From staff-mix to skill-mix and beyond: towards a systemic approach to health workforce management ABSTRACT -Throughout the world,

countries are experiencing shortages of health care workers. Policymakers and system managers have developed a range of methods and initiatives to optimise the available workforce and achieve the right number and mix of personnel needed to provide high-quality care. Our literature review found that such initiatives often focus more on staff types than on staff members' skills and the effective use of those skills. Our review describes evidence about the benefits and pitfalls of current approaches to human resources optimisation in health care. We conclude that in order to use human resources most effectively, health care organisations must consider a more systemic approach - one that accounts for factors beyond narrowly defined human resources management practices and includes organisational and institutional conditions.
METHOD- Our findings are based on a structured review of published literature, including

articles, reviews, comparative studies, observational studies, and dissertation identified through a range of electronic databases: Medline, PubMed, Embase, Current Contents, CINAHL and Google Scholar. Other relevant materials (research reports, administrative reports, and articles) were collected through website searching, reference chaining and contacting experts in the field. The search focused on the literature between 1995 and 2008. However, some key literature prior to 1995 has been included when it was considered to be of particular relevance. The following key-words uncovered many hundreds of 'hits': staff-mix, skill-mix, human resource management, human resource optimisation, workforce performance, human capital, skill management, human resources for health, performance management. All references were reviewed by title and abstract to determine their potential relevance to the review. Letters, comments and editorials were systematically excluded. References that related directly to the subject matter in either the title or the abstract were selected for a more in depth review. In total, we examined full copies of 250 selected studies more thoroughly. The evaluations of the studies and the data extraction were performed manually by the two investigators. Papers were first sorted into two categories: conceptual papers and empirical papers. Conceptual papers were evaluated and sorted according to their theoretical foundations, their comprehensiveness, their relevance and their contribution to subsequent work in the field. Empirical papers were evaluated and classed based on their relevance to the review objective and appropriate criteria of validity (research design, sampling and methods of analysis). We used the technique of interpretative synthesis to collate the findings. This approach involved building a general interpretation grounded in the findings of separate studies and then integrating evidence from across the studies into a coherent theoretical framework comprising a network of constructs and the relationships between them [10]. As for the search strategy, the analysis focused first on evidence and theoretical perspectives drawn from the health care sector; however, as we advanced in the analysis, it has become evident that human resource management is a topic with diffuse boundaries that overlaps with several other fields. Although our selection of articles was clearly focused on human resources in health care, we had to extend our investigation to a wider range of literature in order to fill some gaps of evidence, gain insight from other areas and elaborate the emerging analysis. We particularly draw on theoretical perspectives and empirical work in sociology, economics, management, industrial and labour relations, and psychology that address different aspects of the domain of human resource management. Those works account for 20% of the 250 selected papers.,

DISCUSSION- Personnel deployment conceptualised as a staff-mix issue

Managing human resources in health care involves organising groups of workers with different professional backgrounds, skills, grades, qualifications, expertise and experience in order to achieve optimal patient care. This distinctive feature of health care has become more prominent during recent decades with the emergence of numerous new professions, specialties and occupations. These developments have drawn considerable attention to the concepts of staff-mix and skill mix as policy tools for developing the best combinations of skills across professions and organisations, as well as at the individual level. Increased interest in achieving optimal staffmix also results from pressures arising from both the supply and demand sides of health care. On the supply side, changing the mix of health care staff has often been used as a resourcing strategy to address shortage problems. On the demand side, those changes have been implemented as a means to enlarging the scope of services, fill previously unmet health needs and improve patient care [11,12]. While many regard adequate staff and skill mix to be prerequisites for meeting patients' needs for high-quality care, HR adequacy is, in reality, hard to assess because it relates to many different parameters, including needs, preferences, availability, cost and quality. In this regard, recent reviews have highlighted the diversity of ways in which personnel deployment across teams and organisations is conceptualised [13-15]. Reviews suggest that although the concepts of staff-mix and skill mix are often used interchangeably, the four most prevalent conceptualisations are closer to the notion of staff-mix. We discuss these conceptualisations below. Number of personnel This conceptualisation focuses on the total number of workers in defined occupational groups. It takes into account the volume of work assigned to a given staff member or the amount of direct patient contact a worker experiences over a defined period of time. Common measurements are the number of hours of professional care per patient, per day; and the number of full-time equivalent workers per patient, per day. For pharmacists, the ratio has been defined as the number of prescription orders filled per day. For some physicians, the number of certain procedures performed per year is measured. Research on personnel numbers has focused largely on nurses, and is based on the hypothesis that a lower nurse-to-patient ratio results in a greater workload and poorer quality of care due to time pressures that affect a person's ability to implement best-practice standards. Several empirical studies and systematic reviews support this hypothesis and indicate that the numbers of nurses in a unit and the number of nurses per patient affect patient outcomes, including adverse events, readmissions and mortality [16-22]. One study found that each additional patient in a typical nursing workload situation resulted in an average 7% increase in failure-to-rescue [23]. In another study, hospitals in which nurses cared for an average of eight patients each had riskadjusted mortality rates following common inpatient surgical procedures that were 31% higher than hospitals in which nurses cared for four patients each [24]. Such findings have prompted legislation on safe staffing ratios for nurses in two jurisdictions: California and the state of Victoria in Australia. Yet, there is currently no clear-cut evidence of the effectiveness of such legislated ratios, which may prevent managers from making local decisions about appropriate staffing and are insensitive to many contextual factors (e.g., changes in patient dependency, presence of ancillary personnel or non-nurse providers, technology). In contrast to nursing research, studies of physician resources are based on the premise that higher volumes, rather than hindering the ability to meet patients' needs, lead to improved experience and high-level technical skills [25]. Evidence from recent systematic reviews and

observational studies suggests that higher volumes are, for physicians, associated with lower error rates and lower patient mortality rates [26-28]. Another study that used hospitals as the unit of aggregation showed that facilities with higher case volumes experienced lower complication rates [29]. Such positive findings are, however, balanced by some contradictory evidence. In controlling for institutional factors, some studies have failed to find that physicians who performed high rates of technical procedures experienced lower rates of adverse outcomes, suggesting that improved results reported in other studies may have been due to institutional rather than physician-specific factors [30-33]. Mixing qualifications This conceptualisation focuses on the proportion of highly qualified staff members in the overall pool of professional resources. As yet, there is no indication of the appropriate ratio for any grade on the health care team, although several observational studies support the view that a rich mix of qualified personnel with advanced degrees or specialty certifications is associated with better clinical outcomes. Blegen et al [34] suggest that having a nursing team that is richer in registered nurses contributes to lower patient mortality rates. In a landmark study, Aiken et al [35] found an inverse relationship between the proportion of registered nurses holding undergraduate degrees and patient mortality rates within 30 days of admission: a 10% increase in the proportion of nurses with undergraduate degrees was associated with a 5% decrease in the likelihood of patients dying. Another study found that people cared for in the community by undergraduate degree-level nurses required fewer home visits and had better knowledge and health behaviours than those cared for by nurses without such degrees [36]. Again, it is important to keep in mind that current evidence only suggests some trends; it does not offer clear direction on the most effective skill mix for nurses. Those studies that have found positive associations have reported wide-ranging registered nurse proportions: from a low of 46% to a high of 96% [37-39]. A number of studies have examined the added value of specialty certification among physicians. Evidence suggests that physicians with specialty training have lower rates of adverse outcomes for certain procedures and medical conditions. Researchers have found a significant association between greater prior training by physicians on certain surgical procedures and better results in performing those procedures [40-42]. Similarly, patients with acute myocardial infarction tend to have lower risk-adjusted mortality rates when cared for by cardiologists [43]. In pharmacies, meanwhile, the evidence points in the opposite direction. Studies comparing pharmacists to pharmacy technicians have found similar error rates between the two groups [44,45]. Balancing junior and senior staff members This staff-mix conceptualisation draws attention to the proportion of experienced staff members on health care teams. This proportion is usually measured by the number of years an individual has worked in a particular grade or job category. The most common hypothesis is that longer experience is associated with better patient outcomes. However the evidence is scarce and conflicting. Several observational studies have concluded that more years of surgical experience are not associated with lower rates of post-operative complications [46,47]. Similarly, studies suggest no relationship between years of experience as a registered nurse and patient mortality rates [48]. Conversely, others report that for each additional year of nurse experience on a clinical unit there were four to six fewer deaths for every 1000 acute medical patients discharged (depending on hospital type) [49]. Another study demonstrated that registered nurses' duration of practice was inversely related to rates of medication errors and patient falls [50]. Mixing disciplines

This conceptualisation involves gathering together individuals from different professions and specialties in order to provide well-rounded care. Multidisciplinary teams are commonly used in hospitals or outpatient services. These primary care teams comprise nurses and physicians, and sometimes include specialists. Collaboration is increasing between mental health and primary care workers, and pharmacists are increasingly integrated into primary care teams [51,52]. Increased interest in a 'whole system' approach to care has also contributed to the inclusion of social service staff, community workers and volunteers on primary care teams [53]. There is an extensive body of literature focusing on the potential benefits of multidisciplinary teams and, more broadly, of collaboration amongst professionals from different disciplines as a way to address fragmentation, discontinuity, and lack of receptiveness. In reality, however, the evidence is inconsistent on the effectiveness of multidisciplinary teams compared to care provided by a single group of professionals. A review of 14 systematic reviews and 33 additional randomised trials found that the impact of multidisciplinary teams on quality of life and clinical outcomes varied considerably amongst the studies [54]. Other research indicates that, although multidisciplinary outpatient teams or teams of primary and secondary care personnel working together can improve patient outcomes; this result may vary according to the initiatives undertaken and patients' conditions. A systematic review focusing on people with rheumatoid arthritis found that multidisciplinary outpatient teams may improve functional outcomes more than usual care [55]. Other trials involving elderly people and those who had suffered strokes, however, found no impact on health outcomes [56,57]. Physician-nurse collaboration has particularly attracted researchers' attention. Some studies suggest that a high degree of collaboration is associated with lower mortality and complication rates and with increased patient satisfaction in adult intensive care units (ICUs) [58,59]. Findings about the value of general practitioner (GP) and nurse collaboration in primary care are often less clear. While some studies have found improved clinical outcomes and satisfaction [60], others have discovered no significant improvement over usual care approaches [61,62]. In addition to the conflicting findings, it is difficult to draw clear conclusions from these studies because most multidisciplinary interventions contain several other variables, such as increased follow-up and medication reviews. It is therefore unclear whether multidisciplinary team composition, additional contacts with staff members, or other factors influence outcomes. Similarly, it is uncertain which specific staff members may be more or less useful within multidisciplinary teams. What can we conclude about optimal staff-mix? Health care organisations have a range of options for ensuring a richer staff-mix: Increasing the number of personnel Higher ratios of qualified workers Higher ratios of senior staff members Multidisciplinary teams Despite conflicting findings and the need for further research, a number of studies and systematic reviews suggest that a richer staff-mix may be associated with better outcomes and fewer adverse events for patients. The evidence, however, is highly limited by practical limitations and methodological shortcomings. While many studies have reported positive impacts from enriching staff-mix, they do not offer clear guidance about ideal thresholds in terms of personnel/patient ratios or the proportion of different categories of staff members on teams. More fundamentally, the staff-mix perspective that emphasises numbers and types of personnel gives less attention to the conditions that determine how staff members' skills are used. Despite the

rhetorical use of 'skill mix' to describe the different options for deploying health care personnel, the focus is, in reality, not on skill but on grades, educational qualifications, job titles and duration of experience that are, at best, proxies for skill levels. An effective system of HR optimisation cannot, however, be restricted to the numbers and types of personnel available. Such a system must also ensure that personnel work to their full potential. Doing so requires a more dynamic approach to skill management that goes beyond the mix of available staff members. From staff-mix to skill management Skill management refers to an organisation's ability to optimise the use of its workforce. The focus shifts here from achieving a specific mix of different types of personnel to adapting workers' attributes - such as knowledge, skills, and behaviours - and roles to changing environmental conditions and demands [63,64]. Skill management enables organisations to optimise patient outcomes while ensuring the most effective, flexible and cost-effective use of human resources. A diverse set of interventions have been tested to achieve this dynamic approach to HR optimisation. We divide them into two main dimensions: skill development and skill flexibility. Skill development One of the greatest challenges facing health care organisations in recent years has been how to adjust to the rapid pace of a wide variety of internal and external changes: Environmental changes in consumers' tastes and demands Changes in legal requirements Socio-demographic and epidemiologic changes Technological developments Economic fluctuations. To a large extent, organisations' strategic and practical adjustments depend on their members' capacity to transform. An organisation updates its responses to changes only when its workforce can learn and utilise the skills required to take on new roles and functions. These additional roles and functions may be at higher, parallel, or even lower level [65], and they can come about through two distinct processes: role enhancement and role enlargement. Role enhancement Role enhancement involves expanding a group of workers' skills so they can assume a wider and higher range of responsibilities through innovative and non-traditional roles [66]. Enhancing staff members' roles through new competencies gives to employees the opportunity to acquire new competencies and expand their tasks so that they can take on responsibilities traditionally carried out at higher levels [67]. By altering the content of their work, employees are offered opportunities for individual achievement and recognition. Under this model there is greater work depth because employees are involved in tasks that increase their control or responsibility [68]. Role enrichment is considered a vertical and upward expansion of work because it alters authority, responsibility, level of complexity and assignment specificity [69]. In a specific health care context, role enhancement describes a level of practice that maximizes workers' use of indepth knowledge and skills (related to clinical practice, education, research, professional development, and leadership) to meet clients' health needs [70,71]. Role enhancement does not entail adding functions from other professions. It occurs within a given profession's full scope of practice through the integration of theoretical, research-based and practical knowledge inherent to the development of a discipline [72]. It can also arise from innovative professional activity, new models of health care delivery, and organisational changes

that promote development of new knowledge, skills, and practices. Through experience, continued professional growth and development, and collaboration with colleagues from other disciplines, health care workers can develop new skills, abilities, and techniques they did not obtain during previous clinical preparation [73]. In addition, as health care work expands into new settings, the situational factors that shape service provision in those environments create demands for new skills [74]. In health care, role enhancement has been associated with the potential to increase longitudinal and personal continuity and improve patients' health outcomes by enabling one professional to cover a wider range of care needs or by enabling one patient to be cared for by fewer workers. As a result, many health care professionals such as nurses, pharmacists, and GPs have recently expanded their responsibilities beyond their traditional scope of practice to include more innovative roles. In many cases, these role expansions were initiated in order to ensure that individual professionals would be able to oversee a greater proportion of their patients' care. Primary care and prevention are the main areas in which nurses have taken the lead in delivering expanded services, including health promotion, health screening, and discharge follow-up. Since the 1990s, nurses in UK general practices have been responsible for carrying out well-patient health checks and providing lifestyle counselling and other interventions in accordance with treatment guidelines [75]. Nurses have also expanded their roles by specialising in practice domains and by helping people with particular conditions. Such specialist nurses can be based in either primary or secondary care, and they are particularly active in nurse-led clinics, where nurses assume responsibilities such as managing people with long-term conditions, providing health promotion advice, monitoring and informing patients, and screening for diseases (e.g., cervical screening, cardiovascular screening) [76-79]. Role expansion can also be seen in nurseled outpatient follow-ups, whereby hospital or community-based nurses oversee discharge planning and post-discharge outpatient follow-up [80]. These examples illustrate the expansion of nursing into areas that were often unmet or inadequately addressed. While retaining their generalist background, some GPs have also expanded their roles. In the US and the UK, GPs who hold additional qualifications or training and who focus on particular areas are sometimes known as "GPs with special interests." Such physicians can offer specialist care in the community or work as part of multidisciplinary hospital and primary care teams [81-83]. Similar developments have occurred for pharmacists whose work has expanded far beyond the distribution of medications to include patient education, health promotion, counselling, medication management, health monitoring, and even, in some jurisdictions, prescribing [84-86]. In England, the Medicines Management Collaborative involves 146 primary care trusts and 44 trusts, and it aims to engage all members of the pharmacy team in identifying and addressing patients' unmet pharmaceutical needs [87]. Despite major interest in developing enhanced roles, evidence about the impact of these new roles is limited and has focussed mostly on nursing. Overall, the evidence suggests that health professionals can learn specific advanced skills that fall outside the scope of their routine practice and apply them in clinical settings. However the impact of such role enhancement remains uncertain. Some studies have found improvements associated with organisational innovations that draw on nurses with advanced skills, including nurse-led clinics or specialist nurse-led initiatives [88-91]. Other studies have found fewer or no benefits [92-95]. However there are variations in the nursing interventions in these studies which may lead to inconsistencies in the findings and make it difficult to draw conclusions about the effects of enhanced nursing roles on patient outcomes. We cannot be certain whether any observed

differences are due to the nurses' roles or to other intervention-related factors (e.g., resource intensity, increased follow-up, access to a multidisciplinary team). Thus, although many studies have revealed connections between nurses' role enhancements and safe and effective care or improved patient outcomes, it remains uncertain whether the benefits are due to specific interventions or nurses' roles. Furthermore, the evidence regarding the opportunity costs of such service developments and marginal gains in terms of health outcomes is still scarce and often conflicting.
CONCLUSION- This article has summarised different approaches to optimising HR in health care. We have argued that perspectives that focus on staff-mix, such as those that count the number of personnel needed or focus on generating formulae and algorithms, provide only partial solutions. Wider perspectives, which focus on how human resources can be differently managed either through skill development or skill flexibility, go some way towards conceptualising personnel use in the dynamic and constantly evolving realm of health care. In order to be fully effective, policy-makers, managers, and practitioners need to consider the organisational factors that affect how staff members work. The evidence suggests that no matter which workers are employed or what their roles are, it is only by tackling organisational issues that a fully efficient and effective workforce can be generated. In order to use human resources most effectively, organisations must also consider the institutional environments that frame health care workers' educational preparation, the system of professional regulation, organisational incentives, and the broad range of levers that can be mobilised at both organisational and system levels.

Worksite Wellness Programs for CVD Prevention

ith _130 million Americans employed across the

United States, workplaces provide a large audience for cardiovascular disease (CVD) and stroke prevention activities. Experience has shown that workplace wellness programs are an important strategy to prevent the major shared risk factors for CVD and stroke, including cigarette smoking, obesity, hypertension, dyslipidemia, physical inactivity, and diabetes. An estimated 25% to 30% of companies medical costs per year are spent on employees with the major risk factors listed above.1 Employees and their families share the financial burden through higher contributions to insurance, higher copayments and deductibles, reduction or elimination of coverage, and trade-offs of insurance benefits against wage or salary increases. When programs are successful, their influence extends beyond the individual workers to immediate family members, who are often exposed to their favorable lifestyle changes. Worksite wellness programs that can reduce these risk factors can ultimately decrease the physical and economic burden of chronic diseases, including CVD, stroke, and certain cancers. The societal benefits of a healthy employed population extend well beyond the workplace. As such, comprehensive, culturally sensitive health promotion within the workplace can improve the nations health. The Healthy People 2010 goal is for 75% of all worksites, regardless of size, to develop comprehensive wellness programming.2 However, the development of comprehensive programs takes time and resources, especially for smaller employers. Because program development

and initiation can be resource intensive, the American Heart Association (AHA) supports incremental efforts to achieve a comprehensive worksite wellness program to address CVD and stroke prevention and makes the following recommendations.

Summary of Recommendations
1. Components of Wellness Programs A comprehensive program aimed at improving employees cardiovascular and general health should include the following: Tobacco cessation and prevention, regular physical activity, stress management/ reduction, early detection/screening, nutrition education and promotion, weight management, disease management, CVD education that includes cardiopulmonary resuscitation and automated external defibrillator training, and changes in the work environment to encourage healthy behaviors and promote occupational safety and health. Programming should be integrated into the organizational structure of the workplace by use of the following proven strategies: Health education that relies on existing valid sources and is focused on skill development that is consistent with employees readiness to make behavior changes; initiatives that are incorporated into existing employee assistance programs; and voluntary worksite screening linked with medical care for follow-up on modifiable risk factors. Employers should administer health risk appraisals in combination with organizational health promotion checklists that have already been developed for the worksite before initiating programming so that health needs in the workplace can be identified and employees can learn their risks and health status. Employees health risks must be addressed within comprehensive worksite programs. Research should investigate the effectiveness of wellness programming and how to tailor programming and policies for maximum effect. 2. Environmental Modifications The social and physical environment of the workplace should be designed to be conducive to recommended behaviors. Optimal environmental modifications should promote healthy behaviors while simultaneously minimizing the physical, organizational, and occupational risk in the work environment. Occupational safety and health are integral components of worksite wellness; workplaces should be free from hazards that jeopardize cardiovascular health and employee safety and well-being. 3. Regulations/Policy Approaches The regulatory environment should allow for increased opportunity for employers to reach a greater majority of the employee population and produce health benefits. Employers should adhere to all regulations that address hazards to employee health and safety, providing working conditions that are optimal for cardiovascular

health and well-being. Employers who choose to offer healthy lifestyle behavior incentives in the workplace, such as wellness credits and financial incentives, should provide these directly to the employee. Financial incentives should not be attached to healthcare premiums or health status. 4. Vulnerable/Special Populations Wellness programs must address the needs of all employees at a given workplace, regardless of gender, age, ethnicity, socioeconomic status, culture, job type, or physical or intellectual capacity. Worksite wellness programs should be designed to be culturally sensitive and all-inconclusive, and employers should also consider targeted, complementary interventions for their more vulnerable employees that are specifically designed to engage those who are economically challenged, less educated, or underserved. Worksite wellness programs should help working families balance work and family commitments and incorporate policies around child care, elder/dependent care, telecommuting, and flexible work schedules. Research should be conducted to determine how to improve participation among employees who have the highest risk behaviors.

Wellness Programs
CVD and stroke are the leading causes of death in the United States. The estimated expense associated with all heart diseases combined is $304.6 billion, $24 and $98 billion of which is due to lost productivity from cardiovascular morbidity and mortality, respectively.3 The financial burden associated with stroke is equally weighty; in 2009, an estimated $68.9 billion in direct and indirect costs was spent to diagnose and treat strokes.4 The estimated lifetime cost for hospital stays, rehabilitation, and follow-up care to treat lasting neurological deficits is $140 048 (adjusted to 1999 dollars).4,5 Worksite wellness programs are a proven strategy to prevent major risk factors for CVD and stroke, including cigarette smoking, obesity, hypertension, dyslipidemia, physical inactivity, and diabetes. Historically, wellness programs have included education and screening programs in an effort to increase individual workers awareness of risk factors and suggest strategies to modify health behaviors. Recent evidence from the social sciences and behavioral medicine literature suggests that environmental modification and policy changes and approaches are more successful at producing sustained behavior change that can reach employees across varied socioeconomic groups. Worksite wellness programs represent an opportunity to prevent CVD and stroke in a large segment of the population. The AHA and its regional affiliates have a long history of participating in worksite wellness programs. Because of marked variability in the availability, content, and delivery of wellness programs, the AHA is committed to updating recommendations given current knowledge about effective programs and strategies to produce positive changes.

The Current State of Affairs

Availability
In 2004, the National Worksite Health Promotion Survey conducted 1553 interviews with worksites from different size and industry categories and found that only 6.9% of employers offered comprehensive worksite wellness programming, defined as those programs that incorporated all of the 5 key elements outlined in Healthy People 2010: health education, supportive social and physical work environment, integration, linkage, and worksite screening and education.6 The presence of comprehensive programming varied significantly by worksite size; programs were available at 11.3% of companies with 250 to 749 employees, but only 4.6% of companies with 50 to 99 employees had programs. At least half of the working people in the United States do not have access to health promotion programs because they work in small companies or for employers who have employees distributed in small numbers across multiple sites. Of the 4.9 million firms in the United States, only 0.5% have _500 employees; the majority of firms (99.5%) have _500 employees.7 These larger and smaller firms employ 51% and 36% of the working population, respectively.8 There were also striking disparities in the availability of worksite wellness programs by industry type. Manufacturing and business/ professional services reported having wellness programming 8.7% and 8.3% of the time, respectively, whereas wholesale/ retail (5.6%), transportation (2.9%), finance (2.4%), and agriculture/mining (1.4%) were much less likely to have comprehensive programming.2 Smaller employers face a number of barriers to offering wellness programs. Many of these smaller companies do not have a central human resources function to initiate and organize programs. Moreover, the expense associated with hiring a full-time health promotion staff is difficult to justify in a smaller company. Finally, because health insurance premiums are typically community rated, meaning that premiums for smaller companies are set by the medical utilization experience of their community, reducing their medical care costs by improving the health of employees will not decrease their insurance premiums.8 Consequently, an important financial incentive to develop worksite wellness programs is missing for smaller companies.

Returns on Investment
The payback for investing in worksite wellness programs can be measured in various ways, including decreased direct healthcare costs, improved healthcare utilization, increased performance measures, lower rates of absenteeism, and a reduced prevalence of chronic disease. Financial Returns Employer spending on health promotion and chronic disease prevention and management is a good financial investment when it succeeds in modifying the health of employees. Migration to a lower risk status is estimated to save $53 per employee, savings that recur each year that the employee remains in a low-risk tier.9 Programs have achieved a rate of return on investment that ranges from $3 to $15 for each dollar invested, with savings realized within 12 to 18 months.10 Meta-analyses have shown a 28% average reduction

in sick leave absenteeism, a 26% reduction in healthcare costs, and a 30% decrease in workers compensation and disability management claims costs.11 Other benefits to the companies that offer such programming are recruitment and retention of top employees, as well as an improved corporate image.10 Absenteeism/Presenteeism Employers have to absorb the indirect expense of lost productivity from employees who have chronic illnesses when the employee is absent from the job (absenteeism) or is at the job but impaired because of a health problem (presenteeism).12 Employees with the greatest health risks, poorest emotional health, and higher percentages of adverse behaviors had much higher rates of lost workdays and lower productivity overall.13,14 In a cross-sectional analysis of 2264 employees at a single employer, the rates of absenteeism and presenteeism were estimated to range from 0% to 6.3% and from 1.3% to 25.9%, respectively, among employees with up to 8 risk factors.15 The number of workdays lost was directly associated with the number of risk factors among 2250 employees of a single petrochemical facility; the presence of 0, 1, 2, 3, and 4 or more risk factors was associated with 4.1, 6.4, 8.8, 9.3, and 12.6 days of absenteeism, respectively.13 Bank One attempted to determine whether absenteeism or presenteeism is more costly and estimated that absenteeism represented 6% of total medical costs (direct and indirect), whereas presenteeism was responsible for 63%.16 Observational studies and interventions have shown that changes in health risk factors are directly related to changes in absenteeism and presenteeism. Individuals who reduce 1 health risk factor decrease presenteeism by as much as 9% and absenteeism by 2%.17 Research demonstrates a strong relationship between changes in health risk factors with changes in presenteeism and resultant productivity. Each risk factor increased or reduced was associated with a commensurate change in productivity of 1.9% over time; the savings were estimated to be $950 per year per risk that was reduced.18 Productivity Chronic diseases have a significant adverse influence on productivity; however, it is difficult to quantify productivity in todays postmanufacturing economy, in which so little of what is produced can be measured.16,19 As a result, most productivity estimates are based on questionnaires that can yield widely different estimates of on-the-job productivity gains or losses even when administered in the same setting. 2025 The results of various reviews suggest that on-the-job productivity losses can approximate from 20% to _60% of total health-related costs. It is estimated that health-related productivity losses cost US employers $225.8 billion per year or $1685 per employee per year, of which 71% is due to reduced performance at work.26 Depression alone, a risk factor for new and recurrent CVD and stroke,27 costs US employers approximately $35 billion in lost productivity.28,29 Studies evaluating productivity losses show that such losses are intimately linked to presenteeism and its associated health concerns.14,1922,24,30 Loss of productivity is related to both the severity of dysfunction caused by illness or disease and the summation of health risk factors present.15,3133 Estimates of productivity loss are between 12% and 28% for employees with 0 to 7 or more health

risk factors, respectively.34 As demonstrated in reports evaluating presenteeism, intervention trials aimed at reducing health risk factors have consistently demonstrated significant productivity gains.17,18 Moreover, others have documented the low level of treatment currently provided in the US workforce for many at-risk health conditions, including depression, and the opportunity for substantial productivity gains by undertaking worksite health promotion activities.26,28

Components of Worksite Wellness

Recommendations
A comprehensive program aimed at improving employees cardiovascular and general health should include thelowing: Tobacco cessation and prevention; regular physical activity; stress management/reduction; early detection/ screening; nutrition education and promotion; weight management; disease management; and changes in the work environment to encourage healthy behaviors and promote occupational safety and health (Table). Programming should be integrated into the organizational structure of the workplace by use of the following proven strategies: Health education that relies on existing valid sources and is focused on skill development that is consistent with employees readiness to make behavior changes; initiatives that are integrated into existing employee assistance programs; and voluntary worksite screening linked with medical care for follow-up on modifiable risk factors. Employers should administer health risk appraisals in combination with organizational health promotion checklists that have already been developed for the worksite before initiating programming so that health needs in the workplace can be identified and employees can learn their risks and health status. Employees health risks must be addressed within comprehensive worksite programs. Research should investigate the effectiveness of wellness programming and how to tailor programming and policies for maximum effect. When possible, planning and implementation of worksite wellness programs should optimize use of on-site personnel, physical resources, and organizational capabilities to make it easier for employees to participate.

Content
Tobacco Cessation and Smoking Prevention Direct and environmental (ie, secondhand smoke) exposure to cigarette smoke is associated with substantial morbidity and mortality due to CVD and stroke. Cigarette smokers are 2 to 3 times more likely to die of CVD35 and twice as likely to die of stroke.36 Nonsmokers who are exposed to secondhand smoke at home or at work have a 25% to 30% greater likelihood of developing heart disease.37 Tobacco use in the workplace costs US businesses an estimated $92 billion per year.38 Losses stem from increased healthcare utilization by employees who smoke, decreased productivity, and the exposure of nonsmoking employees and customers to secondhand smoke. On average, smokers miss 6.2 days of work per year compared with nonsmokers, who miss 3.9 days per year.39 The American Productivity Audit of the US workforce reported that tobacco use was a stronger correlate of lost production time among employees than age, alcohol consumption,

family emergencies, or education; lost production time increased in a dose-dependent manner in relation to the amount smoked.39 Studies conducted in the 1970s and 1980s (before the implementation of workplace smoking ordinances) estimated that environmental smoke exposure was responsible for an additional $490 in healthcare expenditures per smoker per year.40,41 A combination of strategies have been used to educate workers about the health consequences of smoking and to help employees stop smoking through interventions that
Table. Components of a Work-Site Wellness Program for Cardiovascular Health
Component Description CVD education Employer organizes and promotes classes and/or provides materials to educate employees about CVDs, stroke, and emergency response. Topics should include the following: (1) Types of CVD and prevalence; (2) risk factors and management; (3) awareness of symptoms and appropriate emergency action; (4) CPR/AED training; and (5) effective use of the healthcare system. Tobacco cessation and prevention Workplace is tobacco free, and employer organizes and promotes services to increase the rate of tobacco cessation and tobacco use prevention among employees and families. Early detection and screening Work site offers employees annual health risk assessment for a range of conditions* and provides feedback and tools to encourage tracking. Other cardiovascular and stroke screening is offered to increase awareness, prevention, treatment, and control of the key risk factors and identify the need for disease management. Weight management Employer offers a safe and effective weight management program that encourages employees to follow a sensible eating plan and engage in regular physical activity. Nutrition Work site provides general nutrition education and/or healthy eating information to the employee population. Examples include a dedicated World Wide Web site, newsletters, e-mail reminders, and point-of-service materials in the cafeteria and/or near vending machines, as well as group classes and individual counseling sessions. Cafeterias and vending machines provide healthy food choices. Physical activity Work site provides accessible indoor or outdoor exercise facilities and programming supporting the adoption of a physically active lifestyle. Examples include an indoor walking path with a mile distance marked off; lighted, attractive stairwells; provision of maps for safe and convenient walking outside the office; and free or markedly reduced access to exercise clubs. Stress management Employers provide education about stress reduction and stress management. Employers work to diminish work-related stressors such as job strain, effort-reward imbalance, long work hours, shift work, and work-family conflict to allow employees the opportunity to improve their work performance and minimize health consequences from stress overload. Environment The worksite should modify the physical and social environment to promote optimal cardiovascular health and

wellness. Examples of physical modifications include improving workplace safety, modifying work stations and office layouts to decrease sedentary behavior, and encouraging physical activity. Social changes include implementing policy changes that build a healthier work culture and appointing members of leadership who are responsible for ensuring commitment and adherence to wellness programming. Occupational safety and health Employers should address all hazards to employee health and safety, providing working conditions that are optimal for cardiovascular health and well-being. CPR/AED indicates cardiopulmonary resuscitation/automated external defibrillator. *Assessment should be considered for the following: blood pressure, body mass index, cholesterol, blood glucose, cigarette smoking/tobacco use, and mental health

have proved effective in other settings.42 A Cochrane review of workplace interventions for smoking cessation identified 51 studies in the literature covering 53 interventions between 1966 and 2008.43 Most of those interventions (n_37) were focused on modifying the behavior of individual workers through group therapy, self-help materials, individual counseling, pharmacological treatment for nicotine addiction, and social support. The remaining studies included interventions aimed at modifying the workplace and included incentive schemes and company competitions. In general, treatments that targeted individual smokers, in particular group counseling and pharmacological agents, were the most successful. Participant quit rates and sustained cessation rates for 6 to 12 months after the intervention were comparable to those when interventions were implemented in other settings.43 Incentive schemes and company competitions did not generate high levels of employee participation, nor did they significantly reduce the prevalence of smoking. Clean indoor air laws have had an important influence on smoking in the workplace. These laws have spread across the country, blanketing most of the working population with smoke-free air and lowering smoking rates. City and statewide legislation prohibiting indoor smoking has decreased smoking prevalence.44 These notable successes suggest that organizational or policy interventions may decrease smoking and have the greatest benefit for the largest number of people. Although smoking inside the workplace is not as common today in the United States as it was in the 1970s and 1980s, only 77% of indoor workers reported that their workplace had policies that restricted smoking behaviors.45 According to the 2004 National Worksite Health Promotion Survey, 40% of worksites completely prohibited smoking on worksite property, and another 56.5% restricted smoking to outside areas only.6 The prevalence of smoking is higher in minorities and persons in lower socioeconomic and occupational classes, and these same groups and women are more likely to be exposed to secondhand smoke.37,46 Blue collar workers are less likely than white collar indoor workers to be covered by smoke-free policies, and workers in certain occupations, such as trucking and fishing, are not covered by smoke-free air laws.47 49

