Accuracy of Medical Certification of Cause of Death at Benha

University Hospitals, Egypt: A One- Year Retrospective

Descriptive Study (2014)
Abdelmonem G. Madboly, M.D.* and Eslam S. Metwally, M.D.*
*
Forensic Medicine & Clinical Toxicology Department, Faculty of Medicine, Benha
University, Egypt

Abstract:
Inaccuracy of the cause of death certification can impair the national health
database and the medico-legal investigations. This study aimed to assess the errors of
medical certification of cause of death at Benha University Hospitals, Egypt, to suggest
necessary corrective measures. Death notification forms (DNFs) of all patients (n=347)
who had expired during the period from January 2014 to December 2014 were assessed.
The demographic characteristics, administrative details and the medical certification of
the cause of death were reviewed using an approved standardized form. Errors in the
DNFs were classified into six categories, from 0 to V according to increasing severity of
errors. The study recorded errors in 100% of DNFs. The high proportion (63.4%) of
DNFs contained major errors (grade V), sufficiently serious to affect the accuracy and
interpretation of the medical certification of the cause of death. There was an obvious
confusion between the underlying cause of death and the mode of death in most (61.1%)
of DNFs. The results of this study underscore the usefulness of the format database
utilized by the ministry of health in Egypt. Educational, managerial and administrative
interventions are urgently needed to improve the standard of DNFs completion.
Keywords: Medical certification; Cause of death; Benha; Egypt; Descriptive study.

INTRODUCTION
An accurate and verifiable system for the recording of causes of death
is an essential prerequisite for meaningful collection of epidemiological
data, and also an ideal safeguard against illegal practice on the part of
individual practitioners (Sington & Cottrell, 2002).
Death certificates actually play a role in medico-legal investigations,
declaration of health events in medical researches, evaluation of mortality in
a community, and it can also be a valuable source of census studies (Swift &
West, 2002 and Lakasing & Minkoff, 2012).
Data of medical certification of cause of death are highly
standardized, because they have been regulated for over 50 years by the

1|Page
World Health Organization (WHO) through the International Classification
of Diseases (ICD). The ICD specifies how the information on cause of death
should be collected (reporting forms), processed (coded), and disseminated
(tabulated). This makes mortality data from death certificates highly
comparable both internationally as well as nationally (WHO, 1993 and
Johansson et al., 2006).
The medical certification of cause of death allows the underlying
cause of death, as well as other significant conditions unrelated to the
immediate cause of death (contributing causes) to be listed (Tormey, 2011).
The underlying cause of death was conceptualized as the disease or
injury that initiated the train of morbid events leading directly to death,
which is recorded in part I of the death certificate (WHO, 1993 and Sington
& Cottrell, 2002).
But, it is increasingly recognized that analysis of underlying causes of
death alone might not be sufficient to represent the complex burden of multi-
morbidity, so contributing causes of death were conceptualized, which
represent consequences or another diseases present at the moment of death
that may have contributed to death, which are listed in part II of the death
certificate (Morton et al., 2000 and Klijs et al., 2014).
The quality of the information provided in death notification forms
(DNFs) death certificates varies according to the personnel responsible for
completing it, as well as on the availability of medical records in the
hospitals (Hernndez et al., 2011).
International reports of inaccuracies in DNFs death certificates
range from 2065%. Inaccuracies can emerge from the initial entry by the
attending doctor or coroner, and the assignment of codes by coders (Slater,
1993; McAllum et al., 2005 and Laurenti et al., 2008).
2|Page
 Many studies have shown inaccuracies in medical certification of
cause of death, as those done by Jansson et al. (1997); Swift & West
(2002); Johansson et al. (2006) and Johansson et al. (2009).
A WHO study put Egypt in the group of low quality death
registration data, as completeness of DNFs death certificates < 70% or ill-
defined codes appear on > 20% of registrations (Mathers et al., 2005).
It is essential to monitor the DNFs death certificates completion, as
it is the primary source of mortality and disease statistics in Egypt, However,
no enough assessments have been conducted to establish the quality of the
data of the medical certification of cause of death in Egypt.
So, this study was conducted to assess the accuracy of completeness
as well as to determine the frequency of errors of medical certification of
cause of death in DNFs at Benha University Hospitals, Egypt, to suggest
necessary corrective measures for improvement of the quality of reporting
the cause of death.
MATERIALS AND METHODS
I- Materials:
This is a retrospective descriptive study that was conducted at Benha
University Hospitals, Egypt. The data of the study were sourced from death
notification forms (DNFs) of all patients who had expired during the period
from the beginning of January 2014 to the end of December 2014.
Benha University Hospitals followed Benha Faculty of Medicine
located in Benha city, Qalubia governorate, Egypt, and provide medical
services for at least 5 million inhabitants.
The study did not primarily validate the cause-of-death diagnoses
because no autopsy information was available. But the DNFs were examined
for inaccuracies in their completion as regard the WHO standard format.
3|Page
 The standard international format approved for the medical
certification of cause of death for hospital in-patient deaths, was used, as
shown in Fig. (1) (Dash et al., 2014).

