-or, they may be the result of dysfunctional hospital committees, and these are worthy of further investigation. The obvious explanation for the increase in utilization rates of revenues from user fees over time is the decline in government allocations to APVVP, potentially necessitating the use of revenues from user fees to make up the deficit, and maintain quality. Information on the utilization patterns of revenues from user fees for 2003-04 suggests that funds have mostly been used for activities that potentially contribute to increased quality of services-payments for contracted personnel (11%), drugs and consumables (14%), maintenance (13%) and electricity (21%). Whether these contributed to increased quality of services relative to the situation before the introduction of user fees, however, appears somewhat questionable in light of the corresponding declines in government allocations. Perhaps the best that can be said is that revenues from user fees helped maintain APVVP service quality in the face of declining contributions from State Governments, at least in the most recent years for which data are available. If all that revenues from user charges did was to maintain quality through filling in for the declines in support from the government, one might naturally expect utilization rates to fall in APVVP facilities on account of such charges. We do not have the actual number of poor in the 'catchment' areas of the sample hospitals in the three regions to calculate utilization rates. However, we do have information on the share of the poor (as indicated by identification cards issued by the government) in total utilization of the health services in APVVP facilities during the period 2001-02 to 2003-04 (Table 4). The data clearly point to the declining share of the poor in total utilization across a broad range of services provided at APVVP hospitals, particularly inpatient care services. This tendency was somewhat less marked in diagnostic services and laboratory tests. Information from the APVVP suggests that overall utilization of inpatient and outpatient care has increased over time. Between 2001-2 and 2003-4, for instance, total utilization at the APVVP facilities in our sample increased at an annual of 26 percent for inpatient stays, and by 19 percent for outpatient visits. However, the declining share of the poor during the same period (Table 4) meant that the utilization of these two types of services by the poor increased much more slowly - by 14 percent and 7 percent, respectively - and in the Rayalseema region, utilization by the poor actually declined over the same period.
269
SECTION IV
Table 4 Proportion of total utilization accounted for by the poor in Andhra Pradesh, by region and type of service, 2001-04
State/region Services 2001-02 2002-03 2003-04
% All Andhra Pradesh Inpatients Outpatients Surgeries Deliveries Laboratory tests Diagnostic tests Andhra Inpatients Outpatients Surgeries Deliveries Laboratory tests Diagnostic tests Rayalseema Inpatients Outpatients Surgeries Deliveries Laboratory tests Diagnostic tests Telangana Inpatients Outpatients Surgeries Deliveries Laboratory tests Diagnostic tests
92 83 82 74 85 64 90 80 72 65 83 72 97 92 84 72 95 65 89 79 95 85 77 56
79 75 79 62 79 62 81 81 75 66 75 73 82 71 63 48 92 63 75 74 95 67 69 52
65 68 74 53 78 63 71 81 67 56 73 67 58 57 56 44 90 66 67 64 92 56 69 56
Source: Authors' estimates using data from the Andhra Pradesh Vaidya Vidhana Parishad.
Implications of user fees in government hospitals in Maharashtra

In contrast to the APVVP, we were able to get much less detailed data for Maharashtra, another state where user fees were introduced in secondary hospitals as part of the reform process supported by the World Bank. User fees were sharply raised in Maharashtra in 1999 and 2001 (personal communication with Ravi Duggal; Duggal 2003). In fact, the average fee paid per patient in the 136 health facilities covered under the World Bank health systems project in Maharashtra more than doubled between 2000 and 2001, with the increase being particularly marked at higher-level facilities such as district hospitals and sub-district hospitals with 100 beds (personal communication with Ravi Duggal). We used the data on health facility utilization from the Department of Health, Maha-
rashtra to assess the impact of this increase on utilization, especially by the poor. Given the sharp increases in user fees between 2000 and 2001, one would expect that utilization by the poor would fall, or at any rate, increase more slowly than richer groups so that their share in overall utilization ought to decline, or else be the same. Table 5 provides us with information on utilization (and the proportion of total utilization accounted for by 'free care') for inpatient stays and outpatient visits in a sample of 55 health facilities-9 community health centres, 35 sub-district hospitals and 11 district hospitals-for the years 1999, 2000 and 2001. Only those facilities were included in the sample that had complete utilization data for the three years. The data suggest that, with one exception, overall utilization declined between 2000 and 2001 for outpatient visits and inpatient care in all four categories of facilities, and the share of the poor in total utilization mostly fell as well. Unfortunately, the recorded data on utilization by families below the poverty line families were incomplete and did not appear reliable. Instead, we used information on the proportion of users of care who obtained the care for 'free' as per the hospital records as a proxy for utilization by the poor. In general, not all 'free' users of public health facilities are poor. An ongoing review of utilization of public facilities in Maharashtra suggests that only about 40% of the 'free' users can be termed poor, with the rest being beneficiaries of various exemptions-government employees, freedom-fighters and the like (personal communication with Ravi Duggal). The use of data on the 'free' users of care can potentially bias our conclusions: for instance, if imposition of user charges is accompanied by better targeting of users in a way that improves health facility access to the poor, then imposition of user charges can be consistent with both improved utilization by the poor and with a decrease in 'free' users of care. However, we do not believe that the bias is a serious one in the case at hand since the user fee regime in Maharashtra itself dates back to before 2000, so any sorting on account of better identification of the poor is likely to have occurred before the hike in user charges in 2000, i.e. we believe that the composition of the poor among 'free users of care' is unlikely to have changed much in the period immediately before and after 2000. Why did utilization by the poor decline? Clearly, the rapid economic growth currently being experienced by Maharashtra and the consequently declining numbers of the poor are confounding elements. However, the anecdotal literature from Maharashtra and elsewhere offers an alternative, perhaps more compelling explanation. First, revenues from user fees in Maharashtra have remained largely unutilized and, therefore, not contributed to quality improvements even when retained by hospital committees at the facility level (Duggal 2003). The underutilization has partly been the result of government orders that have frozen these funds owing to fears of misappropriation. Interestingly, this freeze on fund use has left the collection of user fees unaffected, so that whereas the deterrent effect on utilization of user charges would have remained, it is unlikely that quality of care improved. Second, the exemption scheme for the poor may not have
270
SECTION IV
Table 5 Inpatient and outpatient care utilization in public health facilities in Maharashtra according to facility type and selected years
Facility Total 1999 Free % Total 2000 Free % Total 2001 Free %
Community health centres (CHC) Outpatient visits (000s) Inpatient stays(000s) Sub-district hospitals (50 beds) Outpatient visits (000s) Inpatient stays(000s) Sub-district hospitals (100 beds) Outpatient visits (000s) Inpatient stays(000s) District hospitals Outpatient visits (000s) Inpatient stays(000s)
303.0 28.2 681.5 48.1 714.0 94.1 1339.0 217.3
26.4 14.3 61.2 21.2 86.9 52.1 117.7 52.9
8.7 50.8 9.0 44.1 12.2 55.4 8.8 24.3
326.6 26.7 656.8 51.1 747.5 102.7 1389.6 221.3
28.6 14.7 61.2 23.2 87.0 53.7 130.1 50.4
8.8 55.2 9.3 45.4 11.6 52.3 9.4 22.8
321.7 27.7 565.8 41.8 726.6 112.5 1375.9 229.5
26.6 13.8 57.8 17.7 68.6 58.8 123.0 45.5
8.3 49.9 10.2 42.3 9.4 52.2 8.9 19.8
Note: Data provided by Maharashtra State Department of Health. 'Free' refers to stays or visits provided at no official charge. Data cover 11 district hospitals, 16 sub-district hospitals with 100 beds, 19 sub-district hospitals with 50 beds, and 9 community health centres. Only those hospitals that had a complete set of statistics for the years 1999-2001 were included.
worked as well as envisaged. There is, for instance, evidence from Punjab (another wealthy state with health reforms initiated with World Bank support) that the process for obtaining exemption cards was time-consuming and bureaucratic, making it virtually impossible for a poor person to obtain the benefits associated with such cards (Gupta 2002). Without quality improvements and exemptions, it seems reasonable to support the claim that utilization by the poor must have declined. Declining utilization of services in public health facilities need not be worrying from the standpoint of access to health, if the individuals shift to private sector facilities for health care of comparable quality. However, this argument is unlikely to hold for the poor, who may not be able to afford such care. The more likely outcome is either a shift to self-care or to lower quality providers.
Conclusion
Clearly, neither theory nor empirical analysis offers an openand-shut case on user charges. Provided quality improvements accompany user charges and there are exemptions for the poor or for groups such as children whose health care use is price-elastic, user fees can contribute to improvements in equity. When user fees can contribute to revenues that enable better usage of previously underutilized resources, or when they can be used to guide referrals to higher-end facilities, they can contribute to increased efficiency and quality as well. The optimal strategy on user fees, however, must consider three areas where user fees are especially problematic. The first is in the identification of beneficiaries. The second is their potential impact on the protection offered by public services against the financial risk associated with illness, mostly with
the need for inpatient care. The third is the utilization of funds collected from such fees. As for the identification of beneficiaries, there is some concern that existing methods for this purpose, which have focused on means testing, have not done well in India. Other approaches have also been tried or considered in different countries-by type of service used and by geographical region. All suffer from leakages in some form or the other, and tend to put a large administrative burden on health facility personnel (Gertler and Hammer 1997). For these reasons, a regime based purely on user fees is unlikely to work well. The above discussion also suggests a method of identification and exemption, which may be administratively less burdensome, and simultaneously addresses the second problem of 'insurance against catastrophic health risk'. In particular, some form of community or social insurance, whereby contributions of the poor are undertaken by the government/community may be the way to go. This removes the burden of identifying the poor from health facility personnel, and transfers it to a professional insuring group or communities, who may be able to do it better. An example is the use of village-level management committees (composed of village elders) in community-financing experiments in China who serve to both enroll people into schemes as well as help identify the poor (personal communication with William Hsiao, Harvard University). Of course, for this alternative scheme to work well, insurance must ideally be compulsory-voluntary participation in insurance can potentially lead to adverse selection and risk selection as a response-the original reason for the failure of the free market to provide insurance. On the other hand, some voluntary community-financing schemes have managed to do reasonably well in countries such as China (personal communication with William Hsiao, Harvard University).
271
SECTION IV
An insurance regime that pays health facilities for services provided the precise payment mechanism (capitation basis or other) can be readily combined with a user fee regime that offers many of the benefits discussed above. For instance, insurance may not reimburse expenses when the user visits a high-level facility prior to obtaining referral from a lowerlevel facility, lower co-payments for childhood conditions, and the like.
Acknowledgements
We are grateful to Mr Ravi Duggal of CEHAT for his thoughtful comments that helped improve our analysis, and to Ms Sujatha Rao, Member Secretary of the National Commission on Macroeconomics and Health for encouraging us to work on the study.
272
SECTION IV
References
Arhin-Tenkorang D. Mobilizing resources for health: The case for user fees revisited. Working paper no. #WG3:6. Geneva: World Health Organization, Commission on Macroeconomics and Health; 2000. Andrew C. User fees: They don't reduce costs and they increase inequity [editorial]. British Medical Journal 1997;315:202-3. Duggal R. Whither user charges. Express Healthcare Management 31 August 2003. Gertler P, Hammer J. Strategies for pricing publicly provided health services. Discussion paper. Washington, DC: The World Bank. Available from URL: http://www.worldbank.org/html/dec /Publications/Workpapers /WPS1700series/wps1762/wps1762.pdf 1997. [Accessed on November 25, 2004] Gupta V. World Bank funded health care: Reality or deception. Available from URL: http://www.sikhspectrum.com/062002/world_bank.htm; 2002. [Accessed on November 25, 2004] Gupta I, Dasgupta P. Demand for curative health care in rural India: Choosing between private, public and no care. Discussion paper #14/2000. New Delhi: Institute for Economic Growth; 2000. Hsiao W, Liu Y. Economic reforms and health: Lessons from China. New England Journal of Medicine 1996;335:430-2. Mahal A, Narayana K, Rao S. Expenditures and financing of the department of health, medical and family welfare in Andhra Pradesh: Towards a resource envelope for the period 2003-7. New Delhi: Department for International Development. Reddy S, Vandermoortele J. User financing of basic social services: A review of theoretical arguments and empirical evidence. UNICEF Staff Working Papers. New York: United Nations Children's Fund, Evaluation, Policy and Planning Series; 1996 Paul SR. User fees in sub-Saharan Africa: Aims, findings, policy implications. In: Paul SR, Ainsworth M (eds). Financing health services through user fees and insurance: Case studies from sub-Saharan Africa. Discussion paper number 294. Washington, DC: The World Bank, Africa Technical Department; 1996.
273
SECTION IV
Health insurance in India
OCIAL SECURITY FOR MEDICAL EMERGENCIES IS NOT NEW TO THE INDIAN ETHOS. It is a common practice for villagers to take a piruvu (a collection) to support a household with a sick patient. However, health insurance, as we know it today, was introduced only in 1912 when the first Insurance Act was passed (Devadasan 2004). The current version of the Insurance Act was introduced in 1938. Since then there was little change till 1972 when the insurance industry was nationalized and 107 private insurance companies were brought under the umbrella of the General Insurance Corporation (GIC). Private and foreign entrepreneurs were allowed to enter the market with the enactment of the Insurance Regulatory and Development Act (IRDA) in 1999. The penetration of health insurance in India has been low. It is estimated that only about 3% to 5% of Indians are covered under any form of health insurance. In terms of the market share, the size of the commercial insurance is barely 1% of the total health spending in the country. The Indian health insurance scenario is a mix of mandatory social health insurance (SHI), voluntary private health insurance and community-based health insurance (CBHI). Health insurance is thus really a minor player in the health ecosystem.
Social Health Insurance

Universal coverage has two dimensions: health care coverage (adequate health care) and population coverage (health care for all) and, coupled with the societal values that underpin it, leaves essentially two financing optionsgeneral taxation and SHI. The former implies financing care entirely from general revenue; its viability as the single mechanism to finance universal health coverage is necessarily limited in an environment of competing demands on a severely limited tax base. The SHI is based on income-determined contributions from mandatory membership of, in principal, the entire population with the government subsidizing the financially vulnerable sections. While the SHI is an effective risk-pooling mechanism that allocates services according to need and distributes the financial burden according to the ability to pay (thereby ensuring equity in access), such schemes are difficult and expensive to implement where a majority of the workforce is unemployed or employed in the informal sector.
International experience in SHI: Factors that affect the speed of transition

Achieving universal coverage through SHI is not easy. Evidence from 8 countries with SHI schemes for which sufficient information is readily availableAustria, Belgium, Costa Rica, Germany, Israel, Japan, Republic of Korea (ROK) and Luxembourg shows that the transition period (defined as the number of years between the first law related to health insurance and the latest law enacted to implement universal coverage) is 79 years (Austria), 118 years (Belgium), 20 years (Costa Rica), 127 years (Germany), 84 years (Israel), 36 years (Japan), 26 years (ROK) and 72 years (Luxembourg). These countries embarked on SHI when their economies were still underdeveloped; moreover, coverage is not necessarily a simple linear increase, as some groups are harder to reach than others. For example, moving from 25% to 50% coverage might take less time than moving from 50% to 75% (Carrin and James 2004). International experience suggests the following factors impacting the speed of transition to universal coverage using the SHI financing option: 1. The level of income and structure of the economy (specifically, the relative size of
K. SUJATHA RAO
SECRETARY, NATIONAL COMMISSION ON MACROECONOMICS AND HEALTH, GOVERNMENT OF INDIA, NEW DELHI E-MAIL: ksujatharao@hotmail.com
275
SECTION IV
Box 1 The Insurance Regulatory and Development Act (IRDA) 1999

The IRDA was passed in December 1999 by Parliament. The Act allows for the entry of private sector entities in the Indian insurance sector, including health insurance, and envisages the creation of a regulatory authority. The IRDA is supposed to protect the interests of the policyholders, promote efficiency in the conduct of insurance, regulate the rates and terms and conditions of the policies offered by insurers and direct the maintenance of solvency margins. The IRDA provides sufficient protection for capital and solvency margins. There is an entry requirement of a minimum capital of Rs 100 crore. Then there is a minimum lower bound of Rs 50 crore for the solvency margin along with a requirement of 20% of net premiums or 30% of the average of net incurred claims in the 3 preceding years. The IRDA has wide powers for accounting and auditing insurers. The Insurance Act does not allow the insurers to undertake additional business that is not directly linked to insurance. It discusses the liquidation of a company but does not talk of a Guarantee fund. The IRDA specifies a code of conduct for the insurance agents and also allows for a Tariff Advisory Committee to oversee premium rates, insurance plans and to prevent discrimination. However, there is no specific clause for the consumer, who has to use the CPA of 1986 to redress any complaints. The IRDA does not have much to say about the relationship between the insurer and the provider. Though the Tariff Advisory Committee can make recommendations the IRDA also does not have much to say about rating the premium. The IRDA does not also specify the benefit packages. It however allows for the entry of re-insurers in the market. Its main two functions are maintaining market standards, and overseeing solvency and financial regulations. Conclusion: The legislation concerning health insurance in India is fairly comprehensive even in comparison to a model set of regulations when focusing on auditing, financial controls, investment guidelines and licensing regulations. There is much less regulatory focus on the consumer of insurance products and the overall goals of health policy in the form of regulation that curbs risk selection, protects consumers, promotes HMOs, etc. It also cannot involve in the relationship between insurers and providers (which comes under the MRTP Act) or the expansion of ESIS (which is the ESIS Act). In India health insurance is not given much importance. The IRDA itself contains no reference whatsoever to the health sector or to health insurance. Nor is health mentioned in the nearly 175 pages of the Insurance Act of 1938. This broadly reflects the policy environment in India, where health insurance continues to be neglected. Even in GOIs report on Insurance reforms (1994), there was precisely one reference to health insurance.
Source: Mahal A. Assessing private health insurance in India: Potential impacts and regulatory issues. Economic and Political Weekly 2002:55971.
mine the capacity of SHI schemes to deliver the benefit package. 3. The administrative structure and solidarity in a country determine its ability to actually implement SHI and with legitimacy. In India, its large rural and informal sector accounting for 90% of the population, lack of cohesion and solidarity, and poor institutional capacity to organize them etc. will be constricting factors for the upscaling of the SHI in the near or medium term. The experience with collecting income tax predicts problems in assessing incomes and collecting premiums from small, unregistered firms, unorganized industries and the rural sector. The consumer redressal mechanism may also not function effectively because of the large illiterate population. The SHI is therefore likely to be restricted to the employed population and largely in urban areas, where collection of premium is easier and administrative costs minimal (Annexure). The existing mandatory health insurance schemes in India the Employees State Insurance Scheme (ESIS) and the Central Government Health Scheme (CGHS)were first started as pilot projects in 1948 and 1954, respectively in the context of achieving universal coverage via the SHI. Table 1 summarizes the provisions under these schemes.
Employees State Insurance Scheme (ESIS)

Enacted in 1948, the Employees State Insurance (ESI) Act was the first major legislation on social security in India. The scheme applies to power-using factories employing 10 persons or more, and non-power and other specified establishments employing 20 persons or more, with employees earnings up to Rs 7500 per month being covered, along with their dependants. The current coverage stands at 84 lakh employees and 353 lakh beneficiaries across 22 States and Union Territories (expectedly, the membership is higher for more industrialized States). The benefit package is quite comprehensive in its coverage of health-related expenses, going beyond the cost of medical care to include cash benefits (sickness, maternity, permanent disablement of self and dependant) as well as other benefits such as funeral expenses and rehabilitation allowance. However, the actual package of benefits available is determined more by the type of facility accessed rather than the type of cover. Medical care comprises outpatient care, hospitalization or specialist treatment as well as services of the Indian systems of medicines. These services are provided through a network of ESIS facilities, public care centres, non-governmental organizations (NGOs) and empanelled private practitioners. Corresponding to these arrangements, a variety of payment mechanisms are employed from salaries for ESIS staff to capitation fees for private doctors. The ESIS is financed by a three-way contribution from employers, employees and the State Government. Between 199394 and 199798, the income of the scheme grew substantially (largely due to increases in contributions which now account for 80% of the ESIS income) while medical benefits have actually fallen (from about 50% to less than 30% of the expenditures) and, as a result, the net excess transferred to the ESI fund went up
2.
the formal and informal sector) determine the feasibility of collecting contributions as well as the amounts that may be raised through SHI schemes. Distribution of the population and infrastructure deterFinancing and Delivery of Health Care Services in India
276
SECTION IV
Table 1 Key features of the Employees' State Insurance Scheme (ESIS) and Central Government Health Scheme (CGHS)
Mandatory social insurance schemes Indicators ESIS CGHS
Types of beneficiaries
Factory sector employees (and dependants) with income less than Rs 7500 per month
About 353 lakh beneficiaries in 1998 Medical and other health-related provided through ESIS facilities and partnerships Premiums (financing of scheme) 4.75% of employees' wages by employers; 1.75% of their wages by employees; 12.5% of the total expenses by the State Governments Provider payments Mainly salaries for physicians in dispensaries and referral hospitals. Hospitals have global budget financed by ESIC through State Governments. Administrative costs About 21% of the revenue expenditure. For paying wages for corporation employees, and administering cash benefits, revenue recovery and implementation in new area. Status of finances Contributions: more than 80% of the ESIS incomedouble the expenditure on benefits.
Employees' State Insurance Scheme (ESIS)
Coverage Types of benefits
Employees (and dependants) of Central Government-current and retired, some autonomous and semi-government organizations, Members of Parliament judges, freedom fighters, journalists About lakh beneficiaries in 1996 Medical care through public facilities and restricted private care Varies from Rs 15 to Rs 150 per month based on salaries of the employeesMainly financed by the Central Government funds Salaries for doctors. Treatment in private hospitals is reimbursed on case basis, subject to actual expenditure and prescribed ceilings Direct administrative costs including travel expenditure, office expenses, RRT 5% of the total expenditure. Part of salaries can also be charged to administrative costs. Contributions about 15% of the CGHS income-half of the salary expenditures.
from 14% to 30%. Significantly, the cost of administering the scheme has been steadily increasing as a proportion of expenditure on the revenue account.
Central Government Health Scheme (CGHS)

