Objective: To develop a strategy for evaluating drug efficacy over time that accounts for heterogeneous clinical courses evolving after initiation of therapy and to demonstrate its use in assessing the long-term therapeutic benefit of propranolol after myocardial infarction. Design: Analysis of data from the Beta-Blocker in Heart Attack Trial (BHAT), a randomized, double-blind, placebo-controlled trial that enrolled patients from 1978 to 1980 and followed participants for vital status to April 1982. Setting: Thirty-one clinical centers in the United States and Canada. Patients: Eligible patients included 3297 men and women 30 to 69 years of age who survived 1 year after trial entry. Intervention: Patients were classified as being on treatment at 12 months after randomization if they were receiving beta-blocker therapy at the 12-month visit and off treatment if they were not receiving beta-blocker therapy at that time. Outcome Measure: Vital status evaluated at 720 days of follow-up. Results: A total of 2914 patients (88%) was classified as being at lower risk (strata I and II). For these patients, survival curves by treatment at 12 months were virtually indistinguishable. Among the 383 patients categorized as being at high risk on the basis of recurrent ischemic events, arrhythmias, congestive heart failure, or severe comorbidity during the first 12 months, the use of beta-blockers was associated with a 43% proportional decline in the subsequent risk for death (P = 0.01 by log-rank test). Conclusions: In patients who survived to 1 year with low- to moderate-risk clinical courses, beta-blocker therapy did not have long-term beneficial effect. In contrast, among patients who had a high-risk clinical course during the first year, beta-blockers significantly reduced mortality in the follow-up period.
