Venous thromboembolism as a marker of quality of care in trauma.

BACKGROUND: Venous thromboembolism (VTE) is reported to occur among 7% to 58% of trauma patients. Variability in VTE rates might reflect differences in case mix and quality of care, but also screening practices or data capture. We explored the variation in VTE rates across trauma centers to determine its use as a measure of the quality of patient care. STUDY
DESIGN: The National Trauma Data Bank (version 7.1, admission year 2006) was used to capture a cohort at risk for VTE. Crude and adjusted rates of VTE were determined, and the observed and expected rates were compared across centers. Outlier hospitals were defined as those with considerably more (or fewer) patients than expected. We then assessed the level of concordance between outlier status for deep vein thrombosis (DVT) and pulmonary embolism (PE). Intraclass correlation coefficients (ICC) were calculated to evaluate for the presence of a "center" effect using multilevel modeling.
RESULTS: The 22,421 patients met inclusion criteria from 30 trauma centers. There was marked variability in the rate of VTE across centers, ranging from 0.2% to 13.3%, which was more pronounced for DVT (0.2% to 13.1%) than for PE (0% to 1.7%). There was poor concordance for DVT and PE outlier status. Intraclass correlation coefficient was four times greater for DVT (0.23) than for PE (0.06).
CONCLUSIONS: There was substantial variation in rates of VTE across trauma centers. There was no relationship between DVT and PE outlier status, which is counter to the understanding of the biologic relationship between the two. Lastly, the very low Intraclass correlation coefficient for PE compared with DVT suggests that to a large extent, practice variation has very little impact on PE rates. In light of these findings and concerns about patient ascertainment of DVT, VTE rates might not be a useful measure of quality of care.