In summary, interventions that target individual smokers are successful, but to achieve maximum effectiveness, they should be used in combination with workplace policies, including complete worksite smoking bans. Physical Activity Regular physical activity is recommended to promote and maintain health and to prevent the development of cardiovascular risk factors and related chronic diseases.50 Opportunities for physical activity can be sought during leisure time, can be acquired during active transportation, or can arise in response to occupational duties; however, the likelihood of the workplace serving as a significant source of physical activity has declined, because contemporary work environments are sedentary. Physically demanding work has been reduced or eliminated in many sectors and replaced by labor-saving devices focused on speed, rapid communication, improved efficiency, and increased productivity. Physical inactivity and sedentary behaviors are associated with higher rates of clinical CVD (relative risk of 1.9 for inactive versus active persons in a meta-analysis of 43 studies),51 CVD risk factors,3 and stroke.52,53 Consequently, an important strategy by which employers can lower CVD risk is to provide opportunities for activity in the workplace. There are a number of strategies whereby companies have tried to promote activity in the worksite, including educating employees about the benefits of activity, providing access to safe spaces for activity, and modifying the built environment so that employees can incorporate activity into their work days. A pioneering study of an at-work stair-climbing program in healthy men showed that an appropriate daily training stimulus (approximately 25 flights for a 70-kg man) resulted in a significant increase in aerobic capacity.54 More recently, investigators reported that sedentary adults who exclusively used stairs instead of elevators at work demonstrated increases in cardiorespiratory fitness and reductions in body weight, waist size, and blood pressure.55 The average daily number of floors ascended or descended by each participant increased from 5 to 23 per day. Worksite physical activity counseling has positive effects on daily energy expenditure and cardiorespiratory fitness.56 Studies outside the occupational setting indicate that increasing activity by relatively small amounts can have substantial health benefits in at-risk populations. In the Dose Response to Exercise in Women trial,57 previously sedentary overweight women who met even 50% of the consensus recommendations for physical activity over the 6-month intervention improved their fitness.50 A worksite intervention that provided pedometers to employees to achieve 10 000 steps daily succeeded in increasing physical activity and weight loss and reducing blood pressure.58 Using similar technology, another study reported that the combined use of an accelerometer (a portable watch-sized device to capture movement in both the vertical and horizontal planes compared with just the horizontal plane that a pedometer captures) and World Wide Web site that tracked activity also improved physical activity behaviors in previously sedentary employees.59

The adverse influence of sedentary behavior on health has received increased attention. Hamilton and colleagues60,61 have shown that sedentary behaviors alone, especially sitting, are associated with higher rates of morbidity and mortality, cardiovascular risk factors, type 2 diabetes mellitus, and metabolic syndrome, as well as the physiological derangements that adversely influence lipid metabolism. An employee sitting at a desk is expending 1 metabolic equivalent (1 metabolic equivalent_3.5 mL O2 kg_1 min_1), whereas even the slowest walking (eg, _1 mph) increases an employees metabolic rate to 2 metabolic equivalents.62 Nonexercise activity thermogenesis, the spontaneous physical activities of daily living (including fidgeting while sitting and standing while reading), isa source of energy expenditure for most people.63 When matched with individuals with similar cardiovascular risk profiles, those with highly active ambulatory jobs can have nonexercise activity thermogenesis values _1000 kcal/d higher than their sedentary counterparts.64 Thus, efforts to reduce sitting time through innovations in worksite design and policies can have a significant influence on decreasing sedentary behaviors. Standing workstations and vertical computer desk designs with slow-moving treadmills placed underneath represent innovative workstation designs that can substantially increase nonexercise activity thermogenesis. Treadmill walking at extremely slow speeds (_1 mph) generally does not interfere with a workers ability to use the computer or talk on the telephone. Most importantly, workers reported that they enjoyed using the contemporary workstations and that they supported their use in their own work environment.65 Additional means for increasing physical activity that can be integrated into the workplace include stepping devices,66 basic resistance training equipment, standing workstations, encouraging the use of stairs, centralizing office resources so employees have to walk to access them, encouraging employees to stand while talking on the telephone, walking to deliver messages or have conversations with colleagues versus e-mailling, and holding walking meetings. Many of these activities can be tailored to employees with physical disabilities, highlighting their universal applicability. In summary, habitually sedentary and/or unfit men and women should be counseled to improve their exercise tolerance by starting and maintaining a regular physical activity program that includes structured exercise, increased lifestyle activity, or both. Organizational interventions may include modified workstations that encourage standing or moving and readily available places for activity in the workplace, such as well-lit staircases to promote active ambulation. Stress Management/Reduction Although workplace stress can be attributed to numerous sources, including job insecurity, long working hours, work scheduling, and organizational restructuring, it is most commonly defined as an imbalance between job demands and control (ie, job strain). Work-related stressors that demonstrate robust associations with CVD in the research literature include job strain (ie, high-demand low-control work), effort-reward imbalance (ie, high work efforts combined with low rewards such as support,

respect, security, and income), long work hours, and shift work. These stressors reduce employees ability to work by diverting their attention away from job responsibilities to addressing or coping with the stresses.67 Additionally, high levels of stress have been associated with the development of cardiovascular risk factors and impaired job performance.68 Worldwide, approximately one quarter of working women and 18% of men report high levels of job-related strain.69 71 A 2004 comprehensive review of studies of the association of job strain with CVD risk factors concluded that the weight of evidence suggests that job strain is a CVD risk factor.72 However, a more recent prospective study, which was composed largely of women, did not find a significant association of job strain with ischemic heart disease; rather, low job control is correlated with a significant doubling in the risk of developing heart disease.73 Although further research is required to identify which components of job stress are most strongly associated with heart disease, recommendations to implement stress management programs at both the individual and organizational level are warranted. The 2004 National Worksite Health Promotion Survey reported that one quarter of worksite wellness programs included stress management programming.6 Individualcentered approaches involve teaching employees skills for managing pressures and demands. Such strategies include cognitive behavioral therapy, relaxation techniques, and individual counseling focused on adopting healthy lifestyles.74 A systematic analysis of the literature on job-stress interventions revealed that the greatest impact occurred when the intervention was both organizationally and individually focused. 75 Although individual-centered approaches may favorably modify behavioral issues such as smoking cessation or sedentary behaviors, they are less likely to reduce workplace stress because they do not address the organization of the workplace management approach.76 In a systematic review of organizational-level interventions designed to improve employee control, workers experienced health benefits. Egan et al77 reviewed 18 relevant studies, 11 of which noted improvements in health and none of which reported adverse health effects; however, the authors acknowledged that the organizational interventions were complemented by health education efforts. Approaches that target management within a workplace have proved modestly successful. One study78 used a unique approach and provided a randomized (by worksite) World Wide Webbased supervisor training program on worksite mental health, supervisor support, and psychological distress among subordinate workers. Subordinate workers at the intervention (n_81) and control (n_108) sites completed a brief job stress questionnaire at baseline and at 3-month follow-up.78 Although workplace autonomy and overall job stressors did not differ between subordinate workers at the intervention or the control sites, the item score for a friendly workplace atmosphere increased significantly (P_0.02) at the intervention site, whereas there was no change at the control site. A prospective study of ischemic heart disease events associated with employees subjected to different approaches

to supervision gives insight into potential interventions and the role of organizational policy. Employees whose supervisors provided clarity in goals and role expectation, communicated well and offered feedback about performance, and encouraged employee participation and control were significantly less likely to experience acute myocardial infarction, unstable angina, and cardiac deaths over a 10-year follow-up period.79 There are methodological challenges inherent in these interventional studies. The workplace is frequently changing because of closings, mergers, downsizing, or restructuring,

1730 Circulation October 27, 2009

Downloaded from http://circ.ahajournals.org/ by guest on January 24, 2012

and these changes should be considered when interventional research about job stress is planned. Fundamental restructuring of the workplace influences the effectiveness of the intervention and raises serious questions about the analysis and interpretation of the research results.80 Comprehensive approaches that address both the organizational origins of workplace stress and the behavioral symptoms exhibited by employees are more likely to lead to favorable sustainable outcomes.81 The organizational focus has the added benefit of reaching employees across job classes. Lower-paid and less-educated workers in typical blue collar positions are commonly segregated into low-control/ high-demand positions that are at greater risk of exposure to occupational stress. Historically, these workers have not responded to individual-centered wellness initiatives,74,82 and they are more likely to experience benefits through organizational and occupational changes that make the workplace safer, institute rules and policies about working hours, and provide the protection of worker organizations to advocate for employee rights. Screening and Early Detection Health screening in the employee population requires significant resources but can serve as an investment for employers. Guidelines from the AHA and other public health organizations stress the importance of increasing awareness, prevention, treatment, and control of major risk factors for CVD and stroke.83 Regular recommended screening for heart disease and stroke may identify risk factors such as diabetes, obesity, hypertension, or abnormal blood lipids, allowing for lifestyle/ behavior change and effective use of medications to address these before the manifestation of disease sequelae. If adverse cardiovascular signs or symptoms have developed, there is opportunity for disease management. According to the US Census Bureau, 37 million employed persons were without health insurance in 2007,84 which diminished the likelihood that they would seek preventive health services, including disease screening. Wellness programs that incorporate screening can fill an important national healthcare void by identifying treatable conditions. In order for disease screening to be effective, the diagnostic testing must have high rates of sensitivity and specificity.85 It is important to minimize false-positive results that may lead to further unnecessary and costly evaluations and potential complications. The 2004 National Worksite Health Promotion Survey grouped screenings offered at the workplace and through

employee health plans to estimate and rank the prevalence of such programming in the United States. Blood pressure, cholesterol, and diabetes were the most commonly reported CVD screenings, at 36.4%, 29.4%, and 27.4%, respectively, with alcohol and drug abuse support and cancer screenings also frequently provided.6 Despite the recommendations by the US Preventive Services Task Force for simple screening for obesity,86 adiposity measures were not reported in the National Worksite Health Promotion Survey. Given the prevalence of overweight and obesity and its role in the development of CVD, not assessing the height and weight of employees represents a missed opportunity. Screening programs were most common in the largest worksites. Screening programs were present in 62% of the largest workplaces (_750 employees) and only 16% of workplaces with 50 to 99 employees. According to a recent report on 20 000 employees in 13 US workplaces, self-testing stations for blood pressure and body weight are an attractive screening tool for any size workplace. 87 One fifth (21.7%) of employees used the health station at least once during the first 18 months it was installed, with many returning for repeated visits. The value of such stations is that employees can privately screen themselves for adverse health conditions, and if used in combination with existing wellness programming, employees could self-identify themselves as needing additional lifestyle and/or pharmacological interventions. Additionally, research supports the use of blood pressure measurement outside of clinical offices as an important supplement to these readings to provide a more comprehensive assessment of 24-hour blood pressure modulations.88 In summary, workplace screening for CVD and stroke risk factors has the potential to identify workers who are at risk for disease and who should be encouraged to participate in risk-reduction behavior change programs. The identification of workers with latent disease early in the course of disease provides the opportunity to delay cardiovascular complications and thus decrease the resulting time away from work and the less productive time at work displayed by workers with chronic diseases. Nutrition Education and Weight Maintenance A healthy diet and other lifestyle practices are the cornerstone of the AHAs prevention and treatment activities. The AHAs diet and lifestyle recommendations89 promote healthy diet and lifestyle practices to reduce major CVD risk factors (eg, overweight and obesity, high low-density lipoprotein cholesterol and triglycerides, low high-density lipoprotein cholesterol, and elevated blood pressure and blood glucose levels). Controlling these major risk factors with recommended lifestyle behaviors markedly reduces the risk of CVD and stroke.90 The Centers for Disease Control and Prevention estimated that a 10% weight loss will reduce an overweight individuals lifetime medical costs by $2200 to $5300 by lowering costs associated with the treatment of hypertension, type 2 diabetes mellitus, heart disease, stroke, and high cholesterol.91 Reducing these chronic diseases and underlying risk factors through good nutritional practices may result in economic benefits, decreased healthcare costs, and greater employee productivity in the workplace. The AHA has issued

guidelines with strategies for promoting healthy diets for all adults. Those that can be adopted in the workplace include ensuring access to healthy foods (eg, fruits, vegetables, whole grains, skim milk dairy products, fish, lean meats and poultry, and plant-based meat alternatives); increasing offerings of food choices that are low in saturated fat, trans fat, sodium, added sugar, and calories; and providing nutrition labeling at the point of purchase (eg, in the cafeteria and vending machines).89,92

Carnethon et al Worksite Wellness Programs for CVD Prevention 1731 Two reviews evaluated studies detailing the effects of worksite wellness interventions and reported the following benefits: Availability of nutritious foods, point-of-purchase information, systematic reminders and training of healthcare providers to provide nutrition counseling, and incentives to encourage the purchase of nutritious foods.93,94 The intervention strategies varied from providing health education opportunities, changing the availability of healthy foods, and providing incentives such as lower prices, games, and prizes, most of which were associated with favorable outcomes.94 Other studies have shown that when trying to reach high-risk populations, it is important to address job hazard exposures and other areas of job security to gain credibility, which in turn, increases audience receptivity and participation.9597 A national survey of approximately 3000 employees was conducted online on 2 occasions in July 2007 by HarrisInteractive for the AHA to assess the role that leadership plays in creating an atmosphere in which employees feel free to actively take advantage of worksite wellness programs.98 The participants reported improvements in a number of health outcomes as a result of worksite wellness programs. Those related to nutrition were as follows: Feel better (40%), eat healthier (36%), lost weight (32%), lowered cholesterol (19%), and lowered blood pressure (18%). Employees also reported positive work performance outcomes: Fewer sick days (47%), better productivity (36%), improved quality of work (25%), and higher job satisfaction (21%). In addition, they valued nutritious food choices and effective weight loss programs. Employees reported that nutrition and weight loss programs had the biggest positive influence on health or health habits at 39% and 38%, respectively. These results demonstrate the benefits of worksite wellness programs in individuals who elected to respond to the survey (21% and 11% in the 2 samplings), which leaves questions about the impact in the nonrespondents. However, HarrisInteractive attempted to control for this by demographic weighting using data gathered from previous research. Because most of the adult population is overweight or obese (66.7%), weight loss and weight maintenance programs are an important component of worksite wellness and health promotion.3 Interactions between work, obesity, and occupational safety and health, in cases in which obesity may be associated with adverse work conditions, may increase when employees are in demanding, low-control jobs with long hours.99 Concomitantly, obesity adds to the escalating cost of health insurance and is linked to the number of people who are either uninsured or underinsured.100 Medical spending on obese employees is 37% higher than for people of normal weight.101 Severely obese women experience more than

double the absenteeism of normal-weight women.102 As prevalence data demonstrate, obesity rates are higher in lower socioeconomic groups, an issue that is especially relevant for employers who hire a greater percentage of low-income workers. Employers whose workers fall within the lower socioeconomic classes may pay greater healthcare premiums or anticipate higher healthcare utilization. A recent analysis of employer and employee attitudes toward weight loss programs in the workplace found that both groups view weight loss programs as appropriate and effective. They favored positive financial incentives as motivation for employers and employees to participate in these programs and strongly opposed punitive financial penalties. Employers, especially smaller companies, favored tax incentives for businesses that incorporate weight loss treatment and weight maintenance incentives into worksite wellness.100 In summary, an assessment of worksite wellness nutrition programs indicates that these are generally effective in favorably modifying dietary practices consistent with current recommendations and in reducing major cardiovascular risk factors such as overweight/obesity, hypercholesterolemia, and hypertension. These findings reinforce the need to accelerate worksite nutrition education and weight management activities. Moreover, a Task Force on Community Preventive Services has recommended a combination of nutrition and physical activity programs for worksite based interventions to prevent and control overweight/obesity.103 The implementation of a dietary modification in the worksite can simultaneously target multiple CVD risk factors that are closely linked with dietary intake and obesity.

CVD Education and Automated External Defibrillator/Cardiopulmonary Resuscitation Training

Each year, CVD claims the lives of _864 000 Americans, which makes it the leading cause of death in the United States; there are at least 265 100 Americans each year who experience out-of-hospital cardiac arrests and need emergency care before they reach a hospital.3 Many of these lives can be saved if employees are educated about the chain of survival, in which workplaces are equipped with automated external defibrillators and employees are trained in how to phone 911, begin cardiopulmonary resuscitation, and deliver early defibrillation. The provision of defibrillation on-site with an automated external defibrillator can dramatically increase survival rates for cardiac arrest. Employers should offer classes and/or provide materials that educate employees about CVDs, stroke, and emergency response, including management of risk factors, signs and symptoms of stroke or cardiac arrest, and appropriate emergency response. Delivery The education of employees about risk factors for and signs and symptoms of CVD and stroke is an integral component of worksite wellness programming. Resources for educating employees about risk factors for CVD and stroke are readily available from credible sources, including the AHA104 and the Centers for Disease Control and Prevention.105,106 However, worksites should modify the content as needed to deliver it in a manner that is targeted to the employees level of health literacy.

Education should additionally focus on skills development so that employees have strategies for making healthy changes. To further promote uptake, content should be delivered in stages that are consistent with the employees stage of readiness for behavior change.107 The incorporation of behavioral theories from the behavioral medicine literature that are appropriate to employees background knowledge and readiness for change will ensure maximum success.

Other Aspects of a Worksite Wellness Program

Numerous additional features can improve the implementation and uptake of worksite wellness programs.108 The development of a mission statement for the program can help clarify program goals and desired outcomes. Engaging employees to develop the statement can help with buy-in to the program. The performance of health risk assessments before implementation of a wellness program can be an effective way to raise employees awareness of their health risks and to engage them in a wellness program. Repeating assessments at regular intervals can determine the progress and success of the wellness program in the employee population. To maximize the effectiveness of health risk appraisals, they should incorporate questions about socioeconomic status and education, because these factors are established correlates of poor health and may inform intervention strategy. Health risk assessment surveys can be supplemented with organizational health promotion checklists (eg, Heart Check Lite109) that have already been developed for the worksite. To reassure employees that wellness programs are for their benefit and not related to their job performance or responsibilities, procedures should be in place to ensure confidentiality and privacy. Employers should engage employees in the development and implementation of the program. To reach all employees, wellness programs should have specific policies that address employees who telecommute or work from remote locations. Each program should be an active learning system in which outcome evaluation is an integral component. Regular, timely, personalized communication is an essential component of a program (eg, a powerful Internet interface that registers, engages, tracks, and evaluates each eligible participating member). Employees should have the opportunity to participate in programming individually when possible through self-help modules or group sessions where applicable. Program outcomes should be assessed annually. The administration of additional surveys for employees who express interest in wellness programs to gauge important constructs such as readiness to change, interest in participating in specific programs, health risks, and current preventive care can ensure that programs are tailored to be of the greatest interest to as many employees as possible.

Environmental Interventions
Recommendations
The social and physical environment of the workplace should be designed to be conducive to recommended behaviors. Optimal environmental modifications should promote healthy behaviors while simultaneously minimizing the physical and organizational risk in the work environment. Occupational safety and health are integral components of

worksite wellness; workplaces should be free from hazards that jeopardize cardiovascular health and employee safety and well-being.

Modifying the Workplace to Encourage Positive Behavior Change

Raising awareness about healthy lifestyles through education is paramount, but it is also important for wellness programming to build supportive environments and implement policies to encourage healthy lifestyles. Environmental interventions are defined as those that do not require individuals to self-select into defined educational programs but are implemented for all employees.111 A number of environmental modifications have proved successful at facilitating healthy behaviors and decision making, such as modifying the physical plant to encourage physical activity or ensuring available healthy food options in the cafeteria or vending machines. Engbers et al112 reviewed 13 randomized controlled trials published between 1985 and 2004 that included environmental modifications in health promotion programs at worksites. Four studies focused on cardiovascular risk factor reduction, 8 on cancer risk reduction, and 1 on a healthy lifestyle in general. All of the interventions had multiple components (education and environmental); 3 focused on physical activity; and all studies used environmental modifications to increase consumption of fruit, vegetable, and fiber and to reduce fat intake. The investigators found significant effects of environmental interventions on dietary intake and inconclusive evidence that physical activity was favorably modified. The National Heart, Lung, and Blood Institute has supported environmental interventions to prevent overweight and obesity at worksites.113 One-year results from 1 of those studies suggest that such interventions, when superimposed on existing education programs at the worksite, can lower body mass index. One successful strategy to increase consumption of lower-fat foods is to reduce prices. French114 found that price reductions of 10% to 50% on lower-fat foods (eg, low-fat snacks and vegetables) resulted in a 9% to 93% increase in sales of those items. Numerous studies (for instance, those reported by the Robert Wood Johnson Foundations Active Living Research program) have demonstrated that the built environment influences associated physical activity patterns. For example, open and accessible stairwells and on-site physical activity facilities at worksites increased the physical activity of employees.115,116

Reducing Health Risk in the Work Environment

An increased risk of adverse cardiac events has long been associated with chemical and physical hazards at work. For example, occupational exposure to carbon monoxide (often from gasoline combustion) may cause both angina and acute myocardial infarction. Methylene chloride (used in furniture stripping) is metabolized to carbon monoxide, which can trigger cardiovascular events. Lead exposure and noise have been associated with hypertension in some investigations.117 Environmental tobacco smoke in workplaces such as casinos and bars poses a health risk to employees.118 Environmental studies have long identified exposure to fine particulates as posing cardiac risk in the general population,119,120 and toxicology studies have identified elicits myocardial ischemia in men with coronary artery disease.122 All of these exposures can be monitored and controlled.

Shift work is also associated with sleep disturbances that may increase CVD risk. Unusual bursts of vigorous physical activity can be hazardous, particularly in workers with underlying coronary artery disease. Bursts of activity in untrained or at-risk workers combined with particulate exposure may explain the increased risk of on-the-job cardiac events reported in some studies of firefighters.124

Regulatory Oversight of Worksite Wellness Programs

Policy Recommendations
The regulatory environment should allow for increased opportunity for employers to reach a greater majority of the employee population and produce health benefits. Employers should adhere to all regulations that address hazards to employee health and safety, providing working conditions that are optimal for cardiovascular health and well-being. Regulatory provisions should provide wellness credits for employers who choose to provide healthy lifestyle behavior incentives in the workplace (eg, health promotion services, smoking cessation programs, exercise facilities on site, weight loss programs, or voluntary screenings), and financial incentives can be paid directly to the employee. However, financial incentives should not be attached to healthcare premiums or health status.

Legislative/Regulatory Oversight of Worksite Wellness Programs

Worksite wellness programs are affected by state laws and major federal laws, including the Americans With Disabilities Act of 1990, the Health Insurance Portability and Accountability Act of 1996 (HIPAA), the Age Discrimination in Employment Act of 1967, the Genetic Information Nondiscrimination Act of 2008, the Occupational Safety and Health Act of 1970, the Consolidated Omnibus Budget Reconciliation Act of 1985, the Employee Retirement Income Security Act of 1974, the Mental Health Parity Act of 1996, and the Pregnancy Discrimination Act of 1978. Conversations about health reform have led to a specific review of the HIPAA nondiscrimination provisions and a debate about whether employees who engage in unhealthy behaviors should be held accountable via an increase in their healthcare premiums or an adjustment to their deductibles. HIPAA provisions generally prohibit a group health plan from charging individuals different premiums based on a health factor. However, Congress did not want to prevent the many promising efforts that are using health-planrelated incentives to encourage worksite health promotion, so it permitted group health plans to establish premium discounts or rebates or modification of otherwise applicable copayments or deductibles in return for adherence to programs of health promotion and disease prevention.125 Subsequent regulation delineated between participation-only programs that have no outcome requirements and those programs that require an employee to achieve a certain standard.126 Participation-only programs do not have to meet additional requirements provided they are available for all relevant employees; however, programs that are standard-based must meet additional benchmarks.126,127 The Americans With Disabilities Act regulations are administered

by the US Equal Employment Opportunity Commission, which covers all civil rights and disability legislation affecting employment.126 The Equal Employment Opportunity Commission126 has expressed concern about incentives and other components of worksite wellness programs, including completion of an annual mandatory health risk assessment; the use of monetary incentives connected to program participation or wellness activities; mandatory medical examinations or testing; employers making inquiries about obesity, heart disease, diabetes, or other disabilities; triggers for refusal to participate in disease management or behavior change programs; and employer inquiries concerning prescription drug use. At this time, the Equal Employment Opportunity Commission has not issued formal opinions on these issues. Consequently, employers who develop programming that violates these initial rulings risk Equal Employment Opportunity Commissionimposed prosecution and fines.

Financial Incentives
Behavior change is difficult, especially over the long term, and voluntary programs to encourage lifestyle modification are not consistently effective.128 Employee wellness program managers generally find that once the newness and curiosity about a health promotion/wellness program wears off, employee participation drops off in a dramatic way.8 The proportion of employees taking advantage of such programs falls short of the Healthy People 2010 goal of 75% participation, with average participation rates of only 61%.108 Accordingly, employers are increasingly using incentives to maintain program participation and enhance compliance.129 Incentives provide the employee with an immediate and tangible reward that helps make it easier to modify behaviors that may yield long-term benefits.130 Studies show the favorable outcome associated with the use of financial incentives to foster long-term behavior change, such as quitting smoking or losing weight, especially if the financial incentives are sufficient.131134 Finkelstein and colleagues133 found that a 3-month incentive-based intervention led to weight loss in employees who were provided 2 levels of incentives ($7 and $14 per percentage point of weight loss), which suggests that modest financial incentives can be effective in motivating overweight employees to lose weight. These programs indicate that employers could use incentives (eg, price reduction, monetary incentives, awards, and prizes) to encourage employees to undertake health improvement practices. Other studies have not found incentives to be effective; however, many of these studies are limited by small numbers of participants, cross-sectional designs, and/or very modest awards.135 Traditionally, incentives have been directed toward providers of healthcare services through pay-for-performance programs. There is a growing consensus that incentives should be provided directly to the individual who is engaging in the behavior change.132,136 Additional studies are needed to determine the true efficacy of incentive rewards within worksite wellness programs and whether these promote

1734 Circulation October 27, 2009

Downloaded from http://circ.ahajournals.org/ by guest on January 24, 2012

robust, long-term behavior change, whether incentives should be linked to the associated outcomes, and whether positive or negative incentives are more effective. Accordingly, the

regulatory environment should provide employers the latitude to offer incentives that are not discriminatory and are appropriate for employees, incorporate serial outcome evaluations, determine the influence on disparate populations, and include an analysis of unintended consequences. Although the AHA supports financial incentives paid directly to employees, there is significant concern with incentives tied directly to health insurance premiums or deductibles. Premium surcharges or other cost-sharing measures can make coverage less affordable for those who need it most and increase health disparities among low-income and minority populations. People who cannot afford coverage may have reduced access to therapies and interventions that can help curb unhealthy behaviors (eg, prescription smoking cessation therapy, medically supervised weight loss programs, and medications to control cholesterol and blood pressure). When policies cover a family, the surcharge penalizes every family member. The current regulatory environment allows employers to offer a premium discount or waiver of a cost-sharing requirement based on participation in a program of health promotion or disease prevention. These programs have not been evaluated to determine whether (1) reduced costs result from improved health outcomes or segmentation of the insurance pool, (2) individuals with chronic conditions who may have medical or physical limitations have been treated fairly with regard to these policies, and (3) privacy issues and individual autonomy in the workplace have been protected.

Vulnerable Populations
Policy Recommendations
Wellness programs must address the needs of all employees, regardless of sex, age, ethnicity, socioeconomic status, culture, job type, or physical or intellectual capacity. Worksite wellness programs should be designed to be culturally sensitive and all-inconclusive, and employers should also consider targeted, complementary interventions for their more vulnerable employees specifically designed to engage those who are economically challenged, less educated, or underserved. Worksite wellness programs should help working families balance work and family commitments and incorporate policies around child care, elder/dependent care, telecommuting, and flexible work schedules. Research should be conducted to determine how to improve participation among employees who have the highest-risk behaviors.

Blue Collar/Service Workers

Lower socioeconomic status is an established risk factor for CVD and stroke,137 yet blue collar workers, who predominantly fall into those lower socioeconomic classes, are often overlooked. Blue collar and service workers generally have less access to worksite wellness programs and are at greater risk of practicing lifestyle behaviors that place them at higher risk for coronary heart disease and stroke.138 Lower-income, less-educated, and lower-job-status employees have a higher burden of CVD than their higher-status counterparts.121,133,139 Although social conditions outside of work contribute to their disease experience, so do factors associated with their jobs,

including higher levels of job stress, job insecurity, long working hours, sedentary work, work scheduling issues, shift work, bullying, and harassment.71,140 These factors underscore the need to specifically target employees with lower income and less education in worksite wellness programs.141,142 Blue collar workers and lower-paid workers are, for example, more likely to smoke than those who are white collar or higher paid143 and are less likely to participate in worksite fitness programs.144 The combining of efforts at worksite health promotion with improved worksite protection appears to overcome some of these obstacles. A prospective controlled investigation of smoking cessation in blue collar manufacturing workers demonstrated significantly higher quit rates and cessation maintenance through a program that addressed both individual behavior change and risk reduction in the work environment.82 A controlled investigation of a comprehensive intervention in matched groups of Dutch manufacturing workers showed significantly greater cardiac risk reduction in the cohort that underwent individual- and organizational-level interventions.145

Race/Ethnicity
A limited number of intervention studies have specifically addressed issues unique to race and ethnicity in the working population. One group-randomized study in Hawaii found that after a 2-year intervention, Pacific Islanders, men, and those in managerial positions had a higher body mass index than women and other ethnic groups.146 Cultural barriers and roles within the workplace were considered obstacles to weight loss. In North Carolina, the Centers for Disease Control and Prevention provided Racial and Ethnic Approaches to Community Health 2010 funds to the Eastern Band of Cherokee Indians to develop a community-based intervention to improve the health of a rural, mountainous community with a Native American population that has higher rates of obesity and type 2 diabetes mellitus than the state or the US general population. The program had a significant worksite wellness component for adults. During the first year, team members conducted formative research, formed coalitions, and developed a culturally appropriate community action plan to prevent type 2 diabetes mellitus. Participants in the worksite wellness component met dietary and physical activity goals, demonstrated reductions in body weight, and enjoyed the program. These results led to an expanded worksite wellness initiative to achieve further healthcare cost reductions.147

Women
Women in the workplace are often overlooked as a vulnerable population despite unique challenges posed by

Carnethon et al Worksite Wellness Programs for CVD Prevention 1735

Downloaded from http://circ.ahajournals.org/ by guest on January 24, 2012

pregnancy, family responsibilities, and menopause, a transition associated with heightened cardiovascular risk. A worksite wellness program that focused on middle-aged women working in the healthcare environment148 found that a preponderance of the women had undiagnosed hypertension, abnormal lipid profiles, glucose intolerance, and/or obesity. Thus, an employee population that would typically be well educated about personal health, wellness, and disease prevention was characterized by numerous

unhealthy behaviors, unrecognized disease, inadequate use of preventive health screenings, and numerous obstacles to program participation and follow-up. Time pressures continue to represent major challenges for women (and men) balancing professional, family, and personal commitments and should be considered when worksite wellness programs are developed. In 2002, 77% of single mothers, just over 60% of married mothers with children under the age of 6 years, and 76.8% of married mothers with school-aged children were in the labor force.149 Because working mothers and families balance time spent in their personal commitments and job responsibilities, workplace policies should incorporate child care, flexible schedules, and elder/dependent care as a key feature of workplace health promotion.150

Additional Research
Few data are available regarding the effectiveness of worksite wellness programs in diverse populations. Often, the most disadvantaged workers, who have the greatest need for preventive screenings, health promotion, and/or worksite wellness programs, have the least access and are the most reluctant to participate in these programs.142,151 The fundamental causes of vulnerability are rooted in employees daily lives and are most often beyond the scope of public health.152 It would be beneficial for employers to engage with nontraditional partners to consider ways to reduce health disparities in communities and improve employee well-being overall. Additional research is needed to determine how best to reach and engage underserved populations and optimize worksite wellness for employees of all races, ethnicities, and income levels. Research should focus on whether worksite wellness programs in high-risk and vulnerable populations improve health and reduce health disparities.

Conclusions/Summary
Successful worksite wellness programs engage employees in activities that maximize their potential for health and well-being, grow rapidly in response to their perceived value, and prove sustainable as they establish the business case for their existence. As such, government agencies could have a substantial influence by supplementing private sector investment in large-scale, objective, longerterm studies on programming and outcomes research to better inform the development, implementation, and evaluation of worksite wellness programs.153 Government agencies should also model worksite wellness programs and serve as laboratories for testing research-based lifestyle interventions. The health outcomes for high-risk and health-disparities populations should be particularly scrutinized. The Centers for Disease Control and Prevention recently developed the Healthy Workforce Initiative, a World Wide Web site designed as a resource for worksite health promotion program planners in state and federal government that is also an excellent resource for planners in nongovernment workplaces.154 Federal and state governments should encourage employers to offer programming by providing tax incentives for employers who implement comprehensive worksite wellness programs. Even the most well-designed and well-intentioned worksite

wellness programs are ineffective if employees do not participate. Employers should seek to reduce or eliminate barriers that discourage use of worksite wellness programs to increase participation and employee engagement. One of many obstacles for programs to overcome is the low participation rate among those most likely to have greater health risk. Offering health promotion services such as fitness centers, weight loss programs, and exercise classes on site and providing healthy vending and food choices throughout the workplace environment are small steps. More innovative and forward-thinking employers might consider providing a convenient time and location for exercise and wellness programs during the workday and offering employer-provided paid time off during the workday for exercise, health screenings, or prevention/wellness programs.155 Programs that combine individual and organizational changes boast the greatest success rates in part because combined approaches engender a reciprocal relationship in which employees have a perception that their needs are valued.156 An Institute of Medicine exploration of the design of worksite health programs has embraced this kind of comprehensive approach,157 and the National Institute for Occupational Safety and Health recommendations for effective worksite programs endorse comprehensive efforts that combine health protection and promotion.158 However, continued research is needed to determine the effectiveness of comprehensive programs compared with programs that provide only selective services. Additionally, more work is needed to assess the effectiveness of programs on hard outcomes such as ischemic heart disease and stroke. Visionary employers are looking beyond healthcare costs to consider the total value of health. They understand the importance of establishing work environments that engender fulfillment, improve quality of life, forge positive links with the community, and increase productivity. An effective worksite wellness program can attract exceptional employees, improve on-the-job decision-making and work efficiency, enhance employee morale and organizational commitment, decrease turnover, and reduce organizational conflict.108 Despite the numerous and varied documented benefits of incorporating programming to promote healthy lifestyles in the workplace, such programming has not achieved adequate penetration into the workplaces of America.

Move & Improve: A Worksite Wellness Program in Maine Background We describe the evaluation process and outcomes of Move & Improve, a worksite wellness program in Maine. The evaluation process was based on the Centers for Disease Control and Prevention's Framework for Program Evaluation in Public Health and community-based participatory research principles. Innovative approaches are required to address burgeoning chronic disease trends and risk factors. Worksites are an ideal setting in which to affect working adults and high-risk individuals. Using community-based participatory research methodology increases community capacity for evaluation, dissemination, and use of evaluation results.

