rss
Postgrad Med J 88:690-693 doi:10.1136/postgradmedj-2012-130912
  • Original article

Reporting of incorrect cause-of-death causal sequence on death certificates in the USA: using hypertension and diabetes as an educational illustration

  1. Ichiro Kawachi6
  1. 1Department of Medical Record and Information Management, Occupational Medicine and Neurology, Chi Mei Medical Center, Tainan, Taiwan
  2. 2Department of Occupational Safety, College of Environment, Chia Nan University of Pharmacy and Science, Tainan, Taiwan
  3. 3Department of Internal Medicine, Chi Mei Medical Center, Tainan, Taiwan
  4. 4Department of Leisure, Recreation and Tourism Management, Southern Tainan University of Science and Technology, Tainan, Taiwan
  5. 5NCKU Research Center for Health Data and Institute of Public Health, College of Medicine, National Cheng Kung University, Tainan, Taiwan
  6. 6Department of Society, Human Development and Health, Harvard School of Public Health, Harvard University, Boston, Massachusetts, USA
  1. Correspondence to Dr Tsung-Hsueh Lu, NCKU Research Center for Health Data and Institute of Public Health, College of Medicine, National Cheng Kung University, No 1, Dah Hsueh Road, Tainan City, Tainan 701, Taiwan; robertlu{at}mail.ncku.edu.tw
  • Received 29 February 2012
  • Accepted 14 August 2012
  • Published Online First 18 September 2012

Abstract

Background Little is known about the extent of reporting an incorrect cause-of-death (COD) causal sequence on death certificates.

Objective To determine the frequency of incorrect reporting of hypertension as cause of diabetes on death certificates in the USA.

Methods Multiple-cause mortality files were used to identify death certificates which mentioned both hypertension and diabetes in the USA from 1985 to 2005. The frequency of reporting hypertension on the line below diabetes in part I of the death certificate was calculated.

Results The percentage of cases in which both hypertension and diabetes were included in part I of the death certificate, in which hypertension was reported on the line below diabetes on the death certificate—that is, suggesting that hypertension was a cause of diabetes—increased from 15.5% in 1985 to 36.1% in 2000 and 38.2% in 2005.

Conclusions The frequency of reporting of an incorrect COD causal sequence on death certificates in the USA has increased. Education, training and questioning the opinions of certifying physicians are needed to improve the quality of reporting of COD statements.

Introduction

Cause-of-death (COD) mortality data are commonly used for medical and epidemiological research and an important reference for public health policy making. The tabulation of COD mortality data is based on the underlying COD, which has been defined as the disease or injury that initiated the train of morbid events leading directly to death by the WHO in the International Statistical Classification of Diseases and Related Health Problems (ICD) (p 23).1 To ensure comparability between countries, a standard international form of medical certificate was set in the ICD (figure 1) (p 24).1 One important requirement in reporting COD statements in part I of the death certificate is to report acceptable COD causal sequence between lines a, b, c and d. For example, diabetes (reported on line c) resulted in acute myocardial infarction (reported on line b) and then resulted in congestive heart failure (line a) is an acceptable COD causal sequence (example 2 in box 1).

Box 1

Given the same deceased person, different certifying physicians might report different cause-of-death (COD) statements on the death certificate

In example 1, the first physician thought that hypertension and diabetes contributed to death but did not directly result in acute myocardial infarction (AMI), therefore reported hypertension and diabetes in part II of the death certificate.

In example 2, the second physician believed that the occurrence of AMI was due to long-term poor control of diabetes, hence reported diabetes on the line below AMI.

In example 3, the third physician thought that both diabetes and hypertension were risk factors of AMI, thus reported diabetes on line c and hypertension on line d. This is an incorrect COD causal sequence because hypertension could not result in diabetes according to current medical knowledge.

Example 1

Part I

  • Congestive heart failure

  • Acute myocardial infarction Fx1

Part II

Hypertension, diabetes

Example 2

Part I

  • Congestive heart failure

  • Acute myocardial infarction Fx1

  • Diabetes mellitus Fx1

Part II Hypertension

Example 3

Part I

  • Congestive heart failure

  • Acute myocardial infarction Fx1

  • Diabetes mellitus Fx1

  • Hypertension Fx2

✓ Acceptable COD causal sequence.

× Unacceptable COD causal sequence.

Figure 1

The international form of medical certificate.

However, studies assessing the quality of COD statements suggested that some certifying physicians did not follow the instructions given on the death certificate and reported an incorrect COD causal sequence between lines in part I of the death certificate.2–9 For example, according to current medical knowledge, hypertension could not result in diabetes, therefore the reported COD causal sequence is incorrect in example 3 in box 1—that is, hypertension was incorrectly reported (on the line below) as a cause of diabetes.

Objectives

Most studies examining the quality of COD statements have been hospital-based and cross-sectional.2–9 Little is known about the magnitude of the problem nationally in reporting an incorrect COD causal sequence on death certificates and whether the extent of the problem has increased or decreased during past decades.

