One-year follow-up data on 515 patients who survived hospitalization with MB-creatine kinase-confirmed, acute non-Q wave myocardial infarction were analyzed for factors related to mortality (n = 57) and late reinfarction (n = 64). Twelve of 24 analyzed variables were significantly associated with mortality. Those factors, which were independently predictive of mortality by Cox regression analysis, were persistent ST depression (p = 0.0009), a history of congestive heart failure (CHF) (p = 0.0069), older age (p = 0.0128), and ST elevation at hospital discharge (p = 0.0173). In-hospital reinfarction achieved borderline significance (p = 0.0512). Mortality during the follow-up period was 5.5% in patients with no ST depression, 10.1% in those with ST depression at baseline or discharge, and 22.2% in patients with ST depression at baseline and discharge (i.e., "persistent" ST depression). The age-adjusted risk of mortality for patients with persistent ST depression, discharge-ST elevation, and CHF was 13.99 times as high as was the risk for patients with no ST depression, no discharge-ST elevation, and no CHF. Of the 483 patients with complete electrocardiographic data at both baseline and discharge, 203 (42%) could be stratified into a high risk population with a risk ratio for 1-year mortality more than sevenfold that of patients with no risk factors. Although persistent ST depression was significantly associated with several measures of structural left ventricular damage, the independent significance of ST depression persisted even after adjusting for these factors. The independent predictors of late reinfarction (persistent ST depression, p = 0.0058; Killip class II or III, p = 0.0106; and left ventricular hypertrophy, p = 0.0470) permitted a similar risk stratification. We conclude that 1) easily identified clinical and electrocardiographic factors permit stratification of patients with non-Q wave infarction into high-risk subsets who may benefit from aggressive therapy; 2) ST depression is a highly significant and independent predictor of poor prognosis; and 3) the powerful predictive value of persistent ST depression suggests that non-Q wave myocardial infarction patients with this depression should be viewed as potentially high-risk patients who may be candidates for additional noninvasive testing or early coronary angiography.