Abstract

In Brief Objective: We sought to determine whether hospital variations in surgical mortality were due to differences in complication rates or failure to rescue rates (ie, case-fatality rates in patients with a complication). Background: Wide variations in mortality after major surgery are becoming increasingly apparent. The clinical mechanisms underling these variations are largely unexplored. Methods: We studied all Medicare beneficiaries undergoing 6 major operations in 2005 to 2006: pancreatectomy, esophagectomy, abdominal aortic aneurysm repair, coronary artery bypass grafting, aortic valve replacement, and mitral valve replacement. We ranked hospitals according to risk-adjusted mortality and divided them into 5 equal groups. We then compared the incidence of complications and rates of failure to rescue between the top 20% of hospitals ("best") and bottom 20% of hospitals ("worst"). Analyses were conducted for all operations combined and for each individual procedure. Results: For all 6 operations combined, the worst hospitals had mortality rates 2.5-fold higher than the best hospitals (8.0% vs. 3.0%). However, complication rates were similar at worst and best hospitals (36.4% vs. 32.7%). In contrast, failure to rescue rates were much higher at the worst compared with the best hospitals (16.7% vs. 6.8%). These findings persisted in analyses with individual operations and specific complications. Conclusions: Reducing variations in mortality will require strategies to improve the ability of high-mortality hospitals to manage postoperative complications. Surgical mortality rates vary widely across hospitals. Excess mortality at poorly performing hospitals appears to be attributable to large differences in failure to rescue, but only small differences in the incidence of complications. Improvements in surgical quality will require strategies to improve the ability of poorly performing hospitals to better recognize and manage complications.