Context Move & Improve is an ongoing program that was implemented in 1996. Although evaluation data have been collected since the program's inception, a more systematic evaluation based on community-based participatory research principles was undertaken in 2003 and 2004 with the technical assistance of the Maine Harvard Prevention Research Center and Colby College. Methods The Maine Harvard Prevention Research Center facilitated the development of a program logic model, evaluation questions, data collection instruments, an analysis plan, presentations, and reports. We used a cross-sectional study design with nonparticipant comparison groups. Consequences Data indicate possible program improvement strategies and substantial improvements in lifestyle factors among participants. Interpretation Limitations of the evaluation include participant self-selection, cross-sectional study design, a lack of adequate resources for evaluation, and the challenges of using community-based participatory research methods. Despite these limitations, Move & Improve program staff consider the evaluation of the program a success and have learned ways to improve the program and future evaluation efforts. Overall satisfaction with the process has been nurtured through community-based participatory research methods. This approach also enabled us to meet key evaluation standards. Background Health statistics for both Maine and the United States (1) underscore the need for reducing risk factors associated with cardiovascular disease, diabetes, heart disease, cancer, and obesity. Worksites offer ideal settings for reaching adults, including those at higher risk for chronic diseases. Worksite health promotion programs benefit employees and the organization (2-6). A recent review of worksite health promotion programs recommends that reporting on outcomes for these programs include more information about enrollment, implementation and maintenance, and negative outcomes (2). Move & Improve, a worksite wellness program in Maine, was implemented in 1996 to encourage employees and community members to increase their physical activity. We describe the evaluation process implemented in 2003 and the program outcomes for 2003 and 2004. The Maine Harvard Prevention Research Center (M-HPRC) assisted with the evaluation design and implementation. The program evaluation is based on the Centers for Disease Control and Prevention's (CDC's) Framework for Program Evaluation in Public Health (7) and community-based participatory research (CBPR) principles. The evaluation was designed and implemented with minimal resources. M-HPRC's approach to community-based program evaluation is rooted in the literature on community organization and community building (8-10). This approach is consistent with models for data action

research (11), research translation (12), and community engagement (13). CBPR is an approach to public health research that involves community members as equal partners. CBPR is 1) a participatory process in which power is shared and local expertise is recognized; 2) a cooperative process to which community members and researchers contribute equally; 3) a colearning process for researchers and community members; 4) a process that involves systems development and local community capacity building; 5) a process that empowers participants to increase control over their lives; and 6) a process that balances research and action (14). Context Eastern Maine Medical Center established the Move & Improve program in 1996 to motivate individuals to increase their physical activity and to make healthier lifestyle choices. Move & Improve became a program of Eastern Maine Healthcare Systems in 2004. Move & Improve is a free 12-week program beginning in March each year that is designed to improve health by reducing participants' risk of chronic diseases and obesity. Individuals become involved with the program primarily, though not exclusively, through affiliations with worksites. Other Move & Improve sites include schools and community organizations. Yearly recruitment efforts include reaching out to past and new participants through the mail, statewide newsletters, and collaborating partners statewide who promote the program locally. Under the guidance of volunteer site coordinators who are identified internally by worksites, participants are asked to engage voluntarily in at least 30 minutes of physical activity for at least 4 days per week for a minimum of 8 weeks of the 12-week program. Participants are asked to track their physical activity on a log (either on paper or through an interactive online activity log) and receive encouragement and tips for continued participation and physical activity throughout the program. In addition, the program offers participants community-based stretch breaks at the local mall, statewide monthly walking clinics or clubs, various exercise programs, physical fitness assessments, educational sessions, and other events. Various communication tools have been used over the past 8 years to convey helpful information to participants. These include a quarterly newspaper that features a tip of the week, good-for-you recipes, nutritional information, book reviews, and profiles of program participants and a weekly online newsletter. Move & Improve health promotion tips are shared communitywide and statewide through local newspapers (Bangor Daily News), the local CBS affiliate (WABI TV 5), collaborating partners (i.e., Eastern Maine Medical Center, Sebasticook Valley Hospital, Inland Hospital, The Aroostook Medical Center, and the Healthy Hancock coalition), various program sponsors, Move & Improve coalition members, the Move & Improve Web site, and the e-mail systems of some participating worksites. Move & Improve has collected data about the program since its inception. The number of participating individuals, participating sites, the number of individuals completing the program, and postprogram stage of change have been tracked since 1997. During the first 7 years, the program offered a paperand-pencil evaluation to participants and site coordinators. In the eighth year, Move & Improve began offering all participants and coordinators an online evaluation. Program participation and completion rates have continued to increase. In 1997, approximately 1000 participants registered for the program; in 2003, Move & Improve had more than 9000 participants, and

in 2004 it had more than 11,000 participants. In 2003, almost half were first-time participants, and in 2004, approximately one third were first-time participants. In 2003, physical activity stage of change was measured using a five-stage algorithm (precontemplation, contemplation, preparation, action, and maintenance) based on exercising 4 days per week for 30 minutes per day, adapted from previous stageof-change research (15). In 2004, however, Move & Improve staff became interested in describing participants' physical activity in greater detail and adopted a new measure of physical activity stage of change (using the same five-stage algorithm) which included an additional goal of exercising 5 days per week for 45 minutes per day. The program used current guidelines published by the American College of Sports Medicine to establish the goal levels (available from www.acsm.org). In both 2003 and 2004, a general forward movement through the stages of change was observed. Since the program's inception, participants have reported anecdotally positive effects, such as weight loss, reduced stress, and reduced absenteeism. A technical report compiled by the University of Maine in 2001 (16) reported statistically significant differences in mean systolic blood pressure, total cholesterol, the ratio of total cholesterol to high-density lipoproteins, number of sit-ups in 1 minute, number of push-ups in 1 minute, and 3-minute recovery heart rate between pretest and posttest scores among a self-selected group of participants. In early 2003, the Move & Improve program director approached the M-HPRC to help improve the evaluation design and process with the ultimate goal of contributing to general knowledge about worksite health promotion programs such as this one. The M-HPRC contributed some staff time and a small stipend to help with data analysis. A statistician from Colby College was engaged to help with data entry and analysis. M-HPRC and Colby College used principles of CBPR for carrying out this phase of the evaluation. The objectives of this phase of the evaluation were to 1) characterize Move & Improve participants and participation in the program; 2) learn which worksite and coordinator policies seemed to make a difference to participants; 3) explore whether physical activity and stage of change were affected; 4) learn whether participants experienced other lifestyle outcomes such as weight change, better nutrition, stress reduction, or reduced absenteeism; and 5) explore whether multiyear participation was more likely to sustain greater levels of physical activity. M-HPRC staff led discussions outlining Move & Improve evaluation questions and methodology and were available for technical assistance throughout the process. M-HPRC staff also facilitated the development of a program logic model outlining the program's major activities and desired outcomes. M-HPRC staff facilitated the articulation of the program evaluation questions and evaluation design through negotiation with the Move & Improve staff who would implement it with minimal time and resources. M-HPRC staff drafted the posttest participant and coordinator surveys, which were then discussed and revised with Move & Improve staff input. Initial results were drafted by M-HPRC and Colby staff and presented to program staff for interpretation and revision. Because this evaluation process was community driven, Move & Improve staff made all final decisions about evaluation methodology based on discussions with M-HPRC and Colby staff and available resources.

The 2003 and 2004, evaluations used a cross-sectional study design. Both years' evaluation efforts included nonparticipant comparison groups. In 2003, the comparison site survey was conducted at a worksite in Maine among program participants and nonparticipants and at a nonparticipating corporate partner worksite in Vermont that had comparable demographics. Both comparison groups were suburban, and each had approximately 100 employees. However, the Move & Improve participant group was 63% female, whereas the combined nonparticipant comparison group was 86% female. The comparison groups were also slightly younger. The 2004 program identified a comparable comparison group (from a worksite employing more than 750 people) near the program office in Bangor, Me. Program staff decided to offer a pencil-and-paper format for the comparison group survey in 2004 rather than an online format, which the participants used. A posttest survey was developed to assess participant demographics, level of physical activity, physical activity stage of change before and after participation, change in other lifestyle factors, absenteeism, and years of participation. One thousand randomly selected participants were mailed surveys in 2003 within 1 week of program completion. In 2004, all participants were provided with an opportunity to fill out an online evaluation that followed participation in the program. In 2003, coordinator surveys were developed and mailed to all program coordinators. In 2004, coordinators were offered a survey online. The surveys assessed coordinator demographics, level of physical activity, and strategies coordinators used to motivate participants. Because the information gathered from program participants was generally categorical, hypothesis testing to find associations between them was done using contingency table analyses. Healthy living indicators (e.g., fat intake, soft drink consumption, fruit and vegetable consumption) were recorded as having increased, decreased, or not having changed yielding three nominal categories for each indicator. Change in stage of change was calculated as being the final stage of physical activity (postprogram) minus the initial stage of physical activity (preprogram). Thus, a negative stage of change indicates a decrease in physical activity, and a positive stage of change indicates an increase in physical activity. A three-level categorical variable was used to indicate whether each participant had a decrease in physical activity, no change in physical activity, or an increase in physical activity. This three-level outcome was used in place of the stage of change because of the large number of sparse cell sizes resulting from small numbers reporting very large changes. As expected, few participants selected maintenance (which requires 6 months or more of consistent behavior) or precontemplation (which suggests not yet intending to take action). Contingency table analyses using the Fisher exact test were used to assess the strength of the association between this three-level measure of change in physical activity and the lifestyle factors about which information was obtained. We also explored where movement in stage of change tended to take place between preprogram and postprogram, stratifying by stage. Preprogram stage was determined by participants' recall at the end of the program. Consequences Approximately 43% of participants in 2003 completed the program, and approximately 46% completed the program in 2004, the highest percentage thus far. Of the participants who completed the program in 2003, 317 (31%) responded to the evaluation; in addition, 177 (53%) of the site coordinators, 33 (83%)

of the individuals from the in-state comparison site, and 40 (80%) from the out-of-state comparison site responded to the evaluation. In 2004, 902 (14%) of the 6291 participants who completed the program responded to the online evaluation; in addition, 139 (39%) of 355 site coordinators and 252 (34%) of the 750 eligible employee nonparticipants completed surveys in 2004. Participants from both years were predominantly female (87% in 2003 and 75% in 2004). The substantial decrease in the proportion of female participants in 2004 may be because of a significant program effort in that year to recruit more male participants. The most frequent participant age categories were 45 to 49 years and 50 to 54 years for both years, with approximately 20% of participants in those categories (Table). Age and sex of participants and nonparticipant comparisons were similar (data not shown). Participation in worksites varied greatly and ranged from as little as 15% to more than 50% in some worksites. Two hundred and seventy worksites participated in 2003, and 294 participated in 2004. Table. Age and Sex of Participants in Move & Improve, Maine, 2003 and 2004

Both years of data show that in the 3 months before participating in Move & Improve, more than half of all participants reported no regular exercise or only minimal exercise. As expected, the posttest data from both years show substantial increases in physical activity, with only about 5% of participants remaining inactive or minimally active both years, compared with 27% of comparison group nonparticipants in 2004. In 2004, 61% of all participants increased their physical activity stage of change by one stage or more, and 37% had an increase of two stages or more (Figure 1). In contrast, the majority of comparison group nonparticipants did not increase their stage of physical activity during the same period. Participants beginning in the contemplation stage were more likely to move two stages or more than those beginning in later stages of physical activity (84% in 2004). In 2004, participants beginning in stages 4 and 5 were most likely to report no change (69% of participants in stage 4 and 63% in stage 5). Results were similar for 2003. The lack of movement in later stages is evidence of a ceiling effect: participants who were already more physically active at the beginning of the program were more likely to maintain the same physical activity levels. Figure 1 Percentage of Move & Improve participants in each category of change for physical activity at posttest, 2003 and 2004.

Change in Stage for Physical Activity 2003 Participants, % (n = 317) 2004 Participants, % (n = 902) Decreased No change Increased by one Increased by two Increased by three or more 2 55 31 11 1 3 36 24 22 15

In both 2003 and 2004, a substantial proportion of program participants reported weight loss (41% in 2003, 62% in 2004); increased energy (54% in 2003, 62% in 2004); increased fruit and vegetable consumption (40% in 2003, 50% in 2004); decreased fat intake (33% in 2003, 45% in 2004); decreased television viewing (37% in 2003, 40% in 2004); decreased sugar-sweetened soft drink consumption (20% in 2003, 30% in 2004); decreased stress (33% in 2003, 36% in 2004); and increased water intake (55% in 2003, 60% in 2004) (Figure 2). Each of these improved lifestyle factors was significantly associated with participants' forward movement in physical activity stage of change (P < .001) except for soft drink consumption (P = .08). These results are particularly impressive given that the ceiling effect would likely bias the results toward the null of no association. Figure 2 Percentage of Move & Improve participants with positive change in lifestyle factors at postprogram, 2003 and 2004. All factors were significantly associated with forward movement in physical activity stage of change at P < .001 in both (more ...)

Lifestyle Factor Weight Energy level Fruit and vegetable consumption Fat intake TV viewing

2003 Participants, % (n = 317) 2004 Participants, % (n = 902) 41 54 40 33 37 62 62 50 45 40

Lifestyle Factor Soft drink consumption Stress Water intake

2003 Participants, % (n = 317) 2004 Participants, % (n = 902) 20 33 55 30 36 60

Employee absenteeism seemed to be associated with improvement in stage of change. Employees with a forward movement of two or three stages averaged 1 sick day during the 3 months of the program, whereas employees who stayed the same or regressed in their stage of change averaged 1.5 sick days during the same 3 months in 2003. However, number of sick days reported was small, so we were not able to assess significance. Involvement with Move & Improve for 2 or more years was significantly associated with improvement in stage of change (P = .02). Flextime, group activities, and incentives were the most common practices perceived by participants to make a difference. Coordinators cited group activities and incentives as worksite policies that came about most often as a result of the program. Yet, in contrast to what participants noted would be most helpful, coordinators tried to motivate participants most often using posters, office memos, wellness committees, bulletin boards, and e-mail tips and by registering employees for them.

Interpretation
We evaluated Move & Improve's process and outcomes using CBPR methods. Limitations in our ability to draw conclusions about the program's success include limitations inherent in cross-sectional study design; the small size and location of the comparison groups; the lack of program resources, including staff and funds for long-term follow-up with participants; and challenges that arise from using CBPR methods and local decision making. Move & Improve participants are self-selected and therefore do not represent the general worksite population or any particular high-risk group. Because participants' stage at the start of the program was determined by recall at the end of the program, response bias may have slightly inflated the effects of the program as reported in 2003 and 2004. Move & Improve's challenge now is to increase participation and completion rates and to recruit worksites with higher-risk adult populations. The 2004 evaluation revealed that participants felt that incentives, group activities, and flextime at work made a difference in their participation. Coordinators, however, may not have been able to influence worksite policies to include these factors. Because evaluation data revealed that younger individuals and men tended not to participate at as great a rate as relatively older individuals and women, greater efforts should be made by the program to help coordinators recruit and retain these individuals and to understand barriers to their participation. Perhaps more coordinators should fit this profile to motivate individuals from these groups to participate. Our analyses also revealed that many participants who were already physically active did not change their stage of physical activity over the course of the program. Perhaps some formative data could help elucidate the types of program components that may motivate these individuals to stay

active over time. To motivate and recruit individual worksites to participate, program staff could emphasize the data that indicate less absenteeism for participants. Objective record keeping of numbers of sick days taken by participants should be encouraged. Many final modifications to the evaluation design and instruments were made because of resource constraints by program staff after consultation with M-HPRC and Colby College. These circumstances may have compromised the scientific rigor of the study and our ability to draw objective conclusions from the data. One example is the decision to change the way physical activity stage of change, a key outcome, was measured in 2004. This decision was made by Move & Improve staff to align the outcome measure with the additional program goal of 5 days of physical activity per week for at least 45 minutes per day. Another example is the program's decision to offer paper-and-pencil surveys for the 2004 comparison site when all of the program participants had completed the same survey online. In this case, the employer preferred the pencil-and-paper format. Another modification involved the decision in 2003 to survey in-state nonparticipants for the comparison group from the same worksite as participants. This approach required fewer resources than going to another worksite. With a deeper understanding of the importance of consistency of measurement, Move & Improve program staff plan to use the original physical activity stage of change measure in the future and, because of staffing and funding constraints, discontinue the use of a comparison survey. An advantage of using CBPR was that it increased our adherence to several core evaluation standards (7). The standard of utility was maximized by involving stakeholders so that identification of their needs and intent were not only addressed but were central to the process. Evaluator credibility was enhanced through the relationships that were nurtured. Any findings were first disseminated to key program stakeholders for their review and interpretation. The standard of feasibility was also positively affected through CBPR. Evaluation procedures had to be practical given the resources. Move & Improve staff made all final decisions on how to carry out the evaluation based on their assessment of whether they could get it done in a timely manner. Maximizing feasibility, however, may have also compromised scientific rigor. Move & Improve program evaluation indicates that the program has been a success on many levels. Evaluation data indicate that Move & Improve has a significant impact on participants' lifestyle and risk behaviors and that longer participation in the program may also be associated with greater chronic disease risk reduction. Participants significantly increased their physical activity stage of change compared with nonparticipants during the same time period. Improved lifestyle factors were also significantly associated with forward movement in stage of change. Future evaluation efforts can minimize limitations by adding pretest data collection and keeping measures consistent over time. Longer-term follow-up of participants should also be attempted. The benefits of using CBPR methodology far outweighed limitations in scientific rigor. Move & Improve staff enthusiasm, at least in part because of its integral involvement with every aspect of the evaluation design, helped Move & Improve and M-HPRC staff overcome key barriers. Program staff gained appreciation for how to carry out successful program evaluation with minimal resources and how to improve their program. Several core standards of successful evaluation practice were also maximized.

Effects of a Behavior-Based Weight Management Program Delivered Through a State Cooperative Extension and Local Public Health Department Network, North Carolina, 2008-2009 Introduction Eat Smart, Move More, Weigh Less (ESMMWL) is an adult weight management program developed in response to North Carolina Obesity Plan recommendations to make weight management interventions accessible to underserved populations. ESMMWL was designed to be delivered through the North Carolina Cooperative Extension and North Carolina Division of Public Health. Program coursework included content on evidence-based eating and physical activity behaviors and incorporated mindful eating concepts. The objectives of this study were to describe participant changes in weight and behaviors and to document the effectiveness of the program. Methods In this prospective pilot study, courses were delivered and data collected from January 2008 through June 2009. Instructors provided feedback about implementation. For participants, height, weight, and waist circumference were measured at baseline and completion. Participants completed a questionnaire about changes in their eating and physical activity behaviors, changes in their confidence to engage in weight management behaviors, and their satisfaction with the course. Results Seventy-nine instructors delivered 101 ESMMWL courses in 48 North Carolina counties. Most of the 1,162 completers were white women. Approximately 83% reported moving toward or attaining their goal. The average weight loss was 8.4 lb. Approximately 92% reported an increase in confidence to eat healthfully, and 82% reported an increase in confidence to be physically active. Instructors made suggestions for program standardization. Conclusion This study demonstrated the effectiveness, diffusion, and implementation of a theoretically based weight management program through a state extension and local public health department network. Study of the sustainability of changes in eating and physical activity behaviors is needed. Introduction Eat Smart, Move More North Carolina is a statewide obesity prevention campaign launched in 2001 and administered by the North Carolina Division of Public Health. The campaign engages more than 60 organizations. In 2006, a team representing several state organizations developed an obesity plan (1). The plan recommends making weight management interventions accessible to underserved populations among the state's more than 2 million overweight or obese adult residents. In response to this plan, a multi-agency team recommended the development and delivery of a weight management program through existing infrastructures of the North Carolina Cooperative Extension and North Carolina Division of Public Health. Instructors would be county cooperative extension agents and health department health promotion coordinators (HPCs). As with most extension programs, instructors

would have flexibility in delivering an Eat Smart, Move More, Weigh Less (ESMMWL) course to meet their community needs. Both agencies had experience delivering such programs; extension agents requested an update to their 25-year-old curriculum. A curriculum writing team determined that the ESMMWL course would offer classes focused on the 12 evidence-based eating and physical activity behaviors for weight management (2), use the theory of planned behavior (3), and incorporate acceptance strategies such as "living mindfully" (4-6). The curriculum was peer reviewed by state and local nutrition and physical activity professionals and a family physician. Nineteen lessons focused on known predictors of successful weight management such as eating more fruits and vegetables and being physically active (2). The curriculum includes methods for planning and tracking these behaviors (7). Mindful eating concepts such as acknowledging personal responses to food without judgment and being aware of and reflecting on the effects of eating in response to emotional or environmental issues (6) were included in each lesson. Potential instructors were trained for delivery of ESMMWL. During training, instructors studied factors associated with successful weight management programs, such as use of incentives (2,8). Extension agents and HPCs decided on the number and sequence of lessons, course fee, inclusion of activity breaks during the lessons, food demonstrations, additional handouts, and types of incentives. Details of ESMMWL, including references, PowerPoint presentations, marketing materials, a participant magazine, a food and physical activity diary, a participant evaluation questionnaire, and instructor summary form and training materials, are published elsewhere (9). The objective of this study was to describe changes in weight, body mass index (BMI), and waist circumference, as well as mindfulness and confidence in ability to follow eating and physical activity behaviors that contribute to weight management among participants who completed an ESMMWL course. This study also documents the performance of ESMMWL. Methods Study design and evaluation measures This pilot program included training instructors, implementing ESMMWL in North Carolina counties, and measuring participant changes in weight, waist circumference, health behaviors, and confidence in ability to engage in physical activity and healthful eating. Instructors delivered courses in their own county between January 2008 and June 2009. Start and end dates varied by instructor. The North Carolina State University institutional review board approved the study for the protection of human participants. Participant outcomes At the first meeting, participants recorded their sex, race or ethnicity, age, and goal of weight maintenance or loss. Participants were encouraged to set a goal of losing no more than 2 lb per week. Pairs of participants, guided by instructors, measured their height and beginning and ending waist circumference and weight. BMI was calculated by 1 author (K.J.). Participants who attended the last class and completed measurements and an evaluation questionnaire, provided by the instructor, were considered "completers." The 30-item questionnaire documented self-reported changes in mindfulness

and the 12 eating and physical activity behaviors taught in ESMMWL. It was developed by the writing team and reviewed by an evaluation specialist (9). Using a 5-point Likert scale (very low to very high), participants rated their confidence in engaging in the behavior both before and after the program, and they reported whether changes were a result of program participation. Participants reported their past participation in weight management classes and satisfaction with the ESMMWL course and described their weight as a "lifelong struggle" or a new concern. Data were entered and analyzed by 2 authors (K.J., L.W.). Two summary confidence scores were calculated as measures of program effectiveness. The score for confidence in their ability to engage in physical activity was based on being physically active more or less than 30 minutes per day and participating in strength training, and possible scores ranged from 3 to 15. The score for confidence in their ability to follow healthful eating behaviors was based on responses for 8 distinct healthful eating practices, and possible scores ranged from 8 to 40. Change scores (before and after program) were calculated for BMI, weight, waist circumference, and the 2 confidence scores for comparisons by race and sex. Change in the confidence scores was compared for weight-loss status and history of weight struggle and course participation. Program performance The variables used to describe program performance included the number of instructors who delivered a course compared with the number trained; the number, frequency, and length of classes taught, course location, fees charged, and incentives provided; and participants' satisfaction and instructor comments. A course was "delivered" if instructors returned participant data and the instructor summary form that documented the number, frequency, duration, and site of the classes, fee charged, use of incentives, number of participants enrolled and completed, and instructor feedback. In a follow-up telephone survey of all trained instructors conducted by an author (C.D.), plans for future course offerings were documented. Participant dropout rate was calculated on the basis of the number of participants who completed the measurements and questionnaire on the last class day compared with the number with initial measurements only. The course was made available to adults who wanted "to lose weight, maintain a healthy weight, or learn healthier lifestyle behaviors" (9,10). Fees for the course were established by the instructor, based on the instructor's perception of what the market could bear and, in some cases, factoring in a partial rebate-type incentive for participants who completed all or most of the classes. Instructors used various avenues to market the program, including local newspapers, e-mail to existing clients, or flyers at worksites. Participant characteristics were used to determine whether specific audiences (eg, African Americans vs whites, past participants in weight-loss program vs first-timers) met their goals, were more mindful, or experienced changed confidence in eating or physical activity behaviors. Statistical analysis Data were analyzed using SPSS version 16 (SPSS, Inc, Chicago, Illinois). Frequency distributions and descriptive statistics were used to summarize participants' responses. Independent-samples t tests and Pearson correlations were used to describe the bivariate relationships between BMI changes and participant characteristics and ESMMWL course characteristics to determine what audience might complete a weight management course using this approach. We conducted a multiple linear regression

with change in BMI as the outcome variable and participant and course characteristics as independent variables. Results Participant outcomes Of the 1,162 participants who completed ESMMWL, most were white women. The mean age was 51.8 years. Participants identified an average weight-loss goal of 15.5 lb (range, 0-76 lb); 83% reported that they moved toward or attained their goal. Most completers (87%) lost weight; the average was 8.4 lb (range, 0.1-44 lb). The means for BMI, weight, waist circumference, confidence in ability to be physically active, and confidence in ability to eat healthfully improved significantly after participation (Table 1). Approximately 92% of participants reported an increase in confidence in their ability to eat healthfully, and 82% reported an increase in confidence in their ability to be physically active. Changes in BMI, weight, and waist circumference were significantly different by race (Table 2). Participants who gained weight had significantly smaller changes in confidence scores for physical activity and for healthful eating, on average, than those who lost weight (Table 3). Table 1 Characteristics of Participants Before and After Completion of Eat Smart, Move More, Weigh Less, North Carolina, 2008-2009

Table 2 Changes in Characteristics of Participants, by Race, After Completion of Eat Smart, Move More, Weigh Less, North Carolina, 2008-2009

Table 3 Changes in Confidence Indicators, by Weight Change Category, Among Participants Who Completed Eat Smart, Move More, Weigh Less (N = 1,162), North Carolina, 2008-2009

At the conclusion of their course, most participants reported changes in eating behaviors, physical activity, and mindfulness (Table 4). They also reported increased confidence in their ability to engage in these healthy behaviors.

Table 4 Changes in Behavior Indicators Among Participants Who Completed Eat Smart, Move More, Weigh Less (N = 1,162), North Carolina, 20082009

Program performance Seventy-nine instructors (53 extension agents, 26 HPCs) delivered 101 ESMMWL courses in 48 counties between January 2008 and June 2009. An additional 26 instructors planned to teach in the near future. Fifty ESMMWL courses were delivered at worksites and the rest in community settings such as faith organizations. Instructors taught an average of 16 lessons (range, 8-19) in 15 weeks (range, 8-24); most met weekly (97%) for 1 hour (83%). An average of 24 participants enrolled in a course, and 54% completed it. Participants paid fees ranging from $5 to $150. Approximately 80% paid $25 or less. Instructors reported that they could charge a fee for ESMMWL that would be less expensive than a clinical consult for weight management or fees being charged by commercial weight-loss programs, worksites, or hospital wellness facilities in their communities. Incentive(s) including giveaways, money, and time off from work were offered in 55% of the courses. Variables that were significantly related to change in BMI (incentives, cost, number of lessons attended, confidence to eat healthfully and be physically active, and weight-loss goal) were included in a linear regression model to identify independent predictors (Table 5). Larger improvements in BMI were associated with greater change in confidence in ability to eat healthfully, a larger weight-loss goal, a greater number of weeks participating, and higher cost of the program. Table 5 Independent Participant and Course Predictors of Change in Body Mass Index, Eat Smart, Move More, Weigh Less, North Carolina, 2008-2009 Approximately 60% of participants reported a lifelong struggle with their weight; this was the first weight-loss class for approximately 33% of participants. More African Americans than whites reported that this was their first course ( 2 = 14.22, P < .001). Preprogram BMI was highest among those who had struggled with their weight and been in courses before. Change in confidence in ability to eat healthfully was significant and was greatest among participants who had struggled with their weight most of their lives and who were taking their first course (Table 6).

Table 6 Confidence in Ability to Change, by Experience With Prior Weight-Loss Programs, Among Participants Who Completed Eat Smart, Move More, Weigh Less, North Carolina, 2008-2009

Approximately 97% of completing participants said that ESMMWL met their expectations, and 99% said they would recommend the program to others. Most participants were satisfied or very satisfied with the quality of instructors' presentations, instructors' knowledge, program materials, and overall quality of the program. Instructors responded positively to the ESMMWL magazine and PowerPoint presentations, noting they contained major concepts in a concise, easily understood, colorful, and visually appealing presentation. Some said handouts were needed in addition to the magazine. Instructors reported that it was difficult to engage participants for 19 weeks and the course could be shortened by deleting repetition and combining some topics. They found it difficult to motivate participants to complete the food and physical activity diary for the entire course. They stated that attendance varied because of work conflicts and personal issues. Discussion This evidence-based weight management program included strategies to promote mindful eating and physical activity to a large number of participants through the existing infrastructure of the extension and local health departments. Participants in the ESMMWL course experienced significant and positive changes that contribute to weight management. Most participants who completed a course lost weight at a rate consistent with standards for weight control programs for low-risk patients in North Carolina (11), reduced their waist circumference, and increased their confidence in their ability to eat healthfully and be physically active. Their results were comparable to results reported in self-help, worksite, and nonclinical commercial programs (12-16). Changes in participants' BMI and weight were associated with increased confidence in the ability to eat healthfully and be physically active. Almost all reported being more mindful about eating and physical activity behavior. These results add to the literature that a mindful approach can be effective for weight management, at least in the short term. No recent published reports of comparable noncommercial weight management programs were identified for comparison of program performance. The state agencies found that ESMMWL had an acceptable reach; a large number of agents and HPCs incorporated ESMMWL into their work plan. Reaching consumers in all counties, however, may require increasing the pool of trained instructors to include dietitians, exercise physiologists, and health educators. We report an average dropout rate of 46%, which is higher than the rates reported in studies of commercial weight-loss and clinical programs (12). Instructors commented that participants dropped out for the same reasons as cited in other studies (13), including lack of time, program did not suit them, personal issues, and health limitations. The expected attrition rate for a noncommercial weight management program delivered at the

community level is unknown. ESMMWL performed acceptably both for people seeking their first organized effort at weight loss and those who have been through other programs. Some outcomes differed by race/ethnicity. Further exploration is needed to understand how to reach more men and whether some course content should be tailored by race/ethnicity. On the basis of data from this study, along with instructor feedback and participant evaluation comments, ESMMWL has been standardized as a 15-lesson course, and online training for instructors, based on the live training, has been developed (17). Our study had some limitations. This weight management program was developed on the basis of published reports but was conducted in real-world settings and without a control group. The behavior changes and changes in confidence were based on self-report by course completers. At the time of this pilot, a validated measure of mindful eating was not available (18). We did not have the resources to characterize participants who dropped out nor to determine whether the outcomes were sustained after participation in ESMMWL ended. This study demonstrated the effectiveness, wide diffusion, and implementation of a theoretically based weight management program that included concepts of mindfulness through a state extension and local health agency network. Among participants who completed an ESMMWL course, changes in weight, BMI, and waist circumference were significant, as were changes in confidence in ability to eat healthfully and be physically active. Further study of the sustainability of these changes is needed.