Established in 1954, the CGHS covers employees and retirees of the Central Government, and certain autonomous, semiautonomous and semi-government organizations. It also covers Members of Parliament, governors, accredited journalists and members of the general public in some specified areas. The families of the employees are also covered under the scheme. Total beneficiaries stand at 43 lakh (10.4 lakh card holders, 2003) across 24 cities with membership in Delhi being the highest. Benefits under the scheme include medical care at all levels and home visits/care as well as free medicines and diagnostic services. These services are provided through public facilities (including CGHS-exclusive allopathic, ayurvedic, homeopathic and unani dispensaries) with some specialized treatment (with reimbursement ceilings) being permissible at private facilities. Of the total expenditure, about a third is spent on wages and salaries of the CGHS staff (Table 2) and Figure 1. Table 2 highlights three important points: (i) that 18% of the health departments budget is spent on less than 0.5% of the population; (ii) that most of the expenditure is met by the Central Government as only 12% is the share of contributions.
If the scheme continues in its present form, and contributions stagnate at Rs 50 crore, the proportion of contribution will fall further to 5% of the total over the next five years, given the rising expenditures This calls for steps to ensure that contributions keep pace with expenditure, and perhaps even reduce the subsidy element; and (iii) The period 200104 also witnessed a sharp increase in inpatient expenditures. Coinciding
Fig 1 CGHS expenditure 1999-2004
Source: Ministry of Health and Family Welfare, GOI
277
SECTION IV
visits per card per year, varying between 5.6 visits in Bhopal to 28.4 visits in Bhubaneshwar. The approximate unit cost per visit comes Total expenditure on CGHS (Rs in crore) to a high of Rs 222 in 200203. Similarly inequitous is the payment structure for inpa1999-2000 2000-2001 2001-2002 2002-2003 2003-2004 tient care too. To assess the health-seeking behaviour Establishment 117.1125 123.8712 125.3384 133.1083 139.4496 and the trends towards utilization of health Supplies and materials 106.176 131.2345 165.3858 185.1242 222.9404 facilities after the CGHS opened up to over Professional services 47.8071 51.2002 65.7699 81.9203 140.7256 200 private hospitals for providing care at TOTAL CGHS 271.0956 306.3059 356.4941 400.1528 503.1156 pre-negotiated rates to their members, the TOTAL Department of Health 2132.46 2291.84 2577.04 2625.37 2800.64 NCMH took up a study of the CGHS pay% Share of CGHS 12.7 13.4 13.8 15.2 18.0 ments pertaining to the reimbursements to Total contributions 54.27 52.54 50.65 70.9 60.58 pensioners, hospitals and diagnostic centres. % of expenditure 20.2 17.15 14.21 17.71 12.04 A sample of 1000 claims were examined from Source: Demand for Grants, MOHFW the total bills paid by the Pay and Accounts Office of CGHS, Delhi, during 1999, 2003 with the sharp increase in the membership among retired perand 2004. For 2003 and 2004, all the payments made to sons, this indicates the trend towards adverse selection (Fig. pensioners in the randomly chosen successive months of June 2). and July were taken up for the study. Results of the claims showed an increasing number of cases using private sector facilities, which has budgetary implications for the GovernFig 2 ment, particularly in view of the absence of any regulations regarding prices and the large number of pensioners joining The increasing per person expenditure on the scheme (Table 4). outpatient and inpatient (2001-04) The 1999 sample (July to December) comprising of 104 reimbursement bills showed treatment being taken in government institutions in 58 of the cases. The ratio of the amount spent on government and private hospitals in 1999 was 1:1.25, or 4:5. These ratios changed in 200304 more adversely to government hospitals1:12 in the 2003 sample and 1:8.5 in 2004. Thus, over the 5-year period from 1999 to 2004, there was a sharp rise in the total number of bills, the total expenditure on professional services and payments made to private providers as a proportion of all payments, with government providers claiming just one-tenth of the total payment for provision of professional services in the 2004 sample.
Table 2
Private Health Insurance

Source: Ministry of Health and Family Welfare, GOI
Expenditures that cover outpatient treatment, including medicines for all serving and retired CGHS beneficiaries and inpatient/diagnostic services availed by retired beneficiaries, has thus grown between 12% and 25% per year over the past four years. A gross estimate suggests that another Rs 200 crore would have been incurred on inpatient treatment by serving employees. The maximum increase is seen to have occurred on professional services, i.e. reimbursement to pensioners and direct payments to hospitals and diagnostic centres. The CGHS is a high-cost enterprise with an inequitable spread of service delivery and no control systems for checking market failures such as moral hazard. As can be seen from Table 3, while each dispensary currently caters to an average of 3610 cardholders, varying between a low of 1073 cards per dispensary in Bhubaneshwar to a high of 6662 cards in Pune, the average OPD attendance during 200304 was 14.3 OPD
Since the liberalization of the insurance industry in 2000 India has been promoting private players to enter the health insurance sector. With the enactment of the IRDA, the industry now has a regulatory framework to protect the interests of policy holders. This was followed by another landmark decision in 2001 establishing Third Party Administrators (TPAs) to facilitate speedier expansion by providing an administrativeintermediary structure to the insurance industry. There are, at present, 12 general insurance companies and 25 TPAs. The total number of insurance holders is reported to be 112 lakh with almost 90% enrolled with the four public sector insurance companies. These four companies collected a premium of Rs 1128.64 crore under Mediclaim. Of the 102 lakh enrolled by these four companies (excluding GIC, Employment Guarantee Corporation, AICL), which are permitted to market health insurance products, Mediclaim alone accounts for 97 lakh persons, the rest being enrolled under other insurance
278
SECTION IV
Table 3 City-wise utilization (of allopathy) during 2003-04

No. of cards OPD attendance No. per dispensary Cards per dispensary OPD per dispensary
Ahmedabad Allahabad Bangalore Bhopal Bhubaneshwar Chandigarh Chennai Dehradun Delhi Guwahati Hyderabad Jabalpur Jaipur Kanpur Kolkata Lucknow Meerut Mumbai Nagpur Patna Pune Ranchi Shillong Thiruvananathapuram Total Average
OPD: outpatient department
6672 17794 61409 2627 2147 7762 48156 456468 9243 90262 19534 24504 27439 56426 20068 13626 91379 21274 13407 46631 2789 1771 6155 1047543
118764 279625 592042 14656 60927 103346 486342
106484 949448 227542 380177 529268
724995 508847 390151 70999 12004 98160 5653777
5 7 10 1 2 1 14 1 87 3 14 3 5 9 17 6 6 28 10 5 7 2 1 3 247
1334 2542 6140 2627 1073 7762 3439 5246 3081 6447 6511 4900 3048 3319 3344 2271 3263 2127 2681 6661 1394 1771 2051 83041 3610
Source: MOHFW, GOI
23752 39946 59204 14656 30463 103346 34738
35494 67817 75847 76035 58807
25892 50884 55735 35499 12004 32720 832847 46269
Note: Blank cells indicate data not available and have been excluded in the calculations
The question that arises is whether promoting the private commercial insurance sector will help India achieve its health objecHealth-seeking behaviour and trends towards utilization of tives of equity, efficiency and quality? What health facilities are its implications? Should India consider Government Private Private/ Private as other options, or is this a case of one size institutions institutions Government a % of the fitting all? International experience and eco(in Rs) (in Rs) ratio total nomic theory on private insurance markets however show evidence of widening inequity, 1999- Individual claims 733236 (58) 914897 (46) 1.25 55.5% excessive utilization, adverse selection, 2003- Individual claims 658083 (79) 2018361 (114) increase in inappropriate care, risk selection 2003- Hospital claims 1156281 (33) 16427031 (1425) increasing overall cost of care and in a highly 2003-Diagnostic provider claims 0 2900829 (1287) competitive, voluntary market, high admin2003 Total Paid 1814364 21346221 11.77 92.2% istrative costs, unviable risk pools, under2004- Individual claims 3281255 (305) 7277243 (332) cutting and unrealistic pricing leading to 2004- Hospital claims 1299264 (39) 29227474 (2072) market instability and bankruptcies. Private 2004-Diagnostic provider claims 0 2579414 commercial-led health insurance systems 2004 Total Paid 4580519 39084131 8.53 89.5% resulting in, etc.factors that contribute to Period selected for the study was June-July for 2003 and 2004, and July-December for 1999; Figures in parenthesis are number of cases inflation in costs. Yet of the 39(2001) counSource: NCMH analysis, 2004 tries having private insurance contributing schemes such as Jan Arogya, etc. During 20032004, the claim to 5% of the total health expenditure, 46% were low- and midratio was about 96.34%. The industry, however, believes that dle-income countries where private insurance is perceived as the overall claim ratio is expected to go up from around 130% an important source of health financing (Sikhri 2005), conto 300%350% in the next three years (Table 5). tributing to about 5%20% of the countrys total health spend-
Table 4
279
SECTION IV
Table 5 Premium and claim figures-Mediclaim (1999 to 2003-04)

1) National Insurance Co. Ltd
Year No. of policies issued Number covered Premium received No. of claims reported No. of claims settled Incurred claim amount (Rs in lakh) Incurred claim ratio (%)
1999-2000 2000--01 2001-02 2002-03 2003--04
572308 610571 897480 436273 505260
748508 803742 2497801 2025610 3122536
5210 5668 17614 22533 29802
48653 84392 213313 148963 198573
44760 77643 189595 140274 186110
3630 6045 14572 22037 30471
69.67 117.13 82.70 104.17 102.24
ing. Private insurance in these countries arose in response to increased expectations of affluent classes, covering the healthiest and the wealthiest resulting in limited social gain. Therefore, no country relies on private insurance to resolve the problems of financial risk protection for the poor and the ill. And regulation is required to minimize some of the adverse impacts.
The case of Chile

The USA and Chile are the two best examples of private health insurance. Chile, a middle-income country, consisting of 1.58 crore people, spends US$ 697 per capita (7.2% of the GDP) but has health outcomes that almost equal those of the USA. Table 6 gives a comparative statement of key indicators of India, Chile, China and the USA Chile developed its health system in three phases: the first till the 1980s was focused on reducing the burden of infectious and communicable diseases; the second during the 1980s when the National Health Fund was established to administer the SHI scheme (Fonasa) through a network of 194 hospitals run by the National Health Services System; and the third during the 1990s when health insurance was opened up to the private sector (Isapres). As of date, 67% of the population is enrolled with Fonasa, while 20% are covered under 40,000 private plans with 18 licensed, private Isapres. Insurance is mandatory and all have to pay 7% of their wages for health insurance. Both schemes are regulated by the Superintendence of Isapres, under the Ministry of Health (Government of Chile). Under the Fonasa, care is provided through its own public hospital network and for enhanced contributions, accredited network of private hospitals based on a fixed price reimbursement for specific ambulatory and inpatient medical services. Seventy-five per cent of the Fonasa budget is released to primary health centres that are obliged to provide a predefined package of health services. Isapres, on the other hand, offers a myriad and individually customized, risk-rated premium plans based on the age, health and economic status. They function on a feefor-service basis. The Isapres have the freedom to fix the premium, indicate the content and coverage levels, degree of co-payment and set the limits for reimbursements. Regulation is only on contractual compli-
2) New India Assurance Co. Ltd

(Rs in lakh) Year No. of policies issued Number covered Premium received (Rs) No. of claims reported No. of claims settled Incurred claim amount Incurred claim ratio (%)
1999-2000 2000-01 2001-02 2002-03 2003-04
489150 609255 822534 937012 949648
2163876 2951010 2794510 3086763 2856675
16165 23915 26996 35443 36641
108247 275774 165368 201108 167898
90573 305406 116819 196300 161959
15629.37 20349.96 18853.00 31053.00 30068.12
96.68 85.09 69.84` 87.61 82.06
3) Oriental Insurance Co. Ltd

(Rs in lakh) Year No. of policies issued 1999-00 269288 1077151 Number covered Premium received (Rs) 7450 No. of claims reported 12220 No. of claims settled 11556 Incurred claim amount 6570 Incurred claim ratio (%) 87.13
2000-01
2001-02
376878
502512
1507512
2010047
10553
15075
16386
63166
15420
53617
8870
14188
84.06
94.11
2002-03
2003-04
537061
555858
2148247
2223436
20408
22953
74620
83050
64251
71907
15754
22407
77.19
97.62
4) United India Insurance Co. Ltd

(Rs in lakh) Year No. of policies issued Number covered Premium received (Rs) No. of claims reported No. of claims settled Incurred claim amount Incurred claim ratio (%)
1999-2000 2000-01 2001-02 2002-03 2003-04
322845 105331 140441 245000 305000
904594 361600 482133 772000 845000
9124 11761 14518 21569 23528
60120 32452 30130 40000 50500
54077 27759 25626 37889 42585
7620 8850 15819 22317 25018
83.52 75.25 108.96 103.46 101.92

(Rs in lakh)
GIPSA Companies
Year No. of policies issued Number covered Premium received No. of claims reported No. of claims settled Incurred claim amount Incurred claim ratio (%)
1999-2000 2000-01 2001-02 2002-03 2003-04
1653600 1702035 2362967 2155348 2315768
3924693 5623864 7575427 7885465 9047647
38040 51897 94400 102600 112900
229240 409004 471977 464691 500021
200968 426228 385657 438714 462561
33448 44114 64800 91160 106400
88 85 69 90 94
Source: Department of Insurance, Ministry of Finance, GOI
280
SECTION IV
Table 6 Key indicators of India, Chile, China and the USA

Indicator India Chile
Population in crore Health expenditure per capita (in US$ Public expenditure. on health* as % of GDP/2000 IMR per 1000 live-births 2001 Life expectancy at birth Maternal mortality ratio per 100,000 lakh births*
Source: World Health Report, 2005, WHO, * MDG-UNDP 2002 ,
103.34 96 0.9 68 62 540
1.54 642 3.1 10 77 23
ance but not on the content of the policies. To safeguard the stability of the insurance pool, Isapres are known to follow rigorous procedures of screening out high risks, provide low coverage for high-cost illnesses and expensive procedures, discriminate against or even terminate subscribers with highcost chronic diseases by increasing the premium or contract conditions, forcing the subscriber to opt out. Typically, therefore, the Isapres enrollees have a mean income that is four times more than those enrolled in Fonasa; 70% of the beneficiaries are in the age group of 1564 years with 2.5% above 65 years compared to 62% and 10%, respectively in Fonasa. Such tiered rating is inevitable so as to keep the premium low enough to retain the young healthy subscribers. As in India, Chile has the problem of inappropriate skill mix in public hospitals, not in keeping with the changed epidemiology; the centralized budgeting system giving little discretion, salaried system of provider payments with no incentives to improve efficiencies, resulting in 50% bed occupancy in peripheral hospitals and overcrowded city hospitals. Second, with private sector allowed to provide the same set of services, there is duplication of infrastructure and resultant wastage in the system as a whole. However, since the quality of care is similar in public hospitals, despite a law, 12% of Isapres beneficiaries were found to have availed free care in public hospitals. In other words, the system induces the subscriber to avail ambulatory care in Isapres and move to Fonasa when sick. Besides, due to the short-term character of the contracts and ability to offload patients when ill, the Isapres have no incentive of providing preventive care. This dual system has thus resulted in segmenting the population on the basis of income and risk. With the freedom to fix premiums, the risk-rating system has resulted in a systematic discrimination against fertile women, chronically ill and the elderly through the three stratagems of higher premiums, reduced benefits and refusal to enrol or renew contracts. The lack of uniformity or transparency of insurance plans makes it easier to resort to such tactics. The effect of such a system is seen in the disproportionate share of high-risk persons being discharged onto the public hospitals: HIV/AIDS (82%), cervix cancer (90%), kidney failure (83%) and leukaemia (80%). The system encompasses all the incentives for increased cost of care: fee for service as a basis of provider payment. With mandatory insurance, the competition is on quality, based
on sophisticated technology, which may not always be cost-effective and also puts pressure on the public system to keep pace. Thus, competition is on offering high-technology USA China clinical procedures to low-risk individuals. Third, the need for spending substantial 28.8 128.52 amounts on screening out high-risk patients. 5274 261 Such risk-rated premiums also affect the old 5.8 1.9 or those who fall sick as their option to change 2 31 the insurers is only Fonasa, as no other Isapres 77 71 will accept a high-risk enrollee. The admin8 55 istrative costs of Isapres are 14% and escalation of average fee per visit is 80%, higher than that of Fonasa, which are 1.2% and 50%, respectively. The cost of the whole system is high as despite mandatory payment of 7% of the wage, the out-ofpocket expenditures account for another 35% of the total health spending. Finally, in the event of insolvency or mergers between one Isapres and another, the interests of the enrollee are not protected. In 2002, Chile launched a major health reform process. The key features consist of mobilizing additional resources by earmarking 1% of the value-added tax (VAT) for health; accreditation of facilities and providers in the public and private sector; standardized benefit packages for delivery by Fonasa and Isapres guaranteeing access, opportunity, quality and financial protection; ensuring stability of enrollee interests in case of insolvency of a private insurer; and regulations for preventing risk discrimination and dumping of high-risk enrollees. Of importance is the Standard Benefit Package: access is guaranteed by entitling enrollees to receive care listed in the package at the appropriate level and within reasonable distance; opportunity implies defining a maximum wait period for each service, with the option to get the service from any place of choice to be reimbursed by the plan; quality is ensured by service provisioning by accredited members; and financial protection ensures that none are denied care for want of ability to pay and a ceiling of co-payment to be 20% or not exceeding a patients 2 months wage. For implementing these reforms, organizational and financial restructuring have also been designed with laws protecting enrollee interests and providing for a solidarity compensation fund to compensate private insurers for the enrolment of high-risk persons.
Current status of private health insurance in India

India has lessons to learn from the experience of Chile. India too has a dual system of carea private fee-for-service based sector where the money is paid out-of-pocket by individual households and a tax-based public sector where the providers are salaried. Utilization of insurance under both these systems is partly restricted and rationed by the affordability of the individual household and availability of the budget. On the other hand, insurance as a means of financing is a far more sophisticated mechanism, requiring a comprehensive understanding of the failures that characterize health insurance markets. For example, a problem such as asymmetry in
281
SECTION IV
information puts the patient and the insurer at a disadvantage due to their inability to resist or challenge medical opinion regarding an existing condition or future treatment. Besides, in the absence of knowledge of prices, the provider can shortchange the two by overcharging. Second, cashless insurance creates disincentives to control costs as it appears to be a free good for the patient and the provider, often resulting in excessive treatment by the provider (induced demand) and frivolous use by the patient taking treatment even for a condition which he would normally have ignored or cured with a home remedy (moral hazard). Third, it is only the patients who know their health status. Since it is normally those in need of health care who tend to subscribe to health insurance, this puts the risk on insurance agencies to resort to extensive processes of risk selection, such as medical examination, before being given admittance as an enrollee and focusing on lowrisk groups, such as the young or healthy. Risk selection in individual-based policies however results in increasing the loading fee and consequently the cost of premium. This is one reason for the attractive group discounts being as high as 67%. For these reasons, private commercial health insurance is known to select its customersthe young, healthy, rich, malesleaving the bad risks to the governmentold, poor, young women in the reproductive age group, and the ill. Health insurance in India is usually associated with the Mediclaim policy of the GIC, which was introduced in 1986 as a voluntary health insurance scheme offered by the public sector. The premium based on the age, risk and the benefit package opted for, ranged from a minimum premium of Rs 201 for those <25 years of age, to a maximum benefit of Rs 15,000 with discounts for group memberships. In 2001, there were 78 lakh persons covered under Mediclaim (Gupta 2003). The subscribers are usually from the middle and upper class, especially since there is a tax benefit in subscribing to Mediclaim. The standard Mediclaim policy covers only hospital care and domiciliary hospitalization benefits. Most medical conditions are reimbursed though there are important exclusions, such as pre-existing diseases, pregnancy and child birth, HIV/AIDS, etc. Hospitals with more than 15 beds and registered with a local authority can be identified as providers. The insurance company (or the TPA, where applicable) administers the scheme. Being an indemnity scheme, the patient pays the hospital bills and submits the necessary documents to the company. The company in turn reimburses the patient. A study of 621 GIC claims for the year 199899 by Bhat and Reuben (2001) showed that the average time between submission of documents and reimbursement is 121 days. This study also showed that one-third of the claims were due to adverse selection; 38% pertained to doctors fees and 25% charges for diagnostic services. The provider-induced claims thus accounted for 63%. Yet another interesting insight was that 22% of the total claims were for the treatment of communicable diseases, while 64% were for non-communicable diseases. There is also uncertainty about the amount reimbursed, there are times when the patient is reimbursed only partially, the usual reason being the insufficiency of documentation. The policy is not renewed automatically and is dependent on
the timely payment of premium. Ellis et al. observed that the GIC was more interested in whether the claim pertained to an existing disease or whether the facility was qualified or not, but spent little time on detecting fraud. With claims exceeding 30% a year, more than the household spending, it reflects the problem of moral hazard which requires close monitoring. Second, it was also observed that the GIC sets premium on the filing of claims and not actual amounts settled, giving it a cushion year on year as settled claims amounts are always lower than those filed, an amount that remains unadjusted. During 1994, 4.4% of the insured persons made a claim, of which only 75% of claims were settled. The claims ratio was 45%. However, of late, the claims ratio is growing at a fast rate, allegedly because of collusion between the patients, insurance agents and hospitals. From the above discussion, five features that characterize the health insurance system in India emerge: 1. By and large, the system offers traditional indemnity, under which the insured first pay the amount and then seek reimbursement. Under indemnity, all known diseases or health conditions are excluded and therefore such policies typically have a large number of exclusions. This also means that those most in need of insurance, i.e. the sick, get excluded for any financial risk protection against the diseases they are suffering from. 2. It is a fee-for-service-based payment system. Such a system of payment is advantageous for the provider since he bears no risk for the prices he can charge for services rendered by him. Combined with the asymmetry in information, such a system usually entails increased costs. 3. Policies provide a ceiling of the assured sum. Such a system, and that too within a fee-for-service payment system, results in shortchanging the insured as he gets less value for money, as the provider and the insurer have no obligations to provide quality care and/or over provide/over charge services so long as the amounts are within the assured amount of the insurance policy. 4. The system is based on risk-rated premiums. This again puts the risk on the insured as the premium is fixed in accordance with the health status and age. Under such a system, women in the reproductive age group, the old, the poor and the ill get to pay higher amounts and are discriminated against. 5. The system is voluntary, making it difficult to form viable risk pools for keeping premiums low.
Reasons for poor penetration of health insurance

