The therapeutic efficacy of critical care units. Identifying subgroups of patients who benefit.

The majority of patients are admitted to critical care units for observation and to facilitate intervention if deterioration occurs or complications develop. We attempted to determine if a reduction in mortality in a subgroup of these patients admitted directly to the critical care units could be identified. A new method using the scientific principles of a randomized trial applied to the case-control design was employed. All 1905 patients admitted to the medical service over a three-month period were prospectively evaluated for illness severity and stability. Patients who would not have been eligible for a randomized clinical trial were excluded. Based on the prospective evaluations, four prognostically distinct subgroups of patients were formed. An odds ratio for each of the prognostic groups was calculated, a ratio of greater than 1 indicating a protective effect of direct critical care admission. Only one subgroup of patients, the unstable moderately ill, had an odds ratio greater than 1 (13.3). These results, in association with the results of our previous study, suggest that at the time of admission to the hospital, direct admission to the critical care unit reduced mortality among the unstable moderately ill subgroup of patients.