Hospitals long have been known to vary in their performances. In the last century, it was commonly thought that outcomes were worse when less experienced physicians were rendering care, as often was the case in public hospitals.1 Recent developments in health services research have begun to provide data that may be used as basis for rational policy approaches to this long-standing problem. Investigators have developed creative strategies for analyzing administrative data that give considerable insight into the patterns of mortality among hospitals.2 Others have developed and tested measures of "case mix" (severity and comorbidity) to adjust observed outcomes for patient characteristics at admission.3 Still others have developed precise tools for measuring quality of medical care that predict outcome.4
See also p 2655.
Among factors related to outcome differences, experience is often cited.5 For numerous surgical procedures, mortality can be substantially higher in hospitals performing relatively