The prognostic importance psychosocial factors after acute myocardial infarction (AMI) is still debated. A fourfold increase in risk of mortality after AMI was reported for participants in the beta Blocker Heart Attack Trial who were described as being socially isolated and having high levels of life stress. This study was designed to determine the influence of social isolation and/or life stress on mortality after AMI in an English population. It was a follow-up study of a subset of patients recruited for the Anglo-Scandinavian Study of Early Thrombolysis (ASSET), between November 1986 and February 1988. The study group comprised 1,376 patients with suspected AMI (1,073 men and 303 women), from coronary care units in six English hospitals. Patients who were alive at 7 days and had completed a psychosocial questionnaire within seven days post-infarction were followed up for a median time of three years, the sole outcome measure being death from all causes. All deaths were notified by the National Health Service Central Registry (NHSCR). Cox's regression was used to allow for independent clinical prognostic factors such as age-group, previous documented infarct, complications in hospital, history of diabetes and history of hypertension. Socially isolated patients (in terms of lack of membership of any club or religious group or lack of contact with family and friends) were 49% more likely to die after an infarction than patients classified as not being socially isolated. While this statistic is of borderline significance it does suggest that such patients are at an increased risk of death after AMI. No associations with mortality risk were found with life-stress level, type A behaviour pattern or depression.