Barriers to participation in a worksite wellness program

The leading causes of death in the United States include heart disease, cancer, and stroke [1]; the incidence of these conditions can be reduced by modifying and/or eliminating associated risk factors [2]. The majority (66.3%) of U.S. adults 20 years and older are overweight or obese and 32.2% are obese [3]. A shift towards more sedentary lifestyles and occupations and increased reliance on and use of labor-saving devices [4] may be associated with the increased prevalence of obesity and chronic disease [5]. Obesity is a major issue for corporate society because of the economic impact, negative effects on work performance, and other potentially serious risks and complications associated with obesity [6]. Most adults may spend more time during the day at work than anywhere else. Therefore, it may be important for worksites to be more conducive to employee health and is an excellent place to promote health and wellness. According to the Wellness Councils of America, more than 81% of businesses with more than 50 employees have some type of health promotion program in place [7]. Employee wellness centers have been shown to provide numerous benefits for employees including: weight reduction, increased physical fitness and stamina [8], and decreased stress [9]. Recent research has also discovered that employers benefit from such programs [10] by experiencing reduced healthcare costs, increased productivity, reduced incidence of sickness and absenteeism [11], improved recruitment [10], decreased turnover rates, and enhanced employee

morale [12]. A review of over 70 published research articles on worksite wellness programs found that, on average, employers experience a $3.50-$1.00 savings-to-cost ratio (reduced absenteeism and health care costs compared to program costs) because of wellness programs [7]. On average, research has shown that there is a 28% reduction in sick leave and absenteeism, a 26% reduction in health care costs, and a 30% reduction in worker's compensation claims [11]. Comprehensive worksite wellness programs provide ongoing and integrated programs of health promotion and disease management, with individualized risk reduction for employees as a crucial element [13]. This integrated approach has been found to be much more effective in preventing disease and promoting overall health than addressing each issue separately [14]. Worksites are an ideal setting for health behavior change because they offer access to employees through controlled environmental and communication support systems [15] and a large number of people can be repeatedly reached over an extended period of time [5,16]. By utilizing the built-in social support found in the workplace, recognizing that there are varying levels of influence (intrapersonal, interpersonal, institutional, community), and addressing employee preferences and perceived barriers, the likelihood of achieving and maintaining better health and well-being will be significantly increased [17]. Other factors found to be crucial for successful worksite wellness programs include: long-term commitment, top-level management support, employee involvement, leadership, specified objectives, detailed planning [18], focus on employee needs, resourceful, and a smooth integration into workplace environment [19]. Even if all these factors are present, the true effectiveness of a worksite wellness program is dependent on the characteristics of the target population and the proportion of the population that participates in the intervention [20-22]. If employees are not interested [23], unmotivated, or information is not personally relevant [24], the most well-planned program can fail. The average participation rate among employees for worksite wellness programs is less than 50% [25,26]. McLellan et al. [26] had an overall participation rate of 23%, ranging widely (1086%) among different workgroups [26]. Robroek et al. [25] experienced similar rates with an overall median participation rate of 33%, ranging from 10-64% [25]. Research shows that women are generally more likely to participate in worksite wellness programs than men and, overall, married employees have much higher participation rates than their single co-workers [25]. Other determinants of higher participation rates include: white-collar or secured contract employees [27,28], full-time employees [28,29], older age [26], and small company employees [30]; shift workers [31], lower income, and less education [32] displayed much lower participation rates. Although previous research has indicated what characteristics of employees are most associated with participation in worksite wellness programs, it is still unclear from a qualitative perspective why employees may decide not to participate. Minimal research exists on wellness programs in the university setting, in particular among foodservice employees. The purpose of this research was to determine why employees decided not to participate in an employee wellness program,

Wellness Wednesdays: "Eat & Meet" About Healthy Living, conducted for ARAMARK employees at East Carolina University in Greenville, North Carolina. Subjects and Methods Wellness Wednesdays: "Eat & Meet" About Healthy Living was conducted at East Carolina University, a public, coeducational university located in Greenville, North Carolina. Total enrollment for the fall 2009 semester was 27,654 students, including both on-campus and distance education students, with an additional 1,782 full-time and part-time faculty members [33]. Currently, the university's Campus Wellness Center provides informational tables and events and offers group presentations on health-related topics such as smoking cessation, nutrition, stress management, alcohol use, and physical activity [34]. To date, there are no known established employee wellness programs at East Carolina University. Wellness Wednesdays: "Eat & Meet" About Healthy Living was created by the ARAMARK Nutrition Director in order to provide employees with the opportunity to participate in a worksite wellness program, emphasizing various nutrition- and health-related topics. This program has not been implemented for employees at other ARAMARK locations and is unique to East Carolina University. The program was created in order to address employee health characteristics such as high cholesterol, abnormal glucose levels, hypertension, and overweight/obesity. These health needs were discovered at the 2008 ARAMARK employee wellness screening and served as the basis for the class and the selected topics. Topics covered in the program were those that would benefit employees that are overweight or obese, seeking to lose/maintain weight, have hypertension, high cholesterol, and/or diabetes, or are interested in improving their overall health. The number and length of weekly classes were determined based on optimal employee and management schedules. This was critical in planning because organizers had to ensure that employees would be able to take a break from their responsibilities to attend the weekly classes while still keeping food production on schedule. Collaboration with other organization directors and managers was necessary to make the final decision regarding program duration and length/location of classes. Marketing strategies utilized to promote the wellness program and recruit participants included: discussion of program in foodservice facilities' pre-service meetings (all employees in that shift present) and in employee round table meetings, inclusion of an article in the employee newsletter, and distribution of posters/flyers located in all foodservice locations and on employee bulletin boards. Wellness Wednesdays: "Eat & Meet" About Healthy Living was implemented on September 30, 2009 with its first class and an introduction to how the program operates, the timing and location of classes, topics to be discussed each week, and incentives offered. East Carolina University ARAMARK employees are predominantly service workers (83.7%), with 7.3% administrative support workers, 6.4% officers and manager, and 1.9% other job positions. There are 304 female employees and 177 males. Of these employees, 75.9% are African American, 21.8% are Caucasian, 1.5% are Hispanic, and 0.8% are Asian/Indian. Based

on data from the 2008 ARAMARK employee wellness screening, common health characteristics of employees were obtained. Employees were tested for cholesterol, blood pressure, and glucose readings and body measurements were also taken. Of these employees, 27% of participants had moderate (201-239 mg/dL) to high (> 240 mg/dL) total cholesterol; 21% had high fasting (> 100 mg/dL) or non-fasting (> 140 mg/dL) glucose levels; and 68% had moderate (120/80 to 140/90 mmHg) to high (> 140/90 mmHg) blood pressure readings. Additionally, 27% of employees were classified as overweight (BMI: 25.0-29.9 kg/m2) and 51% as obese (BMI: > 30 kg/m2). Overall, a large amount of ARAMARK employees are overweight or obese and have a moderate to high risk for hypertension, with some having a moderate to high risk for high cholesterol and diabetes. All ARAMARK employees (n = 481) over the age of 18 were eligible to participate. All interested employees (n = 50) obtained clearance/approval from their manager to ensure that schedules permitted participation. Over a 10-week period, 30 minute classes were taught once a week by a Registered Dietitian. Topics (shown in Table 1) included various nutrition- and health-related issues that were intended to increase employees' knowledge and skills on nutrition and healthy eating. The location of the classes alternated between the two dining halls on campus each week to ensure that employees from both ends of campus had the opportunity to attend. The approximate time it took to walk between sites was 15 minutes. Incentives to participate in this program were that employees would receive a $5.00 credit or "wellness bucks" for each class attended. At the end of the 10week period, the total number of classes attended were tallied by reviewing weekly sign-in sheets and given to the payroll/accounting department so payments could be included in employee paychecks. For example, if an employee attended all 10 classes, he/she would receive an extra $50.00 credit in their paycheck A five question knowledge-check quiz was administered to participants at the end of each class to determine the effectiveness of the information and materials presented and the participants' level of knowledge on the topics. Questions were specific to the topic discussed that week. This method of program evaluation was chosen due to the inconsistency of participant attendance. A pre- and post-program evaluation tool would not have provided accurate and valid results because no employee attended all classes. Therefore, only post-class knowledge quizzes were administered for evaluation to gauge level of participant knowledge and retention of the new material. Qualitative interviews were conducted after the completion of the 10-week program. Short, 3-5 minute interviews (n = 19) were randomly obtained from ARAMARK employees (both those who attended, n = 11, and those who did not attend the program, n = 7) and the program organizer (n = 1) with questions aimed at information about attendance, participation, incentives, location, and suggestions. A funnel approach was used in interviews with the broadest questions

asked first (to avoid sensitizing interviewees and leading responses) to more specific prompts for further information. Interview questions are listed in Table 2. Interviews were conducted in East Carolina University dining facilities where ARAMARK employees were working. Selection was completely random and there was no inclusion criteria set in order to obtain unbiased, non-influenced responses. By choosing to use a random sampling strategy and conducting short, informal interviews in the work area, feedback from a wide range of employees was able to be obtained (both participants and non-participants) since employees did not have to stop working or feel pressured to answer in a particular way. Interviews were conducted until saturation of themes occurred. All research protocols were approved by East Carolina University Institutional Review Board Results A total of 50 (10.4%) East Carolina University ARAMARK employees, managers, and leadership team members attended Wellness Wednesdays: "Eat & Meet" About Healthy Living at least once during the 10-week program. Out of the 10 available classes, 50% attended 1 class; 22% attended 2 classes; 14% attended 3 classes; 4% attended 4 classes; and 1% attended 5 classes. No employee attended more than 5 of the 10 classes. It was found that employees from some locations on campus had stronger participation than others. Employees at retail dining facilities located in the middle of campus had the least participation, while the two dining halls (one on each end of campus) and retail facilities on the West End of campus had the highest employee participation rate. Class size varied from week to week and ranged from 4 to 20 people. On average, 11 people attended each class. After reviewing the weekly five question knowledge quizzes administered at the end of each class, it was found that most people scored well on the knowledge assessments. Average scores ranged from 71-100%. Qualitative interviews of random ARAMARK employees after the completion of the program revealed several themes and barriers for not participating. The top three reported barriers to participation (in order from most often to least often reported) were insufficient incentives, inconvenient locations, and time limitations. Employees expressed that offering more money as an incentive would make it more likely for them to attend and the extra money was "always good." Location was an issue due to the distance between foodservice facilities. Some employees would only attend classes held at their location and did not attempt to go to the classes held at the opposite end of campus. Timing and scheduling of the weekly classes were reported to be difficult because they were conducted during the work day and it was often hard to find a time that employees would be able to attend without disrupting their shift schedules and responsibilities. Time was an especially important barrier with retail location facilities which did not have time in between meals to leave and classes were held during one of their busiest times. Scheduling the classes on one morning during the week limited the number of employees that

could attend. Employees who were off on Wednesdays or who worked the night shifts had to make an extra effort to come in during their time off to attend the classes. Some employees reported that they were not interested in the topics discussed and, therefore, did not attend. Topics that were cited as being of interest that were not included in the program or were not covered in-depth included: hypertension, stress management, heart health, shopping on a budget, exercise, and proper child nutrition. Marketing referred to initiatives for the promotion and publication of the program such as flyers, information from location managers during pre-service meetings with employees, and other media outlets. Employees felt that they were not adequately informed and made aware of the program and the timing and scheduling of the occurrence of the classes. Health beliefs were expressed as comments such as not attending because of perceived sufficient health knowledge and having a healthy family. From an organizers perspective, successfully planning and implementing this program was not without its own set of barriers, many of which were similar to those expressed by employees. Scheduling and timing of the weekly classes were reported to be difficult because they were conducted during the work day and it was often hard to find a time that employees would be able to attend without disrupting the shift schedules. Attendance was further hindered when sites were short-handed due to callouts (employee calls to inform managers that he/she will not be at work due to sickness, transportation issues, etc), production schedule being behind target, or by continuous service operations (such as retail outlets versus dining halls that had set meal periods and open/close schedules). To better accommodate the busy schedules of foodservice employees, organizers tried to keep the classes short, 15 to 30 minutes. However, it was reported to be difficult to adequately address all of the information within the short class periods. The location of the weekly classes was another important and sometimes difficult factor for organizers to address. On this college campus, employees were spread out over 14 dining locations. With people in so many different locations, it was nearly impossible to find a location that was convenient for all to attend without having to walk or drive. If employees chose to drive, parking spaces are limited and, therefore, created another problem to factor in with transportation. This issue was addressed by alternating the end of campus that classes were held at, but this still did not capture all dining locations, leaving some employees with a travel time to factor in. Finally, program evaluation on knowledge for Wellness Wednesdays was extremely difficult due to low participation rates and regular attendees. An overall knowledge pre/post-evaluation would not have provided accurate information about the effectiveness of the program because all employees did not attend all 10 classes. With such irregular attendance, it was hard to evaluate how effective the program was in relation to lifestyle factors, weight, and health status as well as preventing maximum knowledge gain due to the inability to build on information from one week to the next.. Discussion According to previous research, the average participation rate among employees for worksite wellness programs is less than 50% [25,26]. McLellan et al. [26] had an overall participation rate

of 23%; Robroek et al. [25] experienced similar rates with an overall median participation rate of 33%. Evaluation of Wellness Wednesdays: "Eat & Meet" About Healthy Eating supported these findings with an overall participation rate of 10.4%. Barriers such as insufficient incentives, inconvenient locations, time limitations, not interested in topics presented, schedule, marketing, health beliefs, and not interested in the program were found and negatively impacted participation rates in this employee wellness program. These barriers also supported previous findings, especially employee disinterest [23] and information presented not being personally relevant [24]. However, the top three barriers reported (incentives, location, and time) had not been previously reported as barriers to worksite wellness program participation. These barriers may be unique to college campus employees and partially related to the physical spread of the work environment. In order to increase participation rates, creative approaches to meeting employees' needs are required. Previous research shows that by addressing employee preferences and perceived barriers, the likelihood of achieving and maintaining better health and well-being will be significantly increased [17]. This information can be obtained prior to the start of the program by distributing a needs and interest survey to all employees and would ensure that the topics presented were relevant and appropriate for the intended audience. Evaluation of Wellness Wednesdays: "Eat & Meet" About Healthy Living provides useful information for future program development of employee wellness programs, especially in work environments that are diverse and physically spread out such as college campuses. Results show that employee wellness programs can be implemented on college campuses and are effective in increasing knowledge and skills of employees on nutrition- and health-related topics. However, program planning that addresses identified barriers including insufficient incentives, inconvenient locations, and time limitations may facilitate higher participation in future worksite wellness program

Impact of the Prevention Plan on Employee Health Risk Reduction

Employers have a compelling cause for concern about employee health and the burden high health care costs place on their competitive positions. According to a 2009 Kaiser Family Foundation survey, average premiums for family health insurance coverage have increased 131% since 1999.1 In 2009, the health portion of our nation's gross domestic product is expected to have increased 1.1 percentage points to 17.3%, the largest single-year increase since 1960.2 Health insurance cost, the second largest expense beyond payroll for many employers, is linked directly to employee health for employers that self-fund or purchase medically underwritten plans. Communityrated groups also feel the impact of employee health through annual rate increases driven by community experience. Poor employee health also impedes profitability by reducing productivity. Compared to a healthy person, an employee in poor health is more likely to be absent from work and less productive while on

the job (presenteeism or health-related performance reduction). Several studies indicate that the financial impact of employee absenteeism and decreased productivity due to poor health is more costly compared to medical and pharmacy claims costs alone.3 7 In fact, in a multiemployer study of over 1,134,281 medical and pharmacy claims across 51,648 employees using the Health and Work Performance Questionnaire developed by Ron Kessler, PhD and the World Health Organization, the average employer health-related productivity costs (presenteeism and absenteeism) were found to be 2.3 times the medical and pharmacy costs alone.3 To absorb escalating costs, most large employers plan to shift a greater share of health care costs to their workers. The worker contribution, or the portion of premiums paid by the employee, has increased 128% since 1999.2 According to the 2009 2010 annual survey by the National Business Group on Health and Towers Watson, employees at many companies will experience significantly higher premiums, deductibles, and co-payments next year. The survey also found that 28% of employers plan to impose spousal surcharges next year, an increase from 21% this year.8 The trend of escalating costs and increased cost-sharing is unsustainable. Continued cost increases without corresponding increases in income impede profitability and many employees cannot afford to shoulder continually increasing portions of their health insurance premiums. However, this health care cost quandary is, in large part, the result of a growing health crisis from an unmitigated growth in the burden of personal health risks leading to chronic illness. In the United States 75% of health care costs stem from the same preventable chronic conditions heart disease, cancer, stroke, chronic obstructive pulmonary disease (bronchitis, emphysema), and diabetes.9 Rising health care costs are more connected to increased health care utilization (due to this unrelenting burden of health risks and illness) than to the increase in unit price for health care services.10 In 2008, a large employer estimated its total health-related costs (medical + pharmacy + presenteeism + absenteeism costs) using the Integrated Benefits Institute Health and Productivity Snapshot (which is based on the Health and Work Performance Questionnaire developed by Dr. Ron Kessler of Harvard along with the World Health Organization).11 As a hypothetical example, if the employer were able to reduce the health-related productivity loss through a comprehensive wellness program by just 1 day per full-time employee per year, it would generate $18.8 million to bottom-line earnings before income tax, depreciation, and amortization. To generate the same impact by increasing top-line revenue, sales revenue would have to grow by $76.6 million. The chief financial officer then translated this figure to $0.84 of value per share improvement, based on the company's current market value.12 Today's reality is that health is a performance driver. The only sustainable way to relieve the economic pressures of rising health care costs is to drain some of the manageable health risks and illness burden from the population.12 According to a Milken Institute study,13 implementing primary (wellness and health promotion to keep healthy people healthy), secondary (screening for earlier detection/diagnosis), and tertiary (earlier evidence-based treatment to reduce complications and disability) prevention in homes and workplaces nationwide would reduce the economic impact of disease by 27% saving $1.1 trillion annually by 2023

and reducing cases of chronic disease by 40 million. The study calculated that 7 chronic conditions (ie, cancer, heart disease, hypertension, mental disorders, diabetes, pulmonary conditions, stroke) cost the US economy more than $1 trillion per year. Based on the growing prevalence of those conditions, it projected an illness-related cost burden of $4 trillion by 2023. However, estimates of gains through improved prevention, detection, and treatment of those 7 conditions would reduce annual treatment costs in the United States by $217 billion and reduce health-related productivity losses by $905 million by 2023. Lowering obesity rates alone could lead to productivity gains of $254 billion and avoid $60 billion in treatment expenditures, according to the study. As shown by Edington in the landmark article, Emerging Research: A View from One Research Center, costs follow risks.14 A high risk for any single risk factor is associated with higher medical costs. When individuals are grouped according to their overall health risk levels and age categories, those with the higher risk levels are more costly to the organization. As risks increase or decrease over time, changes in costs follow in the same direction. Therefore, an organization that succeeds in moving a population from a high-risk category to a moderate-risk category can expect reduced cost levels. Similarly, an organization that succeeds in maintaining a high percentage of employees in low-risk categories can avoid the cost increases associated with moderate- and high-risk categories. The goal of a health promotion program, therefore, should be to move the population into low-risk, low-cost categories and to keep them there.14 For health management programs to succeed in moving people into lower risk categories, however, they must attain a high participation rate from the employee population. Incentives, communications, and organizational commitment through a sustainable culture of health play important roles in growing participation. Employers are increasingly using innovative incentives to increase health promotion participation rates. Incentives are often both extrinsic (financial) and intrinsic (nonfinancial) rewards designed to motivate individuals to modify their behavior by changing the cost and benefits associated with the behavior. Typically, the stronger the culture of health and the better the communication/marketing of the programs, as well as the higher the average dollar value of an incentive, the higher the participation levels.15 Methods The Intervention The eligible population was offered The Prevention Plan , a first-of-its-kind prevention benefit program that identifies an individual's top health risks and designs a customized personal prevention plan to reduce those risks. The Prevention Plan provided a suite of innovative technology, tools, and services encompassing integrated primary prevention (wellness and health promotion), secondary prevention (biometric and lab screening as well as early detection/diagnosis), and tertiary prevention (early intervention and evidence-based chronic condition management) all based on the clinical science of preventive medicine. This study evaluated the impact of The Prevention Plan on employee health risks after 1 year of integrated primary and secondary prevention interventions in a cohort of 2606 individuals.

The Prevention Plan participants completed an online health risk appraisal (HRA), which is a comprehensive questionnaire (totaling approximate 77 questions) about their current health risks, health status, and lifestyle. Their results were posted in a secure personal online storage space for health records on The Prevention Plan Web site. Members also had the option of storing their medical information such as lab results, surgeries, and immunizations and tracking their progress and incentive points in this password-protected area. After completing the HRA, the 2606 cohort participants completed a blood test (at a minimum this included a lipid profile and fasting blood glucose). Professionals from a national laboratory performed blood draws in the workplace or at an off-site clinical lab. Biometric measurements (eg, height, weight, blood pressure) also were taken. Based on the results of the HRA, blood tests, and biometric measures, each member received a customized Personal Prevention Plan. Each individual's top health and lifestyle risks were identified and the individual was provided with step-by-step recommendations to lower his or her health risks and prevent medical conditions from developing. Some members also had the option of selecting a physician review. This included a review of all the identified health risks and the blood test results by a licensed physician who provided the member with written recommendations about his or her health status. The Prevention Plan also made recommendations about screenings, tests, and other follow-up exams based on the age and sex of the member and accepted evidence-based medical protocols and clinical indicators. The report made available to the members included a summary of the preventive benefits covered by the member's health plan (when purchased through the employer). After the initial assessment as described, The Prevention Plan provided members with the support and encouragement necessary to adopt healthy behaviors and lifestyle changes, such as increased exercise, reduced alcohol consumption, or following a low-fat diet. Support services included 24/7 nurse hotlines, one-on-one health coaching, contests, group events, and employer incentives/rewards. Members also received their own customized and personal health dashboards on The Prevention Plan Web site to navigate their health needs. Members had the option of completing action programs, typically 8- to 10week educational programs that addressed key health areas such as nutrition, weight management, smoking cessation, depression, and lowering cholesterol. E-mail and telephone alerts, as well as reminders about needed screenings, were also used to motivate and assist members. A robust health library, symptom checker, medical animations, and daily health news were all made available to members. Within their personal password-protected account, a score meter showcased The Prevention Score for a member, providing an instant snapshot of his or her own prevention efforts as well as progress toward completion of educational tutorials, screenings, and participation in other activities and adherence to recommended programs. As the member engaged in The Prevention Plan during the course of the year via challenges, action programs, activity trackers, registered nurse coaching, among others his or her Prevention Score increased. While protecting employee personal health information, the score level achieved was then linked to customized rewards and incentives such as prizes, gift cards, or health

insurance premium reductions, which further drove both enrollment in The Prevention Plan and engagement in the program during the course of the year. Data The worksite population examined in this study came from 3 employer groups a health services company, a hospital, and a global insurance brokerage ranging in size from approximately 139 employees to 7661 employees with a total eligible population of 10,899. In 2008, The Prevention Plan (HRA, biometrics, and blood tests) was offered to these employees by their employers, resulting in a 52% registration (5667 people). However, because not everyone who registered at baseline completed all 3 steps (HRA, biometrics, and blood tests), and those who did complete all 3 steps the first year may not have completed all 3 steps the second year, in addition to those who did not have all data points in their entirety from the HRA, biometrics, or blood tests in either year the number of employees studied was limited to the 2606 who met all criteria for the study cohort and, therefore, was much smaller than the group that actually participated in the HRA and some of the programs. The final sample for the study consisted of a smaller cohort of 2606 employees, registered in The Prevention Plan , who completed an HRA, blood tests, and biometric screening in both 2008 and 2009, and had a complete set of the 15 health risk data points in both years. Baseline demographics of the cohort can be found in Table 1. Table 1. Cohort Age Breakdown

In order to evaluate the impact of The Prevention Plan over the course of 1 year, we compared the risk transition in the study group to the Natural Flow model developed by Edington.14 The Edington Natural Flow model measured the health risks in a population in which employees did not participate in a health improvement program. An updated version (as of April, 2010) of Edington's research on the Natural Flow transition between risk groups was used (Table 2). Table 2. Natural Flow Risk Transitions (n = 27,555)14

Although ideally we would like to have appropriately adjusted for age and sex, in this analysis our adjustment was limited to matching 15 individual risk factors as much as possible to the ones used by Edington in the 2001 study.14 Variable description Information on health risks, health-related absences, and perception of health were obtained from employees' responses to the HRA questions, and from the blood tests and biometric screenings completed. Edington used the 15 health risk measures and the corresponding high-risk criteria as indicated in Table 3. The current study matched the majority of the measures and criteria except as indicated. Table 3. Health Risks and Behaviors

Adopting similar health risk definitions to those used by Edington,14 Burton et al,16 and Musich et al,17 we were able to assess 15 high risks as shown in Table 3. Our HRA did not assess life or job satisfaction and these were replaced with questions about fatty diet consumption (assessed through HRA questionnaire) and fasting blood glucose (assessed through blood tests). For the former we asked, How often do you eat foods high in unhealthy fats, such as red meats, oils, fried foods, bakery goods, or highfat dairy products? Responses that indicated consumption several times a day, once a day or several times a week were considered high risk. For some of the high-risk evaluations, our HRA had slightly modified questions, which are also indicated in Table 3. Edington's safety belt use criteria assessed anyone using a safety belt less than 90% of the time as high risk. The question in our HRA asked, How much of the time do you buckle your safety belt when driving or riding? and gave the following as response choices: always, usually, sometimes, rarely, or never. We defined high risk as those employees answering, sometimes, rarely, or never. Similarly, for physical activity Edington defined high risk as those who exercised less than 1 time per week. Our physical activity questions were worded, On average, how often do you engage in moderate physical activity for 30 minutes or more? and On average, how often do you engage in vigorous physical activity for 20 minutes or more? Examples of moderate physical activity given to employees were brisk walking, cycling, vacuuming, and gardening. Examples of vigorous physical activity given to employees were heavy lifting, running, aerobics, and fast cycling. The choices available for each of the two questions included: never, 1 day per week, 2 days per week, 3 days per week, 4 days per week, 5 days per week, 6 days per week, or every day. To be assessed as high risk in our analysis the employee would need to have answered never to both of the physical activity questions.

Table 3 shows the rest of the 15 criteria used in our study and how they compared to Edington.14 Following Edington, we define employees with 0 2 health risks as low risk, those with 3 4 health risks as moderate risk, and those with 5 or more health risks as high risk. Procedure We examined trends in health risks, health-related absences, and perceptions of health among 2606 employees who completed an HRA, biometrics, and lab tests in both 2008 and 2009. Both internal and external comparisons were done to identify the impact of the intervention program. The internal analysis compared the change in outcomes for employees who experienced an improvement or reduction in health risks to employees who experienced no change in health risks and, separately, employees whose health risks worsened. The specific outcomes of interest are changes in health risks, health-related absences, and employees' perceptions of their own health. The external analysis compared changes in health risks between the employees in the program and the unmanaged employees from Edington's Natural Flow. We conducted 2-sided hypothesis tests and, depending on the statistic, report significance at the .05, .01, or .001 level. Results Aggregate health transitions Table 1 summarizes the demographics of the baseline and 1-year follow-up of the Prevention Plan participants (n = 2606). A significantly greater percentage of this cohort was female (69.3%); the average age for the cohort was 39 years. In Table 4 we report the distribution of health risk levels among the participants in 2008 and 2009. In 2008, a total of 1452 employees (55.72%) were identified as low risk, 809 (31.04%) were identified as moderate risk, and 345 (13.24%) were identified as high risk. In 2009, the overall risk distribution for these employees improved with 1697 (65.12%) at low risk, 715 (27.44%) at moderate risk, and 194 (7.44%) at high risk. Table 4. Transition of Health Risk Levels in the Cohort Population

We found that the employees showed a significant improvement after exposure to The Prevention Plan program. Relative to 2008, the proportion of low-risk employees in the program in 2009 was 9.40 percentage points higher, the proportion of moderate-risk employees was 3.61 percentage points lower, and the proportion of high-risk employees was 5.79 percentage points lower (P < 0.01, n = 2606). In Figure 1, we show the risk level transitions of the cohort from 2008 to 2009 using the Markov chain analysis. For employees in the high-risk category (5 or more high-risk factors) at baseline, 35.7% remained at high risk, 48.7% moved to moderate risk, and 15.7% moved to low risk. For employees in the moderate-risk category (3 4 high-risk factors) at baseline, 45.9% remained at moderate risk, 46.4%

moved to low risk, and 7.7% moved to high risk. We find that 87.3% of low-risk employees (0 2 high risk factors) in 2008 remained low risk in 2009, 12.1% of low-risk employees moved to moderate risk, and 0.6% of low-risk employees moved to high risk. These results show a strong net population movement from higher to lower risk levels. FIG. 1. Health risk level transitions between 2008 and 2009.

Accepting the Edington data for Natural Flow14 as providing a legitimate benchmark comparison, the actual program risk level transitions are also statistically different from the projected natural flow. In Table 5 we report the health risk level transitions among the cohort compared to the Natural Flow. Modeling the Natural Flow's risk level transitions, 31% (107) in our cohort were expected to move from a high-risk level to a moderate-risk level, 35% (283) from a moderate-risk level to low risk, 6% (21) from a high-risk level to low risk, and 70% (1016) were expected to remain in the low-risk category. Table 5. Health Risk Category Transitions from 2008 to 2009 in the Cohort Population (n = 2606) Compared to Natural Flow14

The actual program risk level transitions were statistically different (P < 0.01) from the projected natural flow distribution with 48.70% (168) in our cohort moving from a high-risk level to a moderate-risk level, 46.35% (375) moving from moderate risk to low risk, 15.65% (54) moving from high risk to low risk, and 87.33% (1268) remaining in the low-risk category. In Table 6 we summarize the Natural Flow and The Prevention Plan comparing the percentage of employees whose health risks improved, remained unchanged, or worsened overall. As can be seen in the table, The Prevention Plan had a statistically significant (P 0.001) risk transition compared to the Natural Flow, with 22.91% having lowered risk, 67.65% with unchanged risk, and 9.44% moving into higher risk levels. Evaluating the percent unchanged further indicates that although 67.65% were in this unchanged category, 71.92% of them were those employees who remained in the low-risk category, 21.10% were those who remained unchanged in the moderate-risk category, and 6.98% were those who remained unchanged in the high-risk category.

Table 6. Summary of Cohort Risk Distribution Based on Risk Categories Identified (Low, Moderate, High) Between 2008 and 2009

In Table 7 we went a step further and analyzed the distribution of risk level change based on the number of individual risks identified in 2008 and 2009. Instead of using the broader category of low (0 2 high risks), moderate (3 4 high risks), and high (5+ high risks), the risk levels were broken down into the number of individual risks identified. Table 7. Distribution of Cohort Based on Number of Individuals Risks Identified From 2008 to 2009

From this table, one can observe who is reducing and adding risks from 2008 to 2009. For example, of those who were low risk in 2008 (n = 1452), 139 (9.57%) had 0 high risks and stayed at 0 high risks in 2009. Similarly, 71 (4.89%) had 0 high risks in 2008 but moved to 1 high risk in 2009, 27 (1.86%) moved to 2 high risks in 2009, and 10 (0.7%) moved to 3 high risks in 2009. Following the same logic in the table, one can further track the other number of individual risks identified (1, 2, 3, 4, and 5+) in 2008 and see their corresponding movement in 2009. Table 8 summarizes this transition and, as can be seen, overall 1100 (42.21%) had a decrease in number of individual risks in 2009, 971 (37.26%) remained unchanged, and 535 (20.53%) had an increase in the number of high risks between 2008 and 2009. Table 8. Summary of Cohort Risk Distribution Based on Number of Individual Risks Identified (1,2,3,4,5+) Between 2008 and 2009 As expected, these numbers are different from those identified in Table 6 because of the more granular analysis. In fact, when we look at the individual risk movements within the risk categories at this magnification, there is a striking increase in the number of people showing health risk improvement 42.21% of the entire cohort of 2606 when we look at individual risk transitions vs. 22.91% of the entire cohort of 2606 when we look at the broader risk categories. Taking this a step further, we also focused on the broader 5+ risk category (5 15 risks). Because this risk level consists of all individual risks 5 15, we expected that individual risks within this level may have shown changes but that these would not have been picked up by when they were combined in the general high-risk category. We concluded that by setting the trigger point for data analysis of the high-

risk category to be individuals with 5 or more high-risk factors, we would miss the potential movement of individual risks within that broad category. Therefore, upon more detailed analysis, we discovered that although it seemed that 123 (35.65%) high-risk individuals from 2008 showed no change in 2009 (ie, they remained in the high-risk level), there actually was some significant risk reduction in individual risk factors within that broader high-risk category. By breaking this group down further into individual risk levels (6, 7, 8, 9, and 10+ individual risks), we were able to measure that 40 people (32.52% of the 123 people who remained unchanged in the high-risk category) actually did lower their number of health risks. For example, this included 20 people who had 6 individual risks identified in 2008, and who moved down to 5 individual risks in 2009. Similarly, 6 people who had 7 individual risks identified in 2008 moved down to 6 individual risks in 2009; of the people who had 8 individual risks identified in 2008, 3 moved to 7 risks and 3 moved to 6 risks in 2009. Only 57 people (46.34% of the 123 people who remained in the high-risk category) remained truly unchanged between 2008 and 2009. Transitions in specific health risks In Table 9, we report the net change in individual high health risks. The table includes the number of employees who had a particular high health risk in 2008 and 2009 and the percent decrease in those risks in 2009 (8 at P 0.01 and 2 at P 0.05). Use of drugs for relaxation, existing medical conditions, smoking, and body mass index (BMI) showed a slightly higher percentage at high risk in 2009 but these were not statistically significant at P 0.05. Of note is a decrease in 2009 of 169 people who had highrisk blood pressure and a decrease of 146 people who had a high fasting blood sugar in 2008 (both at P 0.001). Similarly, people who reported fatty diet consumption decreased by 255 in 2009 (P 0.001). There were other relatively large reductions for stress, alcohol, cholesterol, improvements in physical activity, and perceptions of health. Table 9. Net Change in Individual High Risks

Many studies have found that improvements in diet and exercise contribute to reductions in high blood pressure, high fasting blood sugar, and high cholesterol.18 21 Although we did not see a statistically significant change in BMI over 1 year, an improvement in exercise and/or diet may trigger improvements in obesity over the next few years similar to or greater than the improvements in blood pressure, fasting blood glucose, and/or cholesterol we have already seen. Discussion Reducing health risks is a critical goal for employers because it has been shown to reduce health care costs and improve productivity in the workplace.12,14, 22 25 This study evaluated the impact of The Prevention Plan on employee health risks after 1 year of integrated primary prevention (wellness and health promotion) and secondary prevention (biometric and lab screening as well as early detection)

interventions. The study demonstrates that The Prevention Plan reduces key employee health risk factors and effectively moves employees to lower overall health risk categories. It also reinforces the view that comprehensive, evidence-based primary and secondary prevention programs can begin achieving measurable health improvements in the first year of intervention. The Prevention Plan yielded strong levels of employee participation, personalized prevention recommendations, and health coach advocacy, which may well have had an impact on the level of individual engagement and progress and, therefore, the level of results. This study also provides insight into the health risks that can be most quickly addressed through comprehensive health management. The most significant were a reduction in the proportion of employees with high-risk blood pressure, high-risk fasting blood sugar, and high-risk stress. Of note is the fact that the first two are data points obtained through actual measurements blood pressure reading and a blood glucose test. The significant improvement in the fasting blood test an actual test and not a self-reported questionnaire response is especially indicative of the success of the program over 1 year. In addition, employees showed improvement in cholesterol levels, physical activity, a reduction in fatty diet, and a reduction in heavy drinking. The population also achieved a reduction in health-related illness days and improvement in perception of health. While we did not see a measurable reduction in the high-risk sector on BMI over the 1-year time period of this study, we know that there may well have been weight loss by people who had a much higher BMI at baseline but did not lose enough weight to get under the high-risk threshold of 27.8. We also realize that there can be a lag between improved physical activity and healthy eating and corresponding weight loss outcomes. In this study, we went a step further and analyzed the distribution of risk level change across the cohort of employees (n = 2606) based on both health risk category transitions as well as the number of individual risks transitioned in 2008 and 2009. Therefore, when measuring the risk transitions among the broader categories of low (0 2 high risks), moderate (3 4 high risks), and high (5+ high risks) we also looked more closely at individual risk movements within those risk categories. We observed that while 22.91% of the cohort showed improvement based on the broader low-, moderate-, and high-risk categories, a much higher percent of the cohort (42.21%) had actual reductions in individual health risk factors. We believe this higher level of analysis is important to consider in studies because the broader risk categories may otherwise miss the fact that these individuals in reality have had a decrease in the number of risk factor(s). Similarly, we applied this higher lens of granular analysis to detect health risk movement within the 5+ high-risk category (5 15 health risks). Upon more detailed analysis, we concluded that although it seemed that 123 people (35.65% of the high-risk individuals) from 2008 showed no change in 2009 (ie, they remained in the high-risk level), there actually was some significant risk reduction in individual risk factors within the broader high-risk category. By breaking this group down further into individual risk levels (6, 7, 8, 9, and 10+ individual risks), we were able to detect that 40 people (32.52% of the 123 people who remained unchanged in the highrisk category) did actually lower their number of health risks even though they remained in the highrisk category. For example, this included 20 people who had 6 individual risks identified in 2008, and

who moved down to 5 individual risks in 2009. Similarly, 6 people who had 7 individual risks identified in 2008 moved down to 6 individual risks in 2009 and of the people who had 8 individual risks identified in 2008, 3 moved to 7 risks and 3 moved to 6 risks in 2009. Only 57 people (46.34% of the 123 people who remained in the high-risk category) remained truly unchanged between 2008 and 2009. It is also important to note that keeping the low-risk individuals in that category between years should be considered as much an indicator of success as moving people from high risk to lower risk levels, given the natural flow of the population toward higher risk levels.14 As was evident in this study, The Prevention Plan was successful in having a high number (87.33%) of employees remain in the low-risk category. Research has shown that as prevention and health improvement interventions reduce health risks, reduction in health care costs and improvement in health-related productivity follows. Conversely, those who remain at high risk or have an increase in their risks over time have increased health care costs and experience a decline in their productivity. 25,27 In fact, some studies have shown a linear trend of improved productivity for those employees who reduced the greatest number of health risks and decreased productivity among those who increased health risks.26 Burton et al have concluded that for each risk factor changed, there is a 1.9% change in self-reported productivity loss.26,27 Study Limitations As is the case with most studies, this study has a few limitations. Because participation in the program was voluntary, there may be some selection bias through self-selection. Furthermore, certain groups of employees may be underrepresented in the cohort, which may also impact the results. The small number of certain subsets identified also means that this may not be a true representation of the population. In this type of a study, the Hawthorne effect is always a concern. Participants may report better answers in the HRA for no reason other than the fact that they are being questioned. The fact that our 15 risk factors did not exactly match the ones from the Natural Flow, that some of the questions were asked differently, and that others had different options for answers, may also impact our results. Comparing to the Natural Flow as opposed to the nonparticipating cohort of employees from the 3 employer groups may also have been a limiting factor. Similarly, risk factors such as high blood pressure and high fasting blood sugar may actually represent health conditions and not just health risk factors, which may have been impacted by treatment rather than by lifestyle changes alone. Lack of adjustment for age and sex to the Natural Flow further limits the nature of our conclusions. The short duration (1-year outcome measurements) also means that these changes over 1 year may not necessarily translate into real outcomes that indicate permanent, or at least persistent, health improvements. The study population would need to be followed and evaluated over the next few years. Nevertheless, despite these types of limitations, the literature is filled with studies that show that HRAs are an effective means to assess a population's health risk level and can be used to evaluate and monitor the health of employees. 28 30

Conclusion

Large-scale, population-based changes in health behaviors require a multipronged approach to be successful. Interventions that lead to better health outcomes and subsequent lower health care costs take time. With various stakeholders - from the government to private businesses - all playing a role, much can be accomplished. One of the key factors in tackling lifestyle behaviors will continue to be the education and engagement of the individual in his or her own health. With the power of the Internet, mobile phones, and applications, for the first time in human history we have the ability to directly reach and engage the vast majority of Americans in a literal movement toward better health. The Prevention Plan, an innovative prevention benefit with engaging technology and interactive Web-based tools as well as high-touch outreach by health coaches, provides evidence that personalized prevention recommendations and health coaching advocacy can reduce health risks by engaging individuals to be more proactive about their health. The first year results outlined in this study provide a glimpse into the impact such programs can have even over a relatively short period of time. Furthermore, this study yields more evidence for the business case that prevention is an investment to be leveraged rather than a cost to be justified. It also supports the premise that our health care ecosystem would benefit by focusing as much on the health as we do on the care of people. In that way, we could strengthen our current reactive, illness-oriented sick care system by structuring a more proactive, wellness-oriented health care system built upon the pillars of prevention.

