Could the Coronary Heart Disease Mortality Rates Decline Be Artefactual?Author(s): RÉMI GUIBERTSource: Canadian Journal of Public Health / Revue Canadienne de Sante'e Publique, Vol. 82, No.1 (January/February 1991), pp. 43-45Published by: Canadian Public Health AssociationStable URL: http://www.jstor.org/stable/41989984 .

Accessed: 16/06/2014 01:28

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

.JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact support@jstor.org.

.

Canadian Public Health Association is collaborating with JSTOR to digitize, preserve and extend access toCanadian Journal of Public Health / Revue Canadienne de Sante'e Publique.

http://www.jstor.org

This content downloaded from 195.78.109.162 on Mon, 16 Jun 2014 01:28:27 AMAll use subject to JSTOR Terms and Conditions

Could the Coronary Heart Disease

Mortality Rates Decline Be Artefactual?

RÉMI GUIBERT, M.D., M.Sc.1

The recent decline in Coronary Heart Disease (CHD) mortality rates has been attributed to reduction in risk factors and to improved management . In this article , we review whether artefacts of classification could have played a role as well. Knowledge and information on disease , competing causes of death , death certification accuracy and completeness , advancing age of the population with multiple conditions as well as death certificate coding practice could have affected secular trends of CHD mortality rates. However , the lack of noticeable shifts within the cardiovascular category or with another category makes it difficult for the CHD decline to be artefactual.

Le déclin de la mortalité par maladies coronariennes observé depuis le début des années soixante est attribué à une réduction des facteurs de risque et à une amélioration des traitements . Nous examinons si des artefacts de classification pourraient aussi y avoir contribué.

La connaissance de la maladie , les causes de décès alternatives , la validité et la précision de la certification des décès , le vieillissement croissant de la population avec des conditions pathologiques multiples , ainsi que la codification des certificats de décès auraient pu affecter la tendance séculaire des taux de mortalité par maladie coronarienne. Cependant , une absence de transfert important soit vers d'autres pathologies de la même catégorie , soit vers d'autres catégories de maladies rend les artefacts de classification une explication peu plausible du déclin de la mortalité par maladies coronariennes.

Throughout including the the

USA 1940s

and and Canada,

the 1950s, were considered

some countries, to be including the USA and Canada, were considered to be

experiencing Coronary Heart Disease (CHD) epidemics. In the 1960s, CHD mortality began to decline unexpectedly among some populations and has continued to do so.1 In Canada, between 1969 and 1982, CHD mortality declined by 27% for men and by 31% for women2 (Figure 1).

Population reduction of risk factors and improved therapeutics3 support the general consensus that both reduction of incidence and improvement of case fatality rates contributed largely to the CHD mortality decline.4

But could the CHD mortality rate decline be artefactual? Despite the impressive recent decline in mortality rates, CHD remains the leading cause of death and a major part of

1. Assistant Professor, Family Medicine Department, Université de Montréal and Département de santé communautaire, Cité de la Santé, Ville de Laval, Québec Address for correspondence and for reprint requests: Dr. Rémi Guibert, Groupe de

recherche interdisciplinaire en santé, Faculté de médecine, Université de Montréal, C.P. 6128, succursale "A", Montréal (Québec), H3C 3J7. This research project was supported by National Health Research and Development Project 6605-2502-47.

total health care costs in North America today, so an answer to that question is of interest. We discuss some aspects of misclassification leading to possible artefactual CHD mortality rate variation.

CHD CLASSIFICATION ARTEFACTS

Knowledge of the disease Increased knowledge of CHD stimulated technologic

development and geographic diffusion of diagnostic and monitoring instruments, reinforcing the clinician's ability to classify CHD events. Therefore, a patient who was "correctly diagnosed" may turn out to be misclassified in light of new knowledge at a later point in time. Also, with increased awareness of the disease, individuals consult earlier for their symptoms, giving them a better documented medical history, therefore a better classification.

Competing causes of death An elevation in competing causes of death could have led

to an artefactual reduction in CHD mortality. However,

Canadian Journal of Public Health Vol. 82, January/ February 1991 43

This content downloaded from 195.78.109.162 on Mon, 16 Jun 2014 01:28:27 AMAll use subject to JSTOR Terms and Conditions

Figure 1. Coronary Heart Disease Mortality Rates. Age- standardized, all ages, by sex, Canada, 1951-1982.

mortality rates between 1969 and 1982 for other cardiovascular diseases (CVD) categories declined as well, by 38% for cerebrovascular diseases, and by 25% to 30% for all CVD.2 Thus the CHD decline cannot be explained by a coding shift to other CVD.5

CHD decline cannot be explained either by a coding shift to another cause of death since mortality rates for most other causes of death declined as well, the major exception being the increase of cancers. A coding shift from CHD to malignant neoplasms would be hard to imagine5 6 (Figure 2).

