The purpose of this study was to further our understanding of the effects of nursing-related hospital variables on 30-day mortality rates for hospitalized patients. A retrospective design was used to test the proposed 30-Day Mortality Model. The sample consisted of 75 acute-care hospitals in the province of Ontario, Canada. To develop hospital mortality rates, 46,941 patients discharged from these hospitals who had a most responsible diagnosis of acute myocardial infarction, stroke, pneumonia, or septicemia were included. To develop hospital-level nursing predictor variables, 3,998 responses to the Ontario Registered Nurse Survey of Hospital Characteristics were also included. The findings support a relationship between lower 30-day mortality and 3 predictors: a richer registered nurse skill mix, more years of experience on the clinical unit, and reported larger number of shifts missed. These findings can be used to predict the effects of hospital changes in nursing skill mix and years of RN experience on patient mortality.