Health and Wellness Article Categories

y y y y y y y y

Health Risk Assessment / Health Risk Appraisal Health Promotion and Wellness Program Benefits Wellness Program Challenges, Obstacles and More Obesity in the Workplace Stress Management in the Workplace General Wellness Program Articles Employee Fitness and Wellness Ideas Health Coaching / Wellness Coaching

Health Risk Assessment / Health Risk Appraisal

Health Risk Assessment and the Workplace Wellness Program To design a workplace wellness program, the health promotion intervention strategies need to be based on the unique wellness needs of each companys employee profile. Health Risk Assessments are the best way to determine what kind of workplace wellness program intervention will hold the highest likelihood of success, measure its progress and track the changes in health of the aggregate employee population.... Health Risk Appraisal: Starting a Health Promotion Program A Health Risk Assessment is the name for the process of collecting information from individuals. This instrument is used by employers to identify risk factors directly and indirectly

related to the health of their employees. It is also a way of presenting feedback to people in order to connect them with at least one health promotion program initiative that will mediate health, maintain wellness, and/or avert illness.... Workplace Wellness Health Screenings When it comes down to it, there are many benefits that go along with workplace wellness health screenings. Even though not every company is setting these screenings up, the number of those that are is surely on the way up. This is good to know because workplace wellness health screenings can be quite important to say the least. Not only does it look good for an employer to offer workplace wellness health screenings , but employees also love them because they do not have to leave their office to take part in testing.. Health Risk Appraisal Before implementing a health risk appraisal (HRA) program, you need to consider several things. Do not forget, that you can also seek assistance from a wellness consultant for additional advice and recommendations. Health Risk Assessment A health risk assessment is an assessment tool or questionnaire scientifically designed to identify health risks and outline information to assist person in making healthful changes that impact their health and prevent chronic disease.

Return to top
Health Promotion and Wellness Program Benefits

Wellness Program What is a wellness program? Wellness program information and resources are becoming common, but who do you trust? Around the country companies everywhere are implementing a wellness program in the hopes of improving employee health. There are several advantages to offering a corporate wellness program. Wellness Programs What Are Wellness Programs? Wellness Programs are excellent for waistlines and your organization's bottom line. In today's hectic world, most of us are spending more time at work, and have increasingly less time to look after our health. For a long time, businesses have understood the benefits associated with keeping workers well - increased productivity from reduced absenteeism and reduced disability claims. The Benefits of Health Promotion in the Workplace Health promotion in the workplace benefits not only the company, but the employee as well. Additionally, there is a distinct trickle-down effect as the benefits are transferred from employee to family members. Health promotion in the workplace creates conditions that support and teach

the best possible health for the employee. This gives employees a sense of ownership and they learn to work together to form a coordinated action to improve well being. ... The Benefits of a Tobacco Free Workplace Tobacco related illnesses and deaths account for much of the medical expenses of the United States. The sad thing is that these illnesses and deaths can be prevented, and in effect, smokers are choosing their fate. As an employer, there may be things you can do to make people aware of the benefits of a tobacco free workplace, and one of those things begins with you.... Companies Make Great Gains with Employee Wellness Programs Companies can benefit greatly from adopting and maintaining employee wellness programs. Employee wellness programs can include things like health risk assessments, onsite health screening, health coaching, alcohol and drug counseling, mental health assistance, safety in the workplace, preventing violence in the workplace and diversity education. The tangible benefits of employee wellness program which directly result in economic savings can best be seen in these four areas:... Benefits of Corporate Wellness Most people spend more hours at work than anywhere else in addition to the time they spend commuting each day. In fact, the typical American works approximately 47 hours a week which is at least 164 hours more than the average 20 years ago. Given these statistics, it is easy to see why maintaining a healthy work / life balance is becoming increasingly important. Corporate wellness programs are important tools to establish this balance. Programs that emphasize the benefits of corporate wellness can be implemented in a variety of ways. However, they all share a common goal to promote the well-being of their employees, employers and organization in general. Many companies are starting to realize the tremendous benefits of corporate wellness...... Illness Costs Money but Good Health is Free ...Finding a health promotion program that addresses the unique needs of each company is a complicated and time-consuming task. Not all companies have the manpower or the industryspecific knowledge to take on this challenge, nor can they afford to hire a professional who can provide this personalized service for them. These factors prevent many smaller businesses from accessing the same health care solutions as their larger competitors... Worksite Wellness: Why Employers Should Care About Health and Wellness ...Traditional methods used by businesses to control healthcare costs such as; reducing benefits, increasing employee contributions and the more recent shift to consumer driven health plans are all short-term fixes that fail to address the primary driver of the soaring cost of healthcare namely inadequate investment in health through primary prevention, health risk management and disease management.... Workplace Wellness Programs ...Studies have found that employee absenteeism is directly related to four primary factors. Depression, anxiety and other mental health disorders topped the list at 66% while stress came in a close second at 60%. Negative relationship with a supervisor or manager yielded 44% and co

worker conflict in the workplace was at 28%. These factors can be effectively combated by employing workplace wellness programs. Conflict resolution programs, mental health programs and stress management courses have not only become popular, but have proven to be very effective.... Health Promotion Health Promotion initiatives comes in many different forms, but the main purpose of health promotion is to encourage individuals to take preventive measures to avert the onset or worsening of an illness or disease and to adopt healthier lifestyles. Employers may utilize a wide range of health promotion initiatives, from onsite gyms to simple health promotion newsletters. Workplace Health Promotion Successful workplace health promotion programs must incorporate an outcome-oriented workplace health promotion operating plan. A workplace health promotion operating plan is important because it: Links workplace health promotion initiatives to company needs and strategic priorities; Legitimizes the workplace health promotion program, which increases the likelihood of continued resources and support;... Workplace Wellness Programs Increasing evidence supports the need for workplace wellness programs and more companies than ever are implementing health and wellness strategies to reduce injuries, health care costs and long-term disability. With additional benefits such as reduced absenteeism, higher productivity, reduced use of health care benefits and increased morale and loyalty, its not surprising more and more employers are choosing to implement workplace wellness programs within their companies. Health Promotion and Wellness Programs: Top Health Risks Obesity, tobacco use and stress are the most common hazards to health in the American workforce. In addition to the direct harm they cause to individual wellness, they also provoke many indirect issues related to health care. Company Health and Wellness Programs: Healthy Culture Cultures develop because of shared beliefs, norms, values and traditions. Company cultures are the result of a combination of employee demographics, type of industry and management style. The company culture establishes and controls expectations for healthy behavior and lifestyle choices. Healthy lifestyle behavior is a result of expectations from employees, their families, and the management team. How the wellness program is introduced can create a culture of health or one of anti-health if the employees are forced to participate... Corporate Wellness Companies A comprehensive corporate wellness company program selects and follows goals set out by health care professionals and/or the domestic wellness committee. These goals are based on the results of the health screen. The goals provide direction and facilitate the development of activities to promote their achievement. Usually a responsibility of the wellness committee, the planning of corporate wellness company program activities and promotion should be visible and available for tracking purposes.

Employee Health Screening Tools: Uses and Importance When an employee Health Screening test is administered, it covers standard measurements of vital function. Most often, employees are aware that they have health issues, but have not been tested to determine the level of severity or the risk to their health that these problems represent. The extent to which their health has deteriorated is often an ignored variable. Chronically ill employees have come to accept their present level of illness as normal. The documentation of health risks is necessary in order to be processed for treatment and accessible for admission in employee health care initiatives. Worksite Health Promotion Program ...Gentle persuasion and voluntary involvement will encourage continuous employee participation. To optimize wellness program potential, it should have a range of health solutions that can encompass individual preferences and needs. Good health promotion program initiatives permeate the corporate environment as well as the laborers within it. To give the wellness program constant forward momentum, efforts to improve health should be rewarded and supported with evaluation and follow-up for as long as the employees remain with the company... Worksite Health Promotion Program: Choosing Realistic Goals Goal-oriented worksite health promotion programs need an approximate time specific directive to guide the application of the wellness program components. Generally speaking, the overall goal of a worksite health promotion program is to improve employee health...

Return to top
Wellness Program Challenges, Obstacles and More...

Health Promotion Strategy for the Small Business ...the same companies that do not offer health care also cannot afford to provide on-site environmentally-friendly support and controls. Furthermore, because employee health promotion strategies, health risk appraisals and screening tools are largely provided by health insurance companies, small businesses access to programs that promote wellness at work is greatly diminished. Employee Health Promotion: Making It Work ...As the gap between wellness and illness comes wider, the American public is re-examining company health and wellness programs as a solution to the growing problem of an obese workforce and the resulting unmanageable costs of maintaining this unhealthy lifestyle... Workplace Health and Wellness An Extended Application Workplace health and wellness programs have begun to germinate a growth spurt in health promotion but it has not yet expanded to a size where it can help all employees in any given company. Cohesive design remedies are necessary to extend the application of workplace health and wellness programs to the entire employee population....

Workplace Health Promotion Programs and Staff If health promotion programs are to successfully address the increase in occurrence and reduce the severity of chronic illnesses they should be supervised by staff who understand how to make the design applicable to the changing needs of the workforce. Health and wellness program staff members need to be capable of tailoring health promotion programs to the variety of conditions presented in each corporate environment.... Workplace Wellness Programs: Increasing Employee Participation Unless the employees in the workplace are interested in participating in workplace wellness programs, the addition of these programs to the health care benefit plan will not be effective in reducing corporate health care costs. State-of-the-art fitness equipment only works when it is used and exercise facilities are only beneficial when they are visited regularly by members and employees. Now, employers are faced with the decision of how to increase employee participation and how to make workplace wellness programs more attractive to employees of varying health and wellness through all the ranks.... Wellness in the Workplace: The Challenges One of the biggest challenges for North American corporations over the past 25 years has been to address the explosion in health care costs related to employee benefit plans. Viewed from an international perspective, businesses in the United States spend far more of their profits trying to maintain the health of their staff than any other nation in the world... Finding Health and Wellness Programs for Smaller Companies Despite the contribution of many good workplace wellness programs to the health care of the American public, there are still some companies that fall between the cracks. These are the groups of employees that do not have health insurance or employee assistance plans. Many smaller companies do not have the financial resources to back employee benefit packages; the employees cannot afford to pay outright for their own medical services, nor can they spare the funds for their own private health insurance.... Obstacles to Worksite Wellness Programs The 2004 National Worksite Health Promotion Survey reports that five obstacles impede successful implementation of Worksite Wellness Programs. Lack of employee interest accounted for 63.5%; insufficient staff resources contributed to 50.1%; inadequate funds were responsible for 48.2%; failure to engage high-risk employees added another 48%; and, the inability to elicit the support of upper management resulted in 38% more of the reasons why worksite wellness programs did not achieve their goals for wellness and health. .. Workplace Health Programs: Protecting Employee Rights Among the promise and potential of health promotion initiatives based at the worksite, some restrictions apply. Employee rights are imperative and any workplace health program initiative must have their interests first and foremost. This includes reassuring and taking steps to guarantee that employee confidentiality is respected and protected.... Shifting your Financial Focus: Planning for Group Wellness Designing a strategic plan for including worksite wellness programs in the corporate budget

involves a shift in the perspective from which group wellness is viewed. Planning for group wellness is the equivalent of planning an investment strategy rather than viewing the venture as merely a liability. To facilitate this viewpoint, isolating an outcome or a financial goal and being able to measure it is vital to company planning for worksite wellness programs... Potential Downside to Insurance Company Wellness Programs Employers struggling with the ever-increasing costs for healthcare are wisely turning to worksite wellness programs in search of a solution. For convenience, many employers are looking to their health insurance carrier to provide the appropriate wellness program. However, the convenience of a health insurance provided program may not outweigh the risks. That decision could come back to haunt them. At issue is an insurance carriers ability to use information gathered during a wellness program as justification for increasing an employers rates at renewal. Since no laws prevent an insurance company from using voluntarily provided information for rating purposes, employers should be cautious about the information they provide. Employers should be particularly concerned with how their insurance company uses data collected during a Health Risk Assessment (HRA). ... Corporate Wellness Programs: The No Name Version Is there a generic wellness version for companies who do not have the resources to invest in a large-scale company wellness program, but whose limited resources require it the most? The answer is, yes.

Obesity in the Workplace

Health Promotion Planning: Wellness Plans and Obesity One of the most prevalent health care issues in the United States is the rising rate of obesity. It is the most common, most expensive and least addressed disease among the workforce today. Obesity and tobacco use are the source of almost every preventable illness targeted by wellness plans and single-handedly cost employers most of their health care dollars.... Including Obesity in Group Health and Wellness ...Type 2 diabetes, hypertension, some kinds of cancer, cardiovascular disease, sleep apnea and hypertension, all of which are attributable to obesity, have become the primary causes of death in the United States. The rapid increase in the prevalence of all these conditions has caused medical expenses to explode beyond a level that companies can afford. In contrast, no action has been taken on the part of the employers who pay for the benefits or the health insurance providers that design applications of health care benefits, to address the issue...

Stress Management in the Workplace Stress Management in the Workplace There is no doubt that stress is one on of the leading factors in illness and absenteeism among

employees. Besides lowering a persons immune response, stress makes us want to avoid whatever is causing it. If there is stress at work, workers who feel mildly off will feel even worse and resist coming to work. This costs many hours of productivity, especially when key personnel or production workers are absentin fact, its estimated that $300 billion is lost on stress-caused illnesses and absenteeism....

General Wellness Program Articles Health Promotion in the Workplace: Transforming America The most serious health risks are behaviors that relate to lack of exercise, poor diet, and substance abuse including alcohol and tobacco. These are the toughest health risks to tackle, present the highest cost for group wellness plans, and represent only a small portion of the employee population. Many employers have elected to fund a wellness plan geared toward health promotion in the workplace, but they need help to achieve a higher level of productivity as an end result. ... Workplace Health Promotion ...Workplace health promotion efforts have evolved from offering a health risk screening tool followed by a wellness tutorial and complimented with on-site fitness equipment. In recent years wellness programming has progressed to sophisticated behavioral programming based on indepth and personalized health risk appraisals. Profit margins rely heavily on workplace health promotion to dig them out of the money pit in which their current health insurance plans have them buried. ... Employee Health Promotion Programs: A Second Look ...Who benefits from health promotion programs: the employees whose health is suffering, the company whose profits are sinking under the weight of managing illnesses or the health insurance company whose rates are higher than ever? Based on the expectation that the employees have to exert the most work, and the company has to foot the bill, the employeremployee team should reap the highest level of benefits from the employee health and wellness program.... Worksite Wellness Program Incentives ...Much like sticking to diets or quitting smoking, participation in wellness programs is often fraught with excuses from those who need the program most. Worksite Health Promotion Employees today are facing much higher levels of stress than ever before. The sad reality is that as fast as levels of stress are increasing, levels of fitness are steadily declining. People in all positions and professions are expected to drive longer commutes, perform more tasks, devote more hours and endure more stress. All of these factors contribute to the decline of overall employee health. An answer to this ever growing problem is found in worksite health promotion. ...

Bona Fide Wellness Program ...Generally speaking, a bona fide wellness program must offer some type of discount or limited reward. There are many different types of rewards which can be given in a bona fide wellness program . They include a rebate of the premium, a discount on the price of the premium, a partial waiver of things such as co-pays and deductibles, and many more. As you can imagine, all bona fide wellness program s are different so these rewards will change based on the details of your program... Wellness Programs in the Workplace Have you ever heard of wellness programs in the workplace? If so, there is a good chance that your employer feels strongly about the benefits of this type of program. Health and Wellness in the Workplace: Start-Up Guidelines ...Given the importance of this task you may want to consider a free consultation from an independent wellness consultant, like those at Wellness Proposals, who can help you determine what the corporate objectives should be. Once you have identified your objectives for health and wellness in the workplace, and have established a proposed timeframe within which to meet them, you will be able to choose the most appropriate Health Risk Assessment tool for your specific corporate health care needs... Company Wellness Program A company wellness program is set up in the workplace to offer employees a kind of comprehensive health service. The essence of a company wellness program is to encourage employees to adopt healthier lifestyles and to take measures aimed at preventing the worsening or onset of illnesses. There are various types of company wellness programs and each one may employ different tools and wellness initiatives to encourage active participation from its staff. Corporate Wellness Program With growing focus on employee health, corporate wellness programs are increasingly becoming the order of the day. Research done by American Sports Data shows those individuals who exercise frequently stayed home from work on an average 2.11 days annually compared to 3.06 days for individuals who were sedentary. These statistics alone stress the importance of setting up corporate wellness programs. Employee Wellness Program An employee wellness program is an effective method to promote health and wellness amongst the members of your staff. Programs will encourage awareness of health related issues, improve morale, and often times reduce cost of healthcare throughout the corporation. Health Wellness Program Health wellness programs are increasingly being incorporated by Companies as part of their overall strategy for promoting a healthier workplace. A health wellness program is set up in the workplace to offer employees a kind of comprehensive health service. Wellness Consultant With growing focus on leading healthier lifestyles, health care programs are mushrooming

everywhere. As employers become increasingly aware of the benefits of employing healthy and happy staff, wellness programs are being incorporated into company strategies for promoting a healthier work place. Definition of Wellness How do you define wellness? We continually hear this word during the news, in conversations, at work or read it in newspapers, magazines and the like. Surprisingly, there's no definition of wellness that seems to be universally accepted. Corporate Wellness Corporate wellness programs are designed to nurture wellness in workers, regardless of the work environment. Corporate Wellness programs could be found in factories, corporate offices, large corporations, and small corporations alike. Employee Wellness Having a Employee Wellness program in place can boost morale, improve health and fitness and increase productivity in the workplace. By starting healthful consuming habits, exercise and offering incentives, your workers will not only sign up for the Employee Wellness program, but they will stick with it. Company Wellness ROI Company Wellness programs are a long-term investment. But how long should you wait for results? Finance and the Chief Executive Officer (CEO) want hard numbers to show return on investment (ROI). And wellness ROI is tougher to calculate than, say, a 401(k). Workplace Wellness Keys to Success Workplace wellness programs come in all shapes and sizes. But regardless of plan design there are five common components that set the successful programs apart from the rest. Worksite Wellness How does a worksite wellness program affect a company? Worksite wellness programs undoubtedly benefit both the corporations and staff members. Numerous studies have found a direct link, showing that healthful staff members make happy staff members, hence increasing productivity in the workplace. Workplace Health Promotion Program Workplace health promotion programs are comprised of a number of programs that can include help with spiritual, psychological and physical health and wellness of a person. The absence of wellness has an adverse effect on an individuals performance at work which in turn has a negative impact on the companies bottom line.

Employee fitness program What is Cardio Fitness? We all remember our days in gym class of running laps and climbing ropes. We treasured playing dodge ball and running under a giant parachute. Aside from delighting our childlike

desire to play, these activities were designed to accomplish an important purpose. Gym class and sports practices were developed to get our hearts pumping and to improve our cardiovascular endurance.... Making Time for Fitness Face it, you have children, you have work, and you have a schedule that makes you tired just to THINK about it. It seems as if there just are not enough hours in the day. So how in the world are you going to continue to workout during the all of this rush?

Health Coaching / Wellness Coaching Health and Wellness Coaching Do you find it difficult to stay motivated when trying to make changes to your health? Are you aware that changes must be made in your daily life but you dont know... Health Coach What is a health coach? How can a health coach help you? How is a health coach unique? These questions and more are answered in this article. Low back pain in general practice: cost-effectiveness of a minimal psychosocial intervention versus usual care
Low back pain (LBP) affects a large number of people each year. Lifetime prevalence rates range from 49 to 70% [45]. LBP causes not only great discomfort, but also great economic loss due to work absenteeism [28, 45]. In the UK, LBP is one of the most expensive conditions for which an economic analysis has been carried out [28]. Economic evaluations in which both the costs and clinical outcomes of two or more interventions are compared, are becoming increasingly important, as health care expenditures rise while budgets remain limited [44]. The importance of economic evaluations is illustrated by the fact that some authors suggest that an intervention might be implemented when it is less effective but saves substantial costs [8]. As especially chronic LBP is associated with substantial costs to society [15], a large amount of costs can be saved by interventions that prevent acute LBP becoming chronic. The rationale is that it will be more cost-effective to address a wider target population early with simple low-cost interventions than to expend considerable time and resources on rehabilitating the smaller group of back pain patients who have become incapacitated by chronic pain. As psychosocial factors have been shown to play an important role in the transition from acute to chronic back pain [25, 33], one may assume that early interventions focusing on these factors prevent chronicity. We tested this assumption by conducting a cluster-randomized clinical trial in general practice, comparing a minimal intervention strategy (MIS) aimed at psychosocial factors for patients with (sub)acute LBP to usual care (UC) by the GP, which was not standardized. Our theory on the working mechanisms of MIS was that identification and discussion of psychosocial factors would lead to

modification of these factors, eventually leading to better functioning. Unfortunately, MIS appeared to be no more effective than UC in improving the following clinical outcomes: the degree of functional disability, the recovery rate and the number of patients on sick-leave due to LBP [20]. These findings are in line with Linton and Andersson [26], who showed that their cognitive-behavioral intervention was not more effective than usual care in reducing the degree of back pain and generic function status. However, their intervention was effective in reducing the number of visits to a physician for spinal pain and number of days of sick-leave, implicating time savings for physicians and thus substantial cost savings for society. These results may indicate an increase in coping or self-care with the pain; patients who received the psychosocial intervention, less often visited a physician and had less days of work while they had the same degree of functional disability as the patients who received usual care. These promising results, in combination with the fact that self-care with the pain was also a goal in our MIS, the recent emphasis on health care budgets, and the call for more high quality economic evaluations on the cost-effectiveness of treatments for LBP [43] stimulated us to conduct an economic evaluation from a societal perspective with a follow-up of 1 year. We hypothesized that MIS would be cost-effective compared to UC. Materials and methods Study design The study is designed as a full economic evaluation alongside a cluster-randomized controlled trial and was approved by the Medical Ethics Committee of the VU University Medical Center in Amsterdam, the Netherlands. Randomization and training sessions Randomization took place at the level of the general practice in blocks of four practices, according to a random numbers table prepared before recruitment of general practitioners. General practitioners were informed about their allocation after they had given final consent to participation. Twenty practices (28 GPs) were randomized to the MIS group and 21 practices (32 GPs) to the UC group. The GPs randomized to the MIS group received two training sessions of 2.5 h each which were given by a GP (HvdH) with extensive expertise in development and training of psychosocial interventions. The training consisted of theory, role-playing and feedback on the practiced skills. In addition, a treatment manual was provided. The contents of the training sessions and its evaluation by GPs have been described in more detail elsewhere [21]. Patients and interventions Participating GPs were asked to select ten consecutive patients who consulted them for LBP. Inclusion criteria were age 18 65, non-specific LBP of less than 12 weeks duration (i.e. (sub)acute LBP) or an exacerbation of mild symptoms, and sufficient knowledge of the Dutch language. Exclusion criteria were specific LBP (i.e. LBP caused by specific pathological conditions), LBP currently treated by another healthcare professional, and pregnancy. Patients, but not their GPs, were kept unaware that two different interventions were studied.

Patients received a minimal intervention strategy (MIS) or usual care (UC). The MIS was aimed at identification and discussion of psychosocial prognostic factors. The MIS consultation lasted about 20 min and consisted of three phases: exploration, information and self care. During the exploration phase, the GP explored the presence of psychosocial prognostic factors by asking standardized questions that could be rephrased to fit the style of communication of the doctor and the patient. The following psychosocial prognostic factors were explored: the patient s own ideas on the cause of their LBP, fear avoidance beliefs, worries/distress, pain catastrophising, pain behaviors and reactions from the social environment (family, friends, work). In the information phase the GP provided general information on the cause, course and (im)possibilities of treatment of LBP, thereby giving specific attention to psychosocial factors identified in the exploration phase. Finally, in the self care phase, the GP and patient set specific goals on resuming activities or work. Follow-up consultations were not protocolized, but we advised GPs to make an appointment for a follow-up visit in case they identified obstacles to recovery and suspected an increased risk of chronic LBP. GPs in the UC group provided care as usual. We did not protocolize the content and number of UC consultations, and assumed that GPs would generally follow the guideline for LBP of the Dutch College of General Practitioners [13]. For acute LBP (<6 weeks duration) this guideline advises a wait and see policy. For subacute LBP (6 12 weeks duration) the guideline advises referral for physical therapy in the case of persistent functional disability. Explicit guidance on psychosocial factors is lacking. The contents of both interventions have been described in more detail elsewhere [20]. Data collection Clinical outcomes Baseline data were collected during a home visit by a research assistant, while follow-up data after 12 months were collected using postal questionnaires. Primary clinical outcome measures were functional disability, perceived recovery, and health related quality of life. Functional disability was measured at baseline and after 12 months by the Roland Morris disability questionnaire (0 24) [35]. Perceived recovery was scored by the patient on a 7-point Likert scale (very much/much/slightly improved, no change, slightly/much/very much worse) after 12 months [42]. As a score of at least much improved has been denoted a minimal clinically important change [32], patients were a priori defined as recovered if they reported at least much improvement . Health related quality of life was measured at baseline and after 3, 6 and 12 months by the EuroQol (0 1) [12], covering five domains: mobility, self-care, usual activities, pain/discomfort, and anxiety/depression. The EuroQol scores were transformed into utilities using a representative British sample and time trade-off methods. The utilities were then multiplied by the amount of time a patient spent in this particular health state, with transitions between health states linearly interpolated [7]. This results in quality adjusted life years (QALYs). In order to describe the study population characteristics in more detail and to compare baseline similarity of both intervention groups, five other outcome measures were assessed at baseline: (1) pain severity during the day (0 10) [27]; (2) perceived general health (1 5), using the first question of the

subscale general health perceptions of the short form health survey (SF-36) [47]; (3) fear avoidance beliefs, using the 4-item physical activity subscale of the fear avoidance beliefs questionnaire (FABQ, 0 24) [46]; (4) catastrophising thoughts, using the 6-item subscale of the coping strategies questionnaire (CSQ, 0 36) [36] and (5) distress, measured by the 16-item subscale of the four-dimensional symptom questionnaire (4DSQ, 0 32) [37]. Cost data The economic evaluation was conducted from a societal perspective, indicating that all costs and consequences of the competing interventions are taken into account regardless of who pays for or benefits from them [9]. A societal perspective incorporates direct health care costs, direct non-health care costs and indirect costs due to LBP. All cost data were collected by prospective cost diaries [16] that patients completed for the periods baseline to 3 months, 3 6 months, 6 9 months, and 9 12 months. In addition, GPs were contacted after 12 months follow-up to provide information on follow-up consultations due to LBP in the last year (date, referrals, medication prescribed by physician). GPs used their medical records to complete these registration forms. Table 1 summarizes the cost categories, and the prices and sources used for valuing. Medication in Table 1 includes both the over-the-counter medication and medication prescribed by physicians, while self-prescribed alternative interventions are included in complementary care . Costs of absenteeism from paid labor due to LBP were calculated according to the friction cost approach [23]. This approach is based on a mean income of the Dutch population according to age and gender, and defines a friction period as 154 days [31]. As most cost data were collected during the year 2002, prices were adjusted using consumer price index figures. Table 1 Prices used for valuing resources (year 2002)

Statistical analysis Firstly, baseline similarity was studied. Secondly, we compared baseline characteristics of patients with complete cost data to those with incomplete cost data by using logistic regression analysis. Thirdly, clinical outcomes and total costs were compared. As cost data are characterized by large variation and irregular distributions and as a complete cost dataset was available for 80% of our participants (250/314), we decided that our primary analysis would be a complete case analysis. Differences between groups (MIS minus UC) were calculated for the clinical outcomes: (1) functional disability, by calculating change scores between baseline and 12 months follow-up; (2) perceived recovery at

12 months follow-up; and (3) health-related quality of life gained over 12 months (i.e. QALYs). Students t tests were used to analyze the change scores between the treatment groups for functional disability and quality of life, and a Chi-square test for perceived recovery. To compare costs between the two groups, confidence intervals (CIs) for the mean differences in costs were calculated using bias-corrected and accelerated bootstrapping (2000 replications) [11]. Bootstrapping incorporates drawing samples with replacement and is a preferred method for the analysis of cost data, as it uses the observed distributions of the data without making assumptions about the shape of the distribution [38]. Fourthly, costeffectiveness analyses were performed. Incremental cost-effectiveness ratios were calculated in which the mean difference in total costs (MIS minus UC) was divided by the mean difference in improvement on the clinical outcomes (MIS minus UC). Uncertainty around the ratios was calculated using the biascorrected percentile bootstrapping method (5,000 replications) [5] and plotted on a cost-effectiveness plane. Sensitivity analyses A complete case analysis has some disadvantages. Due to the missing data the power of the analysis is reduced and bias may be introduced due to selective drop-out. Imputation can be used to replace missing data by statistical estimates of the missing values. To explore the robustness of our primary analysis, we performed two sensitivity analyses in which we imputed missing data: in one analysis we imputed all missing cost data, and in the other we imputed only missing days of absenteeism. Imputation was done using the Expectation Maximization algorithm (SPSS 10.1). Results Between September 2001 and April 2003, 314 patients were enrolled in our study: 143 in the MIS group and 171 in the UC group. Table 2 shows that baseline characteristics of GPs and patients were largely similar for the two groups. Less than 9% of all patients withdrew from the study during follow-up. Reasons were no time and no complaints anymore (MIS n = 1, UC n = 4), burden too high due to psychological problems (UC n = 3) or unknown (MIS n = 10; UC n = 8). The flow chart of this study, including information on refusals, exclusions and drop-outs has been published in our previous paper [20]. Table 2 Baseline characteristics of general practitioners and patients

For 116 patients (81%) in the MIS group and 134 patients (78%) in the UC group complete cost data were available. Logistic regression analysis comparing baseline characteristics of patients with complete cost data to those with incomplete cost data showed that patients with incomplete cost data on average were younger and scored higher on distress at baseline.

Clinical outcomes After 1 year follow-up both the groups showed similar improvements in clinical outcomes. Sixty-nine percent of the patients were defined as recovered after 1 year. The difference between both the groups in mean improvement on functional disability was 0.74 points on the RDQ (95% CI, 2.31 to 0.83), and 2% (95% CI, 14 to 10%) on recovery rate. Over the follow-up period of 1 year the mean difference in quality of life was 0.004 QALYs (95% CI, 0.04 to 0.03). All differences favored UC, but were neither clinically relevant nor statistically significant (Table 5). Table 5 Incremental cost-effectiveness ratios for functional disability, perceived recovery and health-related quality of life

Cost data Table 3 lists per patient the mean utilization of resources (i.e. health care, help, absenteeism). In both the groups, resource utilization was low and largely similar. Only two statistically significant differences were found. In the year following randomization patients in the MIS group had more consultations with a GP (MIS 2.7 vs. UC 0.9), excluding the consultation leading to recruitment but including the 20 min consultation aimed at psychosocial measures as specified by the study protocol. Patients in the UC group reported more consultations with a manual therapist (MIS 0.1 vs. UC 0.4) but the proportions of patients who received such a treatment (MIS 2.6% vs. UC 9%) were very low. Table 3 Mean resource use (SD) per patient (n = 250) for MIS and UC during 12 months follow-up, and the percentage of patients who made use of that specific resource

Table 4 shows the mean total costs in both the treatment groups and the difference in costs with 95% CI. Total indirect costs, especially absenteeism from paid work, were the largest contributor to the total costs. The difference in total costs amounted to 490 (95% CI 987 to 92 ) in favor of the MIS group (MIS 799 ; UC 1288 ), but this difference was not statistically significant.

Table 4 Mean costs (SD) in Euros per patient in the MIS and UC group and differences between both the groups during follow-up of 52 weeks

Cost-effectiveness Table 5 shows the incremental cost-effectiveness ratios (ICERs) for the three outcome measures. MIS resulted in less improvement than UC, but saved money. The ICER for functioning was 690 , indicating that per point less improvement on the RDQ MIS saved 690 , while per percent less improvement in recovery rate MIS saved 239 . The difference in QALY s gained during 1 year between both the groups was very small, resulting in a large ICER of 47,348 . The large majority of the bootstrapped ICERs presented on the cost-effectiveness planes are located in the southern quadrants (Fig. 1), indicating that the costs of MIS were lower than the costs of UC. Fig. 1 Cost-effectiveness plane for functional disability (RDQ) in which MIS is compared to UC

Sensitivity analyses Imputation of missing cost data led to a mean difference of 628 (95% CI 1123 to 81 ) in total costs. Imputation of missing data on days of absenteeism led to a mean difference of 545 (95% CI 1031 to 40 ) in total costs. Both differences are statistically significant and in favor of MIS. Discussion The results of our primary analysis showed no statistically significant differences in total costs or clinical outcomes between our psychosocial intervention and UC in patients with (sub)acute LBP in general practice. However, the results of our sensitivity analyses are inconsistent with those of the primary analysis. In this discussion section we will focus on the interpretation of our cost data and the methodological issues involved when interpreting cost data. In previous papers we have discussed several methodological issues involved when interpreting the clinical outcomes (e.g. the quality of the training sessions and interventions) [20, 21 Educating the public health workforce: Issues and challenges Background In public health, as well as other health education contexts, there is increasing recognition of the transformation in public health practice and the necessity for educational providers to keep pace.

