We attempted to determine hospital and pharmacy characteristics associated with mortality rates in 4864 United States hospitals. Data were obtained from the Health Care Financing Administration, the American Hospital Association, and the National Clinical Pharmacy Services survey. Univariate and multivariate regression models were used to determine which hospital characteristics were associated with mortality. A similar regression analysis was performed on 718 hospitals for which detailed pharmacy information was available. In a multivariate regression model, some characteristics of 4864 hospitals associated with reduced mortality rates were high-technology index (R2 = 0.09, p < 0.001), severity of illness (R2 = 0.048, p < 0.001), number of hospital beds (R2 = 0.016, p < 0.001), and medical personnel (R2 = 0.012, p < 0.001). This analysis accounted for 41% of the mortality rate variance. For the 718 hospitals and pharmacies, some of these characteristics were high-technology index (R2 = 0.157, p < 0.001), severity of illness (R2 = 0.07, p < 0.001), number of pharmacists/average daily census (R2 = 0.021, p < 0.001), and combined hospitalwide clinical pharmacy services (R2 = 0.016, p < 0.01). The results of this analysis were similar to those in the only other large study in this area, but that excluded pharmacy characteristics. This is the first study to show a statistically significant association between pharmacist and pharmacy variables and reduced hospital mortality rates.