Figure (1): Standard format used for medical certification of cause of death for
hospital in-patient deaths (Dash et al., 2014).

II- Methods:
The following parameters were studied:
The demographic profile of the deceased: (the full name of the
deceased, age, gender; if female; pregnant or not etc).
The health & administrative data: (date of admission, interval between
onset of disease and death, the certifying medical practitioner (full
name and signature).

4|Page
 The completeness of medical certification of cause of death:(the
underlying cause of death, the antecedent causes, the immediate cause
of death, the mode & manner of death, the contributing causes of death,
and the use of illegible abbreviations in the DNFs).
Errors in the death DNFs were classified into six categories, as these
seemed to be relevant, clear and easy to use, from 0 to V, according to the
increasing severity, Grade IV and V errors were thought to significantly
change the death certificate interpretation, Table (1) (Haque et al., 2013).
Table (1): Grading scale used to assess the DNFs errors (Haque et al., 2013).
Grade Errors
Grade 0 No errors
Errors
Grade IA Incomplete or inaccurate demographics only
Grade IB Inappropriate health and administrative information only
Grade II Comorbidities (antecedent or contributing causes) incomplete
Grade III Comorbidities (antecedent or contributing causes) not listed
Grade IV Inappropriate immediate cause of death or only mechanism (s) or mode of death
Grade V Underlying cause of death incorrectly attributed or placed in improper sequences

III- Statistical design:

The collected data were tabulated and analyzed using statistical
package for the social sciences (SPSS) version 16 for windows (SPSS Inc.,
Chicago, USA). Data were presented as number and percentages. Chi square
test (X2) of significance was used, P value <0.05 was considered significant.

RESULTS
The present study assessed 347 death notification forms (DNFs) of all
patients who had expired during the period from the beginning of January
2014 to the end of December 2014 at Benha University Hospitals, Egypt.
I- Demographic data of the deceased patients:
The present work showed that the majority (55.6%) of DNFs were
issued for male patients, of extremes of age; aged >60 years (29.7%), and
5|Page
<18 years (25.3%). Deaths mostly occurred at the intensive care unit (ICU)
(53.9%) and at pediatric department (21.9%) as showed in Fig. (2) and Table
(2).

Figure (2): Distribution of the studied death notification forms (DNFs) (n = 347),
according to age and gender of the deceased.

Table (2): Distribution of the studied death notification forms (DNFs) (n = 347),
according to the hospital department at which death occurred.
Hospital department No. (n=347) % (100.0)
ICU 187 53.9
Pediatrics 76 21.9

Cardiothoracic surgery 11 3.2

Neurosurgery 7 2.0

General surgery 3 0.9

Internal medicine 20 5.8
Obstetrics/Gynecology 2 0.6
Poison control unit 1 0.3
Cardiology 7 2.0
Emergency department 33 9.5

II- Details of the recorded errors:

The present study illustrated that all the studied DNFs contained
errors, and 96.5% of them had at least two types of errors. All DNFs
contained errors related to the medical certification of the cause of death,

6|Page
96.5% of them had administrative data errors, and 22.2% had demographic
data errors, as showed in Fig. (3, 4, and 5).

Figure (3): Distribution of the studied death notification forms (DNFs) (n = 347),
according to the presence or absence of errors.

