Combined impact of multiple interventions on trends of incidence of surgical site infections in a teaching hospital: a joinpoint regression analysis.

Background : The burden of surgical site infections (SSI) in Brazilian hospitals is high. Both international and local guidelines from governmental agencies advise preventive interventions against SSI. Objective: To measure the combined impact of successive interventions applied in a teaching hospital on the time trends of SSI rates. Methods : Prospective surveillance (both in-hospital and post-discharge) of SSI was performed. A time series of monthly overall SSI rates was obtained for the period from January 2011 through December 2016. During that period, successive interventions, including rules for preoperative care (e.g. hair removal) and intra-operative procedures (e.g., antimicrobial prophylaxis) where implemented. Joinpoint regression analysis was performed on that time series, in order to identify both significant linear trends and trend changes (joinpoints). Results : Overall SSI rate for the study period was 8,5%. Monthly states were stable from January 2011 through April 2013 (trend analysis, beta=0.04; p=0.53). In May/2013 there was a joinpoint (p=0.03) and from that month on a significant trend towards reduction (beta= -0.08; p<0.001). Coherently, yearly rates reduced from 9.9% (in 2011) to 6.9% (in 2016; p<0.001). Discussion : Time series analysis with Joinpoint Regression models was a useful tool for identifying the combined impact of several preventive measures on the incidence of SSI.