Risk-adjusted infection rates in surgery: a model for outcome measurement in hospitals developing new quality improvement programmes

Assessment of healthcare quality is a major challenge in countries such as Hungary where there is limited experience with measurement of patient outcomes. We sought to develop the capacity for valid outcome measurement in Hungarian hospitals using surgical site infection (SSI) surveillance as a model and to identify areas for improvement by comparing SSI rates in Hungarian hospitals to benchmarks published by the United States Centers for Disease Control and Prevention's National Nosocomial Infection Surveillance (NNIS) System. We surveyed the incidence of SSI among 5126 patients undergoing 6006 procedures in 20 public hospitals in Hungary during 1996 using the Hospitals in Europe Link for Infection Control through Surveillance (HELICS) protocol, a protocol consistent with the methods used by the NNIS System. Cholecystectomy, herniorrhaphy, appendectomy, and open reduction of fracture--four of the five most commonly performed procedures in Hungary in 1996--comprised 85% of the procedures analysed. Cumulative SSI rates for herniorrhaphy and appendectomy were comparable to NNIS System benchmarks. Cumulative SSI rates for cholecystectomy were significantly higher in Hungarian hospitals among risk categories that included open procedures. Nearly half of the hospitals had SSI rates for cholecystectomy that were high outliers (>90% percentile) compared to NNIS System benchmarks. Cumulative SSI rates for open reduction of fracture and mastectomy were significantly higher in Hungarian hospitals due to high rates in a few hospitals. The duration of surgery for all procedure types was substantially shorter in Hungarian hospitals compared with NNIS System hospitals. Future work should focus on optimizing prevention strategies for patients undergoing cholecystectomy, open reduction of fracture, and mastectomy. The effect of the utilization of open vs. laparoscopic cholecystectomy, short procedure duration, and procedure volume on SSI rates should be evaluated further. This programme expanded the capacity of Hungarian hospitals to perform surgical site infection surveillance and can serve as a model for hospitals in other countries with limited experience with outcome measurement.