Aim: To describe predictors of death after hospital discharge during 5 years of follow-up in a consecutive series of patients surviving hospitalization for symptoms and signs of a confirmed or suspected acute coronary syndrome. Patients and methods: All patients who between February 15, 1986 and November 9, 1987, were hospitalized at Sahlgrenska University Hospital in Göteborg, Sweden, and fulfilled the above given criteria. Results: In all, 1948 patients were included of whom 731 (38%) had a confirmed acute myocardial infarction (AMI). Independent risk indicators for death were: age (P=0.0001); male sex (P=0.005); a history of previous AMI (P=0.0001), diabetes mellitus (P=0.003) and smoking (P=0.0001); development of AMI during first 3 days in hospital (P=0.0001); in-hospital signs of congestive heart failure (P=0.0001); prescription of digitalis (P=0.001) and diuretics (P=0.02) at hospital discharge. A history of smoking interacted significantly (P=0.02) with the relationship between development of AMI and prognosis. Thus, the difference between patients who did and who did not develop an AMI was more pronounced among non-smokers than smokers. Other factors which interacted significantly with this relationship were a history of angina pectoris, and development of ventricular fibrillation and hypotension while in hospital. Conclusion: Among hospital survivors of a confirmed or suspected acute coronary syndrome predictors of death during 5 years were: age, male sex, history of AMI, diabetes mellitus and smoking, development of AMI and congestive heart failure while in hospital and prescription of digitalis and diuretics at hospital discharge. A history of smoking and angina pectoris as well as development of hypotension and ventricular fibrillation while in hospital interacted significantly with the relationship between development of AMI and prognosis.