The value of measuring severity of disease in clinical research on acutely ill patients.

There are five major factors that determine outcome from disease: (1) disease type, (2) the severity of the disease, (3) the patient's age, (4) his prior health status, and (5) the therapy available. Evaluation of new treatments for various diseases is often done with little information on individual patients' severity. The most widely used method of controlling for acute severity fails to account for interaction among major organ systems and for important threshold effects found within physiologic measurements. To illustrate, we simulated a clinical trial comparing severity and outcome for two groups randomly chosen from 50 consecutive respiratory failure patients. Mean values for a variety of clinical, demographic, and physiologic measures were similar. A severity of disease classification, however, predicted differential mortality (25% vs 37%) that matched actual death rates. Uniform and accurate measurement of acute severity of disease in individual patients could improve the precision of clinical research.