[Usefulness of the statistical bulletin of deaths to identify extrahospital deaths in the context of a myocardial infarction population registry]

This paper aimed to study the usefulness of the Statistical Bulletin of Deaths (SBD) for identifying extrahospital deaths due to acute coronary ischaemia, and to determine the most efficient strategy in the selection of death causes that appear in the SBD, which are the most efficient for the identification.Those extrahospital deaths, which, among the causes of death, recorded a diagnostic code indicating that they might have been caused by coronary ischaemia, were included. To study the usefulness of the death certificate we calculated the sensitivity and the positive predictive value of cardiac ischaemia. To determine the most efficient strategy for selecting causes of death we compared two selection strategies: the first, using only the basic cause of death; and the second using all the causes appearing in the SBD.Of the 395 SBD selected, 161 were classed as acute heart attacks. In those SBD in which cardiac ischaemic disease was given as the basic cause of death, we obtained a sensitivity of 82.6% (CI 95%: 75.9-88.1) and a positive predictive value of 72.7% (CI 95%: 65.6-79). The most efficient strategy in SBD selection proved to be the investigation of death certificates in which cardiac ischaemia appeared as one of all the causes of death, and death certificates in which the basic cause of death was coded as diabetes mellitus, essential arterial hypertension, hypertensive heart disease, cardiac dysrhythmia, and cardiac insufficiency.The information provided by death certificates for extrahospital deaths due to coronary ischaemia is reliable. A sensitive and efficient SBD selection strategy is proposed for the detection of cases.