Figure (4): Distribution of the studied death notification forms (DNFs) (n = 347),
according to the frequency of errors (one or more error types).

Figure (5): Distribution of the studied death notification forms (DNFs) (n = 347),
according to the type of errors.
7|Page
 The present work showed that the most common observed error
among the demographic data was un-recording whether the female in the
child bearing period was pregnant or not (53% %), the most common error
among the administrative data was un-recording of the time interval between
the disease onset and occurrence of death, meanwhile the most common
error in the medical certification of cause of death section in DNFs was
related to the underlying cause of death (either unrecorded at all or it was
recorded but in improper sequence), as showed in Table (3).

Table (3): Distribution of the studied death notification forms (DNFs) (n = 347),
according to the details of each type of errors.
Details of errors No. (n=347) % (100.0)
Demographic data Age 12 3.5
Gender 49 14.1
Pregnancy (n=47)* 25 53.2
Full name of dead 12 3.5
Administrative data Time interval 335 96.5
Certifying doctor 4 1.2
Medical certification Contributing causes 214 61.6
of cause of death Mode of death 16 4.6
Manner of death 37 10.7
Immediate cause 171 49.3
Antecedent causes 201 57.9
Underlying cause of death 216 62.2
Abbreviations used 341 1.7
*47 females in the child bearing period (15-49 years).

III- Grades of errors:

The present study illustrated that the high proportion (63.4%) of
DNFs at Benha University Hospitals contained major errors sufficiently
serious to affect the accuracy and interpretation of the medical certification
of the cause of death; grade V errors, as showed in Fig.(6).

8|Page
 Figure (6): Distribution of the studied death notification forms (DNFs) (n = 347),
according to the grade of errors.

The present work confirmed that there was an obvious confusion

between the underlying cause of death and the mode of death, as in most
(61.1%) of DNFs in which mode of death was mentioned, the mode of death
cardio-pulmonary failure was used instead of a definitive underlying cause
of death, as showed in Fig. (7).

