The classification introduced in 1967 by Killip et al has proved to be a useful method for early risk stratification of patients with acute myocardial infarction (AMI). Over the past 3 decades the overall mortality due to AMI has decreased significantly. The present study evaluates the usefulness of the Killip classification as a method for early risk stratification of patients with AMI in the 1990s. One thousand eight hundred seventy-three consecutive AMI patients were hospitalized in 25 coronary care units operating in Israel, and were followed for 1 year. Higher Killip class was found to be associated with increased in-hospital and 1-year mortality, in thrombolysis- and nonthrombolysis-treated patients (30-day mortality for all patients was 5%, 21%, 35%, and 67% in Killip classes I to IV, respectively). The overall mortality among AMI patients in the 1990s was found to be lower for each Killip class compared with a comparable patient population with AMI, hospitalized in Israel in the 1980s. Thus, the Killip classification is a useful method for early risk stratification of AMI patients in the 1990s.