Traditionally, public health education has been at the postgraduate level; however, over the past decade an upsurge in the growth of undergraduate public health degrees has taken place. Discussion This article explores the impact of these changes on the traditional sphere of Master of Public Health programs, the range of competencies required at undergraduate and postgraduate levels, and the relevance of these changes to the public health workforce. It raises questions about the complexity of educational issues facing tertiary institutions and discusses the implications of these issues on undergraduate and postgraduate programs in public health. Conclusion The planning and provisioning of education in public health must differentiate between the requirements of undergraduate and postgraduate students while also addressing the changing needs of the health workforce. Within Australia, although significant research has been undertaken regarding the competencies required by postgraduate public health students, the approach is still somewhat piecemeal, and does not address undergraduate public health. This paper argues for a consistent approach to competencies that describe and differentiate entry-level and advanced practice. Introduction The growth of undergraduate public health education in Australia has paralleled, but is not necessarily a consequence of, discussions about the requirements for a flexible health workforce to meet contemporary and future health challenges. Health workforce shortages and calls for renewal of the health workforce in Australia have been well-documented. The Australian Government Productivity Commission was appointed by the Council of Australian Governments (COAG in June 2004) to produce a comprehensive research report that outlined the breadth of the trends, issues and challenges in the health workforce in the next 10 years, including efficiencies and effectiveness and the need for innovation. Consultations and submissions were extensive and the Productivity Commission Health Workforce Study was released in January 2006 [1]. Subsequent to this report, "Health Workforce Australia" was established as a committee of the Australian Health Ministers Advisory Council (AHMAC) in late 2007. It claims that "the issues experienced at a national level include workforce shortages, maldistribution, managing changing models of care and maintaining a culture of effective governance and continuous improvement", and its objective is to ensure that our health system most effectively uses a skilled workforce to best support service delivery to all Australians [2]. Discussions about the need for a flexible public health workforce have been propelled not only by the above national developments but also by internal debates within the public health community about a set of competencies that would best meet the changing health profile of Australians. In addition, the evidence of effective public health investments to reduce the burden of premature mortality and morbidity has impacted on public health workforce requirements [3,4].

Furthermore, the new government is committed to a National Preventative Health Strategy, to be developed by a National Preventative Health Taskforce that will focus initially on the burden of chronic diseases, particularly the contribution of alcohol, tobacco and obesity [5]. With the spotlight on prevention within the policy agenda of the new government, discussion is warranted on the skills, competencies and attributes necessary for the public health workforce, as well as an analysis of current educational opportunities needed to fulfil the potential gaps in this workforce. The public health workforce: The changing nature of practice Public health's scope of responsibility is broad and ever-increasing, such that public health practitioners require a variety of skills; many, however, have had insufficient education or training regarding how to address increasingly diverse and emergent public health challenges [6,7]. The skills and proficiencies required include a basic insight as to what public health is, what it does, and how it accomplishes its aims. In addition, core competencies that impart knowledge and expertise regarding all spheres of public health practice are required; as is specialised "know-how", which provides the precise knowledge and expertise necessary for specific programs or functions. Although there are a variety of views as to what these specific competencies are or should be, there has been significant research undertaken by a range of organisations to facilitate the attainment of consensus on the essential skills needed for public health practitioners. The roles and functions of the public health workforce are diverse in nature, as they are influenced by the contexts within which they are set [8]. This varied make-up of the workforce offers both possibilities and challenges for public health training [9]. The challenges result from deficiencies in employee competencies; that is, those necessary for existing duties or those required to deal with evolving problems [9]. The capacity to pinpoint student destinations and utilise relevant approaches to public health workforce development is vital [9,10]. Public health practitioners can come from science, humanities and arts; and this multidisciplinary composition is one of its strengths because the protection of populations requires a comprehension of, for example, the environmental, social and political foundations of health. This is also a limitation, as there are insufficient shared core competencies across different fields [11]. Moreover, the tasks are rarely standardised, entailing distinct combinations of expertise to isolate and assess the origins of public health challenges, provide useful approaches to these, and the capacity to judge their "impact and effectiveness" [8]. Public health education ought to be provided to the entire public health workforce; from the undergraduate level, to in-service training for established public health workers, through to postgraduate level. Furthermore, fundamental public health concepts and skills should be taught to all workers employed in health-related positions, as well as to people working in areas traditionally not seen as being part of the health sector [12] such as town planning and transport. US working groups (Association for Prevention, Teaching and Research, Association of Schools of Public Health, Council of Colleges of Arts and Sciences) go even further, recommending that introductory courses in public health and epidemiology should be available to all undergraduates, not just those in public health [13,14].

The increasing need for practitioners with recognised qualifications in public health has generated the growth in education and training programs. Progress in public health disciplines and an increased range of employment opportunities has led to curricula that meet a more diverse range of needs, as the challenges and problems of public health become progressively more complicated [15]. The quality of public health education is gaining in importance due to the increasingly diverse activities that professionals may have to undertake and the expansion in required skills [16]. A degree in public health serves to enhance graduate skills and the application of those skills, fulfil requirements for promotion and facilitate movement within the health sector. Traditionally, the principal employers of these graduates were the government sector at national, state and local levels; and the non-government sector, such as the Heart Foundation, the Cancer Fund and Diabetes Australia. However, prospects for those with qualifications in public health are increasingly diverse, with a wide range of enterprises employing public health practitioners. Current public health practice has expanded to embrace a range of different environments; for example, voluntary organisations, diverse commercial/industrial sites, community-based groups and health care services. Internationally, authors have discussed the need to increase expertise in the public health workforce if practitioners are to meet the diverse requirements of their roles in protecting and advancing health [16]. Workforce capacity reviews indicate that there are shortages of people with the necessary skills and knowledge [16,17]. However, developing the workforce is a complex task because of the numerous responsibilities that come under the aegis of public health, and the wide-ranging skills that are considered essential [16]. It is timely to consider how public health education meets the challenges of a diverse and increasingly complex workforce. The impact of competencies on curriculum development There are a number of examples, both nationally and internationally, of activity that has occurred to establish core areas of activity or competencies in public health. In 1997, the World Health Organization performed a Delphi study that delineated 37 vital public health functions, and established that it was possible to achieve wide-reaching agreement on these core areas [18]. In Europe, the Association of Schools of Public Health (ASPHER) in the European Region met in Denmark in April 2008 to build on work that had begun within the European Union where projects examined public health courses, competencies and accreditation: "An MPH accreditation document was developed in 2002 by ASPHER and a set of standards and procedures has now led to a set of Accreditation Standards (EU Accreditation of European Public Health Education; MPH Programme Standards)" [19]. This activity was driven in part by the Bologna Declaration. Signed in 1999 by European Union (EU) higher education ministers, the Bologna Process aims to create a European Higher Education Area (EHEA) based on international cooperation and academic exchange that is attractive to European students and staff, as well as to students and staff from other parts of the world [20]. In the United States, core competencies for public health curriculum have been a well-established component of the academic and practice landscape. The Council on Linkages Between Academia and Public Health Practice developed the "Core Competencies for Public Health Professionals" to help

strengthen public health workforce development [21]. This builds on 10 years' work on this subject by the Council and numerous other organisations and individuals in public health academic and practice settings. The list has been compared with the "Essential Public Health Services" [22] to ensure that the competencies help build the skills necessary for service delivery. A consensus set of core competencies for guiding public health workforce development has been achieved in the US. These competencies include analytic/assessment, policy development/program planning, communication, cultural competency, community dimensions of practice, basic public health sciences, financial planning and management and leadership and systems thinking. In the United Kingdom, the Faculty of Public Health identifies the curriculum areas outlining the competencies or learning outcomes that trainees in public health need to attain in order to complete their training. These nine key areas relate to the three domains of public health practice (health protection, health improvement and service quality). These key areas include "surveillance and assessment of the population's health and wellbeing; assessing the evidence of effectiveness of health and healthcare interventions, programmes and services; policy and strategy development and implementation; strategic leadership and collaborative working for health; health intelligence and academic public health" [23]. In Australia, a number of projects have examined the development of core competencies. For example, health promotion competencies were developed in the early 1990s in Western Australia. More recently, in 2006, an extensive consultation process resulted in the development of competencies for health promotion practice in Australia [24]. Another project, managed by the National Public Health Partnership Group, surveyed Australian public health experts regarding their views on defining public health functions. The aim was to help identify essential public health functions and develop Australian public health capacity [25]. PHERP competencies are intended for PHERP-funded universities offering generalist Masters in Public Health. Five broad categories recognise the major themes for public health education and 19 public health units of competency are subsumed under these five broad categories. The categories are: Health Monitoring and Surveillance; Disease Prevention and Control; Health Protection; Health Promotion; and Health Policy, Planning and Management [26]. This work is yet to be agreed upon by PHERP-funded institutions and then brought to a conclusion. The public health education response The majority of public health courses have been at the postgraduate level both in Australia [18] and internationally [27,28]. Accordingly, admission into public health education has generally been open to people with a first degree in a range of professional or academic fields and/or relevant work experience [18,27]. In the past decade in Australia the number of universities offering dedicated undergraduate public health programs has expanded [18]. Mor et al. [29] suggested that increasing attention to public health and health promotion in contemporary society is driving an interest in undergraduate public health

curricula in the United States. In part, this has also occurred in tertiary institutions in Australia. However, there is limited discourse in the Australian context around the implications of these developments some of which are introduced below and covered in detail later in the paper: Does the expansion in undergraduate public health education impact on the nature and scope of the postgraduate public health curriculum? Are the curricula influenced by competencies? Do curricula complement each other or are there clear differences? If there are differences are they a matter of content, level or both? What are the expectations vis--vis undergraduate programs, compared with graduate level programs that have had as their focus the development of public health leadership skills? Are there any differences in the employment prospects for students who complete undergraduate degrees compared with those who complete postgraduate degrees? Are there any differences in graduate capabilities? The research pertaining to undergraduate public health education is limited compared to that of postgraduate public health education, nursing-related public health education and public health training for medical students. There is abundant research and commentary with regard to competencies for Master of Public Health education and training, and some consideration of specific subjects or competencies that should be taught to undergraduates; for example, "cultural competence" [30], and "problem-based learning" skills [31]. However, there is a dearth of literature regarding the differences between undergraduate and postgraduate public health education. Are there implications for postgraduate education for those students who have completed a public health undergraduate degree? "Undergraduate public health education" in this commentary refers to distinct public health education, not other health courses that incorporate public health subjects; for example, nursing. Postgraduate education usually refers to the Master of Public Health. Other postgraduate education in health sciences or health management, for example, includes public health units but does not involve a comprehensive range of specific units in the discipline area. Developments in undergraduate public health education: The international scene The information presented below is a snapshot of a range of tertiary institutions in a number of countries that offer undergraduate education in public health. In the US, undergraduate students' interest in public health is stimulated by a range of global issues including the rise in both pandemic and chronic diseases [11,32], risks to biosecurity [28,32], technological advances requiring new expertise, ecological damage and catastrophes, demographic changes [32], and the high proportion of public health professionals with limited formal public health training [11,28].

Undergraduate public health education can be seen as a natural outgrowth of the rise in interdisciplinary undergraduate education with insights regarding influences on public health having roots in a variety of subjects, including economics, psychology, anthropology and biology [28]. Conversely, the population approach of public health, and the generic expertise and knowledge gained, provide a good basis for a variety of professions, and postgraduate study in a range of disciplines, including social work, law and health policy [28]. For example, there are over 40 undergraduate public health programs in the US [29] that demonstrate a wide and varied curriculum even though there is a new accreditation process for undergraduate programs (managed by the Council on Education for Public Health [CEPH]) should a university choose this option. This new accreditation process has the "potential to radically change the scope of undergraduate public health education" [14]. One of the CEPH's requirements for accreditation is that the university must be focussed on "training students for entry-level public health positions" [14]. By contrast, a survey of Canadian universities retrieved only one undergraduate public health program. None of the five South African universities scanned (Universities of Cape Town, Johannesburg, Pretoria, South Africa and the Witwatersrand) offered undergraduate public health programs; although, the following three universities provided MPH or MA (Public Health) programs Cape Town, South Africa, the Witwatersrand. The Asia Pacific Academic Consortium of Public Health (APACPH), representing public health schools in over 20 countries in the region, has a working party examining undergraduate public health curricula across its member institutions to consolidate commonalities, gaps and competencies in order to integrate public health education with workforce needs to address public health priorities [33]. Implications Without an agreed definition of the fundamental responsibilities of public health practice and the consequent knowledge and proficiencies required of public health professionals, it is difficult to define an appropriate curriculum. A number of issues arise for curriculum designers. First, thoughtful mapping of the curricula and levels of competencies is needed for those students wishing to articulate from an undergraduate public health degree into a postgraduate degree. As discussed earlier in this commentary, this issue is particularly important for examining the core sets of knowledge and skills of epidemiology and biostatistics. Those with undergraduate degrees in public health would be well-placed to tackle advanced epidemiology and a breadth of advanced research studies in qualitative research, social research, health economics, strategic leadership, and change management; as well as retaining a discrete focus on health monitoring and surveillance, health protection, disease prevention and control, health policy planning and management and health promotion. There should be an expectation that students graduate with high levels of analytical and conceptual skills. Second, health care sector engagement is pivotal for advancing curricula, especially in anticipating future needs, and the skills and attributes required of graduates. The use of industry liaison groups,

including international industry partners and course advisory groups to link all graduates to employment may be an important strategy for increasing collaboration between the workplace and the university sector. Third, workforce mobility is an important consideration for graduates of public health programs. For example, a student of an undergraduate public health program may choose not to advance their public health skills, but rather engage in education in management and policy. To date, limited information exists about the employment profiles of students who complete undergraduate programs in Australia to enable tracking of their employment over the past five years to gain further understanding of the choices undergraduate students make. The links between workforce readiness and university education cannot be underestimated, as health employers develop their own competencies. For example, some state health departments have developed entry-level and advanced competencies for public health workers, without reference to university curricula. In addition, a range of other groups have developed benchmarks for graduates' employability skills and attributes. For example, the "Employability Skills for the Future" developed by the Australian Chamber of Commerce and Industry and the Business Council of Australia; Graduate Employability Skills as prepared for the Business, Industry and Higher Education Collaboration Council (2007); and the Graduate Attributes as described by all universities. Conclusion The issues addressed above attempt to answer many of the questions that were raised earlier in this commentary. With the emergence of a number of undergraduate public health programs in Australia one of the challenges is to ensure that students are educated to meet the needs of a flexible and dynamic public health workforce. Employers, professional associations and universities need to work together to ensure that needs are met for entry-level practice with the option of developing more advanced knowledge and skills in areas such as epidemiology and statistics and policy and management at a postgraduate level. Also of importance is the role of the employer in continuing professional development in the workplace, in particular for those public health workers who are entry-level practitioners. Universities and employers need to continue to work together to ensure that the field is well-represented by competent and capable public health practitioners who can work in an everchanging environment to advance the health of the population.

The need for general mental health training

A mental health training infrastructure that can be widely utilized needs to be developed to come closer to the goal of meeting the mental health needs of the community. This infrastructure is dependent upon evaluated mental health training programs. Unfortunately, Australia's mental health care services are fragmented. This fragmentation often leads to poor mental health care. This issue has been identified in numerous state and national reports on mental health services. For instance, in the first report by the Senate Select Committee on Mental Health, it was noted that "Compartmentalization of services (often metaphorically referred to as 'silos') [was] preventing effective care [14]." Mental health workers need to be encouraged out of the 'silos' they have traditionally inhabited. However, most of the current

education and training mental health workers receive reinforces the existing fragmented structures. If the mental health workforce is ever to provide seamless care it is important for general mental health training to be directed at the whole mental health work force. Mental health workers need to develop a common idiom so there is reduced confusion between different sectors. Further, mental health workers from different sectors need to develop a sophisticated understanding of the nature of the services available in other sectors so as to ensure that their clients receive the best possible mix of services. Improved mental health care has been a goal of both State and Federal Australian governments. In response to this need for generalist training the State Government of Victoria through beyondblue: The National Depression Initiative commissioned "the Mental Health Aptitudes into Practice (MAP)" training package which was trialled and delivered throughout Victoria. MAP training program A substantial needs study was undertaken to inform those developing the training program [15,16]. It showed that there was widespread agreement about the need to expand, develop and support all those whose profession meant, that they played significant roles in the treatment of those with mental illness. It was decided not to focus on medically trained doctors as there was a variety of other training already available for them. There was also agreement on the need for training to be focused on roles and tasks that are common across a broad range of primary care and community-based services, such as the ability to identify people with or at risk of depression and related disorders, good referral practices and better care coordination for people with complex needs. Based on these findings, the aim of the training program was established, which was to enhance the overall capacity of the mental health care work force "to acknowledge, accept, and respond effectively to individuals with, and at risk of, depression and related disorders [15]." MAP training was developed firstly for those whose professions were not primarily related to the treatment of mental illness but who were likely to come into contact with many who had mental illness (i.e. ministers of religion, police, personal carers), secondly for health and mental health primary care workers (i.e. community workers, psychiatric disability rehabilitation and support service workers [PDRS workers], aboriginal health workers) and finally for mental health professionals (i.e. psychologists, counsellors, social workers and nurses). To meet the diverse needs of these groups a modular training structure was developed with various modules being considered appropriate for those with different backgrounds and experiences. Curriculum development occurred over a six month period and was followed by a trial period in which the training was piloted and refined. Modules were produced for seven content areas. Although seven training modules were offered only the core modules are evaluated. The core modules comprised: Module 1: Introduction to Mental Health and Mental Illness, Module 2: Depression: Introduction, Module 3: Anxiety and its Treatment, Module 4: Depression: Treatment, Prevention & Relapse. MAP training was completed across Victoria between the 1st of April 2004 and the 4th of September 2005. Two hundred and seventy one days of free training were delivered to 2,043 individuals. Data pertinent to the evaluation was collected routinely as a component of the MAP training package. The

structure of the evaluation was informed by the CIPP (context, input, process and product) evaluation model checklist [17]. This model was chosen as it is particularly focused on evaluating programs that aim to effect improvements that are sustainable in the long term [17]. Evaluation focused on participant changes across five domains that reflected the training aims. The five domains included confidence in dealing with someone with mental health issues, knowledge and skills in relation to mental health issues, mental health literacy, the social distance that they felt it necessary to maintain with those with mental disorders, and community mental health ideology. The last three of these, which are not self explanatory, are described further below. "Mental health literacy refers to the knowledge and beliefs about mental disorders which aid their recognition, management and prevention [18]." It has been repeatedly argued that an impediment to people receiving adequate mental health care is the lack of mental health literacy [18-21]. "Social distance is the degree of proximity an individual is comfortable with in relation to a mentally ill target and it is recognized as a proxy measure of psychiatric stigma [22]." There has been some debate about whether increased mental health literacy leads to a rise in social distance [23]. "The community mental health ideology (CMHI) scale expresses sentiments concerning the therapeutic value of community, the effect of mental health facilities on residential neighborhoods, and the acceptance of deinstitutionalized care [24]." The scale is one of the four scales that make up the CAMI (community attitudes toward the mentally ill) [25]. It was expected that following attendance of the core MAP modules participants would have, 1. improved mental health literacy 2. increased knowledge and skills relating to mental health issues 3. improved confidence in their ability to help someone with mental health problems 4. decreased desire for social distance from those with mental illness 5. raised community mental health ideology Further, it was expected that these changes would be sustained and observable six and twelve months after the training. Methods Study design This study is a prospective evaluation assessing changes in confidence, mental health literacy, attitudes towards effective treatment, knowledge and skills and community mental health ideology following MAP training. These elements were assessed using pen and paper tests prior to the training and then immediately following the training. They were then reassessed 6 months and 12 months after the

training. The evaluation was approved by the Monash University Standing Committee on Ethics in Research involving Humans. Subjects There were 1126 participants in the MAP training program who completed the core MAP training modules. Of these 876 (77.8% of the participants) consented to be involved in this evaluation. The training was designed to be delivered to workers from diverse sectors of the mental health workforce, therefore, the evaluation focused on general skills and attitudes that could be seen to be shared by all those who are likely to work with those with mental illness rather than on specific skills required by those in various mental health professions. Evaluation materials Attempts were made to collect evaluation data on four occasions from each participant: Table Table11 details the extent of data obtained at each stage of the MAP evaluation. The majority of the questions in the MAP evaluation questionnaire were extracted from other mental health training evaluations or general research in mental health. The question accessing the participants' confidence was adapted from a question first reported by Kitchener and Jorm [9]. A higher score on this confidence measure indicates increased confidence. The questions assessing social distance were derived from the work of Link, Phelan, Bresnahan, Seueve and Pescosolido [26]. A lower score on this measure indicates that the trainee desired more social distance. The mental health literacy questions were first used by Goldney [18]. Taylor and Dear [25] developed four scales to measure community attitudes toward the mentally ill; community mental health ideology is one of these scales and was used in this evaluation. The higher the score on this measure the more positive the trainees' view of mental health service provision in the community was. A series of 38 questions asked participants to provide a self assessment of their knowledge and skills. The 18 knowledge questions covered the following areas: understanding of the burden of mental illness, causes of depression and other mental disorders, mental illness risk factors, mental illness treatments, mental health services and resources for the treatment of mental illness. The 20 skills questions covered the following areas: identifying those with mental illness, assessing their needs, engagement and communication skills, referral skills, ability to work collaboratively and dealing with distressed clients. The higher the trainees' score on both the knowledge and skills measure the greater their self assessed skills and knowledge. Table 1 Characteristics of the MAP evaluation participants

The initial evaluation questionnaire (the long version) took participants approximately 20 minutes to complete. All the participants completed this long version of the questionnaire prior to the training (pre)

and immediately following their training (post). Due to concerns about the response rate a shortened version of the questionnaire was constructed. A series of factor analyses and RASCH analyses of pre and post questionnaires from the first 150 participants allowed researchers to determine which questions to retain in the shortened version of the questionnaire. The shortened version of the questionnaire took participants approximately 4 minutes to complete (details of this analysis are not yet published but can be obtained by contacting the first author of this paper). Most of those who completed the evaluation questionnaire six and then twelve months after the training completed the shortened version of the MAP evaluation questionnaire (see Table Table11). Statistical analysis Examining differences amongst those who completed different numbers of the MAP evaluation questionnaires Differences in the ages of those who completed various numbers of questionnaires were examined using an ANOVA. The proportion of males and females completing different numbers of the evaluation questionnaires was examined using a chi-squared analysis. The data collected in the evaluation questionnaire was used to form variables that were either ordinal or continuous. To determine if there were any differences between those who completed one, two, three or four of the evaluation questionnaires on these variables in the initial questionnaire Kruskal-Wallis tests were used on the ordinal data and one way ANOVA was used on the continuous data. Examining Changes over time To maximise the amount data examined, three groups of data were examined: 1. Data from 185 people who completed all four (pre, post, post6, and post 12) of the evaluation questionnaires 2. Data from 358 people who completed the pre, post and post6 evaluation questionnaires 3. Data from the 674 people who completed the pre and post evaluation questionnaires. Changes over time on the ordinal variables were examined using the Friedman test. Changes over time on the continuous variables were examined using repeated measures ANOVA. Results The evaluation participants The participants' ages ranged from 18-81 years (mean = 44.7, median = 46, SD = 10.5) the majority of the participants were over forty. Close to 90% of the participants were female (Table (Table1).Table1).Table ).Table22 lists the occupations of the participants and shows that the participants were drawn from varied occupations. Twenty percent of the participants classified themselves as nurses; the trainers estimate that approximately one third of these nurses were primary care nurses,

one third were working in the aged care sector and one third had other roles in the community. Table Table22 also describes the qualifications of the participants. Table 2 The participant's occupations and qualifications

Differences between those who completed all the evaluation measures and those who didn't The 876 participants in the evaluation completed various combinations of the MAP questionnaires. A one way ANOVA showed that there were no significant age differences between those who completed different numbers of questionnaires F(3, 831) = 2.254 p = .081. A chi-squared analysis showed that there were no significant differences in the proportion of males and females who competed different numbers of questionnaires 2(3,857) = 4.75 p = .191. A chi-squared analysis on a slightly restricted sample (those who had completed as masters degree or higher were excluded from the analysis so as not to violate the assumptions of chi-squared) showed that there was no significant difference in the educational attainment of those who completed different numbers of questionnaires 2(21, 810) = 18.63 p = .609. To examine whether there were any differences in the professions of those who completed different numbers of questionnaires it was necessary to examine a restricted sample as cell sizes were too small for statistical examination of most of the professions. Only the five most prevalent professions (nurses, social worker, personal carer/aged care worker, welfare worker, and youth worker/youth health worker) were included in the analysis. There was no difference in the proportion of these professionals who completed different numbers of the questionnaires 2(12, 466) = 5.23 p = .950. A series of different statistical tests (Kruskal-wallis for ordinal data and one way ANOVA for continuous variables) were conducted and showed that were no significant differences in the participants' initial confidence scores, mental health literacy scores, knowledge and skill ratings and community mental health ideology scores if they completed different numbers of questionnaires. These tests are described in Table Table33. Table 3 Examining differences in those who completed different number of MAP questionnaires

Background There is increasing pressure to tackle the wider social determinants of health through the implementation of appropriate interventions. However, turning these demands for better evidence about interventions around the social determinants of health into action requires identifying what we already know and highlighting areas for further development. Methods Systematic review methodology was used to identify systematic reviews (from 2000 to 2007, developed countries only) that described the health effects of any intervention based on the wider social determinants of health: water and sanitation, agriculture and food, access to health and social care services, unemployment and welfare, working conditions, housing and living environment, education, and transport. Results Thirty systematic reviews were identified. Generally, the effects of interventions on health inequalities were unclear. However, there is suggestive systematic review evidence that certain categories of intervention may impact positively on inequalities or on the health of specific disadvantaged groups, particularly interventions in the fields of housing and the work environment. Conclusion Intervention studies that address inequalities in health are a priority area for future public health research. Keywords: Evidence, health inequalities, interventions, social determinants, systematic review, social inequalities, socioeconomic, unemployment and health

It is well established that health follows a social gradient: better health with increasing socioeconomic position.1 The importance of the social (as opposed to biological or genetic) causes of this gradient for example, housing quality, access to healthcare or quality of work, has also been established.2 3 In turn, this has lead to increasing pressure in research, practice and policy-making environments to tackle these wider social determinants of health, through the implementation of appropriate interventions, and thereby reducing the gradient and health inequalities.2 4 However, there are two concurrent problems. First, the social determinants evidence base is dominated by descriptive, epidemiological studies that, by highlighting associations, are only implicitly able to suggest possible interventions. For example, studies consistently show associations between higher job control and better mental health; by implication, therefore, interventions that increase job control should result in health improvements.5 What is lacking though is further evidence about what sort of interventions might be required or whether they will actually be effective in improving health or reducing the social gradient.

Second, where interventions aimed at reducing health inequalities have been developed and evaluated, they tend to focus on modifying lifestyle factors such as smoking. This may reflect the fact that lifestyle issues are often easier to identify and treat, or it may be indicative of differences in the respective evidence bases; with evidence on tackling the wider social determinants being less apparent and less accessible to policy makers and practitioners. Therefore, what is needed is evidence about what can actually be done to tackle the social determinants of health and health inequalities specifically which interventions are effective and for whom.6 This requires evaluative studies of interventions that address the social determinants of health.3 7 The WHO Measurement and Evidence Knowledge Network for example, noted that it is vital to continue to develop evidence bases about tackling the social determinants of health and health inequalities.8 However, turning this need for better evidence about interventions around the social determinants of health into action requires the identification of what we already know in terms of the effects of interventions and also identifying areas where new studies are needed. This information could then be used to identify priorities for new research. It was in this context that the English Department of Health, Policy Research Programme, via the Public Health Research Consortium, commissioned this umbrella review. Umbrella reviews are an increasingly common way of identifying, appraising and synthesising systematic review evidence.9 12 In addition, umbrella reviews are able to present the overarching findings of such systematic reviews.13 This article therefore synthesises recent systematic reviews on the effects on health and health inequalities of interventions aimed at influencing the social determinants of health. Methods Systematic review methodology was used to locate and evaluate published and unpublished systematic reviews of interventions around the wider social determinants of health ( umbrella review). Search strategy Initially, the Centre for Reviews and Dissemination Wider Public Health database (a web-based database of systematic reviews of public health and related interventions) was manually searched. This consists of evidence from systematic reviews relevant to public health policy and practice and covers the period from 2000 to 2002. To supplement this, the Cochrane Database of Systematic Reviews and the Database of Abstracts of Reviews of Effects (both administrative and public databases) were searched electronically, whereas the Campbell Collaboration Database and the EPPI Centre database of health promotion and public health studies were manually searched from January 2002 to April 2007. Electronic searches of the Criminal Justice Abstracts database (2000 2007) were also undertaken (as it is not covered by any of these databases of systematic reviews). Bibliographies, reference lists and relevant websites were also searched. Experts were contacted and we hand searched four leading journals (American Journal of Public Health, American Journal of Preventive Medicine, Journal of Epidemiology and Community Health, Social Science and Medicine) from January 2002 to April 2007. Full search strategy is in web appendix 1. Inclusion and exclusion criteria

We used the widely cited Dahlgren and Whitehead rainbow model of the main determinants of health (figure 1) as a framework to help to identify the range of social determinants upon which interventions could be based.14 We concentrated on the outer two layers, which included macroeconomic, cultural and environmental conditions in the outermost layer; and living and working conditions and access to essential goods and services in the next layer, specifically water and sanitation, agriculture and food, access to health (and social care) services, unemployment (and welfare), work conditions, housing (and living environment), education and transport. We therefore excluded reviews that only examined interventions based on the inner most layers of the rainbow: individual lifestyle factors and social and community networks. Figure 1 Dahlgren and Whitehead's model of the social determinants of health. Only studies of adult participants (16+) or the general population in developed countries (North America, Europe, Australasia, Japan) were eligible for inclusion. We limited our study to adults because an Institute of Education team was conducting a concurrent umbrella review of child health outcomes.15 In terms of outcomes, we were particularly interested in the impacts on inequalities in health or wellbeing (by socioeconomic status), although we also looked at the overall health effect. We also considered as outcomes the non-health effects (such as employment or income) on people from a disadvantaged group with a pre-existing health condition. Systematic reviews had to meet the two mandatory criteria of Database of Abstracts of Reviews of Effects: (1) that there is a defined review question (with definition of at least two of, the interventions, participants, outcomes or study designs) and (2) that the search strategy included at last one named database, in conjunction with either reference checking, hand-searching, citation searching or contact with authors in the field. Data extraction Two reviewers (CB/MG) independently screened all titles and abstracts identified from the literature search for relevance (n=1694). Full paper manuscripts of any titles/abstracts that were considered relevant by either reviewer were obtained (n=84) and independently assessed for inclusion. Any discrepancies were resolved by consensus and, if necessary, a third reviewer (MP) was consulted. Only studies meeting all the inclusion criteria were data extracted (n=30). Results Thirty systematic reviews of interventions were identified. These are synthesised by domain type in tables 1 4 and in the text below.

Table 1 Summary details of housing and community reviews

Table 2 Summary details of work environment reviews

Table 3 Summary details of transport and access to health and social care services reviews

Table 4 Summary details of unemployment and welfare, agriculture and food, and water and sanitation reviews

Housing and living environment There is a housing evidence base that goes back many decades, including early evaluation studies from the 1930s and a number of controlled trials, and more recently several randomised controlled trials.16 Given this historical focus on the relationship between housing and health, it is probably not surprising that the systematic review housing evidence base is better developed than for other domains. We identified nine systematic reviews focussing on housing and health (table 1)17 24 two were of social changes (rental assistance programmes),17 18 five were of environmental changes to housing (eg, changes in lighting, or physical infrastructure, to reduce risk of falls, or injury)19 23 and two were of wider area-based initiatives.24 25

Reviews of rental assistance (eg, use of rent subsidies to create mixed-income or desegregated housing in poorer US neighbourhoods) suggested that interventions to promote mixed housing may result in increases in perceived neighbourhood safety, perhaps because exposure to crimes against person and property is reduced, along with neighbourhood social disorder. There is tentative systematic review evidence that such housing mobility policies (at least in the USA) do improve health and health behaviours, but the effects are small. Research on the mechanisms is lacking and therefore required. General housing improvement is also associated with positive change in social outcomes, including reductions in fear of crime and improvements in social participation. These interventions ranged from home visits, risk assessments and removal of hazards to reduce the risk of injury, to physical changes to housing structure such as insulation, furniture and more general housing policies. Although two reviews considered the effects on inequalities,22 25 none of the primary studies differentiated their results by socioeconomic status. Work environment There has been a recent shift in focus, from work as a source of occupational diseases to the wider impacts of work on health and well-being.5 This is reflected in the seven systematic reviews we located.26 32 They focussed on four types of intervention (table 2): increased employee control (via participatory health circle staff meetings to discuss ways to improve the work environment, more generic staff participation at work or task restructuring),26 28 changing the organisation of shift work (less nights, shorter shift lengths, etc, or the compressed working week),29 30 privatisation31 and the health and safety regulations.32 Overall, interventions to improve employee control (three reviews)26 28 found consistently positive health effects when job control was actually increased27 (and negative effects when job control decreased).28 The two reviews of changes to shift work29 30 identified some interventions (such as increased control over shift times) that had positive impacts on self-reported (particularly mental) health.30 Conversely, the privatisation review suggested that job insecurity and unemployment resulting from privatisation impacted adversely on mental health.31 The single review of increased health and safety legislation in the construction industry found a decrease in fall-related injuries after the intervention.32 Five of the reviews explicitly looked for evidence of effects on health inequalities and three included studies that reported differences by socioeconomic status (occupation).27 28 31 In one review of participatory interventions,27 one uncontrolled study found improvements in terms of mental health outcomes among manual workers but not managers or clerical employees. In another review of task restructuring,28 an uncontrolled study found that the adverse health effects of a team working intervention were only experienced by the lowest grade of employees. The review of privatisation also identified one study that found that 8 months after privatisation, occupational stress increased only among clerical and administrative staff, and not among manual workers or managers.31 Transport

Transport policies are often cited as a major influence on health and health inequalities, although it is a field where relatively few evaluative studies and reviews have been carried out (at least, ones measuring health outcomes). We located five reviews addressing transport issues (table 3).33 37 Each dealt with a different type of intervention: promoting modal shift from driving to walking and cycling; impacts of new roads; reductions in permissible alcohol when driving, area-wide traffic calming and speed cameras. Despite the differences in intervention type, four of the five reviews included outcomes related to road injuries.33 35 37 In relation to road injury outcomes, the review of legislative interventions to curb alcohol-impaired driving33 found strong evidence to support the reduction of fatal and non-fatal crashes, as did the reviews of traffic calming interventions35 and speed cameras.37 Evidence for the impact of new of road building on injuries36 was less conclusive because whereas out-of-town bypasses delivered reductions in injuries, major new roads did not. There was very limited evidence available on the health effects of interventions aiming to encourage modal transport shift from driving to walking and cycling.34 None of the reviews presented any information relating to impacts on health inequalities. Health and social care services Access to effective healthcare is another determinant of population health. Several different types of access are relevant to the wider social determinants of health, particularly geographic, economic and cultural access. We identified four reviews in the Organisation for Economic Co-operation and Development (OECD) countries (table 3), three of which focused on interventions to improve cultural access (acceptability and appropriateness of services)38 40 and one41 on improving geographic access (location and physical availability of health services) in rural areas. No reviews of economic access (affordability of services) were identified relating to high-income countries. Overall, the evidence evaluating interventions to promote culturally relevant healthcare was generally inconclusive. For example, although positive effects were found for lay health workers in promoting immunisation uptake, there was insufficient evidence to support the use of lay health workers in other contexts.39 Rural outreach interventions improved geographic access to care and self-reported health.41 The reviews all focused on interventions intended to improve access for disadvantaged groups (lowincome and minority populations) and there was some evidence that the interventions were effective for example, ethnic minority patient satisfaction with healthcare services increased after the cultural training intervention.38 However, none of the reviews reported whether impacts of interventions differed for different groups in the population studied. Unemployment and welfare There is considerable observational evidence on the linkages between unemployment and health, which suggests that ill health can be both a cause and a consequence of unemployment (the latter being the so-called direct health selection hypothesis).42 Two of the three reviews we located in this domain were of interventions that aimed to assist those who were prevented from entering the labour market by ill health for example, through supported employment, providing skills and training, and other

mechanisms.43 44 The other review evaluated interventions to increase the uptake of welfare entitlements.45 Details of the reviews are presented in table 4. One review found that although supported employment delivered more positive employment outcomes than prevocational training, there was no significant improvement in comparison with standard care.43 Furthermore, there was little evidence of any impact on health. Similarly, the review44 of welfare to work found some evidence of positive employment effects, although it was not clear to what extent this was due to the influence of contextual confounding factors. This review contained no information on health outcomes. The review of welfare rights interventions45 indicated that there were clear financial effects with a mean gain in income of 1026 per client in the year after the intervention (2004). However, the effects on health outcomes were limited to short-term improvements in mental health. None of the reviews specifically examined differential impacts across socioeconomic groups, although importantly all interventions were targeted at disadvantaged groups. Agriculture and food Agricultural policies affect the quality, quantity, price and availability of food, all of which are important for public health.46 Whereas overall increases in life expectancy may be partly attributed to better nutrition, increases in the prevalence of obesity in many countries point to the contribution food policies also make to overnutrition. Agriculture and food policies and interventions may therefore provide some of the mechanisms for addressing diet-related health inequalities. However, only one review was identified (table 4).47 This focussed on monetary incentives (including price decreases) on low-fat snacks, coupons for farmers' markets, financial rewards and free food provision. All four RCTs included in the review found a positive effect of incentives on the outcomes measured: weight loss, consumption of fruit and vegetables, redemption of coupons and attitudes towards fruit and vegetable consumption. None of the studies differentiated their results by socioeconomic position and none of the reviews focussed on disadvantaged groups. Water and sanitation There are many aspects of water and sanitation likely to impact on population health. Aside from the direct effects of pollution and contamination, other aspects of water management, including abstraction, water metering and the provision of flood defences, may all have potential public health implications. However, there are few available systematic reviews reporting health outcomes and only one that met our inclusion criteria (table 4).48 It focussed on changes in levels of water fluoridation and did not report on the effects on health inequalities. The authors concluded that fluoridation at levels up to 1 ppm had no adverse effects on bone fracture incidence, bone mineral density or bone strength in developed countries. Education There is undoubtedly a strong case for highlighting education as a major determinant of health and health inequalities not least though its interaction with other determinants. For example, Education has traditionally been an important route out of poverty for disadvantaged groups in many countries.

