Do past mortality rates predict future hospital mortality?

Abstract Background This study aimed to determine whether hospitals with higher historical mortality rates are independently associated with worse patient outcomes. Methods Observational study of in-hospital mortality in open abdominal aortic aneurysm repair, aortic valve replacement, and coronary artery bypass graft surgery in a California in-patient database was conducted. Hospitals’ annual historical mortality rates between 1998 and 2010 were calculated based on 3 years of data before each year. Results were adjusted for race, sex, age, hospital teaching status, admission year, insurance status, and Charlson comorbidity index. Results Hospitals were divided into quartiles based on historical mortality rates. For abdominal aortic aneurysm repair, the odds ratio (OR) of in-hospital mortality for hospitals within the highest quartile of prior mortality was 1.30 compared with the lowest quartile (95% confidence interval [CI] 1.03 to 1.63). For aortic valve replacement, the OR was 1.41 for the 3rd quartile (95% CI 1.15 to 1.73) and 1.54 for the highest quartile (95% CI 1.27 to 1.87). For coronary artery bypass graft surgery, the OR was 1.33 for the 3rd (95% CI 1.2 to 1.49) and 1.58 for the highest (95% CI 1.41 to 1.76) quartiles. Conclusion Patients presenting to hospitals with high historical mortality rates have a 30% to 60% increased mortality risk compared with patients presenting to hospitals with low historical mortality rates.