To investigate the efficacy of long term oral anticoagulant treatment in subgroups of patients after myocardial infarction. Analysis of the effect of anticoagulant treatment in subgroups of hospital survivors of myocardial infarction based upon age, gender, history of hypertension, previous myocardial infarction, smoking habits, diabetes mellitus, Killip class, anterior location of infarction, thrombolytic therapy, and use of beta blockers. Participants of a multicentre, randomised, double blind, placebo controlled trial that assessed the effect of oral anticoagulant treatment on mortality as well as cerebrovascular and cardiovascular morbidity in 3404 hospital survivors of acute myocardial infarction. The effect of anticoagulant treatment on recurrent myocardial infarction, cerebrovascular events, and vascular events (the composite endpoint of reinfarction, cerebrovascular event, and vascular death). Long term anticoagulant treatment was associated with a reduction in mortality of 10% (95% confidence interval -11% to 27%), recurrent myocardial infarction of 53% (41% to 62%), cerebrovascular events of 40% (10% to 60%) and vascular events of 35% (24% to 45%). Treatment effect with respect to recurrent myocardial infarction was comparable among all subgroups of patients. Although treatment effect appeared to be somewhat smaller in females than in males (-11% v -45%), and in patients with diabetes compared to those without (-14% v -42%) with respect to vascular events, none of these differences reached statistical significance. In multivariate analysis, more advanced age, previous myocardial infarction, diabetes mellitus, and heart failure during admission were independently associated with increased incidence of cardiovascular complications. The relative benefit of long term anticoagulant therapy in survivors of myocardial infarction is not modified by known prognostic factors for cardiovascular disease.