Variability in reoperation rates at 182 hospitals: a potential target for quality improvement.

Abstract

Reoperation rate has gained increasing attention as a potential indicator of surgical quality. Objectives of this study were to examine the feasibility of assessing reoperation rates at 182 hospitals to identify institutions with outlying performance, to examine potentially modifiable factors that are associated with reoperations, and to determine if a more parsimonious logistic regression model effectively predicts reoperations. Patients were identified who underwent colorectal procedures at 182 hospitals participating in the American College of Surgeon's National Surgical Quality Improvement Program in 2006-2007. Risk-adjusted regression models for reoperation were developed to identify hospitals that had ratios of observed-to-expected events that were substantial outliers. Of 23,098 patients identified, 1,320 (5.7%) required reoperations. Reoperation occurred significantly more often than expected in 16 hospitals and less often than expected in 7 hospitals (p < 0.05). Factors that were associated with an increased risk of reoperation were advanced American Society of Anesthesiologists class, male gender, contaminated wounds, surgical extent, surgical indication, smoking, poor functional status, disseminated cancer, COPD, steroid dependence, anemia, body mass index (calculated as kg/m(2)) >35 or < or = 18.5, and hypertension. Compared with the full logistic regression model, there was a high degree of correlation with the more parsimonious logistic model containing only the first six variables (r = 0.996). There is considerable variability in reoperation rates at American College of Surgeon's National Surgical Quality Improvement Program hospitals. American College of Surgeon's National Surgical Quality Improvement Program data can be used to provide individual hospitals with risk-adjusted self-assessment data on reoperations to potentially identify quality-improvement opportunities.