Journal of Public Health Medicine DOI: 10.1093/pubmed/fdg014
Vol. 25, No. 1, pp. 69–71 Printed in Great Britain
Multiple-cause coding of death from myocardial infarction: population-based study of trends in death certificate data Michael J. Goldacre, Stephen E. Roberts and Myfanwy Griffith
Abstract Background Data on long-term trends in mortality are generally unavailable for multiple-cause coding of deaths. We wanted to know whether multiple-cause coding of deaths for myocardial infarction contributes much to the interpretation of death certificate data on mortality rates for this condition. Methods We analysed all causes of death on death certificates in the former Oxford health service region from 1979 to 1998. Results Of 69 333 death certificates that included myocardial infarction as a cause of death, it was the underlying cause of death in 93.6 per cent. The ratio of ‘mentions’ to ‘underlying cause’ was broadly similar over the study period, during which time there were substantial falls in mortality rates. There were significant changes to the ratios, associated with timing of changes to coding rules; but their effects were small. The ratio of mentions to underlying cause was similar in men and women and in different age groups. Conclusion The underlying cause of death was a robust and almost complete measure of certified deaths for myocardial infarction. Keywords: myocardial infarction, underlying cause of death, multiple-cause coding, trends
Introduction Until recently, in England and in most other countries, only the underlying cause of death has been routinely coded and analysed in national mortality statistics. This means that other diagnostic information on death certificates, regarded by the certifying clinicians as relevant to the death, has been discarded. Furthermore, the rules and methods for selecting the underlying cause of death in England have changed twice in recent years – in 1983 to implement rule 3 of the International Classification of Diseases, and in 1993 with the introduction by the Office for National Statistics of automatic encoding software.1,2 We used a dataset with multiple-cause coding of deaths from 1979 to 1998 to determine whether underlying-cause coding, alone, captured the great majority of deaths attributed to myocardial infarction (MI) over this period; to study trends over time in standardized mortality rates from MI comparing underlying
cause and all ‘mentions’ on the death certificate; and to study sex differences and variation by age in the ratios of mentions to underlying cause.
Methods All causes of death recorded on death certificates were routinely coded in the Oxford record linkage study, which covered the former Oxford National Health Service Region from 1979 to 1998 (population 2.5 million). In describing all causes of death on the death certificate, for simplicity and following conventions, we have used the terms ‘mentions’ to denote a count of all death certificates on which MI was included, and ‘underlying cause’ where MI was recorded as the underlying cause. We compared trends in mortality rates where MI (ICD-9 code 410) was recorded as the underlying cause of death or in other positions on the death certificate. For studying trends over time, mortality rates were standardized by the direct method to the standard European population. We subdivided the deaths into those that, according to the death certificate, occurred in hospital and those that occurred outside hospital. We calculated ratios of mentions to underlying cause of these in comparing certification annually over time, for males and females, for quinquennial age groups (for people aged 35 years and over), and according to whether deaths occurred in hospital or outside.
Results There were 69 333 death certificates that included MI as a certified cause of death. Of these, 64 928 (93.6 per cent) had MI as the underlying cause of death. The population-based standardized
Michael J. Goldacre, director Stephen E. Roberts, statistician Myfanwy Griffith, computer programmer Unit of Health-Care Epidemiology, Department of Public Health, Institute of Health Sciences, University of Oxford, Old Road, Oxford OX3 7LF. Address correspondence to Professor M. J. Goldacre. E-mail:
[email protected]
© Faculty of Public Health Medicine 2003; all rights reserved.
