The Impact of Depth of Infection and Postdischarge Surveillance on Rate of Surgical-Site Infections in a Network of Community Hospitals

Abstract

Objective.To describe the epidemiology of surgical-site infections (SSIs) in community hospitals and to explore the impact of depth of SSI, healthcare location at the time of diagnosis, and variations in surveillance practices on the overall rate of SSI.Design.Retrospective cohort study.Setting.Thirty-seven community hospitals in the southeastern United States.Patients.Consecutive sample of patients undergoing surgical procedures between July 1, 2007, and December 31, 2008.Methods.ANOVA was used to compare rates of SSIs, and the F test was used to compare the distribution of rates of SSIs. Wilcoxon rank-sum was used to test for differences in performance rankings of hospitals.Results.Following 177,706 surgical procedures, 1,919 SSIs were identified (incidence, 1.08 per 100 procedures). Sixty-four percent (1,223 of 1,919) of these were identified as complex SSIs; 87% of the complex SSIs were diagnosed in inpatient settings. The median proportion of superficial-incisional SSIs was 37% (interquartile range, 29.6%–49.5%). Postdischarge SSI surveillance was variable, with 58% of responding hospitals using surgeon letters. As reporting focus was narrowed from all SSIs to complex SSIs (incidence, 0.69 per 100 procedures) and, finally, to complex SSIs diagnosed in the inpatient setting (incidence, 0.51 per 100 procedures), variance in rates changed significantly (P = .02). Performance ranking of individual hospitals, based on rates of SSIs, differed significandy, depending on the reporting method utilized (P = .0006).Conclusions.Inconsistent reporting mediods focused on variable depths of infection and healthcare location at time of diagnosis significandy impact rates of SSI, distribution of rates of SSI, and hospital comparative-performance rankings. We believe that public reporting of SSI rates should be limited to complex SSIs diagnosed in the inpatient setting.Infect Control Hosp Epidemiol 2012;33(3):276-282