U.S. Hospitals with Low Surgical Volume are Excluded from Public Reporting but have Higher Rates of Surgical Site Infection
Background: As the Centers for Medicare & Medicaid Services does not publicly report surgical site infection (SSI) rates for hospitals performing <25 procedures annually, consumers are not able to assess SSI risk for low surgical volume hospitals. We examined SSI rates by annual procedure volume for Medicare patients undergoing total hip arthroplasty (THA) and coronary artery bypass graft (CABG) surgery.
Methods: We used 2010 Medicare Provider and Analysis Review inpatient claims data, including all short-stay acute care U.S. hospitals performing THA or CABG on Medicare patients. We reviewed claims within 90 days of these procedures for codes previously shown to identify deep and organ/space SSI. We used logistic regression to calculate the odds of infection by Medicare surgical volume adjusted for patient age, gender, and coded comorbidities, along with generalized estimating equations to account for clustering within hospitals.
Results: Claims-based SSI rates declined as annual procedure volume increased for both procedures (Table 1). The adjusted odds of SSI coding were significantly higher in the lowest volume hospitals compared with the highest volume hospitals for both total hip arthroplasty (OR 1.5, 95% CI 1.3-1.7) and CABG (OR 1.4, 95% CI 1.1-1.7). Based on Medicare claims data, 54% of hospitals performed <25 THA procedures and 29% performed <25 CABG procedures on Medicare patients during at least one year from 2005 through 2010.
Conclusion: SSI rates assessed by previously-validated claims codes following total hip arthroplasty and CABG are significantly higher for Medicare patients undergoing surgery in U.S. hospitals with lower Medicare surgical volume. This highlights the need to identify and validate methods to assess and compare SSI performance for hospitals with a broad range of surgical volumes.
M. S. Calderwood,
Molnlycke: Conducting a clinical trial for which contributed product is being provided to participating hospitals, Contributed product
M. V. Murphy, None
D. Yokoe, None
R. Platt, None
S. S. Huang, None
See more of: Oral Abstract Session