Original articleThe validity of the diagnosis of acute myocardial infarction in routine statistics: A comparison of mortality and hospital discharge data with the Danish MONICA registry
Introduction
The decline in coronary heart disease mortality in most industrialized countries during recent decades is of major significance for public health [1]. The mortality rate depends on the number of new cases and on the survival among the patients with the disease. The decline in mortality may thus either be caused by a decline in the incidence of coronary heart disease because of healthier lifestyles or improved social conditions in the population or from better survival as a result of better treatment or a changed spontaneous course of the disease [2].
Most countries have two main sources of routinely collected information on trends in heart disease: mortality statistics based on death certificates, and morbidity statistics based on hospital discharge data. Both data sources are based on classification and coding of the clinical diagnosis leading to death or hospital admission according to the International Classification of Diseases (ICD). These data are now increasingly used to compare incidence and mortality over time and among countries 1, 3, 4, 5, 6.
In Denmark, the Danish Registry of Causes of Death include all deaths since 1943 [7] and the National Patient Registry include all hospitalizations since 1978 [8]. These two nationwide registries now allow the analysis of mortality trends for more than 50 years and the morbidity (measured by hospitalization) for more than 20 years.
Most cases of symptomatic acute myocardial infarction (MI) lead to hospital admission or sudden death outside the hospital. Information about hospital discharges and deaths may be combined to identify events of MI by application of appropriate algorithm. This record linkage can be used to estimate the incidence of MI in the population if the registries are carefully validated.
The WHO MONICA (Monitoring Trends and Determinants in Cardiovascular Disease) Project provides a unique opportunity to validate routine mortality and morbidity statistics. The MONICA Project was a multicenter study monitoring the incidence of MI, levels of risk factors, and patterns of therapy over 10 years [2]. The study included 38 well-defined populations in 21 different countries, and all cases of MI and stroke over 10 years were registered according to standardized diagnostic criteria. A drawback of this study is that none of the centers had national coverage, and the degree to which the results can be generalized to the national level in each country is not obvious. In many countries, the routine statistics are nationwide but do not standardize the diagnostic criteria to the same degree as in the MONICA Project. It is, therefore, of great importance to utilize the MONICA data for evaluating the national registries.
Validation studies of routine statistics have been carried out over the years with somewhat heterogeneous results due to differences in methodology or reflecting true differences in the validity of the routinely collected data between countries. Some studies have been carried out comparing community registries with national statistics 9, 10. Other studies have been based on a comparison of routine mortality statistics and hospital statistics and data from the MONICA Project 11, 12. In general, the hospital statistics seem to be fairly valid and the mortality statistics somewhat less valid 4, 13, 14. These findings stress the importance of validating the national registries to determine whether and how they can be used to reflect “true” incidence and mortality.
Thus, this study analyzed the validity of the diagnosis of acute MI in the National Patient Registry and the Danish Registry of Causes of Death by individual-based comparison of the national registries and the registrations in the Danish MONICA study (DANMONICA). We also compared the estimated age-specific incidence of first MI based on routine statistics and on the MONICA registry.
Section snippets
The Danish Registry of Causes of Death
The Danish Registry of Causes of Death includes individual records of all deaths in Denmark since 1943. The record contains the name, personal identification (the unique CPR number assigned to all Danish citizens and other residents of Denmark since 1968), and information from the death certificate including birthplace, date, and place of death and classification of the underlying and contributory causes of death. During the period of the Danish MONICA (DANMONICA) study, the causes of deaths
Results
From 1982 to 1991, 5,539 MI admissions were registered in the National Patient Registry. Most cases (4,768 or 86%) had MI as the primary diagnosis. Record linkage to the DANMONICA registry identified 5,075 (92%) of the admissions with an attack date within 28 days before or after the admission date in the National Patient Registry, and a further 5% were found in the registry with another attack date. The admission could not be found in 3% of the cases.
In the same period, 2,378 deaths were
Discussion
This study showed an overall high agreement between the national routine statistics and the DANMONICA registry. Both the predictive value and the sensitivity were higher for admissions compared with mortality data. The use of MI as either the primary or secondary diagnosis or underlying or contributory cause of death led to a reasonable balance between the predictive value and sensitivity for both hospital admissions and fatal events. Extending the definition by including all deaths with other
Conclusion
In conclusion, we find the combination of the national registries to be a valid and powerful tool for monitoring cardiovascular risk in the general population.
References (17)
- et al.
Contribution of trends in survival and coronary-event rates to changes in coronary heart disease mortality10-year results from 37 WHO MONICA project populations. Monitoring trends and determinants in cardiovascular disease
Lancet
(1999) - et al.
Identification of non-fatal myocardial infarction through hospital discharge data in Western Australia
J Chronic Dis
(1987) - et al.
Estimation of myocardial infarction mortality from routinely collected data in Western Australia
J Chronic Dis
(1987) - et al.
Diagnosis of acute myocardial infarction by MONICA and FINMONICA diagnostic criteria in comparison with hospital discharge diagnosis
J Clin Epidemiol
(1994) International trends in coronary heart disease mortality and incidence rates
J Cardiovasc Risk
(1999)- et al.
Hospital discharge register data in the assessment of trends in acute myocardial infarction
FINMONICA AMI Register Study Team. Ann Med
(1995) - et al.
Attack rate, mortality and case fatality for acute myocardial infarction in Sweden during 1987–95. Results from the national AMI register in Sweden
J Intern Med
(2000) - et al.
Heart statistics
(1999)
Cited by (440)
Cause-specific mortality in patients with alcohol-related liver disease in Denmark: a population-based study
2023, The Lancet Gastroenterology and HepatologyEthnic disparities in out-of-hospital cardiac arrest: A population-based cohort study among adult Danish immigrants
2022, The Lancet Regional Health - EuropeBody mass index trajectories in childhood and incidence rates of type 2 diabetes and coronary heart disease in adulthood: A cohort study
2022, Diabetes Research and Clinical Practice