Figure (7): Distribution of the studied death notification forms (DNFs) in which mode
of death was mentioned (n = 331, 95.4% of DNFs), in relation to the
underlying cause of death.
9|Page
 DISCUSSION
Death certification, a practice with a history dating back to the 12th
century, continues to be a legal obligation for a doctor attending a decedent
during their terminal illness (Swift & West, 2002).
Every physicians duty is bounded to issue the certification of cause
of death. Incomplete or inaccurate entry in these certificates can significantly
impair the precision of a national health information database and the
medico-legal investigations (Armour & Bharucha, 1997; Lahti &Penttila,
2001; Haque et al., 2013 and Dash et al., 2014).
In most countries, death notification forms (DNFs) or death
certificates constitute the largest disease-related source of information and
thus have a significant effect on public health research and policy-making
(DAmico et al., 1999 and lvarez et al., 2009).
The present work showed that of the total studied 347 DNFs the
majority were issued for male patients (55.6%), of age extremes; aged >60
years (29.7%), and <18 years (25.3%). Deaths mostly occurred at the
intensive care unit (ICU) (53.9%) and at pediatric department (21.9%).
This was in accordance with Dash et al. (2014) who reviewed a total
of 151 death certificates at Bhubaneswar, India. The majority of them were
issued for patients in the extremes of the age >60 years (30.5%) and <10
years (27.8%), of them 60.26% were males and 39.74% were females.
Also, Katsakiori et al. (2007) analyzed 487 death certificates in
Greece, of which 51.5% were for males and 48.5% females (median age 82
years, range 5-103 years; and 83 years, range 1-104, respectively).
Meanwhile Burger et al. (2012) assessed 703 DNFs in Cape Town,
South Africa, and found that most (38.0%) of the cases were aged between
14 and 45 years and most (60.7%) of them were males.
10 | P a g e
 lvarez et al. (2009) in Hermosillo, Mexico found that the majority
(58%) of deaths occurred in pediatric departments.
Accurate completion of death certificates is generally poor. Studies
have shown that death certificate error rates are high, particularly in the
developing countries (Myers & Farquhar, 1998; Lakkireddy et al., 2004;
Mathers et al., 2005; Pritt et al., 2005 and Degani et al., 2009).
The present study illustrated that all the studied DNFs contained
errors, and 96.5% of them had at least two types of errors. All DNFs
contained errors related to the medical certification of the cause of death,
meanwhile 96.5% of them had administrative data errors, and 22.2% had
demographic data errors.
This was in agreement with many previous studies that recorded a
high error rates as with Haque et al. (2013) in Pakistan who reported that
99% of reviewed death certificates contained at least one error, about 77%
certificates had 3 or more errors and only 1% certificates were free of errors,
and they confirmed that the most frequent errors pertaining to patients
demographics (92%) followed by medical certification of cause of death
(87%), and errors in administrative data were found in 49% of cases. Burger
et al. (2007) in South Africa recorded errors in 91.7% of DNFs, 43.4% had
at least one major error, most commonly (72.3%) involving the
completeness of medical certification of cause of death, and among the
minor errors irrelevant demographic information in 13% of cases.
Meanwhile other studies recorded a relatively law error rates as with
Swift & West, (2002) in United Kingdom, who revealed that 55% of DNFs
were completed logically and appropriately, however, 35% contained
incomplete data, and nearly 10% were completed to a poor standard, being
illogical or inappropriately completed. Cina et al. (1999) in Wilford Hall
11 | P a g e
Medical Center, Lackland Air Force Base, USA, found that among the
studied 98 death certificates, errors were found in 37% of them. Other earlier
studies recorded errors rates in certification of underlying cause of death
ranging from 16% to 40%, as in studies of Gjersoe et al. (1998) in a medical
hospital department in Denmark, and Jordan & Bass, (1993) in a London,
Ontario teaching hospital respectively.
The recorded high error rates were all in the developing countries like
Egypt, South Africa and Pakistan, meanwhile the recorded law error rates
were issued in the developed countries like USA and United Kingdom, as
they had an established proper system for fulfilling the medical certification
of cause of death. Mathers et al. (2005) in a WHO study found that few
countries have good-quality death registration systems mostly were
developed countries, but completeness of DNFs death certificates in many
developing countries like Egypt was < 70%.
The present work showed that the most common observed error
among the demographic data was un-recording whether the female in the
child bearing period was pregnant or not (53% %), the most common error
among the administrative data was un-recording of the time interval between
the disease onset and occurrence of death, meanwhile the most common
error in the medical certification of cause of death section in DNFs was
related to the underlying cause of death (either unrecorded at all or it was
recorded but in improper sequence).
Dash et al. (2014) at Bhubaneswar India found that in 2.65%
certificates the gender of the patient was missing, in 5.30% the age of the
deceased was not mentioned, the interval between the onset and terminal
event of various conditions was missing in 62.25% of cases. About 26.5%
had inaccurate cause of death (these include various inappropriate terms
12 | P a g e
such as cardiac arrest, cardiac shock, sudden cardiac failure, sepsis,
respiratory failure, respiratory paralysis, and respiratory arrest), incomplete
antecedent & underlying cause of death was found in 23.18% & 55.63%
respectively, and in 43.71% there was use of abbreviations.
Burger et al. (2007) in South Africa found that the time interval
between the onset of disease and death was not mentioned in 81.5% of cases,
while other health professionals details were completed in most cases.
There was no underlying cause of death after mechanism in 13.5%, none
acceptable cause of death in 14.8%, competing causes of death in 15.3%,
incorrect sequencing in 28.7%, abbreviations were used in 23.7%.
Cina et al. (1999) in Wilford Hall Medical Center, USA, found that
the most prevalent error type was the use of a nonspecific diagnosis as the
underlying cause of death (61%). No errors were found in the listed manner
of death.
The present study illustrated that the high proportion (63.4%) of
DNFs at Benha University Hospitals contained major errors sufficiently
serious to affect the accuracy and interpretation of the medical certification
of the cause of death i.e. grade V errors. While grade IV errors involving
immediate cause of death were 17.9%, antecedent or contributing causes
were incomplete in about 10.6% (grade II), and they were not listed in 8.1%
of DNFs (grade III)
Haque et al. (2013) in Pakistan concluded that the most frequent
errors were grade IA errors (92%) involving patients demographics and
Grade V errors regarding underlying cause of death (87%), while grade IV
errors involving immediate cause of death were 62%, comorbidities not
mentioned in about 33% (grade III), and they were incomplete in 18% of

