Medicare's Peer Review Organizations (PROs) now are required to work with hospitals to improve patient outcomes. Which hospitals should be targeted? We used 1988 California discharge data to identify hospitals with higher-than-expected rates of complications in six adult, medical-surgical patient populations. Relative hospital complication rates generally were correlated across clinical areas, although correlations were lower between medical and surgical case types. Higher relative rates of complications were associated with larger size, major teaching facilities, and provision of open heart surgery, as well as with coding more diagnoses per case. Complication rates generally were not related significantly to hospital mortality rates as calculated by the Health Care Financing Administration. Different hospitals may be chosen for quality review depending on the method used to identify poor outcomes.