This study analyses the relationship between hospital outcome, specifically mortality rates, and various hospital and market characteristics. The data are for all general acute-care hospitals in California in 1991 that had at least 50 Medicare cases selected by the Health Care Financing Administration (HCFA) for their computation of 1991 observed and predicted mortality rates. The dependent variable was constructed from these mortality rates. Hospital and market characteristics were obtained from California's Office of Statewide Health Planning and Development 1991 tapes, the American Hospital Association's annual 1991 report, 1990 Census data, and HCFA's 1991 Medicare Hospital Mortality Information Report . Highly significant results (p < 0.01) were found for three of the predictor variables: improved mortality rates were associated with a lower proportion of Medicaid patients, with a larger market, and with a higher prior-year profit margin. Other factors that contributed to improved mortality rates (p < 0.05 or 0.10) were more concentrated hospital markets, a lower percentage of HMO patients, and a higher percentage of residents with at least a Bachelor's degree. Characteristics without significant coefficients were hospital size, the percentage of residents below poverty level, more 'high-tech' services, the proportion of Medicare patients, having a residency programme, and membership in COTH (Council of Teaching Hospitals).