13 | P a g e
certificates (grade II), about 49% of errors involving administrative data
(grade IB), and only 1% death certificates had no errors (grade 0).
Correct certification of the cause of death requires: 1) awareness of
the distinction between "cause" and "mechanism" of death, and 2)
understanding the meaning of the concepts "immediate cause of death" and
"underlying cause of death", both of them included in the standard WHO
format of the medical section of a death certificate (Villar & Perez-Mendez,
2007).
Dash et al. (2014) cleared that cause of death includes any disease or
injury responsible to initiate a chain of events resulting in death, whereas
mode of death is the process of death and may be initiated by the failure of
any of the three vital systems of the body.
Katsakiori et al. (2007) stated that the mechanism of death is a
physiologic derangement or a biochemical disturbance produced by the
cause of death.
These differences are explicitly mentioned in textbooks and literature
and are taught extensively to undergraduates. Although the cause of death
should not be confused with the mode of death, it is disturbingly seen in a
significant proportion of the certifying doctors (Dash et al., 2014).
Megrane et al. (1997) and Katsakiori et al. (2007) recommended that
due to their lack of etiologic specificity, mechanism (s) or mode of death
should not appear on death certificates.
While, Swift & West, (2002) stated that modes of death-for instance
heart failure, respiratory failure, liver failure, or cardiac arrest-are
unacceptable alone. Modes may only be used if qualified by an appropriate
cause of death.

14 | P a g e
 The present work confirmed that there was an obvious confusion
between the underlying cause of death and the mode of death, as in most
(61.1%) of DNFs in which mode of death was mentioned, the mode of death
cardio-pulmonary failure was used instead of a definitive underlying cause
of death.
This was in accordance with Agarwal et al. (2010) in India who
confirmed that the confusion between the cause of death and mode of
death was seen in 86% of the certificates, and Meel (2003) in Umtata,
South Africa who revealed that 79% of DNFs had Cardio-respiratory
failure as a cause of death. This made the WHO to put South Africa in the
group of low quality death registration data (Mathers et al., 2005).
Also, Katsakiori et al. (2007); Lakkireddy et al. (2007) and Degani et
al. (2009) confirmed that the most common error in death certificate
completion committed by physicians was the confusion between cause of
death and mechanistic terminal events mode of death.
Meanwhile El-Nour et al. (2007) in Sudan observed that 47% of
certificates listed mode of death in place of cause of death, and Dash et al.
(2014) in India found that this confusion was evident in 26.5% of death
certificates. On contrast a national study in Taiwan revealed only 7% of such
errors (Lu et al., 2001).
Hanzlick (1996) and Haque et al. (2013) concluded that these errors
originated from a lack of knowledge among physicians on how to identify,
select and differentiate between the underlying cause, direct (immediate)
cause and antecedent cause(s) of death.
While Lu et al. (1996) confirmed that physicians usually confuse the
cause of death with the mechanism (s) or mode of death; probably because
the target of the medical treatment is more often the mechanism (s).
15 | P a g e
 On the other hand Middleton et al. (2011) stated another possible
explanation contributing to these mistakes which was the length of illness
that leads to death. If a person dies after a long, well-characterized illness,
the cause of death on the certificate is likely to be more accurate than a
sudden or unobserved death with lack of adequate information about the
decedents disease history.
Without knowing the reasons for high error rates in death certificate
completion, it is hard to design a relevant intervention program to improve
the quality of death certification (Lu et al., 2001).
Many studies have reported that most problems with reliability of
death certificates arise from the limited formal training and medical
inexperience of personnel in death-certification or disease-coding procedures
as well as perceived lack of certificate importance (Myers & Farquhar,
1998; Morton et al., 2000; Nowels, 2004; Pritt et al., 2005 and Degani et
al., 2009).
In the majority of teaching hospitals, resident physicians are
responsible for death certificate completion, but only a small percentage
receives sufficient formal training (Barber, 1992; Lakkireddy et al., 2007
and Wexelman et al., 2013).
Various studies have reported that a brief educational intervention and
sensitization can significantly reduce the rates of major and minor errors in
the completion of death certificates (Pain et al., 1996; Abos et al., 2006;
Katsakiori et al., 2007; Selinger et al., 2007; Villar & Perez-Mendez, 2007
and Pandya et al., 2009).
Swift & West (2002) and Dash et al. (2014) stated that part of the
legal obligation as well as ethical responsibility of a treating doctor is to
complete a death certificate in accordance with the WHO recommendations.
16 | P a g e
 To ensure correct completion and valid comparisons within and
among countries, the used DNFs death certificates form should follow the
WHO standard form. Failure to follow the requirements of the standard form
would be expected to adversely affect the accuracy and usefulness of cause
of death statistics (Lu et al., 2001).
Counter-signing of DNFs by senior medical staff, multidisciplinary
mortality meetings and querying programs (as used in parts of the USA and
Europe) could also be considered to reduce errors in death registration
(Burger et al., 2007).
Few countries have good-quality mortality registration data. There is
an urgent need for countries to implement death registration systems, or
enhance their existing systems in order to improve knowledge about the
most basic of health statistics: who dies from what? (Mathers et al., 2005).
The study concluded that:
1. All DNFs at Benha University Hospitals, Egypt, contained completion
errors. The high proportion (63.4%) of them was major errors sufficiently
serious to affect the accuracy and interpretation of the medical
certification of the cause of death.
2. There was an obvious confusion between the underlying cause of death
and the mode of death, in most (61.1%) of DNFs.
3. The results of this study underscore the usefulness of the standard format
database utilized by the ministry of health in Egypt.
4. Educational, managerial and administrative interventions are urgently
needed to improve the standards of DNFs completion.
Recommendations:
1. Reporting the data of medical certification of cause of death in a new
DNF designed to ensure registration of medical cause of death according
17 | P a g e
 to the standard format of the WHO guidelines as showed in this article.
This could make mortality data from death certificates in Egypt firstly
accurate and precise, secondly highly comparable both internationally as
well as nationally (N.B.: the already used format in Egypt is insufficient
to fulfill the WHO guidelines, especially in section of medical
certification of cause of death as part I and part II not included at all), as
showed in Fig. (8).