Penetration of health insurance has been slow and halting, despite the huge market estimated to range between Rs 7.520 crores. Some reasons that explain for the slow expansion of health insurance in the country are as follows:
1. Lack of regulations and control on provider behaviour

The unregulated environment and a near total absence of any form of control over providers regarding quality, cost or
282
SECTION IV
data-sharing, makes it difficult for proper underwriting and actuarial premium setting. This puts the entire risk on the insurer as there could be the problems of moral hazard and induced demand. Most insurance companies are therefore wary about selling health insurance as they do not have the data, the expertise and the power to regulate the providers. Weak monitoring systems for checking fraud or manipulation by clients and providers, add to the problem.
eases that are expensive to treat and can be catastrophic. If we take the number of beds as a proxy for availability of institutional care, the variance is high with Kerala having 26 beds per 1000 population compared with 2.5 in Madhya Pradesh.
6. Co-variate risks
High prevalence levels of risks that could affect a majority of the people at the same time could make the enterprise unviable as there would be no gains in forming large pools. The result could be higher premiums. In India this is an important factor due to the large load of communicable diseases. A study of claims (Bhat 2002) found that 22% of total claims were for communicable diseases.
2. Unaffordable premiums and high claim ratios

Increased use of services and high claim ratios only result in higher premiums. The insurance agencies in the face of poor information also tend to overestimate the risk and fix high premiums. Besides, the administrative costs are also high over 30%, i.e. 15% commission to agent; 5.5% administrative fee to TPA; own administrative cost 20%, etc. Patients also experience problems in getting their reimbursements including long delays to partial reimbursements.
Third party administrators

With the entry of TPAs under the IRDA Regulations Act, 2001, the insurance industry is taking a new turn towards Managed Care. The TPAs are required to be registered under the Companies Act, 1956, and licensed by the IRDA, and be contracted by one or several insurance companies for the provision of health services. The original role of a TPA was to provide the back-office administrative set-up to insurance companiesissuing ID cards to subscribers, processing claims, making payments, etc. Taking advantage of the lack of clarity on the specific role and responsibilities of TPAs, some among them are rapidly developing capacity to establish provider networks to service the needs of the insured, collecting and analysing data, fixing and negotiating rates for procedures with providers, contracting providers, processing claims and making direct payment to them and arbitrating any dispute between the subscriber and the provider. This system, often referred to as cashless payment, has resulted in relieving the patients of the psychological stress of having to mobilize resources at short notice. By scrutinizing provider claims, TPAs also help in safeguarding the interests of the insuring company of any fraudulent claims by the providers. For all these services, the insurance companies pay 5.5% of the total amount of premium collected under the policy. In addition, TPAs were also to be given a bonus from insurance companies for reduced claim ratios or for promoting the companies with the insurers. This then would have given them the financial incentives to develop systems for provider control: contracting through predetermined rates for procedures and treatment, utilizations reviews, prior authorization for expensive surgeries, etc. and also ensuring that the patients do not resort to frivolous use of the services. However, with the administrative fee being low and the idea of bonus not operationalized, there is really no incentive for the TPAs to reduce the claim ratios. Secondly, barring a few, for most TPAs health insurance itself is a secondary concern to their main activity of brokerage. The system of TPAs has facilitated cashless payments and expanded access to providers but is yet to show evidence of having been able to control cost or provide appropriate care. As the system of TPAs unfolds there are apprehensions: (i) whether patients will get adequate treatment and appropriFinancing and Delivery of Health Care Services in India
3. Reluctance of the health insurance companies to promote their products and lack of innovation
Apart from high claim ratios, the non-exclusivity of health insurance as a product is another reason. In India, an insurance company cannot sell non-life as well as life insurance products. Since insurance against fire or natural disaster or theft is far more profitable, insurance companies tend to compete by adding low incentive such as premium health insurance products to important clients, cross-subsidizing the resultant losses. With a view to get the non-life accounts, insurance companies tend to provide health insurance cover at unviable premiums. Thus, there is total lack of any effort to promote health insurance through campaigns regarding the benefits of health insurance and lack of innovation to make the policies suitable to the needs of the people.
4. Too many exclusions and administrative procedures

Apart from delays in settlement of claims, non-transparent procedures make it difficult for the insured to know about their entitlements, because of which the insurer is able to, on one stratagem or the other, reduce the claim amount, thus demotivating the insured and deepening mistrust. The benefit package also needs to be modified to suit the needs of the insured. Exclusions go against the logic of covering health risks, though, there can be a system where the existing conditions can be excluded for a time periodone or two years but not forever. Besides, the system entail equity implications.
5. Inadequate supply of services

There is an acute shortage of supply of services in rural areas. Not only is there non-availability of hospitals for simple surgeries, but several parts of the country have barely one or two hospitals with specialist services. Many centres have no cardiologists or orthopaedicians for several non-communicable dis-
283
SECTION IV
ate care; (ii) whether quality of treatment will be compromised with the gradual loss of control and autonomy of the physician on the kind of treatment to be given to his/her patient; (iii) whether costs will go up due to the substantial administrative responsibilities placed on the providers for record maintenance, filling claim and billing formsin USA, where health insurance is organized on a TPA system, doctors spend almost 30% of their time on processing their claims and the administrative costs of the system are 25%30% as compared to 3% in Canada.; (iv) will there be possibilities of collusion between the TPA and some providers in the network, resulting in processing their higher claims even if not justified, affecting the interests of the insurance agency; (v) with TPAs getting organized over time, whether they may acquire monopoly control over the processes and dictate higher administrative fees, since in the current system the TPA bears no risk; and finally (vi) the legal uncertainty of the future in view of the framework regarding the functioning of TPAs being ambiguous and unclear. For example, the IRDA does not supervise or regulate the financial activities of TPAs, the contractual relationships with providers or relationships with the corporate or union health plans. In the light of such limited regulatory oversight, some already combine subscription plans being serviced by a provider network without involving any insurance company such as for example, the Karnataka Police insurance with Apollo-sponsored TPA which provides hospital services in a network of some 35 hospitals. In the absence of any statutory control or obligations imposed by the IRDA, such as networking only accredited providers, or those adhering to certain quality benchmarks, or submitting reports on the qualifications of the provider and performance reports, etc., there is a major lacuna, making it difficult to ensure appropriate accountability in the system. Overall international literature1 does show that the TPA system is expensive (personal communication with Professor ao, Harvard School of Public Health); even when their role is confined to payment of benefits and management of claims, the administrative costs run up to 20%30%. If they are assigned the role of identifying providers then the amount can go even higher to 45%, making insurance products very unaffordable. Besides, such literature also seems to suggest that the TPAs neither have any motivation to undertake the stewardship function to protect consumer interest nor enroll new persons. In this context NGOs could be better agents. Given the complexities of these markets, the key lesson for India is to closely study behavioural responses that such financing systems generate among all the major players and institute appropriate regulatory systems to minimize likely distortions.
Universal Health Insurance Scheme (UHIS)

For providing financial risk protection to the poor, the Government announced a UHIS in 2003. Under this scheme, for a premium of Rs 365 per year per person, Rs 548 for a family of five and Rs 730 for a family of seven, health care for an assured sum of Rs 30,000 was provided. BPL families were given a premium subsidy of Rs 200 per annum. The scheme was redesigned in May 2004 with higher subsidy and restrict-
ing eligibility to BPL families only. The subsidy was increased to Rs 200, Rs 300 and Rs 400 to individuals, families of five and seven, respectively. To make the scheme more saleable, the insurance companies provided for a floater clause that made any member of the family eligible as against the Mediclaim Policy which is for an individual member. Yet in the last two years of its implementation the coverage has been around 10,000 BPL families in the first year and 34,000 in the second year till 31 January 2005. The reasons for failing to attract the rural poor are many. First, the public sector companies who were required to implement this scheme find it to be potentially loss-making and do not invest in propagating it, resulting in very low levels of awareness, reflected in the low enrollment and very poor claim ratios. To meet the targets, it is learnt that several field officers pay up the premium under fictitious names. Second, a major problem has been the identification of the eligible families. Identification became cumbersome as the family needed to have some form of certification, which is difficult to obtain from revenue authorities. Besides, the poor also find it difficult to pay the entire premium money at one time for a future benefit, foregoing current consumption needs. Third, the procedures are cumbersome and difficult for the poorthe premium has to be paid in a lump sum; the paperwork required for enrolment as well as getting claim amounts is very timeconsuming. Fourth, in most places there is a deficit in the supply or availability of service providers, particularly because government hospitals are not eligible. For example, in Uttaranchal, only 17 hospitals could be accredited under this scheme, which could have gone up to 37 if government hospitals were allowed to be included and also expanded access and choice to the enrollees. Besides, in several areas there are just no doctors available. Fifth, there was a set-back due to health insurance companies refusing to renew the previous years policies. Finally, the TPAs are also not willing to implement this scheme at 5.5% of premium amount as their administrative costs of covering rural populations in dispersed villages makes it unviable. During 2004, the Government also provided an insurance product under which for a premium of Rs 120 the sum assured was Rs 10,000. This was, to be available only for self-help groups (SHG). However, the intake is reportedly negligible. The reasons for this poor intake are similar to those cited above. With the Common Minimum Programme (CMP) committed to having a UHIS, there has been much effort and debate to evolve a suitable and sustainable design. To expand the health insurance business, recommendations are also being made to reduce the minimum pre-qualification of Rs 100 crore equity as it will require 15 years to break even. Another set of recommendations is for permitting TPAs and hospitals to introduce health insurance products. There are, however, doubts regarding this model as it may promote conflict of interest. In combining various aspects of provisioning and insuring there could be perverse interests to provide low quality of care over-diagnose or under-treatfor making profits.
284
SECTION IV
Community-Based Health Insurance

Community financing (CF) as a method of raising finance at the community level was initiated by UNICEF under its Bamako Initiative for Africa in 1987. The initiative had the following objectives: (i) to revitalize public health systems; (ii) to decentralize decision-making; (iii) to mobilize resources to cover local operating costs; (iv) to encourage community participation through management of services and locally generated funds; and (v) to define the minimum package of essential health services. (UNICEF 1987). Though the experiment failed in Africa, its concepts are once again gaining recognition as an appropriate strategy for low-income countries which have a weak resource base, poorly developed markets and a vast population having very low threshold of payment capacity (WHO 2002). In community financing, the community is in control of the principal functions of collection and utilization, the membership of the scheme is voluntary and there is willingness to prepay the contributions (Hsiao 2001). The scheme is based on the hypothesis that with greater social capital there will be more willingness to pay and participate. The community has been defined as a group of households living in close proximity or belonging to social, religious or economic organization. The efficacy of the scheme is based on two implicit principles: one, that the community has adequate homogeneity or social coherence that gets easily translated into a capacity to mobilize resources; and two, that the willingness to prepay will be influenced by self-interest when each individual perceives his marginal benefit exceeding his costs, i.e. accessing something of value which can be obtained easily and more in quality through prepayment. Literature reviews of such community-based schemes tend to suggest that they have enabled an increase in the availability of resources; inclusion of the poorest groups on account of government subsidy; enhanced access to health services; and reduced impoverishment on grounds of illness (Jakab 2001). While the CBHI movement is vibrant in Africa, it is slowly picking up momentum in India. Currently, there are about 22 voluntary CBHI programmes in India, initiated and administered by NGOs. Of these about 10 are active (Table 7). In many schemes, the community is also involved in various activities such as creating awareness, collecting premiums, processing claims and reimbursements, and the management of the scheme (deciding the benefit package, the premiums, etc). Devadasan, in his paper identified broadly three types of community health insurance (CHI) schemes and also analyzed their structure and basic features as discussed below: (Devadasan et al. 2004; Fig. 3): Type IThe provider of health care plays the dual role of providing care and running the insurance programme (e.g ACCORD, VHS) Type IIwhere a voluntary organization/NGO is the insurer, while purchasing care from independent providers (e.g. Tribhuvandas Foundation, DHAN Foundation) Type III(intermediary design)The NGO plays the role of the agent purchasing care from providers and insurance
companies (TPA, e.g. SEWA, Karuna Trust, BAIF). The membership of these CHIs scheme varies from 1000 to more than 20 lakh. Most of the schemes operate in rural areas and cover people from the informal sector. Enrolment is usually facilitated by membership of the organizations, e.g. micro finance groups, cooperatives, trade unions, etc. The annual premium ranges from Rs 20 to Rs 120 per individual. The unit of enrolment is an individual and the membership is voluntary in most of the schemes. All the schemes offer hospitalization; this ranges from the classical Mediclaim product to a very comprehensive cover including all conditions and no exclusions. Many NGOs have been successful in negotiating an appropriate insurance package for their members. Most providers are either NGOs or private for-profit organization. The utilization rates range from 6 to more than 240 per 1000 persons insured. The latter obviously indicates extreme adverse selection. The main strengths of the CBHIs schemes are that they have been able to reach out to the weaker sections and provide some form of health security; increase access to health care; protect the households from catastrophic health expenditures and consequent impoverishment or indebtedness. However, sustainability is an issue as these initiatives are dependent on government subsidy or donor assistance. They provide limited protection in view of the very little crosssubsidy between the rich and the poor, resulting in the small size of the revenue pool which also constricts getting a better bargain from the providers. A disturbing factor in these programmes, (barring one or two) is the very low claim ratio, ranging from 0.25 to 0.66, which indicates that the scheme is not able to overcome the barriers that are hindering access or the cover provided is too inadequate or the members too ignorant about their entitlements. It is also seen that the poorest of the poor get excluded on account of their inability to pay their share within the specified time limit. Some NGOs manage the scheme by themselves, which may be illegal within the current IRDA regulations. Also, some of the schemes cover very small numbers and so the potential for scaling-up is restricted. Moreover, many of the schemes see health insurance as an end in itself and do not seek to either promote preventive and promotive health care or extend adequate provider linkages. There is no exhaustive evaluation of the CBHI schemes in India due to the lack of uniformity in MIS. Many questions remain unanswered and need to be researched to see if these models can be implemented and replicated in India. For example, it is not clear how much it costs to administer such schemes, or its impact on strategic purchasing of services, developing provider networks or on the local quack, or the problems for upscaling and finally if the scheme has helped protect the poor from penury and if so, how it can be sustained if NGOs withdraw their support, etc. Of all the schemes in operation, the one that has drawn widespread attention in India is the Yeshaswani, an insurance scheme for farmers, designed and implemented by the Government of Karnataka since 2002. Under this scheme, the Cooperative Department enrolled, through a governFinancing and Delivery of Health Care Services in India
285
SECTION IV
Table 7 Some community health insurance schemes in India

Name and location of the scheme Population covered (target population in 2003) Premium collected (per cent target population covered in 2003) Benefit package
ACCORDGudalur, Nilgiris, Tamil Nadu BAIFUrali Kanchan, Pune, Maharashtra BUCCSBuldhana, Maharashtra DHAN Foundation Kadamalai taluk, Theni District, Tamil Nadu Karuna TrustT Narsipur Block, Mysore District, Karnataka MGIMS HospitalWardha, Maharashtra Raigarh Ambikapur Health Association (RAHA) Raigarh, Chhattisgarh SEWAAhmedabad, Gujarat SHADEKolencherry, Kerala Student's Health HomeKolkata, West Bengal Voluntary Health Services Chennai, Tamil Nadu YeshasviniBangalore, Karnataka
Tribals living in Gudalur taluk and who are members of the AMS union (n =13,000) Women members (between 18 and 58 years) of the micro savings scheme in 22 villages (n =1500). Members of the Buldhana Urban Cooperative and Credit society (n = 175,000). Women members of the micro finance scheme and living in Mayiladumparai block (n =19049) BPL families in T Narsipur Block (n = 278,156) The small farmers and landless labourers living in the 40 villages around Kasturba Hospital (n= 30,000) Poor people living in the catchment area of the 92 rural health centres and hostel students. (n = 92,000 individuals). SEWA Union women members (urban and rural), and their husbands living in 11 Districts of Gujarat (n = 1,067,348) Members of the SHGs operating in Ernakulam district (n = 9000) Full-time student in West Bengal State, from Class 5 to University level. (n =56 lakh students) Total population of the catchment area of 14 mini-health centres (n= 104,247) Members of the District Farmer's cooperative societies and their families (n = 80 lakh)
Rs 25 per person per year (36%) Rs 105 per person per year (58%) NA
Hospitalization cover up to Rs 1500 per person per year Hospitalization cover up to Rs 5000 per person per year Hospitalization cover up to Rs 5000 per person per year Hospitalization cover up to Rs 10,000 per person per year Hospitalization cover up to Rs 2500 per person per year. Includes ambulance services and loss of wages Hospitalization cover up to Rs 1,500 per person per year Primary and secondary health care
Rs 100 per person per year (40%) Rs 30 per person per year. Fully subsidized for the SC/ST population (31%) Rs 48 per family of four (in cash or kind) (90%) Rs 20 per person (58%)
Rs 22.50 per person or Rs 45 for a couple (10%) The Universal Health Insurance Scheme (Rs 548 for a family of 5) (20%) Rs 4 per student per year (23%) Rs 250 per family of five (12%) Rs 120 per person (25%)
Hospitalization cover up to Rs 2000 per person
Hospitalization cover for family up to a maximum limit of Rs 30,000 per family per year Primary and secondary health care
Hospital cover Cover for all surgeries up to Rs 100,000
Source: Devadasan et al. 2004
ment fiat, over 17 lakh farmers within one year and created a corpus of over Rs 15 crore. In the second year, an additional 5 lakh members have been enrolled against the target of 1 crore. The scheme provides financial risk protection against 1600 surgeries offered in 90 accredited hospitals at prefixed rates. Outpatient treatment is free and any diagnostic service resulting in surgery carries a discount of 50%. To keep the premium low at Rs 90, now revised to Rs 120, a Trust chaired by Secretary of the Department of Cooperatives, has been constituted with the premium forming the corpus fund from which the claims are settled. A commercial TPA has been contracted by the Trust at 5.5% of premium collected to
provide ID cards to the members, process the claims and make payments to the service providers. A doctor appointed by the TPA gives prior authorization for expensive surgeries and also scrutinizes correctness of the claims. Within one year of establishment of this scheme, over 27,000 persons were provided outpatient treatment and 4000 surgeries performed. However, sustainability is an issue. The scheme is now facing monetary problems and has a long wait list for surgeries and claims to be paid despite the reimbursement being guaranteed by the Government of Karnataka. Focusing solely on surgical aspects of health can have only a limited appeal and a blind replication of this scheme can give wrong incentives
286
SECTION IV
Fig 3 Types of community health insurance schemes in India
NGO
(ACCORD, JRHIS, SHH, VHS)
Insurance Company
Premium
Health care
Group Premium
Reimbursement
Community
NGO
(SEWA, BAIF, Navsarjan and Karuna) Reimbursement (Karuna Trust, Navsarjan Trust)
NGO
(KKVS, RAHA)
Reimbursement (RAHA) Reimbursement (SEWA, BAIF)
Providers
Premium
Reimbursement (KKVS)
Providers
Health care
Premium
Health care
Community
Source: Devadasan. 2004
Community
for investing in surgery and neglecting other medical needs. However, the scheme has been innovative in demonstrating the benefits of utilizing government resources for the administration of insurance schemes, in bringing down the administrative overheads and facilitating lower premiums. The Yeshaswani model as well as experimentation abroad, seem to clearly point towards the fact that while CBHI is an affordable model of financial risk protection in low-risk settings, it needs institutional support and formal mechanisms for carrying out the critical functions of health insurance collection of premium, settlement of claims, laying down clear rules of entitlements and oversight. Only when such systems are designed and put in place can the CBHI models be upscaled to reach risk pools that are financially viable and provide sustainability.
gatekeeper for referrals. (This portion on China is from Professor Hsaio of Harvard School of Public Health, USA in a personal communication with the author). The scheme details are as under:
Three underpinning concepts