Generally, qualifications improve people's chances of getting a job and of having better pay prospects and the resulting increase in standard of living. This in turn improves opportunities to obtain the prerequisites for health nutritious food, safe housing, a good working environment and social participation. 14 However, perhaps surprisingly, we found no systematic reviews of the health effects of adult education interventions in OECD countries published in the current decade. It should be noted that person-based health education interventions aimed at social determinants in the two inner most layers of the rainbow were excluded from this review. Discussion This project aimed to identify the state of the systematic review evidence base in the current decade in developed countries, addressing the effects on health and health inequalities of interventions targeting the social determinants of health, as well as identifying fruitful areas for future research. The study therefore does what it aims to do, but this is of necessity a very limited answer to the problem of what works in terms of tackling health inequalities as, disappointingly, very few relevant reviews have been conducted. It has already been demonstrated elsewhere that the public health evidence base is sparsely populated,49 and this is particularly true in terms of evaluations of interventions addressing the social determinants of health, especially in relation to health inequalities. Evidence on the differential impacts of interventions by socioeconomic position is largely absent (only 3 of 30 reviews presented results for specified population subgroups), although this is likely to reflect the state of the primary study evidence base rather than that of the systematic reviews.22 25 29 30 34 What we do have however is suggestive evidence that certain categories of intervention may impact positively on inequalities or on the health of specific disadvantaged groups, particularly interventions in the fields of housing and the work environment. In the reviews of work environment interventions for example (such as changes to the organisation of work and privatisation), there is evidence that the effects of change are experienced differently by different levels of employee and that health outcomes differed accordingly. This suggests as noted by Marmot and others50 that the workplace may indeed be an important setting in which inequalities may be addressed. Similarly, there is suggestive evidence that housing change may positively affect physical and mental health, but the actual effects may be small. In the case of transport, the strongest evidence derives from studies of injury prevention, but the wider health impacts of transport policies on inequalities remain to be elucidated further. Given the importance of access to healthcare in potentially helping to reduce health inequalities, it was notable that there is still only limited evidence of effects on health and no direct evidence of impacts on inequalities in health. Similarly, the systematic review evidence base in regards to the other social determinant domains is very limited particularly in terms of the effects of interventions on health inequalities, and in the case of the unemployment and welfare domain on general health, too. We found no reviews on interventions relating to macroeconomic, cultural and environmental conditions (the outermost layer of the rainbow figure 1). These conditions influence the standard of living achieved by different sections of the population, the prevailing level of income inequality, unemployment, job security and so on. Interventions within this category would therefore be aimed at

altering the macroeconomic or cultural environment to reduce poverty and the wider adverse effects of inequality on society, including measures to ensure legal and human rights, healthier macroeconomic and labour market policies, the encouragement of cultural values promoting equal opportunities and environmental hazard control (including upholding international obligations and treaties in this field).51 This gap may be as a result of our focus on intervention studies and it may well be that the evidence base therefore needs to be widened to include reviews of comparative (non-intervention) studies such as those conducted within social epidemiology (such as that by Lynch et al52 on the association between income inequality and population health). Clearly, education is the starkest example of an area in which there can be further development. The reviews that do exist either date from pre-2000 or relate to developing countries. We located no reviews relating to education and adult health outcomes published in this current decade concerning the situation in the high-income countries of the OECD. There are therefore unanswered questions, ripe for review, concerning the relationships between levels of education in a society and/or the nature of educational systems and health outcomes, and how these health outcomes differ by socioeconomic position. Similarly, it was particularly difficult to identify appropriate reviews in the domain of access to health and social care as a social determinant of health. Despite extensive and rigorous searching, we only identified four systematic reviews that met our inclusion criteria. Moreover, the studies in the reviews do not represent the full range or intensity of potential intervention types in this domain. There is for example, a clear need for reviews of the effects of nationwide changes in health systems to improve geographic, economic or cultural access for the population as a whole and for groups in greater need in particular. In terms of the unemployment and welfare domain, there are still areas in need of further research, particularly in terms of the effects on health of welfare to work policies (eg, for lone parents, for the long-term unemployed, for young people), as well as the effects of interventions designed to prevent ill health among people out of work. Similarly, in the transport domain, the effects of policies to promote healthy transport (such as policies to promote walking) require further research.53 More studies are needed in terms of food policies (eg, the effects of the EU Common Agricultural Policy on food pricing and consumption); and in relation to water and sanitation interventions, the effects of water metering, which has been suggested may to lead to poorer families economising on water to the detriment of child health, is an important gap in the systematic review evidence. Limitations The main challenge was simply that there were too few systematic reviews conducted. It was also a challenge to locate the relevant systematic reviews that had been conducted. Searching for studies on the social determinants of health and/or health inequalities is difficult and time-consuming, and the searches can often suffer from a lack of sensitivity and a lack of specificity.54 55 However, to ensure the searches were as extensive as possible, our search strategies were piloted and revised. Furthermore, the searches were conducted by experienced specialist staff at the York Centre for Reviews and

Dissemination. In addition, leading public health journals were hand searched and review authors were contacted. Despite this, as for any review of complex and difficult-to-define social interventions, it is not possible to be sure that all reviews have been located.55 However, there is confidence that the gaps identified, although perhaps surprising, are real. Another important issue to consider with umbrella reviews is the risk of study overlap between the included systematic reviews. However, in keeping with previous public health policy umbrella reviews,9 we found very little overlap for example, in the work environment domain, there were no common studies. A more general limitation of public policy research is also relevant as a lot of the studies included in this umbrella review are from the USA, and there is evidence that the contextual determinants of health act differently in the USA than in Europe due in part to the different welfare systems in place.56 The findings of the USA studies may not therefore be easily transferrable to the European policy context. Conclusion It appears, then, that not only is the public health systematic review evidence base weak in terms of how to tackle the social determinants, but that there are specific areas that appear especially sparsely populated. These are sector-wide policies in education, the health system, food and agriculture, and more generally on the influence of macro-level policies on health inequalities. Although it is now a given that the effects of any interventions on inequalities should be assessed, the systematic review evidence base does not yet allow us to say with any confidence what the effects of interventions on reducing health inequalities are because differential impacts by socioeconomic position are rarely assessed. Nonetheless, one of the positive messages from this umbrella review is that there is a growing systematic review evidence base around housing and regeneration and a significant evidence base on the work environment suggesting that this is indeed a sector with significant responsibility for improving health and reducing inequalities. Given the few intervention studies that address inequalities, it is particularly important to assemble evidence on the mechanisms by which policies may affect health; this will help identify points at which to intervene and will provide a framework for the development of new research.51 For example, the results of systematic reviews that have evaluated the effects of interventions on the determinants of health (but which do not have health as an outcome) could also be examined and their findings extrapolated to tackling health inequalities. This is consistent with the WHO Commission on Social Determinants and the Measurement and Evidence Knowledge Network advice that as evidence comes in many shapes and forms, there is a need to get smarter about synthesising and appraising that evidence.8 What is already known on this topic
y y y

The importance of the social determinants of health inequalities is well established. Therefore, there is increasing pressure to tackle these wider social determinants of health, through the implementation of appropriate interventions. However, there is a lack of evidence about what can actually be done to tackle the social determinants of health and health inequalities.

What this study adds

y y y

This study synthesises recent systematic reviews on the effects on health and health inequalities of interventions aimed at influencing the social determinants of health. It thereby identifies what we already know in terms of the effects of interventions on health and health inequalities, and also where further work needs to be done. Evidence on the differential impacts of interventions by socioeconomic position is largely absent in the systematic review evidence base, although there is suggestive evidence that certain categories of intervention may impact positively on inequalities or on the health of specific disadvantaged groups, in particular, interventions in the fields of housing and the work environment.

Strong association of physical job demands with functional limitations among active people: a population-based study in North-eastern France Study design The initial sample consisted of everyone aged 15 years or more living in 8,000 randomly selected households in the Lorraine region of north-eastern France (2.3 million inhabitants). Only households with a telephone were eligible. Before the initial survey, a 3-month media campaign (television, print, and radio) was conducted in order to raise awareness. The investigation was approved by the Commission Nationale d Informatique et Liberts, and written informed consent was obtained from respondents. The study protocol included: (a) request of participation by means of a questionnaire to ascertain the number of people in the household, then (b) three sendings out standardized auto-questionnaires with a covering letter and a pre-paid envelope to reply at one-month interval. When the number of individuals was unknown, two questionnaires were sent first, and a complementary one was sent later. The standardized auto-questionnaires were completed by the subjects themselves. Measures The questionnaire included: birth date, sex, height, weight, educational level, occupation (coded according to the Insee classification, Paris, 1983), smoking habit, alcohol abuse, living alone, years of employment during the working life, physical job demands, and reported-functional limitations (according to the WHO international classification (Organisation Mondiale de la Sant 1988). Concerning the physical job demands, 12 items were selected, with the following question Please indicate the high job demands for your work : hammer, vibrating platform, pneumatic tools, other vibrating hand tools, handling objects, awkward posture, machine tools, pace of working, working on a production line, standing about and walking, heat, and cold (Yes/No) (Bourgkard et al. 2008; Chau et al. 2005a; Chau et al. 2008b; Lorhandicap group 2004). Those physical job demands showed satisfactory unidimensionality. Indeed, the principal component analysis showed that those job demands are unidimensional: the first eigenvalue (2.19) is much higher than the 2nd and the 3rd eigenvalues (0.56 and 0.30). The summary scale, called cumulated physical job demands (PJD) was defined as the product of

years of employment during the working life with the cumulative number of high physical job demands (range 0 to 12). The following categories of functional limitations were considered: 1. Physical with 20 items: Self-care (2 items: dress yourself, including typing shoelaces and doing buttons; shampoo your hair), arising (2 items: stand up from a straight chair; get in out of bed), Eating (3 items: cut your meat; lift a full cup or glass to your mouth; open a new milk carton), walking (2 items: walk outdoors on flat ground; climb up five steps), hygiene (3 items: wash and dry your body; take a tub bath; get on and off the toilet), reach (2 items: reach and get down a 2.5 kg object from just above your head; bend down to pick up clothing from the floor), grip (3 items: open car doors; open jars which have been previously opened; turn faucets on and off), and activities (3 items: run errands and shop; get in and out of a car; do chores such as vacuuming or yardwork); and 2. Cognitive functional limitations with 4 items: concentration/attention, orientation, problemsolving, and memory. The subjects were asked the following: Indicate the response which corresponds to your abilities during the 8 last days for the following activities . The response was: without difficulty / with some difficulty / with much difficulties / unable to do . For each category of functional limitation (physical or cognitive) a subject was considered with a functional limitation when he/she had responded with some difficulty , with much difficulties or unable to do for at least one of the items concerned. Seven occupational categories were considered: upper professionals (intellectual professionals, upper managerial staff and administrators, medical doctors, independent professionals, engineers), intermediary professionals (managerial staff, school teachers, skilled technicians, foremen, medical and social workers), manual workers (skilled manual workers, farm workers, semi-skilled manual workers and unskilled manual workers), clerks, farmers (farm managers), craftsmen/tradesmen (independent shop or business owners), and other employed people and unknown (Bhattacherjee et al. 2007; Bourgkard et al. 2008; INSEE 2001). Unemployed were categorized according to their last job, considered as the best reflection of their current situation. Educational level was categorized into primary school only versus secondary or above . Obesity was defined as body mass index 30 kg/m2. Alcohol abuse was determined using the DETA questionnaire (at least two positive responses to four items: (i) consumption considered excessive by the subject; (ii) consumption considered excessive by people around the subject, (iii) subject wishes to reduce consumption, and (iv) consumption on waking) (Bourgkard et al. 2008; Guilbert et al. 2001; Khlat et al. 2008). Sample Of the 8,000 households included in the sample, mailings to 193 (2%) were lost (due to addressing error or death). Of 7,807 households contacted, 3,460 (44.3%) participated (all eligible members of the family took part in 86% of those). In total, 6,234 subjects completed a questionnaire; 18 were of unknown sex or age, leaving 6,216 subjects. The distributions of the sample gathered according to age and sex are close with those of the Lorraine population (INSEE 1993) (Table 1). The present study focused on the

subpopulation of 3,368 economically active (either employed or looking for work) subjects aged between 18 and 64 years. Table 1 Distribution according to sex and age of the sample studied and of the general population of Lorraine (INSEE 1993) (%)

Statistical analysis The outcome variables were physical and cognitive functional limitations. This study examined the associations between those dependent variables with physical job demands, and its change when adjusting for the risk factors: age, sex, obesity, smoking, alcohol abuse, low educational level, living alone, and job category. The age was categorized into 5 groups: 18 29, 30 39, 40 49, 50 59, and 60. For the job category, 7 groups were considered: upper professionals; intermediate professionals; manual workers; clerks; farmers; craftsmen and tradesmen; and other professionals and unknown category. First, the associations between each factor and each type of functional limitation was assessed via crude odds ratios (OR) and 95% confidence intervals (95% CI). For the job demands, the PJD score was divided into four categories: 0, 1 29, 30 99, and 100. Two rounds of logistic regression analyses were carried out: We first examined the association between PJD and each type of functional limitation, adjusting for age and sex (model 1). Next, we added in the logistic model obesity, smoking, alcohol abuse, low educational level, living alone, and job category (model 2). The extent of the difference between the estimates arising from the two analyses was interpreted as reflecting the role of those factors in explaining the relationship between the functional limitation and PJD. Results The characteristics of the study sample are in Table 2. The sample included 3,368 subjects (1,799 men and 1,569 women), who belonged to the following groups: upper professionals 13.4%, intermediary professionals 9.4%, manual workers 26.5%, clerks 34.8%, farmers 1.8%, craftsmen and tradesmen 2.1%, and others or unknown 12.0%. The people with low educational level (primary school) represented 18.4%, the obese subjects 6.4%, the current smokers 34.2%, the subjects having reported alcohol abuse 9.0%, and those living alone 9.0%.

Table 2 Relationships between various factors and functional limitations: % and crude odds ratios (OR) and 95% confidence intervals (3,368 subjects)

The subjects suffering from physical functional limitation represented 16.9% and those suffering from cognitive functional limitation 28.6%; 6.6% had physical disability only, 18.4% had cognitive disability only, and 10.2% had both physical and cognitive disabilities. Cumulating several functional limitations were also common: 5.3%, 3.0% and 6.4% of subjects had 1, 2, and 3 or over physical functional limitations, respectively; and 13.0%, 8.1% and 7.6% of subjects had 1, 2, and 3 or over cognitive functional limitations, respectively. Expectedly all three types of functional limitations were strongly related to age. Men suffered less than women from all those functional limitations. Obesity was associated with physical functional limitation. Alcohol abuse and low educational level were associated with both physical and cognitive functional limitations. Living alone was linked with cognitive functional limitation only. Compared to upper professionals, a higher frequency for both physical and cognitive functional limitations was found for all categories of workers except for physical functional limitation among intermediate professionals. Every job demand considered affected between 1.5% and 27.3% of subjects. Nearly all job demands were related to cognitive functional limitation (crude ORs between 1.21 and 2.74). Only using vibrating hand tools, handling objects, cold environment, awkward posture, working on a production line, and standing about and walking were significantly related to physical functional limitation (crude ORs between 1.35 and 1.94). The results obtained with the principal component analysis shows that these job demands are unidimensional and thus validate the calculation of cumulated job demands (PJD). Note that 33.1% of subjects had PJD1-29, 14.7% PJD30-99, and 3.3% PJD 100. The results of two rounds of analyses are presented in Table 3. The first round estimated age and sexadjusted odds-ratios for various PJD levels, using the PJD0 as the reference, and the second round provided odds ratios adjusted for the same variables, with in addition obesity, smoking, alcohol consumption, educational level, living alone and job category as covariates, in order to investigate the role of those factors in explaining the associations between PJD and each category of functional limitation. We found a dose-response association of PJD with physical functional limitation (ORa adjusted for age and sex 1.39, 1.80 and 2.98 for PJD1-29, PJD30-99, and PJD 100, respectively, vs. PJD0) as well as with cognitive functional limitation (ORa adjusted for age and sex 1.32, 1.82 and 2.55 for PJD129, PJD30-99, and PJD 100, respectively, vs. PJD0). We failed to find the individual factors studied in mediating those associations. Among the covariates obesity had significant ORa for physical functional limitation only whereas alcohol abuse and low educational level had significant ORa for both physical and cognitive functional limitations. Living alone had a significant ORa for cognitive functional limitation only. Regarding job category, the ORa were markedly lower than the crude ORs when controlling for all

covariates, especially for manual workers, farmers, craftsmen and tradesmen, and also but less for clerks. Table 3 Relationships between cumulated physical job demands (PJD) and functional limitations: adjusted odds ratios (ORa) and 95% confidence intervals (3,368 subjects)

The relationships between PJD and those functional limitations were clearly stronger when we considered cumulating several functional limitations. The ORa adjusted for all factors studied for PJD 100, vs. PJD0, was 3.34 for 3 or more physical functional limitations and 2.93 for 3 or more cognitive functional limitations (Table 4). Table 4 Relationships between cumulated physical job demands (PJD) and functional limitation: adjusted odds ratiosa (ORa) and 95% confidence intervals (3,368 subjects)

Discussion The present study demonstrates that the physical and cognitive functional limitations were common and that they were strongly related to cumulated physical job demands during the working life. Smoking, alcohol abuse and obesity were included in the analysis to assess their potential mediating role in the association, and education, living alone and job category were included as potential confounders. We found a strong relationships between the cumulated job demands and both physical and cognitive functional limitations, and those were not or very slightly mediated by individual factors such as age, sex, obesity, smoking, alcohol abuse, and neither were they confounded by low educational level, living alone and job category. The selection bias of the sample would be small. Indeed, the households possessing a telephone represented 96%, and those having confidential addresses represented only 16%. According to our discussions before the survey with several associations for persons with disability, this list is not likely to be related to health status or life conditions. The proportion of subjects who participated among those contacted was similar with that reached for surveys with mailed questionnaire in France (Alonso et al. 2004; Lorhandicap group 2000). The distributions according to age and sex of the sample are close with those of the Lorraine population. The quality of the filling in of the questionnaire was very good (nonresponses for various items <4%). Although this study was conducted on a large sample, the interpretation of the results needs some caution due to the presence of a possible selection bias and the use of an auto-questionnaire. However, self-administered occupational health history questionnaire is

reliable and valid (Lewis et al. 2002). The majority of studies rely on self-report as a measure of disability (Newman and Brach 2001). The non-response bias in mailed health surveys is small (Etter and Pernejer 1997; Kant et al. 2003). The use of a self-administered questionnaire based on self-reported responses would be appropriate for assessing functional limitations as consequences of diseases or ageing (OMS 1988). As previously mentioned, all factors studied were validated and used in other studies (Bourgkard et al. 2008; Chau et al. 2005a; Chau et al. 2008b; Guilbert et al. 2006; Guilbert et al. 2001; Khlat et al. 2008). The physical job demands were those reported as highly demanding by the subjects, and have been used elsewhere (Bourgkard et al. 2008; Chau et al. 2005a; Chau et al. 2008b; Lorhandicap group 2004). The calculation of the score PJD was valid because the physical job demands considered were unidimensional The occupational exposure was therefore underestimated because moderate occupational hazards were excluded. Furthermore, many hazards, for example chemical hazards, dust, etc. were not considered (Teschke et al. 2002). It should be noted that most subjects generally have one main job during their working life, so they should be aware of their job demands. Given the large number of statistical tests carried out, type I error may be a concern, but it has to be pointed out that most tests were significant at the 1% level, with very large odds ratios estimates. We found that the ORa adjusted for PJD and other factors were markedly lower than the crude ORs for various job groups (except for intermediate professionals), especially for manual workers, for both physical and cognitive functional limitations. The same pattern was found for educational level and, to a lesser extent for living alone for cognitive functional limitation. This finding was expected because the PJD concerned most jobs, but more particularly manual workers and the least educated. It should be noted that people with lower levels of occupational standing, and particularly manual workers, are less educated and have fewer material resources, and that this profile is associated with poorer health due to the co-occurrence of many factors: unfavourable health behaviors, poor nutrition, occupational hazards, altered living conditions, low income, low educational level, lack of physical activity, barriers to health care (Baumann et al. 2007; Dahlgren and Whitehead 2006; Lochner et al. 2001; Marchand et al. 2003; Mejer 2004). In this sense, our findings illustrate the role of lifestyle factors and adverse working conditions in explaining social inequalities in health (Huisman et al. 2008; Sekine et al. 2006; Warren et al. 2004). Our study found strong relationships between PJD and both physical and cognitive functional limitations, and those associations remained when controlling for sex, age, obesity, smoking, alcohol abuse, educational level, living alone and job category. The physical job demands considered were found to be related to premature mortality (Bourgkard et al. 2008) and also smoking, fatigue, sleep disorders, anxiety, occupational injury, certain diseases (musculoskeletal disorders, cardiovascular disease, for example), physical, sensorial and intellectual impairments (Chau et al. 2005a; Chau et al. 2005b; Chau et al. 2008b; Chau et al. 2008c; Lorhandicap group 2004), which may favour both physical and cognitive functional limitations. Kristal-Boneh et al. (2000) found a hazard ratio of 1.82 of all-cause mortality in workers with a high physical workload compared with the others. We found that the higher prevalence associated with PJD is slightly confounded by smoking, alcohol abuse, low educational level, and job category. These results are consistent with the findings of Diderichsen et al. (1997) who stated the high total burden of diseases of tobacco and alcohol in the European Union (respectively 9.0% and 8.4%;

3.6% for work environment). As already mentioned, job demands could favour drug use that has a great role in disease burden (Dahlgren and Whitehead 2006). In France, one-third of the working population smokes and uses medications or other legal psychoactive substances in order to cope with work-related difficulties, and such use is more common in manual workers (Chau et al. 2008c; Lapeyre-Mestre et al. 2004). Intake of drugs is a leading cause of injuries (Chau et al. 2002; Chau et al. 2005a; Chau et al. 2008b; Gauchard et al. 2003), and smoking and alcohol abuse are common and alter physical and mental abilities of workers, increasing in this way risk of injuries due to job demands (Chau et al. 2002; Chau et al. 2004; Chau et al. 2005a; Chau et al. 2008b; Gauchard et al. 2003). In the present study, the associations found of pneumatic tool use and vibrating platform with cognitive functional limitation and not with physical functional limitation are un bit surprising. In fact, we found relationships of pneumatic tool use with sleep disorders (14.5% vs. 7.6%, p=0.028) and nervousness (18.4 vs. 11.9, p=0.083, close to significance), and also of vibrating platform exposure with chronic pain (15.6% vs. 79.1%, p=0.014) and fatigue (17.9% vs. 12.5%, p=0.078, close to significance) (yet not-published data). The associations between those job demands to cognitive disability may be partly explained by those disorders. The absence of association between these job demands and physical functional limitation may suggest that workers with physical disability would avoid such hazards. It should be noted that PJD 100 was strongly associated with cumulating several functional limitations. Indeed, the ORa (adjusted for all factors studied, vs. PJD 0), was 2.57 for at least one physical functional limitation and reached 3.34 for 3 or more physical functional limitations. It was 2.00 for at least one cognitive functional limitation and reached 2.93 for 3 or more cognitive functional limitations. The subjects with high PJD appeared thus to have a higher prevalence for various types of functional limitations. The gender difference in functional limitations is well known. Indeed, women live longer than men, but men have fewer disabilities than do women (Newman and Brach 2001). A study in Spain showed that men lived more time free of disability than women (Gispert et al. 2007). Our study shows that the gender gap was for both physical and cognitive disabilities, and it was not confounded by the factors studied (when comparing crude and adjusted ORs). Some of the explanations for the disability gap have included reporting bias, higher rates of co-morbidity or chronic health problems, possible physiologic differences, and behavioral factors that could leave women more susceptible to disability than men are (Newman and Brach 2001). The prevalence of pain conditions is higher among women, women report more severe pain, and co-morbidity between pain conditions and psychosomatic problems is higher among women (Baumann et al. 2007; Bingefors and Isacson 2004). Depression is more common in women (Baumann et al. 2007), and sex moderates the relationship between depression and disability, in that when depression is high, women report greater disability than men (Keogh et al. 2006). But it should be noted that there is a higher premature mortality among men than among women which could result in less men surviving with functional limitations (Bourgkard et al. 2008; Newman and Brach 2001). Our study sheds light on the strong associations between physical job demands and both physical and cognitive functional limitations in North-eastern France. We also provide evidence for the association of personal factors such as obesity, alcohol abuse, low educational level and living alone with an elevated prevalence of those functional limitations, and demonstrate that the contribution of those factors to the

PJD disparities in functional limitations is quite limited. Our findings need to be confirmed by other studies on other populations. They may however suggest that intervention policies to prevent functional limitations and limitations of daily living activities, especially at work, should focus on job demands, but proper attention to lifestyle factors should be included to prevent a whole range of physical and cognitive functional limitations.

Association of Workplace Chronic and Acute Stressors with Employee Weight Status: Data From Worksites in Turmoil
Substantial research now exists relating to the psychosocial characteristics of the work environment, as opposed to the physical or chemical work environment, with health outcomes. Pressures and demands at work have been associated with cardiovascular disease morbidity and mortality,1-7 the metabolic syndrome and its components, 8 obesity,9 stroke exhaustion,10 depression and anxiety,11 and selfreported poor health.12 In addition, there is some evidence that these conditions have negative effects on leisure time physical activity,13-14 eating habits,13,15 and the co-occurrence of adverse health behaviors.16 It has been hypothesized that psychosocial working conditions can affect health either directly by its effect on the immune system, biological, and hormonal pathways or indirectly by influencing behaviors that are in the intermediate causal pathway between the psychosocial work environment and health outcomes.9,17 The job demand-control model11,18-19 describes the psychosocial work environment and is perhaps the model that has been used more frequently to examine the effects of working conditions on health across different work settings and countries. The demand-control model was developed for work environments where stressors are chronic and are the product of human organizational decision making. Additionally, the model proposes that the psychosocial job experience promotes new patterns of behaviors and skills over a lifecourse of work experience that have ultimately had an impact on a variety of conditions of adult life. Although there is extensive literature on worksite chronic stressors and health, not much is known about the health effects of acute stressors in the workplace. Although downsizing has become a common feature in American workplaces, little has been said on how downsizing affects remaining employees work-experience and the health of downsizing survivors (employees who have retained their jobs while their companies implemented major lay-offs). It has been observed that downsizing survivors are at higher risk of sickness absence, CVD mortality,20 and self reported morbidity.21 To our knowledge, there is only one article examining the effect of job insecurity on 5-year change in BMI.22 In this paper, we analyze baseline data from a worksite randomized control trial for weight gain prevention, Images of a Healthy Worksite . This study tested interventions attempting to create synergy between the worksite environment and workers food and physical activity choices. The study is being conducted in a manufacturing facility with multiple sites in upstate New York. At the time the study started, the company was undergoing a drastic re-structuring with massive layoffs and building closings. Employees interviewed during the formative research period prior to baseline assessments expressed on-going layoffs that left those remaining doing the work of 5 people, stress eating,23 and

feeling like vegging out instead of engaging in leisure time physical activity (unpublished data). Thus, at the time of baseline assessment, in addition to chronic stressors, the employees were under the effect of acute stressors as well. Given that the primary outcome of the trial is the prevention of weight gain in the population of employees, we consider it necessary to explore whether working conditions, long standing as well as new ones, may influence the main outcome and might have to be considered in the evaluation of the intervention effect as potential confounders or effect modifiers. The goal of this paper is to present the baseline characteristics of the sample of employees and to examine the baseline association between chronic and acute stressors and employee weight status. In addition, we report on the intraclass correlation coefficient (ICC) for body mass index (BMI) overall and by gender in our worksites to aid other investigators in determining the adequate sample size for grouprandomized trials among worksites with characteristics similar to ours. Specifically, our hypotheses are: 1) high job strain, as chronic work stressors based on the demand-control model, and job insecurity, as acute work stressor, are independently and synergistically associated with employee weight status; and 2) the association of high job strain and job insecurity with weight status are at least partly mediated by the measures of physical activity, sedentary behavior, and diet quality. Methods Study Sample The Images of a Healthy Worksite study is a group-randomized control trial that assigned 6 pairs of worksites within a single corporation in upstate New York to a nutritional and physical activity intervention and to delayed intervention. The 2-year intervention consisted in a comprehensive nutrition and physical activity strategy based on participatory research involving the worksite environment (e.g., reduced caloric density in cafeteria meals, half portion offerings, walking routes) and the individual employees (e.g., nutrition education workshops, stress reduction strategies). Employee advisory boards in each intervention worksite provided input in intervention design. Worksites were matched by type of job (white or blue collar) and presence or absence of a cafeteria in the building. We collected baseline data in a cross-sectional sample of employees and we are currently collecting postintervention data on another cross-sectional sample of employees. Since the study lost a pair of worksites due to building closings, we are reporting the analysis of baseline data from the 5 remaining pairs plus an additional worksite that closed after the collection of baseline measurements (11 worksites). More details of the study have been published elsewhere.24 All employees in the 11 worksites were eligible to participate in assessments. We collected baseline measurements in 2,782 (71% of our targeted enrollment). Employees were recruited for baseline assessments through e-mails, flyers, and presentations at team meetings. Following informed consent, project staff took anthropometric measures, and distributed a survey with a self-addressed envelope for employees to complete on their own time. Conceptual Framework Figure 1 represents our hypothesized causal pathway between working conditions and weight status. According to this pathway, both acute and chronic stressors are related to weight status either directly

or indirectly through health behaviors. Directly, adverse psychosocial factors in the workplace may act on employee weight status through the effects of stress and its effects on the neuroendocrine system resulting in weight gain and abdominal fat accumulation.17,25 For example, stress may cause an increase in adrenal corticoid and therefore accumulation of abdominal fat as is seen in patients with Cushing s syndrome.25 Also, stress may decrease sex hormones leading to weight gain and abdominal obesity as it is observed among menopausal women.25 Indirectly, adverse working conditions may affect weight status through unhealthy eating behaviors such as the consumption of fatty and sweet foods or not engaging in leisure time physical activity due to long work hours.26-27 Figure 1 Hypothesized conceptual framework of the causal pathways linking working conditions with employee weight status. Outcome Measures We examined two outcome measures related to weight status: Body mass index (BMI) and proportion of overweight/obese employees. Body weight was measured using Tanita BWB 800S scale without shoes and wearing light clothing and height was measured without shoes using Shorr Infant/Child/Adult Height/ Length Measuring Board stadiometer. Body mass index (BMI) (kg/m2) was then calculated. Overweight/obesity was defined as BMI greater than 24.9 and healthy weight/underweight was defined as a BMI equal or less than 24.9.28 Main Predictors Assessment of chronic and acute stressors at work was based on a modified Job Content Questionnaire.29 Chronic work stressors were defined following the demand-control model. The Demand/Control model,11,18-19 describes two dimensions of the psychosocial work environment: the psychological demands of work and decision latitude, which is a combined measure of task control and skill use. The psychological demand dimension relates to pressures and heavy demands ( how hard one works ); the decision latitude dimension refers to the worker s ability to control his or her own activities and skill usage (who makes the decisions? who does what? ). The decision latitude scale has two components: task authority employee s control over decision making relevant to his/her work tasks and skill discretion employee s control over the use and development of his/her skills. When the psychological demand and the decision latitude dimensions are cross-tabulated, four quadrants of the psychosocial work environment are defined (figure 2). The high demand-low decision latitude quadrant, high job strain, is considered the most related to illness. The combination of high psychological demands and high decision latitude is defined as the active quadrant in which employees can cope with high demands because they can make more relevant decisions (this situation corresponds to psychological growth). The low-demand-high decision latitude quadrant, low job strain, is the ideal condition while the low-demand-low decision latitude quadrant, passive, seems to be associated with risk of loss of skills and psychological atrophy.19 Decision latitude was assessed by eight questions and psychological demands were assessed with 5 questions from the Job Content Questionnaire (JCQ).29 To create high

and low variables for the 4 quadrants of the demand-control model, the scores for the two dimensions (demands and latitude) were divided at the median point of the overall sample distribution. Cronbach s alpha for the decision latitude scale for men and women were 0.83 and 0.80 and for the psychological demands were 0.63 and 0.62, respectively.29 Figure 2 Psychological demand decision latitude model. Source: Karasek

Acute work stressors were operationalized by the three-question job insecurity scale from the JCQ: five options for how steady is your job? (regular and steady to both seasonal and frequent lay-offs); four options for my job security is good (strongly disagree to strongly agree); and 4 options for how likely is it that in the next couple of years you will lose your job (not at all likely to very likely). The score ranges from 3 to 13, the higher the score, the higher the job insecurity. Cronbach s alpha for job insecurity was 0.53 and 0.41.29 Potential Confounders To assess the independent association between high job strain and job insecurity with the outcomes we included the following potential confounders: age in years, gender, race (White, African American, Other), income ( $29,999, $30,000-$59,999, $60,000), education (secondary or less, undergraduate, graduate), and smoking status (never smoked, current, ex-smoker). Potential Intermediate Variables Three potential intermediate variables of the association between chronic and acute stressors and weight status were included in the model (figure 1): physical activity, sedentary behavior, and diet quality. Physical activity was operationalized using the Godin questionnaire that provides a total leisure time physical activity score.30 The questionnaire assesses how many times, on average, an employee has performed strenuous, moderate, and mild physical activity for more than 10 minutes during free time. The Pearson correlation between the questionnaire and objectives measures of physical condition was 0.24 (p<0.001). Two-week test-retest reliability coefficients were respectively 0.84, 0.46, and 0.48 for strenuous, moderate, and light leisure time physical activity.30 We combined strenuous, moderate, and light leisure time physical activity to obtain a total Godin score. Sedentary behavior was operationalized as the number of hours a day the employees spend watching TV (TV viewing: < 2 hrs/day, 2-3 hrs/day, 4 hrs/day). It has been proposed that TV viewing contributes to weight gain by providing passive entertainment through television and, therefore, making physical activity less attractive.31 Hours watching TV has been found to be negatively associated with cardiorespiratory fitness and moderateintensity and hard-intensity physical activity and positively associated with light-intensity physical activity and BMI among adults.32 Diet quality was operationalized as the number of servings of fruits and vegetables per day using a 7-question food frequency questionnaire (FFQ). This FFQ asks about the

frequency and types of fruits and vegetables consumed in the past month fruit juices, fruits, salad, fried and not fried potatoes, and other vegetables. French fries are excluded from the calculation of the number of servings of fruit and vegetables. This short FFQ is an abbreviated version of a larger FFQ instrument.33 It was assumed that the more servings of fruits and vegetables the better quality of the overall diet. There is evidence that healthy eating (as defined by various scoring methods) helps maintain a healthy weight34 since, for example, a liberal intake of fruits and vegetables as well as wholegrain breads and cereals are markers for nutrient-dense diets providing adequate levels of dietary fiber.35 Statistical Analysis We described the baseline characteristics by intervention and control status and the statistical comparisons of means and proportions. To test the hypotheses, we modeled the relationship of BMI with the quadrants of psychosocial work environment, job insecurity and other covariates by treating BMI as a continuous variable and as a two level categorical variable separately. When fitting models with BMI as a continuous variable, we used the linear mixed-effects models by considering the worksites as a random effect. When fitting models with BMI as a two-level categorical variable, we used the generalized linear mixed-effects models (GLIMMIX procedure in SAS) with a logit link function. We first modeled the relationship between BMI and psychosocial work environment, job insecurity and other covariates by treating BMI as a continuous variable. Psychosocial work environment was modeled as a 3-level variable (high job strain, low job strain, active, and passive as the reference category). We fit a mixed-effects model for the data with the worksites as the random effects. We modeled the relationship of BMI with quadrant of the demand-control model and job insecurity separately in model 1A and model 1B, respectively, and then we fitted them together jointly in model 2. We added other potential confounders such as age, gender, race, income, smoking, and education with quadrants of the demand-control model and job insecurity in model 3. In the last model, we removed the insignificant terms from model 3 and added three potential intermediate covariates in model 4. We also tested for interactions between the quadrants of the demand-control model and job insecurity. To further examine the effect of psychosocial work condition and job insecurity on employees weight status, we also considered the model where BMI is a two-level categorical variable (obese/overweight versus healthy weight/underweight). We then modeled the logit of falling into the obese/overweight group given the quadrants of the demand-control model, job insecurity, and other potential confounder and intermediate variables. We followed the same model building procedure as in the continuous outcome case. We fitted a generalized linear mixed-effects model with a logit link function to the data with the worksites as the random effects. This model estimates odds ratios (OR) with the associated confidence intervals. We modeled the relationship of BMI with quadrant and job insecurity separately in model 1A and model 1B and then jointly in model 2. We added the same potential confounders with the quadrants of the demand-control model and job insecurity in model 3. In the last model, we removed the insignificant terms from model 3 and added the three potential intermediate variables in model 4.