Death certification For each death, only one condition must be reported as

the "underlying cause of death". With today's increased life expectancy, people are likely to die at an advanced age with multiple conditions. Therefore, in the presence of another important condition, CHD may not be chosen as the underlying cause of death, even if CHD partially contributed to the death of the individual.

Figure 2. Deaths by cause as percentage of total deaths in the United States, 1968-1982. Causes of death for 1968 classified by the eighth revision, and for 1982 by the ninth revision of the International Classification of Diseases adapted for use in the United States. Data for 1982 are provisional; for 1968, final.

The death certificate has been recognized as an inaccurate record of the cause of death7 89 when compared with autopsy reports81011 or with hospital records.121314 When compared with autopsy reports, death certificates for the circulatory system showed a net overestimate of 7%8 and of 10%, 10 increasing with age, up to 50% over 75 years of age.13 Comparing death certificates and hospital records shows similar discrepancies: concordance declining with patient age and length of stay.121314 Gittelsohn12 showed that concordance between death certificates and clinical information varied little for the period 1969-1975.

Death certifícate coding Variation in mortality rates can also be due to: 1) the

International Classification of Diseases (ICD) system, 2) the nosologist (competency, preferences), and 3) key entry errors.15

The Ninth Revisions of the ICD did not affect the coding of Acute Myocardial Infarction (AMI) (ICD-9 No. 410), with a comparability ratio 0.99 to 1.01 (95% confidence limits).16 However, the comparability ratio was .88 for all CHD and .75 for chronic CHD.17

With the Ninth Revision, the separation of CVD Unspecified (ICD-9 No. 429.2) accounts for a large part of the 12% reduction of the overall CHD mortality rates observed in 1979 in the United States.16 This difference is partially explained by the medical examiner preferences for generalized cardiovascular terms rather than specific terms.17 For unknown reasons, the CHD mortality rates data from Statistics Canada do not show that difference in the yearly trend for Canada or for Canadian regions.

Inter-nosologist variation occurred, 86.5% agreement rates for different causes of death (three-digit ICDA code agreement among three nosologists).18 Intra-nosologist agreement varied from 94.6%18 to 98. 8%19 for CHD.

In Canada, because there are only one or two nosologists per province, coding differences among provinces may arise. Any important systematic differences between them would lead to apparent important differences in mortality statistics between provinces. A quality assessment study reported an overall 7.18% coding error rate for "Cause of Death" for the years 1979-80 varying from 2.2% to 18.1% between provinces.15 Our study showed no statistically significant differences in error rates of death certificate coding between the years 1970 and 1984 for two Canadian provinces for AMI.19

DISCUSSION

It is difficult to ascertain to what extent population knowledge of the disease affected the CHD mortality rates. Given the widespread awareness and knowledge of the disease over the last two decades, better knowledge and diagnostic procedures would have reduced the false positive rates especially in the increasing over-75 age group, therefore artificially reducing CHD mortality rates.

44 Canadian Journal of Public Health Vol. 82

This content downloaded from 195.78.109.162 on Mon, 16 Jun 2014 01:28:27 AMAll use subject to JSTOR Terms and Conditions

Figure 3. Ischemic Heart Disease 1931-1980. Annual change in life expectancy at age 35 due to acute ischemic heart disease, 1931-1980 (Data for ICDA revision years not shown).

Competing causes of death cannot explain the CHD mortality decline; in fact, after controlling for changes in competing causes of death and for ICD revisions, a .7-year increase in life expectancy at birth could be attributed to the reduction of death from Acute Myocardial Infarction (AMI) in the 60s and the 70s20 (Figure 3).

At the death certification level, multiple conditions could artificially reduce the CHD mortality rates; the increasing over-75 age group would favour an artificial increase.

ICD revisions occur at a point in time and could not explain the continuous and gradual decline in CHD mortality rates since the 60s. Even if death certificate coding error rates are important, they did not vary for AMI between 1970 and 1984.