(a) People by themselves, especially those living in rural areas and the poor, are unlikely to be able to raise enough money for such schemes to be fully self-financing, necessitating public subsidy; (b) Even if financing could somehow be organized, there is the issue of how services are to be delivered, since existing services are neither efficient nor effective in terms of quality; (c) Issues related to governance and management: how is one to organize and manage such schemes and who will perform the stewardship (or oversight) functions? These include overseeing the financial functioning and health of such schemes, the functioning of medical care providers, contracting (if any) with providers.
Chinas model of Community-Based Health Insurance

In China, a model is under implementation on a pilot basis their combins certain design features to address the rotten health system in the rural areas. The model, based on a partnership between the government and the community, uses the village-based barefoot doctor as the key provider and
287
SECTION IV
Fig 4 Chinas health insurance system
Regular budget County and Town Government

Public health and prevention
County hospital
Government subsidy
Reimbursement for services and
Fund Management Office
for village doctors' salaries and bonus after THC find their
Township health center

Salary and bonus payments to village doctors
Premium
work meet quality standards
Co-insurance
Enrollees
Village Health Post
Source: Hsiao, HSPH, USA
Co-insurance
Outline of the scheme

Premium
The premium is Rs 100 per person plus the government subsidy (Fig. 4). The contributions of each enrolled member are roughly equal, except for the very poor, from whom no premium is charged. There is an upper limit of benefits. Enrolment is on an annual basisthe first month of each year for renewals or new enrolmentsand after that no enrolments are allowed to prevent adverse selection.
scheme upon examination of the prescription. Roughly 230 drugs can be prescribedincluding both modern and traditional medicinesbased on some type of essential drugs list. In addition, in case of referral to the subdistrict facility (Tehsil level), the patient is reimbursed 50% of the expenditures incurred; 20%30% of all expenses for hospitalization at higher levels.
Referral
The patient can go to a higher order facility only if he gets a referral from the village doctor, who in turn has to take notes and justify why he is referring (Fig 5).
Jurisdiction of the scheme Government subsidy

Typically, each scheme covers several villages. The experience has been that 93% of the people covered under the scheme support it, but only 60% actually join it. Even this is sufficient to ensure a deep enough pool of members6000 and above. A sum of Rs 110 per person is provided to those communities willing to set-up community financing organizations with a minimum number of members (about 70%). This acts as an incentive as well as subsidy.
Benefit package Provisioning

Free consultation visits to the village doctor as he is salaried under the scheme, but there is payment for the drugs prescribed. Fifty per cent of the amount of payment is reimbursed by the Typically in China, every village has one or a maximum of two village doctors. The villagers decide which doctor is to
288
SECTION IV
be involved with the scheme. This village doctor is employed on a salaried contract that lasts for only one year at a time. The village doctor gets paid in two parts: (i) a salary as per contract, and (ii) a performance-linked bonus. The salary is just enough to cover subsistence to ensure that he is interested in the bonus. The bonus depends on three factors: (i) the demand for the service of the doctor as reflected in the number of visits by the villagers; (ii) careful keeping of medical records (patient details, age, sex, number of visits, diagnosis, prescription); and (ii) regularity in attending continuing medical education and passing medical exams.
common health conditions, and on recognizing when to refer patients to higher-level facilities.
Medicines
The village doctor purchases the medicines from subdistrict level storage facilities operated by some agency, which he is allowed to sell at no more than 20% mark-up of the cost price. The price lists are prominently announced and displayed on notice boards.
Administration Training
The barefoot village doctor (who is like the RMP/quack in India) is provided training. Those practising for over 5 years are exempted from training in the first year of the scheme, but in subsequent years they have to take continuing education courses or pass exams to remain a part of the scheme. For all others, a three-year course is necessary for qualification to work in the scheme, followed by annual examinations, and short continuing education courses on an annual basis. The training (and continuing education) of village-based doctors is focused on promoting the ability to treat the eight most A manager and a clerk handle the day-to-day operation of each scheme. The manager reports on the functioning of the scheme to a management committee; the clerk keeps premium receipts and payment records. The Management committee is organized at several levels. First, each village under the scheme has a management committee of 5typically composed of retired teachers, retired local officials, etc. Their functions are: (i) overseeing the functioning of the local doctors clinicmaking sure there are no complaints, no price gouging, etc.; (ii) maintaining a suggestion box where complaints can be placed; and (iii) organ-
Fig 5 Model for provision of health care services in China
County Hospital
Referral
Township Health center

Patients who decided to bypass referral (they get less reimbursement) Emergency service Referral
Patients
Normal
Village Health Post
289
SECTION IV
izing enrolment under the scheme. From each village-level management committee one person is nominated to represent them at the Board of Directors at the subdistrict level (the level at which the scheme is organized). Because of the large number of likely members of the Board, it meets only four times a year. However, a standing committee of 7 that works on behalf of this Board, meets more frequently and oversees the manager and the clerk along with other functions. The standing committee may change from year to year. No payments are made to the standing committee and management committee members. The standing committee, based on simple contracts, employs the village doctors, etc. (Fig 6).
(a) Auditing of accounts; (b) Checking drug quality through laboratory testing and random checks. A broader approach to analyse/assess CBHI schemes is needed through examination of two policy issues: (i) coordination of CBHI and government risk pools, and (ii) equity implications of CBHI schemes and the role of government subsidies in such schemes. There is a strong need for empirical work to explore how CBHI schemes and the broader health care financing system interact. Even if individual schemes achieve their objectives (in terms of equity, efficiency, etc.), it does not necessarily imply that such objectives will be achieved at the system level.
Oversight and stewardship by government

In technical matters for which the committees at the village and subdistrict levels are not equipped, the government provides the support:
What are the lessons for India?

The lessons that emerge from the China Model and discussions in the earlier paragraphs are that given the huge size, diversity and levels of development in India, it is important
Fig 6 Regulation, monitoring and supervision of community-based health care system in China
County People's Congress Party Secretary
Supervise and Monitor
Regulate and
County and Town Government

Audit and Subsidize
Monitor
County hospital
Supervise
Town People's Congress

Report
Board of Directors
Regulate and monitor
Supervise and train
Executive Committee
Township Health Centre

Pay, supervise and train Monitor clinical work
Elect 1 members
Fund Management Office

Financial report
File claim records
Report
Village Health Post Board of Directors
Enrolled peasants
Elect 5 members
290
SECTION IV
to adopt a four legged strategy for affording real risk protection to the poor: (i) Bring down covariate risk in the community and address the containment of infectious and childhood diseases by intensifying public health programmes; (ii) strengthen government facilities to enable them to provide equally good quality care to the poor this is the cheapest and most affordable option for government in the short run as a well functioning public health system has great potential to protect the poor from risk; (iii) experiment with different models of financing to spread risk and reduce the burden on the government. Such models will imply designing the features and implementing them on a pilot basis before coming up with a final policy framework; and (iv) in places/states where there are networks of self help groups and evidence of solidarity experiment the China Model as it would: (a) strengthen participation and that of local bodies and Panchayat Raj institutions; (b) incorporate the RMP into the system; (c) drastically reduce costs; (d) enable providing healthcare within the village itself. Such designing needs to be based on, first, being clear as to what the objectives of public policy areis it deepening insurance markets, or extending financial risk protection against illness, or increasing FDI to India? Second, the size of the risk pool and lowered risk factors are critical for low premiums that could be affordable for the majority of people. Thirdly if the system is to be based largely on a fee for service system of payment, with zero cost at the point of service, then it will entail putting in place a set of prerequisites, such as standardized treatment protocols and unit costs; regulations to control the provider, disease classification using ICD-10 and/or grouping of diseases under Diagnostic Related Group for payment system, pricing controls, putting sub-limits to expenses rather than having a cap of assured sum to contain charges, making it mandatory for data returns, standardization of claim forms etc. Besides, in any insurance system it is equally necessary to have regulations for quality and cost; mechanisms for accreditation and certification, issue of unique ID cards for members, different premium structures, controlling prices etc.
Implementation of the Package - Restructuring Institutional Mechanisms and Reorganizing Relationships

Most importantly, for the actual implementation of UHIS, a critical institutional player needs to be inducted on the demand side commanding considerable market power to negotiate the best possible care at the most affordable prices for the patients. The concept is based on the assumption that organizational structures are normally shaped to suit the objectives of financing systems. If expenditure control is the overriding objective then there is usually a tendency towards centralization of all spending decisionscosting, sanctioning, releasing, accounting, mandating referrals through gatekeepers, etc. But in a prepaid insurance system the key actor is the consumer on whose willingness to contribute rests the whole system. Since individual patients cannot be expected
to bargain prices on account of their vulnerability and the superior strength of the provider who has more information on their needs, Enthovens idea of a sponsor, as an instrument to strengthen demand side to tilt the market to the advantage of the consumer has force (Enthoven 1983, 1993). It implies that the purchasing function needs to be centralized into one entity large enough to make a difference to the practice and earnings of the providers. The concept also draws from the power of a single payer being able to negotiate better terms as in Canada than in a multipayer environment as in US. This concept is then the theoretical basis for proposing a Social Health Insurance Corporation as the sponsor and reinsurer for independent health insurance companies. One option for the establishment of such a SHIC in Indian conditions is a) by the merger of the ESIS (medical side) with the CGHS; and b) by steadily moving towards a mandatory health insurance paradigm. For this the starting point could be mandating all public servants working in the government or government owned entities to compulsorily pool their contributions to the SHIC. Combined with a broadened ESIS membership, this alone will increase the corpus amount over four to five times to the existing Rs. 1100 crore of premium collected for Health Insurance. This corpus can further be widened when the premium subsidy for the poor is also pooled in here. Such a mechanism will provide the required volume and velocity required to trigger establishment of health insurance companies, professional provider networks, mutual fund cooperatives like weavers and fishermen cooperative societies, a federation of CBHI schemes, HMOs by hospitals having more than 500 beds and ability to establishment own provider network etc. All these entities subject to meeting solvency rules etc could be the vehicles to access the poor in the rural hinterlandlike the commercial banks reach out through the grameen bank networks. For their operations the SHIC can act as the reinsurer. Secondly, in the future years, as this system settles down, the SHIC can also establish an equalization fund as in Chile. In Chile the equalization fund is made up of a proportion of the premiums collected by all the insurance companies being pooled into the fund. Subsequently, this fund reimburses the companies in accordance with the risk profile of the insured. This then acts as a positive incentive to adhere to the guidelines of not denying insurance to any one on grounds of risk and having exclusions of any kind. The success of this model will however depend upon our ability to bring in a high caliber of professional management to the SHIC and other financing entities having capacity to collect premiums, issue IDs, process claims, reimburse in a timely fashion, accredit and develop provider networks, negotiate rates etc. Secondly, the SHIC will have to be a financing instrument and not a provider. This then means that own hospitals and dispensaries by the ESIS, CGHS and PSUs will need to be converted into Trust hospitals available for their members at dedicated times and for general public at other times. The advantage of this measure will be two: a) that the SHIC will be able to accrue more for the corpus as the administrative costs now pass onto the hospitals units which become self
291
SECTION IV
financing, mobilizing its money from insurance policies/ user fees; and b) the general public as well as the employees and members of ESIS, CGHS and PSUs all get a wider access and choice of hospitals. At present the CGHS dispensaries have an average of 14 OP patients per day and the ESIS hospitals have an average occupancy rate of 50%, with some having even as low as 10%. Likewise, in several remote areas where public infrastructure is weak the PSUs have excellent medical facilities with capacity to serve the local populations. Thus over 2000 facilities under ESIS, CGHS and PSUs can be opened up to the general public with an incremental amount of additional budget. This would also increase access and feasibility of insurance reaching the poor. Economy of scale that comes from a large risk pool is an important consideration for covering hospitalization. By laying down a minimum level of say 75-80% population coverage at the village panchayat level, or a risk pool of 1015,000 members as the eligibility criteria for receiving government subsidies, it could be ensured to have a more representative membership the healthy and the better paying sections along with sick and poor. And in order to achieve the willingness to pay and optimal participation, substantial subsidies targeted to the poor, a need based standardized benefit package and linkage to provider networks need to be incorporated as the three core elements of the design. The flip side of the SHIC model is that it may entail high administrative costs due to the several layers of intermediation. The SHIC will also need to be run on professional lines with highly skilled and trained persons, which will again increase costs. The issue then is whether such a structure will be suitable and affordable and other alternatives to reduce administrative costs need to be examined. In the absence of any experience it is incumbent to try out on pilot basis some of the financing systems. To monitor and regulate this huge initiative, there will also be an urgent need to have an independent health regulator and a body of laws that address various issues related to health insurance markets and the serious distortions that the current trends are creating and will be difficult to remove later. In the absence of such a roadmap for reform and clarity of vision, the goal of having a Universal Health Insurance will not be realized. This goal was first articulated in 1954 by the then Prime Minister Jawaharlal Nehru. It is indeed a tragedy that even after five decades such an important goal continues to be an aspiration.
large size of out of pocket expenditures provides an opportunity to pool these resources and facilitate spreading risk from households to government and employers on a shared basis which will be a more equitable financial arrangement. The dimension of equity is of particular concern as the inelasticities of demand for acute care, are resulting in over 33 lakh persons being pushed below poverty line, every year. In short the social benefits of instituting social insurance as a financial instrument to replace user fees, outweighs the possible risks of moral hazard and increased costs, typical outcomes of prepaid insurance. How to minimize these two market failures are of concern and need to be addressed by developing a well thought out strategy taking international evidence into account so we build on existing knowledge and learn from others experiences. It is argued that it is not advisable for governments to intervene in health insurance markets in a piecemeal mannerinsurance for pensioners by the Department of Personnel; for weavers by the Department of Textiles, for fishermen by the Department of Agriculture, for farmers by the Department of Cooperatives, poor women by the Department of Rural Development etc., as such attempts fragment risk pools. In other words, resorting to insurance as a financing instrument must be an act of a deliberate strategy that addresses the market failures in order to ensure that inequities do not widen and the poor are not marginalized two typical outcomes of private, fragmented insurance systems. In conclusion it is reiterated that given the fiscal constraints for government to provide universal access to free health care, insurance can be an important means of mobilizing resources, providing risk protection and achieving improved health outcomes. The critical need is to experiment with the wide range of financing instruments available in different scenarios and have adequate flexibility in the design features, the structures and processes, institutional mechanisms and regulatory frameworks, so that a viable balance can be achieved for minimizing market distortions so that the outcomes do not make the cure worse than the disease (Enthoven 1983, 1993). Unregulated markets are inefficient and inequitable, requiring governments to intervene to ensure no segmentation in the system (Bloom, 2001). For this, the burden of building partnerships and managing change is on the government, which in turn needs to base its strategy on sound research.
Acknowledgements
I gratefully acknowledge the assistance of Dr Somil Nagpal for the analysis of the CGHS and Dr Alaka Singha, WHO, Geneva. I am most indebted to Dr. William Hsiao, Harvard School of Public Health, USA for so generously sharing his experiences of China.
Conclusion
The present system of financing and payment systems raise several important concerns on the suitability of the structure to meet current day problems and future challenges. The
292
SECTION IV
Annexure 1 Number of workforce by employment category, income status and industry classification (1999-2000)
Industry High 1. Organized Sector 1.a. Government 1.b.1 Agriculture 1.b.2 Manufacturing, etc. 1.b.3. Services, etc. 2. Unorganized Sector 2.1. Regular salaried 2.1.a. Agriculture 2.1.b. Manufacturing, etc. 2.1.c. Services, etc. 2.2. Self-employed 2.2.a. Agriculture 2.2.b. Manufacturing, etc. 2.2.c. Services, etc. 2.3. Casual employed 2.3.a. Agriculture 2.3.b. Manufacturing, etc. 2.3.c. Services, etc. (1+2) Total Workforce Rural Middle Low High Urban Middle Low High 0.14 0.14 0.00 0.03 0.04 3.12 0.80 0.05 0.26 0.65 1.38 0.99 0.19 0.39 0.44 0.34 0.17 0.01 3.27 All areas Middle Low 1.80 1.09 0.03 0.34 0.18 13.21 2.55 0.12 0.69 1.61 7.87 5.26 0.86 1.74 3.35 2.17 0.72 0.32 15.01 0.87 0.72 0.05 0.17 0.07 17.30 1.78 0.20 0.51 1.03 8.35 5.14 1.22 1.81 7.10 5.15 1.36 0.65 18.16 Total 2.81 1.94 0.09 0.54 0.25 33.62 5.13 0.37 1.46 3.30 17.60 11.39 2.27 3.94 10.90 7.66 2.26 0.98 36.44
2.23 0.40 0.04 0.11 0.28 1.19 0.96 0.10 0.22 0.38 0.34 0.17 0.01 2.23
10.15 0.87 0.11 0.18 0.47 6.56 5.18 0.55 0.80 3.05 2.11 0.52 0.20 10.15
12.68 0.43 0.19 0.11 0.21 6.23 4.85 0.64 0.67 5.95 4.94 0.77 0.30 12.68
0.90 0.40 0.01 0.15 0.37 0.19 0.03 0.09 0.16 0.06 0.00 0.00 0.00 0.90
3.06 1.68 0.01 0.51 1.14 1.31 0.08 0.31 0.94 0.31 0.06 0.20 0.12 3.06
4.62 1.35 0.01 0.40 0.82 2.12 0.29 0.58 1.14 1.16 0.21 0.60 0.35 4.62
Note: The number of workforce has been measured by the current daily status (CDS). Figures are reconciled using Table 51 of the NSS Report No. 458 and tables extracted from unit-level record data of the 55th Round along with a table from Economic Survey (2003-04). Data on break-up of urban-rural organized employment are not available. High, middle and low denote the household monthly per capita expenditure class. Source: Unit Level Records of Employment and Unemployment Survey, 55th Round, National Sample Survey (NSS), 1999-2000
293
SECTION IV
REFERENCES
Bhat R, Reuben E. Analysis of claims and reimbursements made under Mediclaim Policy of GIC, W.P.No. 2001-08-09. Ahmedabad: IIM; 2001. Bhat R. Characteristics of private medical practitioners in India: A providers perspective. Health Policy and Planning 1999;14:2637 London, OUP. Bhat R. The private /public mix in health care in India Health Policy and Planning ,8:11 43-56 OUP, 1993. Bhat R. Public private partnerships in the health sector: Issues and prospects, May 1999, W.P No. 99 05-06 Ahmedabad: IIM. Bhat R. Public private partnerships in health sector: Issues and policy options, January 2000, Paper prepared for DFID, Delhi. Bloom Gerald, Equity In Health In Unequal Societies: Meeting Health Needs in contexts of Social Change, Health Policy, 2001 Carrin G, Chris J. Reaching universal coverage via social health insurance. Discussion paper 2, Papers on Health Financing and Contracting, WHO, 2004. Devadasan N, Kent Ranson, Wim Van Damme, Bart Criel, Community Insurance in India : An overview, EPW, July, 2004 Dyna Arhin-Tenkorang Health Insurance for the Informal Sector in Africa: Design features, risk protection and resource mobilization, Geneva: CMH; 2001. Ellis Randall P, Moneer Alam, Indrani Gupta, Health Insurance in India : Prognosis and Prospects, EPW, Jan 22, 2000. Enthoven A Managed Competition in Health Care and the Unfinished Agenda, Health Care Financing Review, 1986. Enthoven A. The History and Principles of Managed Competition, Health Affairs, 1993. Enthoven A. Managed Competition of Alternative Delivery Systems, Journal Of Health Politics, Policy & Law, 1988. Fuchs V. Who Shall Live? Health Economics & Social Choice, 1974. Garg C. Implications of current experiences of health insurance in India. In: Private health Insurance and Public health goals in India: Report on a National Seminar, New Delhi: World Bank; 2000. Garg C. Is health insurance feasible in India: Issues in private and social health insurance? Health security in India (unpublished). Government of India. National Sample Survey Organization. Household Consumer Expenditure Survey, 55th Round (19992000). Government of India. Private Health Insurance And Public Health Goals In India Report on a National Seminar, GOI, 2000 Gupta I, Dasgupta P. Demand for Curative Health Care in Rural India: Choosing Between Private, Public And No Care, NCAER, 2002 Working Paper Series No. 82. Gupts Indrani, Private Health Insurance and Health Costs: Results from a Delhi Study, EPW, vol XXXVII July, 2002. Hsaio WC Abnormal Economics in the Health Sector, Health Policy 32, 1995. Hsaio W. Unmet health needs of 2 billion: Is community financing a solution. Working Paper for CMH, 2001. Jakab M, Krishnan. Literature Review on Community Financing, Washington: World Bank, 2001. Kent RM, Devadasan N, Acharya A, Ruth AF. How to design a community based health insurance scheme: Lessons learnt From the Indian experienceReport to the World Bank, 30 June, 2003 Mahal A. Assessing private health insurance in India: Potential impacts and regulatory issues. Economic and Political Weekly 2002:55971. Dr. Marcelo Tokman and Ms. Consuelo Espinora Marty. Report on Health Insurance in Chile submitted to NCMH in December 2004. Pauly M. Is cream skimming a problem for the competitive medical market? Journal of Health Economics 1984;3 Pauly M. A primer on competition in medical markets. In: Frech HE III (ed). Health care in America. 1988. Purohit BC, Siddiqui TA. Cost recovery in diagnostic facilities, EPW, July 1995 Rao S. Health insurance: Concepts, issues and challenges. EPW, August 2004 Sen PD. Community control of health financing in India: A review of local Experiences, October 1997. Sekhrie N, Savedoff W. Private health insurance: Implications for developing countries; policy and practice. Bulletin of the World Health Organization 2005;85:12734.
294
SECTION IV
The World Health Report, 2005, Make Every Mother and Child Count. Geneva: WHO. World Bank. Better health systems for Indias pooranalysis, findings and options. 2001
Xingzhu Liu et al. The Chinese experience of hospital price regulation. Health Policy and Planning 2000.
295
SECTION IV
Resource Devolution from the Centre to States: Enhancing the Revenue Capacity of States for Implementation of Essential Health Interventions
P
M. GOVINDA RAO
DIRECTOR NATIONAL INSTITUTE OF PUBLIC FINANCE AND POLICY 18/2 SATSANG VIHAR MARG, NEW DELHI 110067 INDIA EMAIL: mgr@nipfp.org.in
MITA CHOUDHURY
ECONOMIST NATIONAL INSTITUTE OF PUBLIC FINANCE AND POLICY 18/2 SATSANG VIHAR MARG, NEW DELHI 110067 INDIA EMAIL: mita@nipfp.org.in
MUKESH ANAND
SENIOR ECONOMIST NATIONAL INSTITUTE OF PUBLIC FINANCE AND POLICY 18/2 SATSANG VIHAR MARG, NEW DELHI 110067 INDIA EMAIL: manand@nipfp.org.in
ROVIDING ACCESS TO ADEQUATE HEALTH CARE SERVICES IS AN IMPORTANT component of empowering people with human capital. This, however, can be achieved only when the spending on health care is adequate and delivery systems efficient. Ensuring adequate outlay on health services and efficient use of allotted expenditure are important not only to improve the productivity and earning capacity of the population, particularly the poor, but also their health status. Not surprisingly, improving health indicators is an important component of the Millennium Development Goals (MDGs) set by the United Nations. There are also important targets on health status achievements set for the Tenth Plan. The Common Minimum Programme of the ruling UPA government also seeks to increase the public expenditure by the Centre and States on health from the present level of less than 1% to 2%-3% of the gross domestic product (GDP). The provision of health and family welfare services falls in the realm of concurrent responsibility of the Centre and the States, but the latter have a predominant role in the delivery of these services. However, fiscal pressures at the State level led to compression of expenditures by the State Governments partly compensated by an increase in Central financing of these services, particularly for some prioritized programmes implemented through Central sector and Centrally sponsored schemes. In general, over 85% of the public expenditure on medical and public health is incurred by the State Governments. This paper identifies the resource gap between the desired and the actual health expenditure in 15 major States in India (14 large, non-special category States and Assam), and highlights the extent to which the gap can be reduced by augmenting resources at the State level. Further, it estimates the resource gap that cannot be met through States own resources and therefore requires Central transfers. The design of Central transfers needed for meeting the required health expenditure of various States is also discussed. The principal motivation for this paper is the concern for achieving the targets set for improving the health status of Indias population, particularly the poor and the vulnerable. While fulfilling the targets for improving the health status set by both national and international agencies (Tenth Plan goals and MDGs) requires considerable augmentation in expenditures, the deteriorating fiscal situation at the State level has imposed severe constraints in financing them. In particular, there has been a decline in social sector expenditure as a percentage of both gross state domestic product (GSDP) and total expenditure in a majority of States in the 1990s (Dev and Mooij 2005). The combined expenditure of States on medical and public health, sanitation, water supply and family welfare declined from 8.4% of the total expenditure in 1990-91 to 7.2% in 2001-02. As a proportion of GSDP, the decline was from 1.5% to 1.3% during the period. In the context of deteriorating finances of the States, the decline in health expenditure is a matter of concern. This is more so because the share of public expenditure in GDP in the case of health expenditure is much lower in India for the level of per capita income. In 2000, the total expenditure by the Centre and the State Governments in India was about 0.9% of GDP. In comparison, Bangladesh and Bhutan, with lower per capita GDP, spent 1.4% and 3.7% of GDP, respectively on health. Advanced countries such as the US and UK spent substantially higher amounts-5.8% and 5.9% of the GDP, respectively. With the prevailing level of public expenditure on health and its declining trend in the 1990s, it appears difficult for India to achieve the health targets of MDGs and the
297
SECTION IV
Resource Devolution from the Centre to States
Tenth Plan objectives. As per the provisional estimates, the infant mortality rate (IMR) in India stood at 66 in 2001 (Sample Registration System Bulletin 2002), which was much higher than the Tenth Plan target of 45 by 2007. Similarly, the maternal mortality ratio (MMR) was much higher than the target. The problem is exacerbated by the fact that there are significant variations in the IMR, MMR and life expectancy at birth (LEB) between different States. In fact, health sector outcomes in the poorer States are extremely low. Similarly, there is considerable catching up to do in the health status of women, scheduled castes and tribes. This calls for substantial increases in the resources allocated to the public provision of health, targeting of health expenditure to areas and groups of population with low health indicators and focusing on the delivery of health services to transform public expenditure into improved outputs and outcomes. This study attempts to estimate the expenditure required between 2005-06 and 2009-10 for meeting specific health goals and explores the possible means of meeting the expenditure requirement in 15 selected States.1 The choice of the terminal year 2009-10 for estimating expenditure requirement is driven by the fact that the MDGs have to be met by 2015. To arrive at the desired outcomes by 2015, incurring appropriate expenditure during the period 2005-06 to 2009-10 is crucial.
Expenditure Requirement for Health and Related Sectors

