Benchmarking trauma center performance in traumatic brain injury

BACKGROUND: Trauma centers (TCs) generally use mortality to gauge performance. However, differences in mortality outcomes might reflect different approaches or philosophies toward end-of-life care. We postulate that discharge home (DH) as a proxy for functional outcome may be a more useful measure of quality and may have significant implications on the assessment of TC performance and external benchmarking efforts. METHODS: Data were derived from the National Trauma Data Bank (2007-2009). We included patients (18 years or older) with isolated, severe blunt head injuries who were admitted to Level I and Level II TCs. Observed-to-expected (O/E) mortality ratios were calculated and used to rank TC performance by mortality and then DH. Concordance between performance measures was calculated using a weighted kappa statistic. RESULTS: In total, 19,705 patients in 240 TCs were identified. Crude mortality ranged from 4% to 60%, whereas rates of DH ranged from 3% to 66%. When O/E ratios for mortality were evaluated, five centers were identified as high performers. Of these five centers, only two were also high performers for DH. The concordance of outlier status and correlation across O/E ratios between mortality and DH high performers was 0.16 (poor). CONCLUSION: Centers that are characterized as high performers when evaluating mortality are not high performers for functional outcome as evaluated by DH. DH may provide an alternative way of assessing quality of care delivered to patients with traumatic brain injury. LEVEL OF EVIDENCE: Care management study, level III.