We examined the effect of beta-blockers on outcome in patients enrolled in the Multicenter Study of Myocardial Ischemia. We started prospective enrollment of patients after an acute coronary event (acute myocardial infarction [MI] or unstable angina) in North America, Israel, and Japan from mid-1988, and followed-up the 1042 patients for 6-43 months (average, 26 months); 47.2% of the patients were receiving beta-blockers at baseline. Overall, beta-blockers were associated with neutral effects on the chronic outcome: the cardiac event (cardiac death or recurrent nonfatal MI) rate was 6.8% for patients on beta-blockers and 7.5% for those who were not on beta-blockers (P = 0.72). However, a significant interaction (P = 0.018) was found between event type and the use of beta-blockers in regard to the cardiac event rate. That is, beta-blockers were associated with a lower risk of the cardiac event for patients whose index event type was MI (risk ratio, 0.59) in contrast to those whose index event was unstable angina (risk ratio, 1.92). The chronic use of beta-blockers in patients recovered from unstable angina may be associated with an increase in risk.