OBJECTIVE: To describe the change in the use of medication after acute myocardial infarction (AMI) and discuss its possible impact on risk and risk indicators for death. PATIENTS: All patients discharged alive after hospitalization for AMI at Sahlgrenska Hospital (covering half the community of Göteborg, i.e. 250,000 of 500,000 inhabitants) during 1986-1987 (period I) and at Sahlgrenska Hospital and Ostra Hospital (covering the whole community of Göteborg, 500,000 inhabitants) during 1990-1991 (period II). METHODS: Overall mortality was retrospectively evaluated during 5 years of follow-up. RESULTS: In all, 740 patients were included in the study during period I and 1448 during period II. The 5-year mortalities were 44.1% for period I patients and 39.3% for period II patients (P = 0.036). The relative risk of death for period II patients was 0.78 [95% confidence interval (CI) 0.67-0.89, P = 0.0005] after adjustment for differences at baseline. There was a significant interaction with a history of congestive heart failure; improvement in duration of survival was found only for patients without such a history. During period I, only 3% of patients were administered fibrinolytic agents, compared with 33% of patients during period II (P