Figure (8): Format used for medical certification of cause of death for hospital in-
patient deaths in Egypt (Template 32a, Ministry of health, Egypt, 2015).

2. This study suggested a need for improvement of the training of death

certifiers, particularly resident and house officer physicians, as specific

18 | P a g e
 training (interactive workshops & regular death review meetings) offered
to them can lead to dramatic improvement in the accuracy of completion
of DNFs.
This study had some limitations; firstly, its retrospective design.
Secondly, it was conducted in a single teaching hospital, and it only
reflected a single institutional practice. Further studies needed at a national
level to correctly evaluate this medical issue.
ACKNOWLEDGEMENT
The authors express their gratitude to the general manager, physicians
and all employees at Benha University Hospitals, Egypt, for providing this
study with the clinical data needed to perform the statistical analysis.
www.fmed.bu.edu.eg

REFERENCES
Abos, R.; Perez, G.; Rovira, E. and et al. (2006): A pilot program to
improve causes of death certification in primary care of Catalonia,
Spain. Gac. Sanit., 20:450-456.
Agarwal, S.S.; Kumar, A.G.V.; Kumar, L. and et al. (2010): A study on
appraisal of effectiveness of the MCCD scheme. J. Indian Acad.
Foren. Med., 32(4): 318-320.
lvarez, G.; Harlow, S.D.; Denman, C. and et al. (2009): Quality of cause-
of-death statements and its impact on infant mortality statistics in
Hermosillo, Mexico. Am. J. Pub. Health, 25 (2): 120-127.
Armour, A. and Bharucha, H. (1997): Nosological inaccuracies in death
certification in Northern Ireland. The Ulster Med. J., 66: 13-17.
Barber, J.B. (1992): Improving the accuracy of death certificates. J. Natl.
Med. Assoc., 84(12): 10071008.
Burger, E.H.; Groenewald, P.; Bradshaw, D. and et al. (2012): Validation
study of cause of death statistics in Cape Town, South Africa, found
poor agreement. J. Clin. Epidemiol., 65: 309-316.
Burger, E.H.; Merwe, L. and Volmink, J. (2007): Errors in the completion
of the death notification form. S. Afr. Med. J., 97: 1077-1081.

