To assess the relative importance of ancillary properties (ie, intrinsic sympathomimetic activity (ISA), beta 1-selectivity, membrane stabilizing activity, and lipophilicity) in the effect of beta-blockers on mortality, a meta-analysis of all available secondary and primary prevention trials was performed. Seventy-one trials evaluating the effect on mortality after myocardial infarction (MI) were identified. The overall relative risk (RR) of mortality during beta-blocker treatment versus placebo was 0.89 (95% confidence interval [CI] 0.84-0.93), with a trend according to time of intervention: very early intervention RR 0.94 (95% CI 0.87-1.01), early intervention RR 0.91 (95% CI 0.81-1.01), and late intervention RR 0.80 (95% CI 0.73-0.88). Results were similar or even more marked for the end points one-week mortality, reinfarction, and sudden death. beta 1-selectivity, lipophilicity, absence of membrane stabilizing property, and absence of ISA were associated with a greater risk reduction compared with beta-blockers with the opposite ancillary property. When the effect of the three most frequently used beta-blocking drugs (atenolol, metoprolol, and propranolol) were compared, the drug with the combination of ancillary properties showing the most pronounced beneficial effects (metoprolol) had the most marked effect on survival. A similar trend was observed when the five published primary prevention trials comparing beta-blockers and diuretics in patients with hypertension were considered, but the number of studies was too low to allow for definite conclusions. We conclude that beta-blockade after MI leads to a substantial reduction in mortality. The so-called 'class-effect' of beta-blockers, however, can be questioned, because ancillary properties appear to play an important role in the efficacy of these drugs.