Health outcomes are determined not only by direct expenditure on the health sector but also by expenditure on related sectors such as safe drinking water, sanitation, nutrition, primary education and roads (Shiva Kumar 2005, Deolalikar 2004). Expenditure requirement in this analysis is, therefore, viewed as a package of expenditures required in each of these sectors rather than the health sector alone.2 Such expenditures in different sectors mutually reinforce each other and have been argued to be important in the context of assessing budgetary allocations for achieving health goals (Shiva Kumar 2005). This study estimates the input deficiencies in each of these sectors from the specified national norms/targets in the States and transforms these into the expenditure needs for the respective States.
Health Sector (Medical, Public Health and Family Welfare)

In this paper, expenditure requirement in the health sector is estimated such that it is adequte to provide a minimum level of access to health care facilities, both in terms of physical facilities and manpower. In particular, the study focuses on
1 2 3
the national norms related to rural primary health care institutions such as subcentres (SCs), primary health centres (PHCs) and community health centres (CHCs) and estimates the resource requirements for meeting the national norms related to these institutions. As per the national norms, there should be one SC for every 5000 population, 1 PHC for every 30,000 population and 1 CHC for every 120,000 population in the plains. The corresponding figures for tribal/difficult terrains are 3000, 20,000 and 80,000, respectively. While the norms may be inadequate to achieve the desired outcomes in many States, they aim to ensure the provision of the minimum level of health infrastructure in each State. The expenditure requirement for the health sector during 2005-06 to 200910 is given in Table 1. The need for increased expenditures in the health sector arises from the fact that the existing infrastructure of SCs, PHCs and CHCs is grossly inadequate in many States. As per the Bulletin on Rural Health Statistics (2002), none of the 15 major States under study have achieved the required level of provision in all the three categories of SCs, PHCs and CHCs. While some States have achieved the norms in terms of SCs and PHCs, none of the States have achieved the targets with respect to CHCs. The number of States meeting the norms in individual categories reduces as one moves from SCs to PHCs and CHCs. Even where the norms are met in terms of the number of facilities required, many of them are non-functional due to lack of equipment and need for civil works. The mere meeting of norms in terms of the number of facilities is therefore not enough. Many of these facilities also suffer from shortage of manpower. The estimates therefore include the cost of upgrading the equipment facilities, civil works and manpower in the existing facilities, apart from setting up new facilities, to fulfil the national norms. The total requirement of expenditure in the health sector comprises the capital and the revenue components. The capital component of expenditure requirement further consists of two parts. The first is the cost of building new facilities for fulfilling the national norms for SCs, PHCs and CHCs, and the second is the cost of upgrading the civil works and equipment in the existing SCs, PHCs and CHCs. It is assumed that this capital expenditure will be carried out in a phased manner over a period of five years between 2005-06 and 200910 to eliminate all gaps in the physical infrastructure by 2010. In addition to the cost of covering up the existing gap, the estimate of requirements also includes the cost of providing health care services for the additional population in each year between 2005-06 and 2009-10.3, 4 The revenue expenditure requirement in the health sector also comprises two parts. First, in addition to the expendi-
The 15 States are Andhra Pradesh, Assam, Bihar (including Jharkhand), Gujarat, Haryana, Karnataka, Kerala, Madhya Pradesh (including Chhattisgarh), Maharashtra, Orissa, Punjab, Rajasthan, Tamil Nadu, Uttar Pradesh (including Uttaranchal) and West Bengal. The National Commission on Macroeconomics and Health (NCMH) has identified expenditure on these sectors as important for achieving health goals. Information on the existing SCs, PHCs and CHCs has been taken from the Bulletin on Rural Health Statistics, which provides the figures updated till 2001-02. Unfortunately, the information on the addition to infrastructure in the States between 2001-02 and 2005-06 is not easily available. To account for some likely increase in infrastructure between 2001-02 and 2004-05, capital expenditure on SCs, PHCs and CHCs in individual States, provided in the States' Finance Accounts, along with the unit cost of building these facilities were used. For water supply and sanitation, a 10% increase in access between 2001-02 and 2004-05 has been assumed. The National Commission on Macroeconomics and Health (NCMH) estimates the unit cost used for building SCs, PHCs and CHCs to be Rs 24.5 lakh for a PHC, Rs 80.5 lakh for a CHC and Rs 2 lakh for an SC. The cost of upgrading the civil works and equipment in the existing facilities has also been provided by NCMH based on a facility survey carried out in 1999.
298
SECTION IV
Table 1 Expenditure requirement for the health sector between 2005-06 to 2009-10
State 2005-06 As percentage of GSDP 2006-07 2007-08 2008-09 2009-10 2005-06 Real per capita (in Rs) (2005-06 prices) 2006-07 2007-08 2008-09 2009-10
Andhra Pradesh Assam Bihar (including Jharkhand) Gujarat Haryana Karnataka Kerala MP (including Chhattisgarh) Maharashtra Orissa Punjab Rajasthan Tamil Nadu UP (including Uttaranchal) West Bengal
0.92 1.35 2.04 0.70 0.64 0.95 0.98 1.17 0.71 1.47 0.94 1.34 0.93 1.18 0.95
0.93 1.37 2.18 0.67 0.65 0.91 0.95 1.16 0.69 1.43 0.93 1.29 0.92 1.22 0.99
0.93 1.38 2.31 0.65 0.65 0.87 0.92 1.16 0.67 1.39 0.92 1.25 0.90 1.26 1.02
0.94 1.39 2.43 0.62 0.66 0.83 0.90 1.15 0.65 1.36 0.92 1.20 0.88 1.30 1.04
0.94 1.40 2.54 0.59 0.66 0.80 0.87 1.13 0.63 1.32 0.91 1.16 0.86 1.33 1.06
251 234 229 234 248 269 303 213 275 233 373 286 285 178 238
253 239 247 225 253 267 303 213 267 229 372 278 281 187 250
255 243 265 217 257 267 303 213 260 224 370 269 275 195 260
257 246 282 208 260 267 303 213 260 220 368 259 269 203 269
258 249 298 199 263 267 303 213 260 214 365 250 261 210 276
Note: Projections of GSDP were made using the prescriptive growth rates suggested by the Twelfth Finance Commission
Fig 1 Per capita additional requirement of resources for the health sector, 2009-10
ture being currently incurred to run the existing SCs, PHCs and CHCs, it includes the salary expenditure required to provide these existing facilities with manpower as per the norms. Second, it includes the expenditure that would be required to run the new SCs, PHCs and CHCs to be built between 200506 and 2009-10 with adequate number of health workers.5 The additional requirement of resources is estimated as the extent of resources required over and above the actual expenditure as a percentage of GSDP incurred in 2001-02.6
5 6 7
The estimate indicates that an additional amount of about Rs 26,439 crore (at 2005-06 prices) is required to provide a minimum level of access to health care facilities in the States (Table 2).7 Nearly 60% of this amount is needed in Uttar Pradesh (UP) and Bihar alone. On average, Madhya Pradesh, West Bengal and Orissa account for about 20% of additional expenditure. The States of Maharashtra and Karnataka require less than 1% of this amount, while Kerala requires no additional expenditure. As a percentage of GSDP and in per capita terms, Bihar, Uttar Pradesh, Assam, Madhya Pradesh and Orissa require the highest increase in expenditure in the health sector. It may be noted that the existing levels of per capita expenditures in these states are among the lowest in the country (Table 2). Table 2 also shows that the ratio of health sector expenditure to GSDP is very high in these States. This, however, is merely a reflection of low GSDP in these States. If one examines the States that lie above the average level of per capita additional requirement of resources during 200910, the five low-income States of Bihar, UP, Assam, MP and Orissa are included (Fig. 1). The requirements for Haryana and West Bengal are also relatively high. It must be noted that among the selected States, the amount of GSDP devoted by Haryana towards the health sector is the lowest. As the incomes of West Bengal and Haryana are higher than average, as a percentage of GSDP, their additional requirements are relatively low. In general, additional requirements both in terms of per capita as well as GSDP, indicate that Bihar, UP, Assam and
For calculating the salary requirements in each year, the Central Government pay scales for different levels of medical personnel at SCs, PHCs and CHCs were used. The population projections for the years 2005-06 to 2009-2010 provided by the Registrar General of India were used for the estimations. Data on State Finances of India published by the Reserve Bank of India were used for the actual expenditures. An average inflation rate of 7% was assumed throughout the study.
299
SECTION IV
Table 2 Additional requirement of resources for the health sector between 2005-06 to 2009-10
Additional As percentage of GSDP Current level State (2001-02) 2005-06 2006-07 2007-08 2008-09 2009-10 Current level (2001-02) 2005-06 2006-07 2007-08 2008-09 2009-10 Real per capita (in Rs) (2005-06 prices) resources required (2005-10) (Rs in crore 2005-06) prices
Andhra Pradesh Assam Bihar (including Jharkhand) Gujarat Haryana Karnataka Kerala MP (including Chhattisgarh) Maharashtra Orissa Punjab Rajasthan Tamil Nadu UP (including Uttaranchal) West Bengal Total
0.86 1.04 1.27 0.58 0.54 0.94 1.01 0.80 0.67 1.02 0.85 1.10 0.84 0.74 0.92
0.06 0.31 0.77 0.12 0.10 0.01 0.00 0.37 0.04 0.45 0.09 0.24 0.09 0.44 0.03
0.07 0.33 0.91 0.09 0.11 0.00 0.00 0.36 0.02 0.41 0.08 0.19 0.08 0.48 0.07
0.07 0.34 1.04 0.07 0.11 0.00 0.00 0.36 0.00 0.37 0.07 0.15 0.06 0.52 0.10
0.08 0.35 1.16 0.04 0.12 0.00 0.00 0.35 0.00 0.34 0.07 0.10 0.04 0.56 0.12
0.08 0.36 1.27 0.01 0.12 0.00 0.00 0.33 0.00 0.30 0.06 0.06 0.02 0.59 0.14
232 176 140 193 207 267 303 148 260 161 338 236 256 111 229
19 59 88 41 41 2 0 65 15 72 35 50 30 66 9
21 63 107 33 46 0 0 65 8 68 33 42 25 75 20
23 67 124 24 50 0 0 65 0 63 32 33 19 84 30
25 70 141 15 53 0 0 65 0 58 30 23 13 91 39
26 73 157 6 56 0 0 65 0 53 27 14 6 99 47
944 976 7,150 634 554 10 0 2,983 223 1,210 405 990 612 8,463 1,286 26,439
MP occupy the top four positions. The health sector in these States therefore needs a special focus.
Safe Drinking Water and Sanitation