As the prevalence of both hypertension and diabetes is increasing,10 ,11 physicians are expecting to issue more death certificates for decedents with both hypertension and diabetes. Both hypertension and diabetes could initiate cardiovascular morbid events causing death and it is sometimes difficult for the certifying physician to determine the single main COD causal sequence.12 Using hypertension and diabetes as an educational illustration, this US nationwide population-based study aimed to determine the changes in the frequency of reporting of

  1. both hypertension and diabetes on the death certificate;

  2. both hypertension and diabetes in part I of the death certificate;

  3. hypertension as a ‘cause’ (on the line below) of diabetes in part I of the death certificate from 1985 to 2005 in the USA.

Methods

Source of data

Multiple-cause mortality files compiled by the National Center for Health Statistics13 were used to identify deaths in which both hypertension and diabetes were mentioned on death certificates for the years 1985, 1990, 1995, 2000 and 2005. The ICD-9 codes were used for the years 1985, 1990 and 1995 and the ICD-10 codes were used for the years 2000 and 2005 in the USA. For hypertension, the ICD-9 codes were 401–404 and the ICD-10 codes were I10–I15. For diabetes, the ICD-9 code was 250 and the ICD-10 codes were E10–E14.

Identifying the relative positions of hypertension and diabetes

The entity-axis format records in the multiple-causes mortality files provide the ICD-9 and ICD-10 codes and also information about the exact position—that is, the first or second or third diagnosis on lines a or b or c or d in part I or part II of the death certificate, reported by the certifying physician. We used this information to identify cases in which both hypertension and diabetes were reported in part I of the death certificate and cases in which hypertension was reported on the line below diabetes. The detailed algorithm used to determine the relative position of the different diagnoses was described in a previous study.14

Analysis

We calculated three kinds of percentages for each study year: (1) the percentage of deaths in which both hypertension and diabetes were reported on the death certificate among all deaths; (2) the percentage of deaths in which both hypertension and diabetes were reported in part I of the death certificate among deaths in which both hypertension and diabetes were mentioned; (3) the percentage of deaths in which hypertension was reported on the line below diabetes in part I of the death certificate among deaths in which both diabetes and hypertension were reported in part I of the death certificate. To ensure comparability across years, the percentages were age-adjusted (0–44, 45–64, 65–74 and ≥75 years) using the age structure of the decedents in 2000 as the standard.

Results

Reporting of both hypertension and diabetes on the death certificate

The number of death certificates in which hypertension and diabetes were both reported increased between 1985 and 2000 and 2005 from 25 940 to 60 305 to 82 711. This equates to an increase in the inclusion of hypertension and diabetes as a proportion of all death certificates from 1.2% to 2.5% to 3.4% (table 1).

Table 1

Reporting of both hypertension and diabetes on death certificates in the USA, 1985–2005

Reporting of both hypertension and diabetes in part I of the death certificate

The percentage of cases in which both hypertension and diabetes were mentioned in the death certificate in which they were both included in part I increased from 20.9% in 1985 to 24.2% in 2000 and 24.7% in 2005 (table 1).

Reporting of hypertension on the line below diabetes

The percentage of cases in which both hypertension and diabetes were included in part I of the death certificate in which hypertension was reported on the line below diabetes on the death certificate—that is, suggesting that hypertension was a cause of diabetes—increased from 15.5% in 1985 to 36.1% in 2000 and 38.2% in 2005 (table 1).

Discussion

Main findings

We found an increase in US death certificates in which diabetes and hypertension were included on death certificates between 1985 and 2005. This is in keeping with known increases in prevalence of both these conditions. However, we found an alarming increase in incorrect recording of hypertension on the line below diabetes in part I of the death certificate—suggesting that many certifying physicians were unaware that diseases or conditions reported on lines a, b, c and d in part I of the death certificate should have an acceptable COD causal relationship. This implies that there is a need for education of certifying physicians in how to record underlying and coexisting and causal conditions on death certificates. Assuming that inaccurate recording of coexisting conditions is not confined to hypertension and diabetes, the impact of this lack of understanding of how to complete a death certificate would be to skew information essential in the use of COD mortality data.

Strengths and limitations

One of the strengths of this study was that it was a nationwide population-based study spanning a 25-year period. A second strength was the use of multiple-cause mortality file data, which provided information about the relative positions of diagnoses reported on the death certificate, information that is not available in traditional COD mortality files.

The main limitation of this study was that information about certifying physicians or hospitals was not available and therefore we were unable to further analyse the characteristics of the physicians or hospitals associated with a high frequency of incorrect reporting. The information is important for educational interventions.