19 | P a g e
Cina, S.J.; Selby, D.M. and Clark, B. (1999): Accuracy of death
certification in two tertiary care military hospitals. Mil. Med.,
164(12): 897-899.
DAmico, M.; Agozzino, E.; Biagino, A. and et al. (1999): Ill-defined and
multiple causes on death certificatesa study of misclassification in
mortality statistics. Eur. J. Epidemiol., 15(2): 141148
Dash, S.K.; Behera, B.K. and Patro, S. (2014): Accuracy in certification of
cause of death in a tertiary care hospital- a retrospective analysis. J.
Foren. Legal Med., 24: 33-36.
Degani, A.T.; Patel, R.M.; Smith, B.E. and et al. (2009): The Effect of
Student Training on Accuracy of Completion of Death Certificates.
Med. Educ. Online, 14: 17.
El-Nour, A.A.M.; Ibrahim, Y.A.H. and Ali, M.M. (2007): Evaluation of
death certificates in the pediatric hospitals in Khartoum state during
2004. Sudan J. Pub. Health, 2(1):29-37.
Gjersoe, P.; Andersen, S.E.; Molbak, A.G. and et al. (1998): Reliability of
death certificates. The reproducibility of the recorded causes of death
in patients admitted to departments of internal medicine. Ugeskr.
Laeger., 160:50305034.
Hanzlick, R. (1996): Protocol for writing cause-of-death statements for
death due to natural causes. Arch. Intern. Med., 156: 2526.
Haque, A.S.; Shamim, K.; Siddiqui, N.H. and et al. (2013): Death
certificate completion skills of hospital physicians in a developing
country. B.M.C. Health Services Research, 13: 205-209.
Hernndez, B.; Ramrez-Villalobos, D.; Romero, M. and et al. (2011):
Assessing quality of medical death certification: concordance
between gold standard diagnosis and underlying cause of death in
selected Mexican hospitals. Popul. Health Metrics, 9: 38-47.
Jansson, B.; Johansson, L.A.; Rosen, M. and et al. (1997): National
adaptations of the ICD rules for classification da problem in the
evaluation of cause of death trends. J. Clin. Epidemiol., 50: 367-375.
Johansson, L.A.; Bjorkenstam, C. and Westerling, R. (2009): Unexplained
differences between hospital and mortality data indicated mistakes in
death certification: an investigation of 1,094 deaths in Sweden
during 1995. J. Clin. Epidemiol., 62: 1202-1209.
Johansson, L.A.; Westerling, R. and Rosenberg, H.M. (2006):
Methodology of studies evaluating death certificate accuracy was
flawed. J. Clin. Epidemiol., 59: 125-131.
Jordan, J.M. and Bass, M.J. (1993): Errors in death certificate completion
in a teaching hospital. Clin. Invest. Med., 16:249255.
20 | P a g e
Katsakiori, P.F.; Panagiotopoulou, E.C.; Sakellaropoulos, G.C. and et al.
(2007): Errors in death certificates in a rural area of Greece. Rural
and Remote Health, 7: 822.
Klijs, B.; Nusselder, W.J. and Mackenbach, J.P. (2014): Nationwide
individual record linkage study showed poor agreement of causes of
death and hospital diagnoses at individual level but reasonable
agreement at population level. J. Clin. Epidemiol., 67: 160-168.
Lahti, R.A. and Penttila, A. (2001): The validity of death certificates:
routine validation of death certification and its effects on mortality
statistics. Foren. Sci. Internat., 115: 15-32.
Lakasing, E. and Minkoff, S. (2012): Uncertainties in death certification.
Br. J. Gen. Pract., 62 (605): 658659.
Lakkireddy, D.R.; Basarakodu, K.R.; Vacek, J.L. and et al. (2007):
Improving death certificate completion: a trial of two training
interventions. J. Gener. Int. Med., 22: 544-548.
Lakkireddy, D.R.; Gowda, M.S.; Murray, C.W. and et al. (2004): Death
certificate completion: How well are physicians trained and are