According to the 2001 Census, only about 67% of households in the selected States have access to safe drinking water and in States such as Assam and Orissa, it is less than 50%. Similarly, less than 30% households have access to toilet facilities in Bihar, UP, Orissa, MP and Rajasthan. The percentage is as low as 15 in Orissa. We now highlight the resource requirements for providing all households in the States with access to safe drinking water and toilet facilities by 2010.8 Table 3 provides the expenditure requirement for water supply and sanitation in the period 2005-06 to 2009-10. The resource requirements for water and sanitation also have a capital and revenue component. The capital component includes the cost of providing all households not having access to safe drinking water and toilet facilities with these facilities. This requirement of expenditure is spread over a fiveyear period between 2005-06 and 2009-10. It also includes the cost of providing the additional population in each year
8 9
between 2005-06 and 2009-10 with these facilities. The revenue component includes the expenditure associated with the increased coverage. In the case of safe drinking water, an additional 10% of the capital cost is also included for maintenance of the water supply systems. Estimates indicate that an additional amount of Rs 17,593 crore will be required for providing safe drinking water and toilet facilities to all households (Table 4). Of these, four States Kerala, Maharashtra, West Bengal and Orissa account for more than 60% of the requirement. The high requirement of Kerala may be attributed to the low access to safe drinking water.9 Only 20% of households in the State have access to safe drinking water. In contrast, Tamil Nadu, Gujarat, Haryana and Andhra Pradesh do not require any additional expenditure. As a percentage of GSDP, excluding Kerala, Assam and Orissa require the highest increase. Interestingly, Kerala and Karnataka spent the lowest amount of their GSDP on water supply and sanitation in 2001-02 among the selected States. Given the low level of spending in Kerala and the low access to safe drinking water, marked increases in expenditure, both as a percentage of GSDP and per capita are required in the State. While Karnataka also requires a substantial increase
It is assumed that 30% of the uncovered population will have access to piped water and the remaining 70% will have access to handpumps. The unit cost of providing piped water was taken to be approximately Rs 1200 per capita and that of handpumps Rs 140 per capita. A unit cost of Rs 1000 was taken for building a toilet per household. Partly, the high requirement in Kerala despite it being a high rainfall State is due to the definition of safe drinking water. A large proportion of the population in the State uses the well water, which is considered unsafe. Requirement of the State is an overestimate as the unit cost of providing safe water in Kerala would be lower due to high density of population.
300
SECTION IV
Table 3 Expenditure requirement for water supply and sanitation between 2005-06 to 2009-10
0.29 0.82 0.68 0.21 0.56 0.43 0.54 0.72 0.30 1.13 0.29 1.18 0.26 0.49 0.43
0.30 0.88 0.69 0.21 0.58 0.44 0.65 0.73 0.30 1.22 0.29 1.25 0.26 0.48 0.44
0.30 0.93 0.70 0.21 0.59 0.44 0.74 0.73 0.30 1.29 0.28 1.30 0.25 0.47 0.45
0.30 0.98 0.71 0.20 0.61 0.44 0.83 0.73 0.30 1.36 0.28 1.35 0.25 0.47 0.45
0.30 1.03 0.72 0.20 0.62 0.44 0.91 0.73 0.31 1.41 0.27 1.39 0.24 0.46 0.46
81 146 76 88 278 126 176 131 118 180 117 275 143 75 112
81 159 77 88 278 131 218 134 120 197 118 275 143 74 117
81 170 78 88 278 92 255 135 122 212 114 280 143 73 123
81 182 80 88 278 92 294 136 123 228 114 291 143 73 125
81 195 81 88 278 92 332 137 130 241 110 302 143 72 131
Table 4 Additional requirement of resources for water supply and sanitation between 2005-06 to 2009-10
0.30 0.46 0.57 0.26 0.73 0.32 0.18 0.51 0.19 0.57 0.25 1.28 0.47 0.36 0.27
0 0.36 0.11 0 0 0.11 0.36 0.21 0.11 0.56 0.04 0 0 0.13 0.16
0 0.42 0.12 0 0 0.12 0.47 0.22 0.11 0.65 0.04 0 0 0.12 0.17
0 0.47 0.13 0 0 0.12 0.56 0.22 0.11 0.72 0.03 0.02 0 0.11 0.18
0 0.52 0.14 0 0 0.12 0.65 0.22 0.11 0.79 0.03 0.07 0 0.11 0.18
0 0.57 0.15 0 0 0.12 0.73 0.22 0.12 0.84 0.02 0.11 0 0.1 0.19
81 79 63 88 278 92 52 94 73 90 101 275 143 55 66
0 67 13 0 0 34 123 37 45 89 16 0 0 20 47
0 80 14 0 0 39 165 39 47 107 17 0 0 19 52
0 92 15 0 0 0 202 40 48 122 13 4 0 18 57
0 104 17 0 0 0 242 41 50 137 13 16 0 18 59
0 116 19 0 0 0 279 43 57 150 9 26 0 17 65
0 1,349 897 0 0 415 3,532 1,842 2,455 2,336 175 300 0 1,834 2,459 17,593
301
SECTION IV
Fig 2 Per capita additional requirement of resources for water supply and sanitation, 2009-10
Fig 3 Per capita additional requirement of resources for providing nutritional supplements, 2009-10
in per capita terms, given its income level, the required increase as a percentage of GSDP is relatively moderate. Apart from these States, West Bengal requires a marked increase both in per capita terms as well as a percentage of GSDP (Table 4). The two low-income States of Assam and Orissa require special policy focus.
Nutrition
One of the primary causes of infant and child mortality in India is maternal and child malnutrition. Keeping this in view, policy stance in recent times has focused on providing nutritional supplements to pregnant and lactating mothers and undernourished children. However, the coverage of the provision of nutritional supplements has not yet been universal. We estimate the resource requirements for making this universal. Specifically, we estimate the requirement of resources for providing nutritional supplements to all malnourished children in the age group of 6-71 months, and all pregnant and lactating mothers below the poverty line. The unit cost of providing nutritional supplements to children in the age group of 6-71 months under the ICDS scheme is Rs 3.10 per child per day. Similarly, the unit cost of providing nutritional supplements to severely malnourished children is Rs 3.81 per child per day and to pregnant and lactating mothers Rs 3.41 per beneficiary per day. These unit costs are used to estimate the expenditure requirements. It may be noted that the above-mentioned unit cost for providing nutritional supplements is abysmally low. However, as these norms have been specified by the Government of India, they have been used to estimate the expenditure requirements. Table 5 gives the expenditure requirements for providing nutritional supplements in the period 2005-06 to 2009-10. A total of Rs 56,383 crore is additionally required for providing nutritional supplements to all malnourished children between the age of 6 and 71 months, and pregnant and lactating mothers below the poverty line (Table 6). Of this, more than 50% is required in Bihar and UP alone. Uttar Pradesh
alone calls for more than 30% of this required expenditure. Madhya Pradesh, West Bengal and Rajasthan along with UP and Bihar account for almost 80% of the requirement. In contrast, Tamil Nadu and Andhra Pradesh spend a substantial amount of their GSDP on nutrition and therefore do not need any additional expenditure. As a percentage of GSDP, Bihar, UP, MP, Orissa and Rajasthan occupy the top five positions in terms of requirement. Andhra Pradesh and Tamil Nadu have the highest expenditure both as a percentage of GSDP and in per capita terms, and therefore do not require any increase in expenditure. Kerala, Maharashtra, Punjab and Haryana require an increase of less than 0.2% of their GSDP. In terms of per capita, Bihar, UP, Orissa, MP and Rajasthan require substantial increase (Table 6). Thus, Bihar, UP, Orissa, MP and Rajasthan call for a special policy focus.
Primary Schooling
We now estimate the expenditure requirement for providing primary schooling to all children in the age group of 5-14 years in selected States. It is important to note that universalizing primary education is not only important for achieving health outcomes, but also has various other positive externalities. In fact, bringing all children to school is an MDG as well as a Tenth Plan goal by itself. The expenditure required for universalizing primary education therefore should not be seen as a requirement for achieving health outcomes alone. The capital cost for universalizing elementary education was estimated based on the report of the Expert Group on Financial Requirements for Making Elementary Education a Fundamental Right (GOI 1999). The estimates provided in the above study were modified for the number of schools built between 1993 and 2002 (based on the Sixth and Seventh School Education Survey). The revenue expenditure requirement was calculated based on an estimate of an average expenditure per child in primary school provided by the National Commission on Macroeconomics and Health (NCMH). As ear-
302
SECTION IV
Table 5 Expenditure requirement for providing nutritional supplements between 2005-06 to 2009-10
0.23 0.53 1.49 0.28 0.18 0.28 0.13 0.87 0.25 0.79 0.13 0.64 0.20 0.98 0.35
0.24 0.55 1.56 0.28 0.19 0.29 0.14 0.91 0.26 0.82 0.13 0.67 0.20 1.03 0.36
0.25 0.58 1.64 0.29 0.19 0.30 0.14 0.95 0.27 0.85 0.14 0.69 0.21 1.07 0.37
0.29 0.66 1.88 0.33 0.22 0.33 0.16 1.08 0.30 0.96 0.15 0.78 0.23 1.23 0.41
0.27 0.63 1.81 0.31 0.21 0.31 0.15 1.03 0.28 0.92 0.15 0.74 0.22 1.17 0.39
94 96 172 95 74 85 45 152 100 126 53 134 97 149 101
94 102 183 98 80 91 50 163 105 134 54 144 97 160 107
94 110 196 104 82 26 52 174 111 142 60 152 97 170 114
94 128 229 122 97 26 60 203 127 165 66 177 97 200 132
94 124 225 118 95 26 58 198 120 162 67 173 97 195 130
Fig 4 Per capita additional requirement of resources for primary schooling, 2009-10
Note: Resource requirements for Gujarat are high due to data problems
Fig 5 Per capita additional requirement of resources for construction of roads, 2009-10
lier, requirement of capital expenditure is distributed over a five-year period between 2005-06 and 2009-10 (Table 7). An additional amount of Rs 106,008 crore is required to provide all children with primary schooling (Table 8). Of these, the States of Bihar, UP and Gujarat account for the largest share. The high requirement of Gujarat is on account of a substantial decline in the number of primary schools in the State reported by the Sixth and the Seventh All India School Education Surveys.10 Apart from these States, MP and West Ben10 This is likely to be due to data problems.
gal call for a substantial increase in expenditure. The five States of Bihar, UP, Gujarat, MP and West Bengal account for more than 90% of the requirement. On the other hand, Tamil Nadu, Karnataka, Kerala and Maharashtra do not require any additional expenditure (Fig. 4). However, even in these States, all children are not in school and therefore one might need to identify the possible reasons for why these children have remained out of school and spend on appropriate heads required to bring these children to school.
303
SECTION IV
Table 6 Additional requirement of resources for providing nutritional supplements between 2005-06 to 2009-10
0.35 0.11 0.05 0.10 0.05 0.09 0.00 0.10 0.16 0.11 0.00 0.15 0.32 0.00 0.04
0 0.42 1.44 0.18 0.13 0.19 0.13 0.77 0.09 0.68 0.13 0.49 0 0.98 0.31
0 0.44 1.51 0.18 0.14 0.2 0.14 0.81 0.1 0.71 0.13 0.52 0 1.03 0.32
0 0.47 1.59 0.19 0.14 0.21 0.14 0.85 0.11 0.74 0.14 0.54 0 1.07 0.33
0 0.55 1.83 0.23 0.17 0.24 0.16 0.98 0.14 0.85 0.15 0.63 0 1.23 0.37
0 0.52 1.76 0.21 0.16 0.22 0.15 0.93 0.12 0.81 0.15 0.59 0 1.17 0.35
94 18 7 34 20 26 1 18 63 17 0 31 97 0 10
0 78 165 61 54 59 44 134 37 109 53 103 0 149 91
0 84 176 64 60 65 49 145 42 117 54 113 0 160 97
0 92 189 70 62 0 51 156 48 125 60 121 0 170 104
0 110 222 88 77 0 59 185 64 148 66 146 0 200 122
0 106 218 84 75 0 57 180 57 145 67 142 0 195 120
0 1,379 11,204 1,979 736 703 910 7,365 2,471 2,478 775 3,876 0 17,814 4,693 56,383
Table 7 Expenditure requirements for providing primary schooling to all children between 2005-06 to 2009-10
1.49 3.14 5.60 4.12 1.17 1.51 1.06 3.15 1.26 3.37 1.23 2.94 1.14 3.61 2.19
1.45 3.07 5.48 3.93 1.13 1.45 1.03 3.07 1.21 3.27 1.20 2.83 1.09 3.52 2.10
1.41 3.00 5.36 3.74 1.10 1.39 1.00 2.98 1.17 3.18 1.17 2.73 1.04 3.44 2.01
1.37 2.92 5.25 3.56 1.07 1.33 0.97 2.90 1.13 3.09 1.14 2.63 0.99 3.36 1.92
1.34 2.85 5.14 3.39 1.03 1.28 0.94 2.82 1.08 2.99 1.10 2.53 0.95 3.29 1.84
416 531 632 1386 450 439 419 567 707 537 494 628 392 546 595
406 527 624 1356 433 439 419 560 707 526 488 610 392 538 581
397 527 616 1322 429 439 419 550 707 515 481 593 392 531 566
386 527 609 1288 429 439 419 542 707 504 474 575 392 523 550
379 527 602 1254 429 439 419 533 707 491 462 555 392 516 535
304
SECTION IV
Table 8 Additional requirement of resources for providing primary schooling between 2005-06 to 2009-10
Additional As percentage of GSDP Current level State (2001-02) 2005-06 2006-07 2007-08 2008-09 2009-10 Current level (2001-02) 2005-06 2006-07 2007-08 2008-09 2009-10 Per capita (in Rs) (2005-06 prices) resources required (2005-10) (Rs in crore 2005-06) prices
1.08 3.12 2.62 1.41 1.12 1.54 1.40 1.56 1.82 2.25 0.63 2.10 1.29 1.97 1.01
0.41 0.02 2.98 2.71 0.05 0 0 1.59 0 1.12 0.6 0.84 0 1.64 1.18
0.37 0 2.86 2.52 0.01 0 0 1.51 0 1.02 0.57 0.73 0 1.55 1.09
0.33 0 2.74 2.33 0 0 0 1.42 0 0.93 0.54 0.63 0 1.47 1
0.29 0 2.63 2.15 0 0 0 1.34 0 0.84 0.51 0.53 0 1.39 0.91
0.26 0 2.52 1.98 0 0 0 1.26 0 0.74 0.47 0.43 0 1.32 0.83
291 527 290 466 429 439 419 289 707 358 251 452 392 296 250
124 4 342 920 21 0 0 277 0 179 243 176 0 249 345
115 0 334 890 4 0 0 270 0 167 237 158 0 241 332
105 0 326 856 0 0 0 261 0 157 230 141 0 234 316
95 0 319 822 0 0 0 253 0 146 223 123 0 226 300
88 0 312 788 0 0 0 244 0 132 211 103 0 220 285
4,338 11 18,782 23,037 54 0 0 11,963 0 3,006 2,956 4,321 0 23,728 13,811 106,008
Table 9 Additional requirement of resources for connecting all habitations by road between 2005-06 to 2009-10
State 2005-06 As a percentage of GSDP 2006-07 2007-08 2008-09 2009-10 Real per capita (in Rs) (2005-06 prices 2005-06 2006-07 2007-08 2008-09 2009-10 Additional resources required (2005-10)(Rs in crore 2005-06) prices
0.06 1.90 1.69 0.16 0.00 0.05 0.01 2.96 0.09 3.33 0.03 0.81 0.07 0.82 0.57
0.06 2.02 1.79 0.17 0.00 0.05 0.01 3.11 0.09 3.53 0.03 0.85 0.07 0.86 0.59
0.07 2.12 1.88 0.17 0.00 0.05 0.01 3.25 0.09 3.71 0.03 0.88 0.07 0.90 0.61
0.07 2.22 1.97 0.18 0.00 0.05 0.01 3.38 0.10 3.87 0.03 0.90 0.08 0.93 0.63
0.07 2.30 2.04 0.18 0.00 0.05 0.02 3.48 0.10 4.02 0.03 0.92 0.08 0.96 0.64
18 355 194 54 0 16 3 517 37 532 12 170 22 125 167
19 385 209 60 0 16 4 557 38 579 12 184 23 134 179
22 413 224 62 0 17 4 597 39 626 13 197 24 143 193
23 442 239 69 0 18 4 637 45 672 13 209 29 152 208
24 469 252 72 0 18 8 674 47 718 13 221 30 160 220
873 6,061 12,902 1,714 0 489 77 27,419 2,063 12,065 166 6,085 855 14,513 8,485 93,765
305
SECTION IV
Fig 6 Per capita additional requirement of resources in health and related sectors, 2009-10
Note: The requirements for Gujarat are affected by problems in data on schooling.
Fig 7 Per capita additional requirement of resources in health and related sectors and per capita income of states
Roads
Recognizing the importance of roads, the Prime Ministers Gram Sadak Yojana (PMGSY) was introduced in December 2000. This scheme aims to connect all rural habitations by roads. The cost is based on information provided by PMGSY. As in the case of primary schooling, apart from the positive impact on health outcomes, expenditure on roads has other positive externalities too. This expenditure therefore should not be treated as an expenditure that is exclusively directed towards health outcomes. As of March 2004, Bihar, UP, MP and West Bengal had the highest number of unconnected habitations among the 15 States. Table 9 shows that these four States account for approximately two-thirds of the total requirement of resources. If one includes the requirement for Orissa, the total share of resources required in these States increases to more than 80%. For all selected States taken together, a sum of about Rs 93,765 crore is required for connecting all habitations by roads. While in absolute terms, the five States of Bihar, UP, MP, West Bengal and Orissa account for the largest share, as a percentage of GSDP, the States of Orissa, Assam, Bihar and MP require significant increases. Table 10 shows the total additional requirement of resources in different sectors from 2005-06 to 2009-10. For health, water, sanitation and nutrition alone, a total of Rs 100,415 crore is required over the next five years. If one adds up the expenditure for primary schooling and roads, the requirements almost triple. The total combined requirement of all sectors is of the order of Rs 300,188 crore. The requirement for primary schooling alone is more than the combined requirement of health, water, sanitation and nutrition. If one focuses on the low-income States of Assam, Bihar, Orissa, MP and UP, which is just around the average of the selected States, the requirements are of the order of Rs 199,730 crore (Table 10). Even if one focuses only on health, water, sanitation
and nutrition, the requirements are about Rs 70,000 crore. Figure 6 shows that excluding Gujarat, whose requirements are primarily determined by high requirements for primary schooling (mainly due to data problems), MP, Orissa, Bihar, Assam, West Bengal and UP occupy the top positions in terms of additional per capita requirements in the terminal year 2009-10. In fact, UP, MP and Bihar account for more than 50% of the additional requirement of resources (Table 10). Even if one focuses only on health, water, sanitation and nutrition, Bihar, UP and MP occupy the top positions. These States therefore require a special policy attention. The requirement of Kerala is primarily determined by its high requirement for safe drinking water. In contrast, Karnataka, Tamil Nadu, Haryana, Andhra Pradesh and Maharashtra occupy the lowest positions in terms of additional requirements. It is interesting to note that the additional expenditure requirements are particularly high in States with low per capita GSDP. Figure 7 indicates the association of the additional per capita expenditure requirements in the year 2009-10 with current (2002-03) per capita income of States. Low-income States are also the ones with high poverty (the correlation of per capita income with poverty in States is more than 0.8). Thus, in general, lower the income level of a State, higher is its expenditure requirement for health outcomes. This point underlines the importance of expenditures on anti-poverty programmes including employment creation and incomegeneration activities, particularly in States with higher concentration of poverty. The estimated additional requirement has to be met either with additional mobilization of resources at the State-level or through Central transfers. The next section assesses the extent of resources that can be mobilized at the State level.
306
SECTION IV
Table 10 State-wise additional requirement of resources in health and related sectors between 2005-06 to 2009-10 (Rs in crore) at 2005-06 prices
I States Health sector Water andsanitation Nutrition Total(I) Primary schooling II Roads Total(II) Total(I+II)
944 976 7,150 634 554 10 0 2,983 223 1,210 405 990 612 8,463 1,286 26,439
0 1,349 897 0 0 415 3,532 1,842 2,455 2,336 175 300 0 1,834 2,459 17,593
0 1,379 11,204 1,979 736 703 910 7,365 2,471 2,478 775 3,876 0 17,814 4,693 56,383
944 3,704 19,251 2,613 1,290 1,128 4,442 12,190 5,149 6,024 1,355 5,166 612 28,111 8,438 100,415
4,338 11 18,782 23,037 54 0 0 11,963 0 3,006 2,956 4,321 0 23,728 13,811 106,008
873 6,061 12,902 1,714 0 489 77 27,419 2,063 12,065 166 6,085 855 14,513 8,485 93,765
5,211 6,072 31,684 24,751 54 489 77 39,382 2,063 15,071 3,122 10,406 855 38,241 22,296 199,773
6,155 9,776 50,935 27,364 1,344 1,617 4,519 51,572 7,212 21,095 4,477 15,572 1,467 66,352 30,734 300,188
Mobilization of resources by States to meet additional resource requirements for health and related expenditure
To identify the extent to which resources can be mobilized at the State level to meet health requirements, two possibilities have been explored: first, reprioritization and reallocation of the existing resources towards health and second, generation of additional revenues.
Reallocation of resources
To examine the extent of reallocation possible, expenditures in States are classified into two groups: committed and discretionary (non-committed). Committed expenditures are those for which the States are assumed to assign high priority and are committed to spend on. Although all expenditures may be considered discretionary in the long term, the distinction is legitimate in the short and medium term and the policy-makers are always confronted with this distinction. For the purpose of this analysis, it is assumed that States are committed to meet the expenditure on wages and salaries, interest payments and pensions on a priority basis and that the resources used up for meeting these expenditures cannot be reallocated. The residual revenue that remains with States after meeting expenditures on wages and salaries, interest payments and pensions is termed as discretionary and is assumed to be available for reallocation towards health. An analysis of the extent of discretionary resources available
11 Data on salary expenditure between 1994-95 and 2002-03 have been taken from the TFC.
with States calls for an evaluation of the extent of expenditure on pensions, interest payments and salaries that would be incurred in each State during the next five years. Such an assessment has been carried out for interest payments and pensions by the Twelfth Finance Commission (TFC) using various assumptions on the States capability to contain these expenditures. While the projections of interest payments and pensions in the States by the TFC may seem to be on the lower side relative to what it would be if the past rate of growth of these expenditures continued, the TFC estimate provides a benchmark for these expenditures, which the States should strive to achieve. Given the objective of this exercise to arrive at the maximum discretionary resources available with the States, these benchmark estimates for interest payments and pensions have been used in this analysis. The salary expenditure, however, is likely to be difficult to contain in the immediate future. Although the TFC has suggested that States should attempt to achieve the ratio of salary expenditure to revenue expenditure at 199697 levels, these levels may be difficult to achieve in the next five years. Salary expenditures in this analysis are therefore projected based on their growth rate between 1994-95 and 2002-03.11 An examination of the share of committed expenditures in total revenues over the next five years (Table 11) indicates that in many cases, a large portion of States revenues will be used up for meeting the committed expenses, leaving very little for discretionary expenditure.12 States such as Assam, Orissa, Bihar, Punjab and West Bengal are unlikely to have any resources available for discretionary expenditure in the next five years. In five
12 Projections of total revenues in States were based on the past growth rate of revenues between 1993-94 and 2002-03 based on data provided by the TFC
307
SECTION IV
Table 11 Committed expenditure as percentage of the total revenues in States between 2005-06 and 2009-10
State 2005-06 2006-07 2007-08 2008-09 2009-10
Andhra Pradesh Assam Bihar (including Jharkand) Gujarat Haryana Karnataka Kerala Madhya Pradesh (including Chhattisgarh) Maharashtra Orissa Punjab Rajasthan Tamil Nadu Uttar Pradesh (including Uttaranchal) West Bengal
70.38 105.58 101.69 46.41 79.28 71.17 96.36 76.37 92.80 105.26 117.02 95.79 83.15 86.64 138.88
68.50 108.01 101.25 45.29 79.75 70.87 95.63 77.75 94.08 104.97 117.60 95.90 82.93 85.41 139.17
66.69 110.54 100.86 44.21 80.11 70.59 94.92 79.22 95.44 104.76 118.33 96.09 82.72 84.25 139.60
64.95 113.17 100.49 43.17 80.43 70.32 94.23 80.78 96.87 104.62 119.21 96.36 82.53 83.13 140.18
63.27 115.90 100.16 42.15 80.72 70.06 93.56 82.44 98.38 104.54 120.25 96.69 82.34 82.07 140.91
Table 12 State-wise additional resources that can be directed towards health, family welfare, water supply, sanitation and nutrition by reallocating 5% of discretionary resources
Percentage of GSDP State 2005-06 2006-07 2007-08 2008-09 2009-10
Andhra Pradesh Assam Bihar (including Jharkand) Gujarat Haryana Karnataka Kerala Madhya Pradesh (including Chhattisgarh) Maharashtra Orissa Punjab Rajasthan Tamil Nadu Uttar Pradesh (including Uttaranchal) West Bengal
0.20 0 0 0.37 0.13 0.18 0.02 0.20 0.04 0 0 0.03 0.12 0.10 0
0.22 0 0 0.37 0.13 0.18 0.03 0.18 0.03 0 0 0.03 0.12 0.11 0
0.23 0 0 0.38 0.13 0.18 0.03 0.17 0.02 0 0 0.03 0.11 0.12 0
0.25 0 0 0.38 0.13 0.18 0.04 0.16 0.02 0 0 0.02 0.11 0.12 0
0.27 0 0 0.38 0.13 0.18 0.04 0.14 0.01 0 0 0.02 0.11 0.13 0
out of the remaining ten States, committed expenditure will use up more than 80% of their resources in the recent future. It is disturbing to note that States which have a high requirement of health expenditure are particularly stressed in terms of availability of resources for reallocation towards health. Whatever discretionary resources are available, this analysis assumes that at the most 5% of resources available for discretionary expenditures in the years 2005-06 to 2009-10 can be reprioritized towards health. This would mean that 5% of the discretionary resources would have to be extracted from non-health sectors and reallocated towards health. Diversion of resources from non-health to health sectors however would require a detailed cost-benefit analysis of expenditure on various sectors and needs to be carefully worked out.13 The additional resources that can be reallocated towards health, family welfare, water supply and sanitation through 5% reallocation of discretionary expenditures are shown in Table 12. It is evident that the extent of additional resources that can be directed towards health, family welfare, water supply, sanitation and nutrition through reallocation of discretionary resources are limited at the moment. This, however, does not mean that over time, it is not possible to take appropriate measures to reprioritize expenditure in favour of these sectors. The TFC has pointed out that the debt situation is particularly bad in Bihar, Himachal Pradesh, Kerala, Orissa, Punjab, Rajasthan, UP and West Bengal, which has led to high interest payments in these States. Debt rescheduling and a reduction in interest rates would provide some relief to the State governments. The incentive-based debt write-off too will help the State governments in exercising fiscal prudence and reducing the revenue deficits. Effects of such measures, however, are likely to fructify only in the long run.
Generation of additional revenue