Interpretation of the findings in relation to previous studies

As expected, the number and percentage of reports of both hypertension and diabetes on death certificates has increased continuously during the past 25 years, which is consistent with the increasing trends in the prevalence of hypertension and diabetes.10 ,11 However, we did not find a concomitant increase in the percentage of reports of both hypertension and diabetes in part I of the death certificate during the study period. In other words, most physicians still thought that hypertension and diabetes were distal causes of cardiovascular diseases which directly cause death and should be recorded in part II of the death certificate.

One possible explanation of the notable increase in the percentage of improper reporting of hypertension on the line below diabetes between 1985 and 1990, is the addition of a fourth line (line d) in part I in the 1989 revision of the US standard death certificate, which was adopted by all states.15 A previous study also revealed a compensatory decrease in the reporting of multiple conditions per line from 1985 to 1990.14 It is plausible that hypertension might often have been reported on the same line as diabetes before 1989 and was then shifted to the line below diabetes after 1989.

Diabetes often coexists with hypertension among elderly people. It is often difficult for certifying physicians to determine one major train of morbid events. A survey study indicated that different subspecialties of internists had different opinions about the reporting of diabetes as the underlying COD depending on whether the patient died from ischaemic heart disease, cerebrovascular disease or infectious disease.12 As indicated in box 1, given the same deceased person, different certifying physicians might have different opinions in determining which diseases or conditions should be included in part I of the death certificate.

Implications for physician education

It is reasonable to infer that the erroneous listing of hypertension on the line below diabetes in part I of the death certificate arises from a lack of awareness of the correct practice for COD certification, rather than any lack of understanding of the pathophysiology of hypertension and diabetes. Many physicians appear to ignore the instructions, such as the words ‘due to or as a consequence of’ immediately below lines a, b and c on the standard death certificate (figure 1).

A review of educational interventions targeted at improving the quality of COD certification suggested that printed educational material is the intervention with the least impact and interactive workshops are the most effective intervention.16 Another way to improve the quality of COD statistics is to question the certifier who reported an incorrect COD causal sequence on the death certificate to obtain clarification.17

One important step to avoid the reporting of an incorrect COD causal sequence in part I of the death certificate is to determine the major (and minor) train of morbid events leading directly (contributing) to death. Only disease or conditions in the major train should be included in part I of the death certificate and disease or conditions in the minor train should be recorded in part II of the death certificate. The determination of ONE major train of morbid events is based upon the attending physician's best judgement. Therefore, given the same deceased person, different certifying physicians might have different opinions about the major train of morbid events leading directly to death (box 2).

Box 2

Given the same deceased person, different certifying physicians might have different opinions about the major train of morbid events leading directly to death

According to the guideline set by the World Health Organisation,1 diseases, injuries or complications leading directly to death should be reported in part I of the death certificate. Other significant diseases or conditions contributing to death but not resulting in the underlying cause-of-death should be reported in part II of the death certificate.

The determination of ONE major train of morbid events is based upon the attending physician's best judgement. Therefore, given the same deceased person, different certifying physicians might have different opinions about the major train of morbid events leading directly to death

Example 1 Part I

  • Sepsis

  • Ascites

  • Liver cirrhosis

Part II

Minor train: chronic bronchitis→pneumonia↘

Major train: liver cirrhosis→ascites→sepsis→death

Minor train: diabetes→urinary tract infection↗

Example 2

Part I

  • Sepsis

  • Pneumonia

  • Chronic bronchitis

Part II

Diabetes and chronic bronchitis

Liver cirrhosis, diabetes

Minor train: liver cirrhosis→ascites→peritonitis↘

Major train: chronic bronchitis→pneumonia→sepsis→death

Minor train: diabetes→urinary tract infection↗

Main messages

  • The percentage of incorrect reporting of hypertension on the line below diabetes on death certificates increased from 15.5% in 1985 to 36.1% in 2000 and 38.2% in 2005.

  • Many certifying physicians completing cause-of-death statements were unaware that diseases or conditions reported on lines a, b, c and d in part I of the death certificate should have an acceptable causal relationship.

Current research questions

  • What is the magnitude of the problem in reporting an incorrect cause-of-death causal sequence on death certificate in the USA?

  • Did the size of problem increase or decrease during past decades in the USA?

Footnotes

  • Contributors T-JC and T-HL researched the data and wrote the manuscript. IK and C-YL reviewed/edited the manuscript and contributed to the discussion.

  • Funding This study was partially funded by the National Science Council of Taiwan (NSC98-2314-B-006-015-MY2), the Department of Health of Taiwan (99Z4001) and the Chi-Mei and National Cheng Kung University joint programme (CMNCKU9916). The funding sources had no role in the design and conduct of the study; the collection, management, analysis and interpretation of the data; or the preparation, review and approval of the manuscript.

  • Competing interests None.

  • Provenance and peer review Not commissioned; externally peer reviewed.

References


Free sample
This recent issue is free to all users to allow everyone the opportunity to see the full scope and typical content of PMJ.
View free sample issue >>

Don't forget to sign up for content alerts so you keep up to date with all the articles as they are published.