Surgical volume and the risk of surgical site infection in community hospitals: size matters

OBJECTIVE: To determine if surgical volume affects the risk of surgical site infections (SSI) in community hospitals. BACKGROUND: The utility of public reporting and the optimal methods to employ when reporting SSI rates remain controversial and contentious issues. Studies examining the association between surgical volume and SSI risk have included few community hospitals and have reported conflicting results. METHODS: A prospective study of surgical procedures performed at 18 community hospitals from January 1, 2004 to December 31, 2005, was performed. Hospitals were separated based on average surgical volume per year: small (<1500 procedures), medium (1500 < or = procedures < 4000), and large (> or =4000 procedures). The risk of SSI for each category was determined using multivariable Poisson regression. RESULTS: Prospective surveillance identified 1434 SSIs after 132,111 surgical procedures (prevalence rate = 1.09/100 procedures). In unadjusted analysis, the risk of SSI was almost twice as high at small hospitals [prevalence rate ratio (PRR) = 1.9 (95% CI 1.78-2.05)] and large hospitals [PRR = 1.79 (95% CI 1.70-1.90)] compared with medium hospitals. After adjusting for differences between hospital category and important confounders, the risk of SSI at small hospitals was still 1.5 times higher than medium hospitals [adjusted PRR = 1.49 (95% CI 1.39-1.60)], whereas the risk at large hospitals was substantially decreased compared with medium hospitals [adjusted PRR = 1.29 (95% CI 1.22-1.36)]. OUTCOMES: The relationship between hospital surgical volume and rates of SSI in community hospitals is important and complex. As public reporting of SSI rates expands, improved methods for risk-adjusting infection rates are needed.