In a large population of patients (n = 3666) who were discharged from the hospital after acute myocardial infarction and followed up for 1 year, factors associated with recurrent nonfatal (n = 171) or fatal (n = 74) infarction were identified. Also, the effects of combining various end points (recurrent nonfatal or fatal infarction and other cardiac death) in multivariate analyses, a practice common in many small studies that evaluate the predictive-value of various treatments or special tests, was examined. In univariate analyses, patients with nonfatal recurrent infarction did not differ with respect to age or gender from infarct-free survivors, but they more often had a history of previous myocardial infarction, congestive heart failure, angina pectoris, and diabetes; more severe pulmonary congestion was present on chest x-ray during the admission, and a non-Q wave index infarction was more frequent. Patients with either a fatal or nonfatal recurrent infarction had more angina pectoris during follow-up (55% to 60%) compared with 27% in event-free survivors and 31% in patients who died of other cardiac causes in whom this factor could be assessed before death. In multivariate analyses, historical and clinical prognostic factors were ranked differently for fatal or nonfatal reinfarction and other cardiac causes of death; angina pectoris at follow-up was highly related to recurrent infarction (fatal or nonfatal), along with a history of diabetes, and a non-Q wave index infarction. These factors were not independently related to other causes of cardiac death.(ABSTRACT TRUNCATED AT 250 WORDS)