Rates of surgical site infection after hip replacement as a hospital performance indicator: analysis of data from the English mandatory surveillance system.

OBJECTIVE: To describe rates of surgical site infection (SSI) after hip replacement and to use these data to provide a simple mechanism for identifying poorly performing hospitals that takes into account variations in sample size.

DESIGN: Prospective surveillance study.

SETTING: A total of 125 acute care hospitals in England that participated in mandatory SSI surveillance from April 1, 2004 through March 31, 2005.

PATIENTS: Patients who underwent total hip replacement (THR) or hip hemiarthroplasty (HH).

METHODS: A standard data set was collected for all eligible operations at participating hospitals for a minimum of 3 months annually. Defined methods were used to identify SSIs that occurred during the inpatient stay. Data were checked for quality and accuracy, and funnel plots were constructed by plotting the incidence of SSI against the number of operations.

RESULTS: Data were collected on 16,765 THRs and 5,395 HHs. The cumulative SSI incidence rates were 1.26% for THR and 4.06% for HH; the incidence densities were 1.38 SSIs per 1,000 postoperative inpatient days for THR and 2.3 SSIs per 1,000 postoperative inpatient days for HH. The risk of infection associated with revision surgery was significantly higher than that associated with primary surgery (2.7% [95% confidence interval, 2.0%-3.5%] vs. 1.1% [95% confidence interval, 1.0%-1.2%]; P=.003). Rates varied considerably among hospitals. Nineteen hospitals had rates above the 90th percentile. However, the use of funnel plots to adjust for the precision of estimated SSI rates identified 7 hospitals that warranted further investigation, including 2 with crude rates below the 90th percentile.

CONCLUSIONS: Funnel plots of rates of SSI after hip replacement provide a valuable method of presenting hospital performance data, clearly identifying hospitals with unusually high or low rates while adjusting for the precision of the estimated rate. This information can be used to target and support local interventions to reduce the risk of infection.