States can generate additional revenue either through tax or non-tax sources. In this section, the potential for generating additional tax revenue is examined. The issue of tax potential has attracted the attention of researchers in the past. At one level, there are some scholars such as Colin Clarke who preferred to make judgements about tax revenue that should/could be mobilized and he suggested that the ratio of 25% of GDP as the norm. In contrast, e.g. Musgrave has suggested that absolute taxable capacity is a myth and specifying this involves making arbitrary judgements. Therefore, the scholars should be concerned with optimal budgets which meant that each country should determine deci-
Note: Projections of GSDP have been made using the prescriptive growth rates suggested by the Twelfth Finance Commission
13 The 5% of discretionary resources that can be reallocated towards health is over and above the discretionary resources already allocated towards health.
308
SECTION IV
sions to raise revenues depending on the degree of market failure and the extent of state intervention envisaged. Here again, Musgrave suggests the need to make a crucial difference between public provision and public production of services (Musgrave 1973). While absolute taxable capacity is difficult to conceptualize and impossible to measure in any objective sense, Musgrave (1959) emphasized the relevance and importance of relative taxable capacity. This can be estimated by comparing different countries or sub-national units in a federation. Thus, two countries or sub-national units in a country which are similar in economic circumstances should be able to generate equal amount of revenue and the differences could then be attributed to the differences in their preference patterns. Thus taxable capacity of different units in a federation can be estimated by estimating the average behaviour of the States in raising revenues after controlling for economic factors that can cause differences in taxable capacity. Thus, taxable capacity of a country/State is defined as the revenue it can generate if it levied an average effective rate of tax on its base (Bahl 1971, 1972). Alternatively, one can also specify and estimate taxable capacity with respect to the highest effective tax rate or any other exogenously specified effective tax rate. Given that the ability to raise tax revenues may be more than proportionately higher in a more developed country/State, the effective tax rate will have to be determined with respect to the development of a particular State and a simple average would not serve the purpose. This, therefore, has to be estimated using statistical techniques to take account of the non-linear relationship between the level of development and taxable capacity. Variations (variance) in tax revenues between different States (t2) may be due to variations in their capacity to raise revenues (tc2) or variations in the efforts put in by them (te2). t2 = tc2 + te2 ..............................(1) If one were able to identify all the factors that contributed to taxable capacity variation, it would be possible to estimate it. Alternatively, if one controlled for variations in tax effort among States, it would be possible to derive their taxable capacity. There are three alternative methods employed to estimate taxable capacities of the States. These are: (i) aggregate regression (AR) approach; (ii) representative tax system (RTS) approach; (iii) tax frontier (TF) approach. Appropriateness of a method to be employed to estimate taxable capacity depends on the availability of disaggregated data, the extent to which the relationship between taxable capacity and the variables representing it are perceived to be non-linear, and the degree of interdependence of the tax base with tax rate. It is useful to discuss the three methods used in some detail.
tions in taxable capacity. Thus, tax-GSDP ratio or per capita tax revenue of the States are regressed on taxable capacity variables. Taxable capacity variables essentially represent the variables representing the tax bases or their proxies. This can be done in a cross-section model or, in order to get greater degrees of freedom, by combining cross-sections in a covariance model. The estimated parameters of the equation provide behavioural relationship between tax-GSDP ratio (or per capita tax revenue) and various capacity factors estimated in the equation. If it is hypothesized that the taxable capacity is a non-linear function of taxable capacity variables, it is possible to make the hypothesized functional specification in the model. Once the behavioural relationship is estimated, it is easy to estimate the taxable capacity by substituting the actual values of the taxable capacity variables in the equation. The estimated coefficient for each capacity variable gives the average behavioural relationship and substituting the actual capacity variables provides the estimate of taxable capacity of each State. The estimation of tax capacity above assumes that the coefficients of the respective bases (which indicate the average effective rate at which the bases are used across States) represent the normative rates at which States ought to raise taxes. The residual term, which is the difference between the actual tax revenue and the estimated tax capacity, is then used to indicate the tax effort of the respective States. There are a number of shortcomings in this approach. First, it may not be able to include exhaustible list of taxable capacity factors and, therefore, the unexplained variation, which is attributed to tax effort may actually be due to omitted variables. Second, even if it is assumed that all taxable capacity factors are included, the residual variation is the combination of variations in tax effort and the random error term and to attribute it entirely to tax effort may not be appropriate. Finally, some variables may impact on both taxable capacity and tax effort and it may not always be possible to isolate the effect of capacity from effort variables. Thus, higher per capita GSDP or urbanization in State may also represent better organization of the economy and ensure greater effort. Later studies have tried to improve upon this implicit assumption by separating out the effect of tax effort of individual States from the random error element by combining crosssection observations over time and introducing State-specific fixed effects in the regression specification using panel data (First Report of the Ninth Finance Commission 1988, Condoo et al. 2000). However, it is important to note that the State-specific (fixed) effect may also be due to a variety of other factors and not entirely due to tax effort. Any omitted variable that is specific to the State and changes slowly (or does not) over time will also be captured by the State-specific fixed effect. Hence, what portion of the State-specific fixed effect can be attributed exclusively to tax effort may be an arguable issue.
Aggregate regression approach Representative tax system approach

In the AR method, the actual tax revenue (termed as tax performance) is regressed on all factors representing variaThe representative tax system (RTS) approach to measuring
309
SECTION IV
taxable capacity was first employed by the Advisory Commission on Intergovernmental Relations (ACIR) in the United States. In this approach, taxable capacity is estimated for each of the taxes levied by the States. The taxable capacity of each tax is estimated by applying the representative rate to the tax base of the State. The representative rate is the average effective rate of each of the taxes levied in States. This is estimated by dividing all States revenue collection from the tax with the sum of the value of the tax base over all the States. As in the AR approach, this assumes that the average effective tax rate of the States is the normative rate at which the States ought to levy. The taxable capacity of different taxes is summed to arrive at the aggregate taxable capacity of a State. The ratio of actual tax collection to the tax capacity (as estimated above) then provides an indicator of the relative tax efforts of different States. The major shortcoming of this approach is that it assumes that individual tax bases are independent of each other (Second Report of the Ninth Finance Commission). Second, the approach assumes that tax bases and rates are independent of each other and the average effective rates adequately capture the non-linear relationship between the tax bases and rates (Sen and Tulasidhar 1988). Besides, the data requirement for applying this approach is large and in most cases disaggregated data on various tax bases or even their close proxies are simply not available. The method is also suitable only when there is significant homogeneity in the tax structures (Chelliah and Sinha 1982).
Tax frontier approach

In the tax frontier (TF) approach, the taxable capacity of States is conceived as a production frontier and the distance from the frontier is considered as the tax effort. Thus, technical efficiency is interpreted as the tax efficiency of States or the tax effort. The main difference of the TF approach with the AR and the RTS approach is in the way in which the normative rate for estimating tax capacity is indexed. While in the TF approach the normative rate is equated with the highest rate, it is the average rate that is used as the norm in the AR and RTS approaches. The TF approach has however been criticized on the grounds that the formulation of tax capacity as a production frontier is ill-conceived. It is argued that unlike firms, whose objective is to maximize profits, the primary objective of States is not to maximize tax revenue (Coondoo et. al. 2000). Thus, all the existing methods to measuring taxable capacity and effort have shortcomings. In addition, there is a serious problem in the States tax system in India which prevents the objective assessment of the taxable capacities of the States. It must be noted that States sales taxes, which contribute to about two-thirds of own tax revenues, are not destination based. The system of cascading sales taxes coupled with the levy of inter-State sales tax results in significant inter-State tax exportation (Rao and Singh 2005). When there is full for-
ward shifting of the tax, inter-State tax exportation is from the richer to poorer States. Thus, tax revenues collected by State Governments include collections from non-residents. In this exercise, we have used the AR approach to measure taxable capacity of the States with some modifications. As the emphasis is on generating additional revenue to create fiscal space for financing incremental expenditure in the health sector, the study first tries to project tax revenues at average effort and then tries to measure the revenue gains through increase in the effort itself. As mentioned earlier, the relative taxable capacity using the regression approach is estimated by regressing the variables representing the tax bases and their proxies on the taxGSDP ratio of the States in cross-section regression. Apart from tax bases, it also requires the identification of other factors that facilitate revenue collections, particularly those representing organization of the economy. Earlier studies have used various indicators to estimate tax performance. The most common indicator that has been used in almost all studies on the issue is the State income (Nambiar and Rao 1972, Sen 1983, Oommen 1987, Finance Commission 1988, Coondoo et al. 2000). Along with the State income, Oommen (1987) also used its components such as the proportion of income from agriculture, proportion of income from manufacturing and proportion of income from hotels, trade and commerce to explain variation in tax performance. However, due to the inclusion of individual components of State income, the variable for aggregate State income was insignificant (possibly due to multicollinearity problems) and was later dropped. Oommen (1987) argued that income from hotels, trade and commerce would affect the sales tax revenue while income from manufacturing would affect both the sales and excise tax revenue. Nambiar and Rao (1972), Sen (1983) and Finance Commission (1988) also used non-agricultural income and non-primary sectoral SDP in addition to State income to explain tax performance. However, these variables are components of State income causing multicollinearity problems.14 Sen (1983) also used the percentage of population below poverty line. Also, Coondoo et. al. (2000) used per capita bank deposits and per capita power consumption of States in addition to State income. Apart from these variables, Nambiar and Rao (1972) and Sen (1983) used the degree of urbanization, Finance Commission (1988) used inequality of consumption expenditure (indicated by Lorenz ratio) and Coondoo et. al. (2000) used the proportion of SC and ST population to explain tax performance across States. Based on the above studies, our model employs the four commonly used determinants of taxable capacity namely: per capita State Domestic Product (SDP), share of manufacturing SDP, headcount measure of poverty and urbanization. Per capita SDP has been used in almost every study on taxable capacity. Given the level of per capita SDP, the share of nonprimary sector SDP or manufacturing SDP has been used to capture the effect of industrialization. The inclusion of poverty
14 Finance Commission (1988) included both State income and non-primary sectoral SDP in the regression equation. Possibly due to the multicollinearity, they found that while the coefficient of
State income was significant, the coefficient of non-primary sectoral SDP was insignificant.
310
SECTION IV
Table 13 Regression results using panel data from 1995-96 to 2002-03

Model 1 Model 2 Model 3
Per capita GSDP Interaction (per capita GSDP* group_dummy) Urbanization West Bengal Uttar Pradesh (including Uttaranchal) Tamil Nadu Rajasthan Punjab Orissa Maharashtra Madhya Pradesh (including Chhattisgarh) Kerala Karnataka Haryana Gujarat Bihar (including Jharkhand) Assam Andhra Pradesh F-test for no fixed effects
**Significant at 1% *Significant at 5%
0.081(20.096)** 26.948(2.328)* -1326.502(-4.415)** -885.430(-3.687)** -876.039(-2.284)* -979.688(-3.696)** -1241.804(-3.871)** -765.172(-4.413)** -1350.57(-3.082)** -999.614(-3.548)** -798.430(-2.641)** -892.642(-2.516)* -907.476(-3.320)** -1136.767(-2.915)** -681.52(-4.206)** -744.811(-5.888)** -1013.684(-3.193)** 29.279**
.087(22.537)** -0.037(-4.893)** 30.053(2.866)** -954.297(-3.382)** -648.983(-2.917)** -1120.289(-3.197)** -642.722(-2.577)* -1505.611(-5.103)** -506.074(-3.058)** -1641.411(-4.096)** -709.784(-2.713)** -1021.920(-3.686)** -1117.584(-3.448)** -1146.720(-4.551)** -1391.781(-3.904)** -514.970(-3.426)** -457.366(-3.556)** -1215.839(-4.192)** 13.169**
1.109(18.921)** 0.945(3.696)** -7.323(-10.215)** -6.906(-10.348)** -6.937(-9.040)** -6.937(-10.022)** -7.115(-9.602)** -6.692(-10.948)** -7.283(-9.145)** -7.000(--9.992)** -6.789(-9.385)** -6.902(-9.188)** -6.859(-9.641)** -7.107(-9.209)** -6.727(-11.414)** -6.569(-11.671)** -6.990(-9.595)** 59.398**
has been primarily to measure income distribution. Urbanization has been used to denote the organization of the economy and the extent of monetized transactions that could be taxed. While these four indicators were used as explanatory variables in the model, either the tax-GSDP ratio or per capita tax has been employed as the dependent variable. Given that the objective of this exercise is to make future projections of tax revenue, per capita tax revenue (which gives a better fit of the model) is used as the dependent variable.15 Of the various capacity variables, after the 1990s, data on poverty ratio is available only for 1993-94 and 1999-2000. Further, regression estimates for these years showed that poverty was highly correlated with GSDP and the share of manufacturing sector GSDP with total GSDP. The model with only GSDP and urbanization had the highest explanatory power. Therefore, a pooled model using data for the period 1995-96 to 2002-03 was estimated using State-specific fixed effects. While GSDP figures were available from the TFC (based on CSO), actual figures of urbanization were not readily available. However, projected urbanization estimates of the Registrar General (Census of India 1991) were employed to estimate the model.16 The specification of the panel data model including the cross-section observations for the years 1995-96 to 200203 was as follows:
15
Per capita tax revenue = + 1 (per capita GSDP)it + 2 (urbanization)it + uit where 1 = State-specific effect for the ith State As in the OLS model, results in the pooled model including State-specific fixed effects indicated that both per capita GSDP and urbanization had a significant effect on per capita tax revenue (Model 1 in Table 13). The above regression specification was further modified keeping in view the first report of the Ninth Finance Commission, which highlighted that the slope coefficients of the tax function were homogeneous within similar income groups but not across groups. States were classified into relatively high and low income groups and an interaction term of per capita GSDP and the dummy variable distinguishing the two groups was included in the regression specification to account for any differences of slopes between the two groups. The dummy variable assumed the value of 1 if a State belonged to the lower income group and zero otherwise. Results indicated that the effect of per capita GSDP on tax revenue was higher for States with relatively higher income (Model 2 in Table 13). To take into account the non-linearity in the relationship, the model was re-estimated in the log linear form. The model in the log form was
The source of per capita SDP was CSO, poverty figures from Sen and Himanshu (2004) and urbanization figures from NSSO. rank correlation is 1.
16 Analysis of the projected values of urbanization compiled by the Registrar General and the actual census figures of 2001 show that the correlation between the two was about 0.97 and the
311
SECTION IV
Table 14 Comparison of own tax revenue projections (as percentage of the GSDP)
Present study State 2005-06 2006-07 2007-08 2008-09 2009-10 2005-06 Twelfth Finance Commission 2006-07 2007-08 2008-09 2009-10
8.0 5.4 5.1 7.6 9.4 9.3 10.8 8.0 8.2 7.1 7.6 7.4 9.8 6.9 4.8
8.4 5.6 5.2 7.5 9.7 9.4 11.3 8.1 8.2 7.3 7.7 7.4 9.8 7.0 4.9
8.8 5.7 5.3 7.4 9.9 9.5 11.8 8.2 8.3 7.5 7.7 7.5 9.8 7.2 5.0
9.2 5.9 5.4 7.3 10.2 9.6 12.3 8.3 8.3 7.7 7.7 7.6 9.7 7.3 5.1
9.6 6.0 5.5 7.2 10.5 9.8 12.9 8.4 8.3 7.9 7.7 7.6 9.7 7.4 5.1
8.0 5.9 6.3 7.8 9.4 9.9 9.5 8.1 8.3 7.2 8.0 7.6 10.1 7.1 5.8
8.1 6.0 6.4 8.1 9.6 10.3 9.8 8.2 8.6 7.4 8.3 7.8 10.3 7.3 6.0
8.3 6.1 6.5 8.3 9.9 10.6 10.1 8.4 8.8 7.5 8.5 8.0 10.6 7.5 6.2
8.5 6.2 6.7 8.6 10.2 11.0 10.4 8.6 9.0 7.6 8.8 8.2 10.8 7.6 6.5
8.6 6.4 6.8 8.9 10.4 11.4 10.7 8.8 9.3 7.8 9.1 8.3 11.1 7.8 6.8
used for projecting future tax revenues, specifically for the period 2005-06 to 2009-10 (Model 3 in Table 13). The projection of taxable capacity from 2005-06 to 200910 was made by substituting the actual values of taxable capacity variables in the equation. For the same period, projections of own tax revenues were also made based on the past trend from 1993-94 to 2002-03. The higher of the two estimates was used to indicate the likely generation of own taxes across States between 2005-06 and 2009-10. It may be noted that at the past rate, four States-Gujarat, Kerala, Karnataka and West Bengal will fall short of the projections made through the regression model and will have to generate additional taxes to reach the levels predicted by the model. A comparison of these projections with the TFC projections shows that, in general, the latter are on the higher side (Table 14). In particular, this is true for States for which the requirement of resources for health expenditure is particularly high. Given our objective to estimate the maximum own tax revenues that the States can possibly generate, one may be hopeful of achieving the higher of the two projections, i.e. the TFC projections of own taxes. We therefore use the TFC projections of own tax revenue to calculate the additional own tax revenues that can be generated in the States from 200506 to 2009-10 (Table 15). It is important to note that States, which will be unable to meet the committed expenditures in the projected period, will have to generate additional revenues to meet their committed liabilities in addition to their revenue generation for health expenditures. The additional own tax revenues generated in the States will be distributed across different sectors and therefore cannot be entirely allocated towards health and health-related sectors. The National Health Policy 2002 has set a goal of
spending 7% of State budgets to the health sector. Based on this, we assume that 7% of the additional own tax revenues generated can be directed towards health. Another 3% of this additional revenue is assumed to be directed towards primary schooling. The resources out of additional own tax revenues that can be allocated towards health and related sectors and the corresponding deficits at 2005-06 prices are shown in Table 16 and Table 17. As the requirement of resources for roads deals with the requirements of PMGSY, which is a Centrally Sponsored scheme, we do not assume any additional allocation towards roads at the State level. Figures indicate that a total amount of Rs 38,758 crore can be additionally allocated between 2005-06 and 2009-10 towards health and related sectors. This is about 13% of total requirement of resources in the above period. Even if one concentrates on the requirement of health, water, sanitation and nutrition alone, the total amount that can be additionally allocated at the State level is about Rs 31,557 crore. Given the constraints on resources, if one wishes to focus only on these sectors in the six States whose per capita additional requirements were relatively high, viz. Bihar, Assam, Orissa, MP, West Bengal and UP, the deficit at the State level is around Rs 66,812 crore (Table 17). The States of Bihar, Assam, Orissa, UP and MP not only have a high requirement of health expenditures, but also have a relatively low capability of generating additional revenues and therefore have a high deficit. Possibly recognizing this, the TFC has provided additional grants for health expenditures specifically to the States of Assam, Bihar, Jharkhand, MP, Orissa, UP and Uttaranchal to equalize the health expenditures within the special and non-special category States. The TFC has also allocated additional grants for equalizing education expen-
312
SECTION IV
Table 15 Additional own tax revenue projections

