Trauma quality improvement using risk-adjusted outcomes

PURPOSE: The National Surgical Quality Improvement Program has improved the quality of surgical care by tracking risk-adjusted patient outcomes. Unlike the National Surgical Quality Improvement Program, the trauma center verification program of the American College of Surgeons (ACS) focuses on availability of optimal resources, not outcomes. We hypothesized that significant variations in outcomes exist across similar level ACS-verified trauma centers despite availability of similar resources. METHODS: The National Trauma Data Bank was used to identify adult patients (age 16-99 years) who were treated at ACS-verified Level I trauma centers that submitted at least 1,000 patients during the 5-year study period (264,102 patients from 58 trauma centers, excluding dead upon arrival). Multivariate logistic regression was used to analyze expected survival for each patient, adjusted for age, gender, race, injury mechanism, transfer status, and injury severity. Observed-to-expected survival ratios (O/E ratios with 95% confidence intervals) were used to rank trauma centers as high performers (O/E ratio significantly larger than 1), low performers (O/E ratio significantly less than 1), or average performers (O/E ratio overlapping 1). RESULTS: Almost half the centers performed significantly different from their risk-adjusted expectation. Fourteen were high performers, 11 were low performers, and 33 were average performers. CONCLUSIONS: The trauma center verification process in its present form may not ensure optimal outcome across all verified centers. If further validated, these findings suggest significant room for trauma quality improvement by replicating structures and processes of high performing trauma centers.