We need objective and reliable ways of measuring the severity of disease of hospitalized patients. This paper demonstrates the international predictive accuracy of a severity of disease measure on 1504 consecutive, unscheduled intensive care admissions to 14 hospitals in the United States, France, Spain, and Finland. Using laboratory data gathered within 24 hours of ICU admission, the Acute Physiology Score of APACHE (Acute Physiology and Chronic Health Evaluation) was a strong and stable predictor of hospital survival and concurrent therapeutic effort. In ordinary least squares and logistic multiple regression analysis, the impact of the Acute Physiology Score (APS) was highly significant (p less than 0.001) and of virtually identical magnitude in the United States and European hospitals. The use of this severity of disease measure should help researchers gain insights concerning the efficacy of medical services and the characteristics of physician decision making by permitting more precise prognostic stratification of severely ill patients.