cardiovascular causes overstated? Am. J. Foren. Med., 117: 492-498.
Laurenti, R.; De-Mello Jorge, M.H.P. and Gotlieb, S.L.D. (2008):
Underlying cause-of-death mortality statistics: considering the
reliability of data. Rev. Panam. Salud. Publica., 23: 349-356.
Lu, T.H.; Shau, W.Y.; Shih, T.P. and et al. (2001): Factors associated with
errors in death certificate completion. A national study in Taiwan. J.
Clin. Epidemiol., 54: 232-238.
Lu, T.H.; Shih, T.P.; Lai, H.S. and et al. (1996): Analysis of formative
errors and validity of cause-of-death diagnosis in a teaching hospital.
Chin. J. Public Health, 15: 373381.
Mathers, C.D.; Ma Fat D.; Inoue, M. and et al. (2005): Counting the dead
and what they die from: an assessment of the global status of COD
data. Bull. World Health Organ., 83: 171- 177.
McAllum, C.; George, I. S. and White, G. (2005): Death certification and
doctors dilemmas: a qualitative study of GPs perspectives. Br. J.
Gen. Practice, 55: 677683.
Meel, B.L. (2003): Certification of deaths at Umtata General Hospital, South
Africa. J. Clin. Foren. Med., 10: 13-15.
Megrane, B.P.; Billiland, M.G.F. and King, D.E. (1997): Certification of
death by family physicians. Am. Fam. Physician, 51: 14331438.
Middleton, D.; Anderson, R.; Billingsly, T. and et al. (2011): Death
certification: issues and interventions. Open J. Prev. Med., 1(3):167
170.
21 | P a g e
Morton, L.; Omar, R.; Carroll, S. and et al. (2000): Incomplete and
inaccurate death certification- the impact on research. J. Pub. Health
Med., 22 (2): 133-137.
Myers, K.A. and Farquhar, D.R.E. (1998): Improving the accuracy of death
certification. C.M.A.J., 158(10): 13171323.
Nowels, D. (2004): Completing and signing the death certificate. Am. Fam.
Physician, 70(9): 1813-1817.
Pain, C.H.; Aylin, P.; Taub, N.A. and et al. (1996): Death certification:
production and evaluation of a training video. Med. Educ., 30: 434-
439.
Pandya, H.; Bose, N.; Shah, R. and et al. (2009): Educational intervention
to improve death certification at a teaching hospital. Natl. Med. J.
India, 22(6): 317-319.
Pritt, B.S.; Hardin, N.J.; Richmond, J.A. and et al. (2005): Death
certificate errors at an academic institution. Arch. Pathol. Lab. Med.,
129: 1476-1479.
Selinger, C.; Ellis, R. and Harrington, M. (2007): A good death certificate:
improved performance by simple educational measures. Postgrad
Med. J. 83:285-286.
Sington, J.D. and Cottrell, B.J. (2002): Analysis of the sensitivity of death
certificates in 440 hospital deaths: a comparison with necropsy
findings. J. Clin. Pathol., 55: 499-502.
Slater, D.N. (1993): Certifying the cause of death: an audit of wording
inaccuracies. J. Clin. Pathol., 46: 232234.
Swift, B. and West, K. (2002): Death certification: an audit of practice
entering the 21st century. J. Clin. Pathol., 55 (4): 275279.
Tormey, W.P. (2011): Accuracy of death certification. B. J. Psych., 199:
432-433.
Villar, J. and Perez-Mendez, L. (2007): Evaluating and educational
intervention to improve the accuracy of death certification among
trainees from various specialties. B.M.C. Health Serv. Res., 7:183.
Wexelman, B.A.; Eden, E. and Rose, K.M. (2013): Survey of New York
City resident physicians on cause-of- death reporting. Prev. Chronic
Dis., 10: 120288.
World Health Organization (WHO) (1993): International statistical
classification of diseases and related health problems. 10th Rev.,
Vol. 2, Geneva.

22 | P a g e


, :
()4102
*
. * , .
*
, ,

.

" " .
(
) , , .
743 ( 73) ,
3104 ,2014

. ( ) ,
1 5 . )0( :
( ) , ( )47.4 ( ),
.
( )40.0
( ) .
,
,
.

23 | P a g e