Meaningful comparison of patient outcomes requires an assessment of the severity of illness for the patients being compared. The more severe the underlying illness, the worse the expected outcome. We studied several severity of illness indicators derived from different methodologies in a medical intensive care unit. We compared the Acute Physiologic and Chronic Health Evaluation II, the accepted benchmark indicator for intensive care units, with one complex indicator, Computerized Severity Score, and three simpler indicators, Comorbidity, McCabe-Jackson, and American Society of Anesthesiologists. We found that all correlated well with a comorbidity index. We conclude that the Acute Physiologic and Chronic Health Evaluation II, the Computerized Severity Score, and the McCabe-Jackson scoring systems appear to be comparable predictors of comorbidity in a medical intensive care unit. Selection of a severity indicator will depend on the resources available and the intended uses.
(Arch Intern Med. 1991;151:2201-2205)