The patterns of decline for CHD mortality have shown national and regional inconsistencies, in particular in the USA and Canada.215 The declines also varied by gender, age, ethnicity, and socio-economic status. For example, the CHD mortality decline was particularly pronounced for young, white males of higher socio-economic status.5 The geographic and chronologic patterns of onset of the decline in CHD mortality rates appear to be more similar to other patterns of social changes (diet, smoking) and shifts in productive economic activities than to diffusion of medical care.9 These inconsistencies in the patterns of decline do not favour artefacts of classification. Which phenomenon could explain misclassification which would be specific to a given age group or gender or socio-economic status?

Secular trends in disease diagnosis, death certification and coding practices could have affected, to some extent, CHD mortality rates, however insufficiently to bring any doubt about the reality of the CHD mortality rate decline phenomenon.

REFERENCES 1. Walker WJ. Coronary mortality. What is going on? JAMA 1974; 227:

1045-6. 2. Nicholls E, Nair C. Cardiovascular diseases in Canada. Health and

Welfare Canada 1986 , Catalogue 82-544. 3. Walker WJ. Changing U.S. life style and declining vascular mortality:

a retrospective. N Engl J Med 1983; 308 (11): 649-51. 4. Stamler J. The marked decline in coronary heart disease mortality

rates in the United States, 1968-1981: Summary of findings and possible explanations. Cardiology 1985; 72: 11-12.

5. Stern MP. The recent decline in ischemic heart disease mortality. Ann Int Med 1979; 91: 630-40.

6. Pyorala K, Epstein FH, Kornitzer M. Changing trends in coronary heart disease mortality: possible explanations. Cardiology 1985; 72: 5-10.

7. Medical Services Study Group of the Royal College ol Physicians oí London. Death certification and epidemiological research. Brit Med J 1978; 2: 1063-5.

8. Kircher T, Nelson J, Burdo H. The autopsy as a measure of accuracy of the death certificate. N Engl J Med 1985; 313 (20): 1263-9.

9. Glasser JH. The quality and utility of death certificate data. PH 1981; 71(3): 231-3.

10. Engel LW, Sträuchen JA, Chiazze L et al. Accuracy of death certification in an autopsied population with specific attention to malignant neoplasms and vascular diseases. Am J Epidemiol 1980; 1 1 1 (1): 99-112.

11. Cameron HM, McGoodan E. A prospective study of 1152 hospital autopsies: inaccuracies in death certification. Brit J Path 1981; 133: 273-83.

12. Gitelsohn A, Senning J. Studies on the reliability of vital and health records: 1. Comparison of cause of death and hospital record diagnoses. Am J Public Health 1979; 69 (7): 680-9.

13. Shekelle RB, Shryock A, Paul O et al. Diet, serum cholesterol, and death from coronary heart disease. N Engl J Med 1981; 304 (2): 65-9.

14. Heliovaara W, Reunanen A, Aromaa A et al. Validity ol hospital discharge data in a prospective epidemiological study on stroke and myocardial infarction. Acta Med Scand 1984; 216: 309-15.

15. Nagmur D, Curne S, Taylor E et al. Quality Assessment Study ot Cause of Death Coding Data Years 1979 and 1980. Vital Statistics and Disease Registries Section, Health Division, Statistics Canada 1981.

16. National Center for Health Statistics. Estimates of selected comparability ratios based on dual coding of 1976 death certificates by the Eighth and Ninth Revisions of the International Classification of Diseases. Monthly Vital Statistics Report 1980; 28 (11): 12.

17. Sorlie P, Gold E. The effect of physician terminology preference on coronary heart disease mortality: an artefact uncovered by the Ninth Revision ICD. Am J Public Health 1987; 77 (2): 148-52.

18. Curb JD, Babcock C, Pressel S et al. Nosological coding of cause of death. Am J Epidemiol 1983; 118 (1): 122-8.

19. Guibert RL, Wigle DT, Williams JI. Decline of acute myocardial infarction not due to cause of death coding. Can J Public Health 1989; 80(6): 418-22.

20. Smith DP, Slater C. The pattern of ischemic heart disease 1931-1980. Statistics in Medicine 1985; 4: 397-412.

Received: November 18, 1988 Accepted: February 22, 1989

January/February 1991 45

This content downloaded from 195.78.109.162 on Mon, 16 Jun 2014 01:28:27 AMAll use subject to JSTOR Terms and Conditions