The ICC is the fraction of the total variation in the data that is attributable to the unit of assignment. In this case, we are reporting the fraction of the total variation in BMI that is attributable to the clustering of employees in the same worksite in comparison with employees of different worksites. The ICC overall was computed using the linear mixed-effects model with the worksites as the random effects. ICC by gender was computed similarly within the same gender group. All the models were fitted using SAS 9.1 windows version (Cary, NC). The study was approved by the University of Rochester Research Subjects Review Board. Results Descriptive Statistics (table 1) The intervention and control groups are well balanced with respect to relevant characteristics with the exception of gender, the psychosocial work environment, and the educational level of the employees. This is a sample of middle aged employees (mean age of approximately 47 years), mostly white (>90%), mostly male (56% and 69% for control and intervention, respectively), married (more than a quarter of the employees), highly educated (76% and 81% of control and intervention employees, respectively, have undergraduate education or more) and relatively well-paid (67% and 72% of control and intervention employees, respectively, earn more than $60,000 a year), who have been working in the company for an average of almost 22 years. The employees weight status is in the overweight category (mean BMI of almost 29 for both groups) and the prevalence of overweight/obesity is 72.1% and 75.6% for control and intervention groups, respectively. Most of the employees work on low strain jobs (36% of the control and 32% of the intervention groups). The job insecurity score is approximately 7 for both groups with a range of 3 to 13 (data not shown). The mean Godin score (total leisure time physical activity) for both groups is 34.7 with a range of 0 to 169 (data not shown) and more than 65% of the employees watch 2 or more hours of TV per day. Table 1 Images of a Healthy Worksite. Baseline Descriptive Statistics (N=11; n=2782)

Baseline Association between Chronic and Acute Stressors and BMI (table 2) High job strain was the only quadrant of the demand-control model independently and positively associated with BMI in the crude analysis (model 1A) while job security was not (model 1B). Employees working in a high job strain environment have almost 1 BMI unit more than those working in a passive one ( =0.88, SE=0.35). These associations hold when fitting the two predictors together in model 2 and controlling for potential confounders in model 3. In addition, age, African American race, and income between $30,000 and $59,999 are positively associated with BMI while employees of other race, and females have a lower BMI than their reference values controlling for all other variables in the model.

The effect of high job strain on BMI did not hold once the potential intermediate variables were included (model 4). Of the hypothesized intermediate variables, Godin score and TV-viewing were found to be associated with the outcome in the expected direction. For each unit change in Godin score, BMI decreases by 0.02 units ( =-0.02, SE=0.00) and employees who watch between 2-3 hours and equal or more than 4 hours of TV per day have a BMI 2.37 and 1.35 units larger than employees who watch less than 2 hours a day, respectively. Number of servings of fruits and vegetables per day had no statistical significant effect on BMI ( =-0.22, SE=0.06) although the effect is in the expected direction. The interaction term between the quadrants of the demand-control model and job insecurity are not statistically significant and therefore they were dropped from the models (data not shown). Table 2 Images of a Healthy Worksite. Baseline association between chronic and acute stressors with BMI (N=11; n=2782)

Baseline Association between Chronic and Acute Stressors and the odds of overweight/obesity (table 3) High and low job strain (OR=1.49, CI=1.13,196; OR=1.30, CI=1.01,16.7, respectively) and job insecurity (OR=1.06, CI=1.01,1.12) increased the odds of being overweight/obese in the crude analysis (models 1A and1B). The effect of job insecurity and low job strain did not hold in any of the subsequent models. High job strain, however, remained independently and positively associated with overweight/obesity in model 2 (OR=1.44,CI=1.09,1.91) and its effect did not appear to be confounded by the variables in model 3 (OR=1.45, CI=1.09,1.95). In the latter model, the active quadrant of the demand-control model became a significant predictor of the odds of overweight/obesity (OR=1.45, CI=1.09,1.94) compared to the passive quadrant. Age and African American race significantly increased while Other race and graduate education significantly decreased the odds of overweight/obesity in this sample. The inclusion of the hypothesized intermediate variables in model 4 removed the effect of high job strain while holding the effect of the active quadrant (OR=1.50, CI=1.12,2.03). In this model, Godin score decreased (OR=0.99, CI=0.97,0.99) and TV-viewing increased the odds of overweigh/obesity. Employees who report watching 2-3 hours a day of TV increased their odds of being overweight/obese by 77% while watching more than 4 hours a day increased the odds by 150% compared to those who watch less than 2 hours. Servings of fruit and vegetables per day was not associated with the outcome. Also, in these models, there was not interaction between the quadrants of the demand-control model and job insecurity (data not shown).

Table 3 Images of a Healthy Worksite. Baseline association between chronic and acute stressors with the odds of being obese/overweight1 (N=11; n=2782)

ICC for BMI In our sample, the ICC for BMI overall is 0.0195. Once stratified by gender, the ICC for BMI for males is 0.0193, ICC for BMI for females is 0.0346. Discussion This cross-sectional study based on the baseline assessment of employees participating in a grouprandomized trial for weight gain prevention suggests that high job strain, as a construct for chronic stressors in the work environment, and not job insecurity, as an acute stressor, is associated with weight status. Furthermore, there is no synergistic effect of chronic and acute stressors on the outcome. In this sample of employees it appears that high job strain has an effect on weight status through the adoption of unhealthy behaviors such as not engaging in leisure time physical activity and practicing sedentary behaviors (e.g., watching TV) since the effect of job strain all but disappears once the behavioral variables are added to the models. Diet quality, as operationalized by the intake of fruit and vegetables, does not seem to mediate the effect of a high job strain environment on weight status. These results are fairly consistent regardless of the way weight status was modeled. Unexpectedly, employees in an active work environment are more likely to be overweight or obese and its effect does not seem to be mediated by any of the postulated intermediate variables. A fair comparison of the results of the current study with previous literature is difficult since there are differences in the way the psychosocial work environment is defined, the confounders included in the analyses, and the measurement of weight status (BMI, overweight, weight gain). In a recent review9 of 18 investigations examining the components of the demand-control model and body weight published between 1989 and 2006, 15 examined the effect of high job strain, of which 12 studies were crosssectional and 3 were prospective. According to this review, the evidence is inconsistent since in 3 of the studies the association between high job strain and a measure of body weight was fully confirmed, in one was partially confirmed, and in 10 there was no association. In a subsequent comparative study of pooled cross sectional data from three worksite cohorts in Britain, Finland, and Japan, the high strain, passive, and active quadrants of the demand-control model were not associated with obesity36 when compared with the low strain quadrant neither in the unadjusted nor in the fully adjusted models. On the contrary, in the prospective Whitehall II study,37 repeated measures of work stress over a period of 19 years showed a dose response relationship with incident overall obesity (BMI >=30kg/m2) and central obesity once adjusted for a variety of health behaviors (alcohol consumption, smoking, daily intake of fruit and vegetables, fiber consumption, exercise). Work stress, however, is defined as high job strain and lack of work social support and, thus, it is not fully comparable to the effect of job strain alone as

examined in the study presented here. The inconsistency of the relationship between job strain and body weight might be explained by the findings of two prospective studies included in the previous review.9 These two studies among British civil servants38 and Danish men22 indicated that the longitudinal effects of job strain on weight differ according to initial BMI. Whereas high job strain predicted weight gain among overweight or obese employees at baseline, high job strain predicted weight loss among the leanest ones. Other potential explanations for the conflicting results is that different population of workers may have diverging coping strategies allowing them to handle high job strain in different ways. For example, age, social support in and outside the workplace, work-home interface,39 and the existence of a social safety net may influence the way workers respond to a stressful work environment. In addition, different individual responses to stress through eating and physical activity have been reported in the literature26,40-41 and may explain the lack of consistency across populations and study designs. Our study adds more evidence of a lack of cross-sectional association between high job strain and body weight. Although the magnitude of the association was relatively strong after adjusting for confounders, measures of leisure time physical activity and sedentary behavior remove the effect of job strain on weight status. High job strain was associated with low leisure time physical activity among men in London and women in Helsinki14,36 but was not related to measures of exercise among workers in Minnesota.42 To the best of our knowledge, there are no studies of the effect of high job strain on TV viewing as a measure of sedentary behavior. There is, however, evidence of an association of hours of TV viewing with weight gain31 and overweight and obesity35 among adults. Thus, further investigations of sedentary behaviors as measured by TV-viewing and high job strain are needed to confirm the mediating role of the latter behavior with weight status. We hypothesized that physical activity, sedentary behaviors, and the number of servings of fruit and vegetables per day as a proxy for diet quality would be one of the mechanisms for which high job strain would have an effect on measures of body weight. The hypotheses were partially confirmed since our measures of physical activity and sedentary behaviors were associated with weight status in our sample and modify the magnitude of the association between job strain and weight status. In contrast, this study did not confirm our hypothesis regarding diet quality. It is possible that the fruit and vegetable FFQ used was not an appropriate measure of diet quality since it does not include other components of a healthy diet such as whole grain breads and cereals and the quality of the fat intake (saturated versus unsaturated fat). Although several studies have found no relationship between high job strain and diet quality measured as the consumption of fruit and vegetables,23,43-44 one Japanese cross-sectional study found that high job strain ratio was related to a higher intake of fat among men.15 Again, the varied tools used to measure diet quality preclude an accurate comparison of the findings. Interestingly, Jeffery and French31 found that energy intake and percentage of energy from fat were also positively associated with TV viewing. It does raise the possibility that in our study, TV viewing might have inadvertently served as a proxy for caloric intake and diet quality and may have counteracted the effect of our proxy for diet quality. Finally, it is important to note that characteristics of the psychosocial work environment have been hypothesized to have a direct effect on health outcomes including overweight and obesity independently of their effect on individual behavior.9

An unexpected finding was that active work environment was associated with overweight or obesity. Among Minnesota workers, active job environment was not associated with BMI.42 Since the effect of an active job environment was only observed in the adjusted analysis and was not mediated by other variables, we suspected that the relationship might be confounded by age since employees in an active work situation have more resources to cope with high psychological demands and the latter may be related to seniority. However, in our data there is no difference in mean age between workers in high job strain situations and all other quadrants included active environments (data not shown). In synthesis, further investigations of the characteristics of employees who work in active job environments in our sample will help understand this association. Given that at the time of baseline assessments in the current study, the company was implementing massive lay-offs, we hypothesized that job insecurity would work as an acute stressor in the workplace and would act directly or indirectly on measures of weight status.22 Also, we expected to find an interaction between acute and chronic stressors. In our sample, we were unable to confirm our hypotheses. Health effects on lay-off survivors are generally understudied.20-21 It is plausible that our measurements were done too soon to be able to capture any effects on employees weights. This study has several limitations. One of which is the impossibility of providing causal inferences given the cross-sectional nature of the data. Also, we do not know how representative our sample is of our worksite population since the company lacks information for relevant comparisons. Finally, the 3 item subscale assessing job insecurity as an acute stressor does not necessarily represent current job insecurity. One item of the subscale represents job insecurity in general, another represents current job insecurity, and the third represents future job insecurity. We expected, nonetheless, that in our particular context this subscale would capture the current situation. Although employees concern about their job security might have existed for a long time, the situation at the moment of baseline data collection went beyond what the company had previously experienced as evidenced by the unprecedented magnitude of the layoffs and the fact that entire worksites and their operations were decommissioned. The strengths of the study are that we count with measured height and weight and self-reported instead of imputed job strain measures. These data suggest that wellness programs in the worksite should target health enhancing behaviors to minimize the health effects of psychosocial work conditions. At the same time, worksites should examine their organizational and personnel development in order to prevent stressful work environments. For example, future research should investigate mechanisms that could reduce employees demands or increase their control over their jobs such as supportive supervision and changes in the structure of jobs. Further studies on the effect of the psychosocial work environment on weight status should be prospective, examine the potential bidirectionality of the effect of job stress by stratifying for baseline weight status, include measures of individual psychological features and work-home interface that can moderate the effect of working conditions. More evidence is needed on the health of lay-off survivors, in particular under the current economic downturn.

Self-Reported Disability in Adults with Severe Obesity Obesity is the most frequent metabolic disease worldwide and can progressively lead to a spectrum of comorbidities, including type 2 diabetes, hypertension, dyslipidaemia, cardiovascular disease, liver dysfunction, and osteoarthritis [1 3]. Preventing obesity-related co-morbidity relies on effective weight loss interventions; however, it is becoming evident that there is also a further need to focus on the daily living and well-being of obese patients. Obesity is still associated with high early mortality, but advances in the treatment of cardiovascular risk factors and acute coronary syndromes are now offering better cardioprotection options and prolong life expectancy [4]. Current data support the notion that in developed societies an increasing number of obese patients are expected to live more than previously estimated, despite failing to reduce their body weight [4, 5]. Furthermore, demographic and epidemiological projections predict growing and progressively ageing obese populations in the Western world [6 8]. These populations are expected to exhibit an escalating burden of obesity-related disease, particularly regarding complications which were previously underestimated or underexpressed due to earlier mortality, such as mobility problems and disability [9]. Longstanding and/or progressive obesity can eventually impair the physical ability of the patient to function in everyday life. Obesity-associated disability has been shown to correlate with body mass index (BMI) and the presence of comorbidities [4, 10 13]. Difficulty to perform simple everyday tasks (e.g., walking, climbing steps, driving, and dressing) may complicate the daily life of obese patients to the extent of inability to engage in usual social activities and employment [4]. The consequent impact on quality of life is devastating and may lead to a vicious cycle where obesity progressively causes physical inactivity, functional limitations, and mental distress (e.g., anxiety, depression) and vice versa [14, 15]. Furthermore, the economic burden posed by obesity-related disability on healthcare systems is alarming with reports suggesting that relative medical spending for the obese may be up to 100% higher than for normal-weight adults [16 18]. Evidence from the World Health Report shows that overweight and obesity is responsible for 8 15% of disability-adjusted life years lost in Europe and North America [19]. Notably, data from the National Longitudinal Survey of Youth in the USA indicate that being obese raises the probability of receiving disability income by 6.92 percentage points for men and by 5.64 percentage points for women, which is the equivalent to the effect of losing 15.9 and 16.7 years of education, respectively [16, 20]. Investigating the relationship between obesity and disability, identifying individuals at greater risk and improving their functional capacity are increasingly recognised as important steps in the care of obese patients. We present an observational study aimed to explore associations between increasing BMI and self-reported disability in adults with severe obesity. 2. Patients and Methods The study cohort was recruited from adults with BMI 35 kg/m2 followed at specialist outpatient obesity clinics at the Warwickshire Institute of Diabetes, Endocrinology and Metabolism (WISDEM, University Hospitals of Coventry and Warwickshire NHS Trust) and at the Birmingham Heartlands Hospital (Heart of England NHS Foundation Trust). Exclusion criteria included obesity secondary to endocrine or systemic

disease (e.g., Cushing's syndrome) and disability attributed to systemic disease other than obesity (e.g., rheumatoid arthritis, neurological disorders) or to previous injuries/accidents. Patients with disability due to further secondary complications of cardiometabolic disease (e.g., diabetic foot ulcers, symptomatic diabetic neuropathy, Charcot's arthropathy, symptomatic ischemic heart disease, and heart failure) were also excluded. The study was approved by the local ethics committee, and all participants provided informed consent. A total of 262 patients (183 females/79 males; mean age: 44.9 10.5 years) completed the study questionnaire, and all data were collected according to protocol, including demographic and comorbidity data obtained from reviewing the patients' medical charts. 2.1. Assessments 2.1.1. Anthropometry Body weight and height were measured at the outpatient obesity clinics in participants without shoes and heavy clothing. Weight was measured to the nearest 0.5 kg using a digital platform scale suitable for morbidly obese patients with a capacity of 300 kg (Seca 675, Seca, Hamburg, Germany). The Harpenden stadiometer was used to measure height to the nearest 0.1 cm. BMI was calculated as body weight in kilograms divided by the square of the height in meters. For the purposes of this study participants were categorized into three BMI groups: Group I: 35 39.99 kg/m2; Group II: 40 44.99 kg/m2; Group III: 45.0 kg/m2. 2.1.2. Health Assessment Questionnaire The Stanford Health Assessment Questionnaire (HAQ) was used as a validated self-report measure of functional ability in daily life [21, 22]. The HAQ has been widely applied in research, and, although initially developed for use in rheumatology, it is considered a generic instrument rather than disease specific [22 24]. Briefly, disability is assessed by the HAQ disability index through 20 questions regarding the degree of difficulty in performing two or three specific activities in eight distinct categories. These categories are (1) dressing and grooming; (2) arising; (3) eating; (4) walking; (5) hygiene; (6) reach; (7) grip; (8) common daily activities. Four possible grades of difficulty are provided for answering each question, which are rated as Without ANY Difficulty, With SOME Difficulty, With MUCH Difficulty, and UNABLE To do and are assigned a score of 0, 1, 2, and 3, respectively. In addition, each category has a companion variable for aids/devices that documents if any type(s) of assistance is required for the respective daily activities. For these variables the patient is also asked to report whether he/she (1) needs no assistance; (2) uses a special device in his/her daily activities; (3) usually needs help from another person; or (4) usually needs both a special device and help from another person. A complete copy of the instrument and instructions on its use can be downloaded from http://aramis.stanford.edu. Provided that the participant has given answers for at least six categories, the average score of the completed categories determines the final HAQ score (Standard HAQ disability index score) which ranges from 0 to 3 (0: no functional disability; 3: worst functional disability). Because healthy individuals consistently score zero on the HAQ [23], for the purposes of this study participants were also divided based on their final HAQ score to patients with no disability (HAQ score: 0) and patients with at least some degree of disability (HAQ score > 0).

2.1.3. Self-Reported Difficulty in Standing and Walking Independently of the HAQ, participants were also asked about having difficulty: (1) in standing unaided for 2 to 3 minutes and (2) in walking more than 100 metres (if necessary with aids). Four possible grades of difficulty were provided for each of these two questions: Without ANY Difficulty, With SOME Difficulty, With MUCH Difficulty, and UNABLE To do, assigned a score of 0, 1, 2, and 3, respectively. For the purposes of this aspect of the study participants were dichotomised based on their responses regarding difficulty in standing and in walking more than 100 metres to either having no difficulty (score = 0) or having at least some difficulty (score > 0). 2.2. Statistical Analysis The Statistical Package for the Social Sciences, SPSS, version 17.0 (SPSS Inc, Chicago, IL, USA) was used to analyze data. Results are expressed as percentage, mean standard deviation or median (range). The Kolmogorov-Smirnov and the Shapiro-Wilk tests were used to determine whether each study variable had a normal distribution. Based on these tests, BMI and HAQ score distributions in this study were nonparametric. Thus, comparisons between study groups were performed with the Mann-Whitney Utest or the Kruskal-Wallis test, and correlations were tested by the Spearman's rank correlation coefficient. Prevalence rates of disability between patient groups were compared and tested for statistical significance by chi-square test. Logistic regression was applied to test the relation of disability with various covariates. A P value of <0.05 was considered statistically significant. 3. Results A total of 262 obese patients consented to participate in the study out of 434 patients that were invited to participate, representing a 60% response rate (183 female (F) and 79 male (M) patients; with approximately 61.5% and 58% response rate in women and men, resp.). Mean age of the study participants was 44.9 10.5 years (women: 44.2 10.5 years; men: 46.7 10.4 years), with a mean BMI of 46.84 8.5 kg/m2 (women: 46.4 8.1 kg/m2; men: 47.8 9.2 kg/m2). Approximately 35% of the participants had type 2 diabetes (T2DM), 37% metabolic syndrome (metabolic syndrome as defined by the International Diabetes Federation definition, IDF, [25]), and 29% a diagnosis of clinical depression. Patient characteristics and distribution of participants by BMI group, gender and presence of comorbidities are presented in Table 1. Table 1 Selected patient characteristics and patient distribution by body mass index (BMI) group, gender, and presence of comorbidities.

A non-parametric distribution was noted for BMI and HAQ score in the study cohort. HAQ scores by BMI group, gender, and comorbidities are presented in Table 2. The median value of the HAQ score for the entire study cohort was 0.375 (range: 0 2.65) and an increase in the HAQ score was noted with

increasing BMI. The Kruskal-Wallis test revealed statistically significant difference in the HAQ score between the three BMI groups (P < 0.001). The Mann-Whitney test was used to compare HAQ scores between the different pairs of BMI groups showing that (1) Group II had a significantly higher HAQ score compared to Group I (P = 0.004); (2) Group III had a significantly higher HAQ score compared to Group I (P < 0.001); (3) Group III had a significantly higher HAQ score compared to Group II (P < 0.001) (Figure 1). Spearman's correlation showed that there was a significant correlation between HAQ score and BMI (r = 0.420, P < 0.001), as well as between HAQ score and age (r = 0.208, P = 0.001). After controlling for age, the correlation between HAQ score and BMI remained significant (P < 0.001). Table 2 Health Assessment Questionnaire (HAQ) disability index scores by body mass index (BMI) group, gender, and comorbidities.

Figure 1 Health Assessment Questionnaire (HAQ) scores (median, range) across the three body mass index (BMI) study groups (Group I: 35 39.99 kg/m2; Group II: 40 44.99 kg/m2; Group III: 45.0 kg/m2). *Group III versus (more ...) Healthy individuals consistently score zero on the HAQ [23], thus, an analysis was performed by dichotomizing the study cohort based on the HAQ score to patients with no disability (HAQ score: 0) and patients with at least some degree of disability (HAQ score > 0). Of all study participants, 72.5% had HAQ scores higher than zero, reporting at least a mild degree of difficulty in activities of daily living. The prevalence rates of this degree of disability (HAQ score > 0) between the different patient groups were compared and tested for statistical significance by chi-square test. Based on Pearson chi-square test, the prevalence rate of disability (HAQ score > 0) was related to BMI and to the presence of T2DM, metabolic syndrome, and depression, while it was not related to gender. Indeed, the prevalence rate of disability (HAQ score > 0) was significantly higher: (1) among participants with a higher BMI (51.7% for Group I versus 66.2% for Group II versus 85.5% for Group III, P < 0.001, Table 3); (2) among participants with T2DM (81.5% versus 67.6% in nondiabetic participants, P = 0.016); (3) among participants with metabolic syndrome (79.6% versus 68.3% in participants without metabolic syndrome, P = 0.047); (4) among participants with clinical depression (85.75% versus 67% in nondepressed participants, P = 0.002). Table 3 also presents the distribution of participants across the three BMI study groups when the HAQ score is categorized into 4 grades: (1) HAQ: 0 (no disability); (2) HAQ: 0.1 0.99 (mild to moderate difficulty); (3) HAQ: 1 1.99 (moderate to severe disability); (4) HAQ: 2-3 (severe to very severe disability). Logistic regression modelling in this study cohort showed that BMI was associated with an odds ratio (OR) of 1.128 (95% CI: 1.075 1.184; P < 0.001) for disability (HAQ score > 0) adjusted for age and with an OR of 1.127 (95% CI: 1.073 1.185; P < 0.001) adjusted for both age and depression. T2DM and metabolic syndrome when entered as covariates into the logistic regression model did not have a

statistically significant effect to the model. These obesity-related comorbidities were considered intermediaries in and not confounders to the association between increased BMI and disability, since based on the exclusion criteria of this study, patients with disability attributed to further secondary complications of cardio-metabolic disease (e.g., diabetic foot ulcers, symptomatic diabetic neuropathy, Charcot's arthropathy, symptomatic ischemic heart disease, and heart failure) were excluded. Table 3 Prevalence rates of degrees of disability based on the Health Assessment Questionnaire (HAQ) score across the three body mass index (BMI) study groups (no disability (HAQ score: 0); at least some degree of disability (HAQ score > 0); mild to moderate (more ...)

Finally, the prevalence rates of having at least some difficulty in standing (score > 0) and in walking more than 100 meters (score > 0) between the different patient groups were also compared and tested for statistical significance. Based on Pearson chi-square test, the prevalence rate of difficulty in standing (score > 0) was related to BMI and to the presence of T2DM, while it was not related to gender, metabolic syndrome, and depression. Thus, the prevalence rate of having at least some difficulty in standing was significantly higher: (1) among participants with a higher BMI (8.3% for Group I, 27.1% for Group II, 24.8% for Group III, P = 0.016); (2) among participants with T2DM (30.8% versus 16.7% in nondiabetic participants, P = 0.008). Furthermore, the prevalence rate of difficulty in walking more than 100 meters (score > 0) was related to BMI and to the presence of depression, while it was not related to gender, T2DM, and metabolic syndrome. Indeed, the prevalence rate of having at least some difficulty in walking more than 100 meters was significantly higher: (1) among participants with a higher BMI (18.6% for Group I, 44.1% for Group II, 53.2% for Group III, P < 0.001); (2) among participants with depression (58.6% versus 36.5% in nondepressed participants, P = 0.001). 4. Discussion The primary objective of this study was to explore associations between obesity and self-reported disability in adults with BMI 35 kg/m2. The data from our cohort of patients with severe obesity showed that self-reported disability, as expressed by the HAQ score, correlated with BMI, age, and the presence of T2DM, metabolic syndrome, and clinical depression. This finding agrees with data from the literature that have documented the burden of disability in general and obese populations, as well as in various other patient groups [26 39]. The mean HAQ score for our study population was 0.607 (95% CI: 0.528 0.686) with a median value of 0.375 (range: 0 2.65). Krishnan et al. have reported normative values for the HAQ disability index in the general population in Finland, documenting a population mean HAQ score of 0.25 (95% CI: 0.22 0.28), with 32% of respondents having at least some disability (HAQ score > 0) [26]. Given that healthy individuals consistently score zero on the HAQ [23] and that the HAQ score distribution is not Gaussian, we also applied zero as a cut point for the HAQ score in order to dichotomize our study cohort into patients without disability and patients with at least some difficulty in activities of daily living. In this

analysis, 72.5% of our study participants had HAQ scores higher than zero. This is in accord with the analysis of the data from Finland showing that within the studied general population, which included approximately 20% obese participants (BMI >30 kg/m2), individuals with BMI 30 kg/m2 had a significantly lower prevalence rate of disability compared with obese individuals (HAQ score > 0: 28.4% versus 51.7% for nonobese and obese individuals, resp.) [26]. Of note, in our study the prevalence rate of disability (HAQ score > 0) was 51.7% in Group I, 66.2% in Group II, and 85.5% in Group III, further documenting a gradient of increasing self-reported disability as BMI increases over 35 kg/m2. Indeed, comparing the HAQ score for the three study BMI groups, we found that Group III had significantly higher HAQ score compared with the other two groups, while Group II had also significantly higher HAQ score compared with Group I. The recognition of this gradient may be useful in clinical practice for identifying obese individuals with greater difficulty in performing everyday tasks and could also allow further stratification of patients in order to intensify interventions and prioritize the use of available healthcare resources. It must be noted that different cut points can be used to define or categorize disability based on the HAQ disability index score [40, 41]. HAQ scores up to 1 are generally considered to reflect mild to moderate difficulty in daily life activities, while scores between 1 and 2 represent moderate to severe disability, and scores of 2 to 3 indicate severe to very severe disability [40]. A study by Walter et al., exploring the effects of obesity on mortality and disability in the older population in The Netherlands, has used a cut point of 0.5 to define a participant as at least mildly disabled [27], as previously applied for participants in the Rotterdam study cohort [33]. The results of this study in older adults (age 55 years and older) also documented that BMI was related to self-reported disability, with more years lost to disability with increasing body weight, supporting our study findings. Of note, applying zero as a cut point for the HAQ score in the context of obesity provides a distinct measure to dichotomize patients for the presence of disability, which is not affected by the dispersion of the HAQ score within different BMI categories and hence may be suggested as a less controversial method than using any other cut-off value. Furthermore, this cut point can also be regarded as a treatment goal for weight loss interventions. Thus, a notion of recovery from obesity-related disability could be advocated similarly to remission/resolution of T2DM with weight loss. Interestingly, studies in patients with rheumatoid arthritis have documented that HAQ disability index scores needed to improve by approximately 0.22 units before participants stopped rating themselves as about the same (minimally clinically important difference) [41]. In patients with severe obesity, further research is required to evaluate at what extent weight loss is associated with clinically meaningful differences in physical functioning/disability scores and whether certain weight loss interventions (e.g., diet, exercise, cognitive behavioral therapy, pharmacotherapy, bariatric surgery, and their combinations) might differ regarding such functional/health status outcomes [42]. In addition to BMI, the HAQ score in our study also correlated with age, which is consistent with available data from the general population [26, 39]. However, in this study cohort no relation was noted between the HAQ score and gender, contrary to published evidence indicating that activities of daily living may be more affected in women, especially in older populations [27, 43 45]. This could be partly attributed to a referral bias of obese patients with higher disability independently of gender. Krishnan et

al. also reported that disability among women increased at a faster rate compared to that among men and that women had a higher estimated mean HAQ score (0.28 versus 0.18 in men) [26]. Yet, also in this study, which included 1530 adults, this gender difference was no longer statistically significant after adjustment for age. Finally, in our cohort of severely obese patients the HAQ score was related to the presence of T2DM, metabolic syndrome, and clinical depression. Central obesity is a prerequisite for the IDF metabolic syndrome definition [25], and T2DM is pathogenetically linked to obesity [46]. In addition, for the purposes of this study patients with disability due to secondary complications of T2DM and/or metabolic syndrome (e.g., diabetic foot ulcers, symptomatic diabetic neuropathy, Charcot's arthropathy, symptomatic ischemic heart disease, and heart failure) were excluded. Hence, these comorbidities were regarded as intermediaries in and not confounders to the association between increased BMI and disability and when entered as covariates into the logistic regression models of this study did not have a statistically significant effect. Contrary, the published data regarding the association between obesity and depression is less strong and mixed [47 51]. Therefore, depression was entered as a covariate in the applied logistic regression model which showed that, adjusted for both age and depression, BMI was associated with an OR of 1.127 (95% CI: 1.073 1.185; P < 0.001) for disability (HAQ score > 0).