70
JOURNAL OF PUBLIC HEALTH MEDICINE
(a)
Deaths in hospital and deaths outside
Standardized mortality rate (per 100 000)
120 Died outside hospital - all mentions Died outside hospital underlying cause Died in hospital - all mentions
100
Died in hospital - underlying cause 80
60
40
20
0 1978
1980
1982
1984
1986
1988
1990
1992
1994
1996
1998
Year
(b)
Men and women
Standardized mortality rate (per 100 000)
250 Men - all mentions Men - underlying cause Women - all mentions Women - underlying cause 200
150
100
50
0 1978
1980
1982
1984
1986
1988
1990
1992
1994
1996
1998
Year
Figure Standardized mortality rates for myocardial infarction (1979–1998) certified as the underlying cause of death and all mentions in any position on the death certificate: (a) for deaths in hospital and deaths outside; (b) for men and women.
CODING OF DEATH FROM MYOCARDIAL INFARCTION
mortality rate for MI as the underlying cause of death declined from 154 per 100 000 population (95 per cent confidence interval (CI) 149–158) in 1979 to 70 per 100 000 (67–73) in 1998, and the mortality rate for MI recorded in any position on the death certificate fell from 165 per 100 000 (160–170) to 77 (74–80) (Figure). In the period 1979–1982, before the rule 3 change, the ratio of mentions to underlying cause was 1.052; from 1983 to 1992 it was 1.066; and from 1993 to 1998, after the second change to methods of selecting underlying cause, it was 1.088. The ratios for each period are based on very large numbers and the changes between time periods were significant (2(1) 32, comparing 1979–1982 and 1983–1992; 2(1) 54, comparing 1983–1992 and 1993–1998). Assuming that changes in selection of underlying cause account for the changes in the ratios, their effect was slightly to exaggerate the decline seen in mortality rates from data based on underlying cause alone. However, the effect was small. There was a larger decline in mortality rates from deaths occurring outside hospital than for deaths in hospital (Figure). Over the 20 year period the ratio was slightly higher for deaths occurring in hospital (1.092) compared with those outside (1.047). The ratio of mentions to underlying cause across the 20 year period was similar between quinquennial age groups. There was no particular gradient across ages and the ratio varied between a high of 1.078 for those aged 35–39 and a low of 1.055 for those aged 50–54. The ratio was also very similar for men (1.065) and women (1.072) (see the Figure).
Discussion The large reduction in mortality rates from MI in recent decades is well documented,3,4 but little is known about trends in multiple-cause coded death rates. Our data show that the use of underlying cause of death, as distinct from the analysis of all certified causes, is robust for MI. The ratio of mentions to underlying cause varied a little over time, probably because of
71
changes to the procedures for selecting the underlying cause of death. However, the changes were small and the addition of data on mentions adds little to the overall profile of the substantial decline in mortality. Their addition also adds little to the profiles of mortality comparing men and women and comparing age groups. None the less, given the increasing availability of multiple-cause coding in national statistics nowadays – available nationally in England since 19932 – it would be prudent to check periodically that this is invariably so.
Acknowledgements S.R. receives funding from the Department of Health as part of its funding for the National Centre for Health Outcomes Development; the views expressed in this paper are those of the authors and not necessarily those of the Department of Health. The Unit of Health-Care Epidemiology is funded by the South East Regional Office of the NHS Executive. There are no competing interests.
References 1 World Health Organization. International Classification of Diseases. Manual of the international, statistical classification of diseases, injuries and causes of death, Vol. 1. Ninth revision. Geneva: WHO, 1977: 702–706. 2 Office for National Statistics. Mortality statistics, series DH2 no. 21. Cause 1993 (revised) and 1994. London: HMSO, 1996: v–viii. 3 Tunstall-Pedoe H, Kuulasmaa K, Mahonen M, et al. Contribution of trends in survival and coronary-event rates to changes in coronary heart disease mortality: 10-year results from 37 WHO MONICA project populations. Monitoring trends and determinants in cardiovascular disease. Lancet 1999; 353: 1547–1557. 4 Tunstall-Pedoe H, Vanuzzo D, Hobbs M, et al. Estimation of contribution of changes in coronary care to improving survival, event rates, and coronary heart disease mortality across the WHO MONICA Project populations. Lancet 2000; 355: 688–700.
Accepted on 16 September 2002