As percentage of GSDP State 2005-06 2006-07 2007-08 2008-09 2009-10 2005-06 At current prices (Rs in crore) 2006-07 2007-08 2008-09 2009-10
0.70 1.39 1.83 0.42 1.01 1.39 1.57 2.37 0.58 1.62 1.35 1.37 1.13 1.62 1.54
0.80 1.49 1.93 0.72 1.21 1.79 1.87 2.47 0.88 1.82 1.65 1.57 1.33 1.82 1.74
1.00 1.59 2.03 0.92 1.51 2.09 2.17 2.67 1.08 1.92 1.85 1.77 1.63 2.02 1.94
1.20 1.69 2.23 1.22 1.81 2.49 2.47 2.87 1.28 2.02 2.15 1.97 1.83 2.12 2.24
1.30 1.89 2.33 1.52 2.01 2.89 2.77 3.07 1.58 2.22 2.45 2.07 2.13 2.32 2.54
1712 738 2330 749 910 2438 1836 3664 2312 979 1383 1710 2333 4780 3845
2172 878 2728 1448 1221 3542 2427 4266 3929 1221 1876 2211 3098 6011 4900
3013 1040 3185 2087 1706 4665 3126 5152 5401 1430 2335 2811 4283 7467 6163
4014 1227 3884 3122 2291 6269 3949 6188 7169 1670 3012 3529 5424 8772 8027
4,827 1,523 4,504 4,388 2,849 8,207 4,916 7,396 9,911 2,037 3,809 4,183 7,121 10,744 10,267
Table 16 Additional revenues that can be directed towards health and related sectors (Rs in crore) at current prices
Total (through 5% reallocation and directing part of From additional tax revenues State 2005-06 2006-07 2007-08 2008-09 2009-10 2005-06 the additional own tax revenues) 2006-07 2007-08 2008-09 2009-10
171.2 73.8 233 74.9 91 243.8 183.6 366.4 231.2 97.9 138.3 171 233.3 478 384.5
217.2 87.8 272.8 144.8 122.1 354.2 242.7 426.6 392.9 122.1 187.6 221.1 309.8 601.1 490
301.3 104 318.5 208.7 170.6 466.5 312.6 515.2 540.1 143 233.5 281.1 428.3 746.7 616.3
401.4 122.7 388.4 312.2 229.1 626.9 394.9 618.8 716.9 167 301.2 352.9 542.4 877.2 802.7
482.7 152.3 450.4 438.8 284.9 820.7 491.6 739.6 991.1 203.7 380.9 418.3 712.1 1074.4 1026.7
660 74 233 735 208 560 207 676 391 98 138 208 481 773 385
814 88 273 889 253 710 282 737 527 122 188 263 589 964 490
994 104 319 1071 318 868 356 843 640 143 234 329 717 1190 616
1238 123 388 1285 394 1080 459 964 829 167 301 389 868 1374 803
1485 152 450 1536 469 1332 563 1077 1054 204 381 459 1080 1676 1027
ditures to the States of Assam, Bihar, Jharkhand, MP, Orissa, Rajasthan, UP and West Bengal. The total grant under these two heads at 2005-06 prices is about Rs 13,927 crore. The State-wise grants and the deficit even after the TFC grants for health and education are shown in Table 18. It must be noted that the release of additional grants from the TFC has been tied to various conditions. In general, the release of grants for health and education has been made con-
ditional on States meeting the Commissions projections for non-plan revenue expenditure (NPRE) on health and education. Given that these projections of NPRE is higher than what would be achieved if the past growth rate of NPRE on these sectors continued, fiscally stressed States may find it difficult to actually access these grants. Even if one assumes that the concerned States will be able to access the TFC grants, there is still a substantial amount
313
SECTION IV
Table 17 State-wise deficit of resources in health and related sectors, 2005-06 to 2009-10 (Rs in crore) at 2005-06 prices
State Requirement Total Additional allocation Deficit Requirement Health* Additional allocation Deficit
*Includes, health, water, sanitation and nutrition
6,155 9,776 50,935 27,364 1,344 1,617 4,519 51,572 7,212 21,095 4,477 15,572 1,467 66,352 30,734 300,188
4,433 463 1,427 4,721 1,401 3,879 1,585 3,710 2,923 629 1,054 1,409 3,191 5,114 2,819 38,758
1,722 9,313 49,508 22,643 0 0 2,934 47,862 4,289 20,466 3,423 14,163 0 61,238 27,915 261,430
944 3,704 19,251 2,613 1,290 1,128 4,442 12,190 5,149 6,024 1,355 5,166 612 28,111 8,438 100,417
4,033 324 999 4,426 1,174 3,243 1,171 3,024 2,199 440 738 1,040 2,626 4,146 1,973 31,557
0 3,380 18,252 0 116 0 3,271 9,166 2,950 5,584 617 4,126 0 23,965 6,465 77,892
Table 18 Additional (conditional equalization) grants provided to individual States for meeting health and education expenditures by the Twelfth Finance Commission (TFC) and the deficit after using the TFC grant (Rs in crore) 2005-06 prices
State Deficit before TFC transfer for health* Deficit before TFC transfer (Total) Health Education Total Deficit after TFC transfer for health Deficit after TFC transfer (Total)
*Includes, health water, sanitation and nutrition
0 3,380 18,252 0 116 0 3,271 9,166 2,950 5,584 617 4,126 0 23,965 6,465 77,892
1,722 9,313 49,508 22,643 0 0 2,934 47,862 4,289 20,466 3,423 14,163 0 61,238 27,915 261,430
0 829 1881 0 0 0 0 169 0 163 0 0 0 2068 0 5110
0 960 2891 0 0 0 0 398 0 280 0 88 0 3861 340 8818
0 1,789 4,772 0 0 0 0 567 0 443 0 88 0 5,928 340 13,927
0 2,551 16,371 0 116 0 3,271 8,997 2,950 5,421 617 4,126 0 21,897 6,465 72,782
1,722 7,524 44,736 22,643 0 0 2,934 47,295 4,289 20,023 3423 14,075 0 55,310 27,575 247,503
314
SECTION IV
of deficit in the requirement of resources for meeting health goals (Table 18). These additional resources have to be met by other Central transfers. The next section discusses the nature of Central transfers that would be required and the norms that should be followed for Central transfers to States for meeting the resource gap.
Central transfers to States

The above analysis shows that State Governments will have to augment considerable resources through better tax effort and release more resources for the social sectors through better fiscal management and reprioritization. Even so, this can meet the requirements only partially and significant additional resources will have to be committed to health and allied sectors. Thus, achievement of the MDGs as well as the Tenth Plan Goals in the health sector will crucially depend on additional resources made available through the transfer system and better targeting of these transfers. Central transfers to States fall into three categories. The first is the statutory transfers comprising tax devolution and grants, which are given on the basis of the recommendations of the Finance Commission. The second is the plan assistance given by the Planning Commission on the basis of the consensus formula approved by the National Development Council (NDC). The third source is the transfer given by various Central ministries for the Central sector and Centrally sponsored schemes. Analytically, transfers can be given for general purposes, to offset the general fiscal disabilities of the States, or for specific purposes. While the former is given to enable every State to provide a given level of public services at a given tax price, the latter is given to ensure minimum standards of specified services. The latter are given as these services are considered to be meritorious and, therefore, everyone is entitled to a minimum level of their consumption. These transfers are targeted to spend on specified purposes/sectors and they have to be targeted to those States with shortfalls. The statutory and formula-based transfers of the Planning Commission are essentially general-purpose transfers. These are meant to offset the general fiscal disabilities of States arising from the low revenue-raising capacity or higher unit cost of providing public services. These are formula-based transfers meant for general augmentation of resources and cannot be pre-empted for spending on health and allied sectors. The important exception to this is the upgradation grants for health and education recommended by the TFC. As discussed in the previous section, while these are useful supplements, they would be inadequate to meet the requirements for achieving the set goals. Besides being general-purpose transfers, it is doubtful whether the Centre would be able to augment them much to bridge the gap in States resources.for the health sector for at least two reasons. First, given the compulsions of restoring fiscal balance at the Central level itself and given further the compulsions of meeting the fiscal targets set by the Fiscal Responsibility and Budget Management Act, it would be difficult for the Centre to make additional resources available for
general purpose spending. Second, the TFC has already made recommendations with marginally increased transfers from the last Commission and these recommendations will be valid during 2005-06 to 2009-10. Under this no additional resources would be available. This would imply that any increase in the transfers will have to be for specific purposes under the Central sector and Centrally sponsored schemes. At present, a large number of schemes are being administered by various ministries resulting in the thin spread of resources, multiplication of bureaucracy and often, poor targeting. To meet the shortfalls in the health sector, it is necessary to significantly augment specific transfers to enhance the resources for health spending. In addition, it would be necessary to consolidate various schemes under the broad heads of basic education, healthcare, maternal and child health, nutrition, water supply, sanitation and rural roads. This will target the transfers to augment spending in desired sectors. Equally important is proper designing of the transfer system. It is useful to have purpose-specific grants. To have the system with right incentives and to ensure that the additional resources provided by the Centre are used for incremental spending and not merely to substitute States own spending, it is useful to mandate the States to make matching contributions. Of course, matching requirement places poorer and resource constrained States at a disadvantage, but the matching requirement itself can be varied with the level of per capita incomes in the States (Feldstein 1975). Thus, highincome States may be required to contribute, 50%, middleincome States 30% and low-income States a mere 10%. Such a design of the transfer system for specific purposes will preserve the incentives, impart a sense of ownership and participation by the States and help to augment resources for the desired sectors.
Conclusion
This study analyses the resource requirements for meeting certain targets of the health sector and analyses the gap between the required and the actual expenditure in 15 major States in India. It highlights the extent of resources that can be mobilized at the State level to meet the resource gap and estimates the residual gap that has to be met by Central transfers. Estimates indicate that the additional expenditure required for meeting the specific norms/targets in health and related sectors (which include safe drinking water, sanitation, nutrition, primary schooling and roads) is about Rs 300,168 crore. One can argue that the expenditure on primary schooling and roads has various other positive externalities and are not exclusively incurred towards health. Although, not exclusively towards health, these expenditures have a significant bearing towards health outcomes and cannot be ignored if one has to reach the health targets. Even if one focuses only on medical, public health, safe drinking water and sanitation, which are directly incurred towards health outcomes, the total requirement is about Rs 100,415 crore. In general, there is a deficit of about Rs 247,503 crore at the State level. The requireFinancing and Delivery of Health Care Services in India
315
SECTION IV
ments are particularly high in States with low per capita income and high poverty levels. These are also the States where the productivity of expenditure and delivery of services are particularly poor. If one is constrained on the resource front, these aspects have to be specifically focused upon. Improving the productivity of expenditure and delivery systems in these lowincome States can actually reduce the resource requirement. However, it would be too optimistic to expect any appreciable improvement in the productivity of healthcare expenditure in the near future. In particular, it may be noted that the level of productivity and delivery systems are often affected by a number of social, cultural and historical factors which change slowly over time. These improvements therefore cannot act as a substitute for increased allocation of funds in the short run. One therefore has to find resources to make increased allocation to healthcare expenditure in the next five to ten years. Increased allocation to healthcare expenditure can be done by (i) raising more resources; (ii) reprioritizing the expenditure allocation in favour of medical and public health, water supply and sanitation; and (iii) targeting the expenditures to States and regions where the health indicators are poor and have considerable catching up to do. The possibility of raising additional resources has been discussed at length. We have compared our estimates of taxable capacity with the estimates made by the TFC. The ability of the States to contribute additional resources to the health sector critically depends on their effort in raising revenues close to their capacity. In this context, two points are important. First, our estimates show that there is a possibility of raising revenues, particularly in some States where the actual revenues raised are below their capacity. Second, the capacity estimation itself is relative to other States and not in the absolute sense. In other words, if there is a general undertaxation by all States, it does not show up in the estimates. For example, although the States have been assigned the power to levy tax on agricultural income and wealth, they have mainly for political reasons desisted from this and even the land revenue collections have declined over the years. It is certainly possible to raise the bar through better tax administration. The most important initiative in this regard is the introduction of value added tax (VAT) by most States with effect from 1 April 2005. Although this is expected to be revenue neutral, it is expected to increase the revenue productivity of the tax system in the long term. The extension of the tax net to the retail stage would broaden the base and is expected to more than offset the loss of tax base due to giving credit to inputs. More importantly, the self-enforcing nature of the tax is expected to significantly improve the tax compliance and this could improve revenue productivity. Thus, it should not be difficult to improve the revenues by at least by 1% to 1.5% of GSDP over the next five years. The second way to release more resources to the health sector is to reprioritize the expenditure in favour of the sector. This might have to be done within the limits imposed by the Fiscal Responsibility and Budget Management (FRBM) Act
by each State mandated by the TFC. Although the TFC has expressed its concerns over the increase in expenditures on salaries and wages it should be ensured that these do not impose any reduction in expenditure on education and health sectors. Infact it is important to ensure that CHCs, PHCs and subcentres are properly staffed to provide the required services. The debt rescheduling and reduction in interest rates recommended by the TFC would allow additional fiscal space to the States. Besides, the TFC has recommended incentivebased debt write off to the tune of Rs 32,198 crore over the same period. It is important that the additional fiscal space created by these recommendations is used for human development. Another important strategy to be adopted to improve the effectiveness of health expenditure is to target the allocations to States where the health outcomes are poor. It is precisely for this reason that the TFC has recommended equalizing grants to those States with less than average per capita expenditures within the revenue account. Although the TFCs equalization does not entirely cover the shortfall in per capita expenditures, this type of targeting expenditures could help to improve the health outcomes precisely in States with large shortfalls from the norms. The recommendations of the TFC, however, can cover only a partial requirement of the States. Achieving the MDGs and the Tenth Plan targets would require significant additional resources and improved productivity in spending to focus on outcomes rather than outlays. Much of the intervention in this area will have to come by way of consolidation of a plethora of Central schemes prevailing at present, and augmentation of specific purpose transfers for broadly defined purposes. This paper argues that the appropriate design for targeting, preserving the incentives and to ensure participatory provision is to have a specific purpose transfer with matching contributions from the States, the latter varying with the level of their development. From the above discussion, it would be reasonable to summarize that significant additional allocation to health sector is within the realm of possibility. This, however, would require that States be clear in their assignment of priority. Investment in human capital is critical to both accelerating growth, enhance productivity and empowering the poor. Improving the health status of population is a critical component of human development and the States will have to reassign their priorities in favour of the health sector in the interest of development.
Acknowledgements
The authors would like to place on record the help, guidance and advise from Ms Sujatha Rao. They had the benefit of interaction with her at every stage of the study. She not only helped in the conceptualization and in evolving the methodology for the study but also went through the successive drafts with meticulous care. Thanks are also due to Dr Ajay Mahal whose advice and guidance in the initial stages of the study were extremely useful. The authors are also indebted to the staff of the National Commission on Macroeconomics and Health
316
SECTION IV
for discussions on various matters relating to the study. The assistance provided by Ms Gita Bhatnagar is gratefully acknowledged. The authors own full responsibility for the views expressed and for errors of omission and commission.
Note: The Governing Body of the National Institute of Public Finance and Policy does not take any responsibility for the views expressed in this report. The responsibility belongs to the researchers who conducted the study.
317
SECTION IV
References
Bahl Roy W. A regression approach to tax effort and tax ratio analysis. International Monetary Fund Staff Papers 1971;18:570-611. Bahl Roy W. A representative tax system approach to measuring tax effort in developing countries. International Monetary Fund Staff Papers 1972;19:87124. Chelliah Raja J, Sinha N. Measurement of tax effort of state governments 1973-76. New Delhi: National Institute of Public Finance and Policy; 1982. Coondoo D, Majumdar A, Neogi C. Tax capacity function: A note on specification, estimation and application. In: Srivastav DK (ed). Fiscal federalism in India: Contemporary challenges: Issues before the 11th Finance Commission. New Delhi: National Institute of Public Finance and Policy; 2000. Deolalikar A. Attaining the Millennium Development Goals in India: Role of public policy and service delivery. Washington: The World Bank; 2004. Dev MS, Mooij J. Patterns in social sector expenditure: Pre- and post-reform periods. In: Parikh KS, Radhakrishna R (eds). India development report 2004-05. New Delhi: Oxford University Press; 2005. Feldstein M. Wealth neutrality and local choice in public education. American Economic Review 1975;65:75-89. Government of India (GOI). First Report of the Ninth Finance Commission (for 1989-90), July 1988. Government of India. Second Report of the Ninth Finance Commission (for 1989-90), December 1989. Government of India. Expert Group Report on Financial Requirements for Making Elementary Education a Fundamental Right. New Delhi: Department of Education, Ministry of Human Resource Development, Government of India; 1999. Government of India. Report of the Twelfth Finance Commission (2005-2010), 2004. Jha R, et al. Tax efficiency in selected Indian states. Empirical Economics 1999;24:641-54. Lotz Jorgen R, Elliot RM. Measuring tax effort in developing countries. International Monetary Fund Staff Papers 1967;14:478-99. Musgrave RA. The theory of public finance: A study in public economy. New York: McGraw-Hill; 1959. Musgrave RA. Fiscal systems. Yale University Press; 1973. Nambiar KV, Govinda Rao M. Tax performance of states. Economic and Political Weekly 1972;1036-8. Oommen MA. Relative tax effort of states. Economic and Political Weekly 1987;XXII. Rao Govinda M. India: Intergovernmental fiscal relations in a planned economy. In: Bird RM, Vanllaucourt F (eds). Fiscal decentralisation in developing countries. Cambridge University Press; 1998:78-115. Rao Govinda M, Singh N. Political economy of federalism in India. New Delhi: Oxford University Press; 2005. Reddy KN. Inter-state tax effort. Economic and Political Weekly 1975. Sarma JVM. Panel data models and measurement of states tax effort in India. Working Paper No.9/89. New Delhi: National Institute of Public Finance and Policy;1989. Sen T. Relative tax capacity and tax effort of Indian states. Working Paper. New Delhi: National Institute of Public Finance and Policy; 1983. Sen T. Relative tax effort of Indian states. Working Paper. New Delhi: National Institute of Public Finance and Policy; 1997. Sen T, Tulasidhar VB. Taxable capacity and tax effort of states of India. Working Paper. New Delhi: National Institute of Public Finance and Policy; 1988. Sen and Himanshu. Poverty and inequality in India-I. Economic and Political Weekly Vol XXXIX, No 38, September 18-24, 2004. Shiva Kumar AK. Budgeting for health: Some considerations. Economic and Political Weekly 2005:1391-6. Thimmaiah G. Revenue potential and revenue efforts of southern states. (A study sponsored by the Planning Commission of India). New Delhi: Oxford and IBH Publishing Co; 1979.
318

Financing and Delivery of Health Services NCMCH

Загружено:

Сведения о документе

Исходное описание:

Авторское право

Доступные форматы

Поделиться этим документом

Поделиться или встроить документ

Параметры публикации

Этот документ был вам полезен?

Это неприемлемый материал?

Авторское право:

Доступные форматы

Financing and Delivery of Health Services NCMCH

Загружено:

Авторское право:

Доступные форматы

Background Papers of the National Commission on Macroeconomics and Health

Financing and Delivery of Health Care Services in India

National Commission on Macroeconomics and Health

MINISTRY OF HEALTH AND FAMILY WELFARE GOVERNMENT OF INDIA, 2005

NCMH Background Papers

Financing and Delivery of Health Care Services in India

Ministry of Health & Family Welfare, Government of India

Cover design: Quote Design Studio

Financing and Delivery of Health Care Services in India

Financing and Delivery of Health Care Services in India

Sujatha Rao Secretary, NCMH

Financing and Delivery of Health Care Services in India

List of Contributors and Reviewers

Financing and Delivery of Health Care Services in India

Financing and Delivery of Health Care Services in India

iii v 1 3 19 37 39 43 65 77 89 125 135 151 153 175 185 213 226

237 239 256 265 275 297

SECTION I Health, Poverty and Economic Growth in India

Health, Poverty and Economic Growth in India

Health, Poverty and Economic Growth in India

Economic growth and health: A review of cross-country and regional studies

Financing and Delivery of Health Care Services in India

Health, Poverty and Economic Growth in India

Health, Poverty and Economic Growth in India

Financing and Delivery of Health Care Services in India

Health, Poverty and Economic Growth in India

Health, poverty and economic growth: Inter-State descriptive analysis

Table 1 Basic characteristics of the States included in the study

Financing and Delivery of Health Care Services in India

Health, Poverty and Economic Growth in India

Fig 2 Relationship between initial income and growth rate

Fig 1 Per capita income and growth rate by States, 1970-2000

Fig 3 Trends in life expectancy at birth (LEB), 1970-99, India

Financing and Delivery of Health Care Services in India

Health, Poverty and Economic Growth in India

Model specification and estimation issues

Econometric model and empirical analysis

Health, Poverty and Economic Growth in India

Empirical results: Estimates of economic growth and health equations

Financing and Delivery of Health Care Services in India

Health, Poverty and Economic Growth in India

0.174 (3.01) 1.265 (2.35)

0.078 (1.85) 0.156 (0.39) 0.069 (6.74)

-33.223 -31.659 -33.690 0.346 0.329 0.377 14 14 14 42 42 42

14.162 (2.47) -9.837 (1.37) -5.880 0.304 14 42

Financing and Delivery of Health Care Services in India

Health, Poverty and Economic Growth in India

Panel data estimates of the aggregate Cobb-Douglas production function

Log Labour Log Capital Log LEB Years of schooling

0.461 (7.15) 0.414 (7.58) 1.134 (4.51)

Constant -7.639 R2 0.934 F statistics 244.00 Number of States 14 Number of observations 56

0.678 (10.86) 0.355 (5.66) 1.960 (5.74)

-13.793 0.909 180.48 14 56

Economic growth, poverty and health: Theory and empirical evidence

Financing and Delivery of Health Care Services in India

Health, Poverty and Economic Growth in India

Log labour Log capital Log LEB Years of schooling

0.910 (6.82) 0.318 (4.91) 1.510 (2.71)

0.607 (8.24) 0.451 (4.78) 1.817 (3.73)

-13.344 0.910 85.37 3.23 (0.36) 14 56

Fig 6 Growth rate and poverty by States, India

Health, Poverty and Economic Growth in India

Fig 7 Per capita income-poverty association, panel data, 1970-71 to 1999-2000