The prognostic value of QRS score (Selvester), ST depression, ST elevation, extrasystoles, P terminal force in V1, and QTc derived from the predischarge 12 lead electrocardiogram was assessed after myocardial infarction in 474 patients without intraventricular conduction defects, ventricular hypertrophy, or atrial fibrillation. The usefulness of these results in risk assessment was compared with that of other clinical data. During follow up 45 patients died. Logistic regression analysis showed that QRS score, ST depression, and QTc were independently predictive of cardiac mortality. When multivariate analysis was applied to clinical and electrocardiographic data together, however, the 12 lead electrocardiogram did not provide independent information additional to that provided by other routine clinical findings and laboratory tests such as a history of previous myocardial infarction, clinical signs of persistent heart failure, indication for digitalis or antiarrhythmic drugs at discharge, and enlarged heart on chest x ray. In conclusion, the electrocardiogram has important prognostic value; however, it is not powerful enough to further improve the risk assessment of post-infarction patients.