Mila H. Ju1, Jeanette W. Chung1, Mark K. Eskandari2, David J. Bentrem1, Karl Y. Bilimoria1
1Surgical Outcomes and Quality Improvement Center, Department of Surgery, Northwestern University, Feinberg School of Medicine, Chicago, IL; 2Division of Vascular Surgery, Department of Surgery, Northwestern University, Feinberg School of Medicine, Chicago, IL.
OBJECTIVES: Venous thromboembolism (VTE) remains a major cause of postoperative morbidity and mortality. Oversight agencies publicly report hospital performance on VTE process of care (SCIP-VTE 1 and 2: prophylaxis order and administration) and outcome metrics (risk-adjusted PSI-12 VTE rate). Our objective was to exam the association between hospital structural characteristics and these measures.
METHODS: Hospital PSI-12 and SCIP-VTE data from the CMS Hospital Compare program were linked with American Hospital Association data, which contains information on hospital characteristics. The association between structural factors (hospital size, accreditations, participation in quality improvement initiatives) and PSI-12 and composite SCIP-VTE 1/2 performance was examined. Bivariate and multivariable comparisons focused on hospitals designated as "worse than national rate" for PSI-12 and those in the bottom quartile for SCIP-VTE 1/2 performance.
RESULTS: Hospital performance on SCIP-VTE-1 and SCIP-VTE-2 process measures was highly correlated: 93.3% (p<0.001). However, composite SCIP VTE 1/2 performance was only very weakly correlated with PSI-12 VTE outcomes: 4.9% (p=0.009). Hospitals in the bottom quartile for SCIP-VTE were less likely to be large hospitals (≥300 beds), disseminate quality reports, participate in ACS NSQIP, and have accredited programs (JCAHO, ACGME, Commission on Cancer, Level 1 trauma, burn and transplant). Interestingly, hospitals designated with a "worse than national rate" for PSI-12 were more likely to have these structural factors. As the number of accreditations or quality participations increased, adherence to SCIP-VTE increased but PSI-12 performance worsened.
CONCLUSIONS: Increased adherence to VTE process measures was not associated with better VTE outcome. Large hospitals with more accreditations, which are typically associated with better outcomes, were instead more likely to have higher VTE rates despite having better adherence to SCIP-VTE process measures.
AUTHOR DISCLOSURES: D. J. Bentrem: Nothing to disclose; K. Y. Bilimoria: Nothing to disclose; J. W. Chung: Nothing to disclose; M. K. Eskandari: Nothing to disclose; M. H. Ju: Nothing to disclose.
Posted April 2013