Two years of surgical site infection surveillance in Western Australia: analysing variation between hospitals

Surgical site infections (SSIs) following hip and knee arthroplasty are significant adverse events. The aim of this study was to present the first 2 years of aggregated and hospital-based SSI rates from the Healthcare Infection Surveillance Western Australia (HISWA) surveillance system. HISWA collects risk-adjusted data based on standardised collection methodologies and definitions. In total, 4131 hip and 4858 knee procedures were monitored at 10 public and private hospitals. There were significant variations in reported SSI rates following arthroplasty at WA hospitals (mean 1.82, range 0.67-3.83 SSIs per 100 hip procedures; mean 1.5, range 0.47- 3.67 SSIs per 100 knee procedures). Funnel plots identified hospitals with significantly high and low SSI rates, in particular hospitals with high operative volumes and significantly better outcomes. Significant variation was observed when analysis was performed by operative volume, United States National Nosocomial Infection Surveillance (NNIS) risk group and infection type (deep/organ space v. superficial). Investigations did not reveal marked differences in surveillance methodology between sites, but other causes of variation, including patient factors that are not incorporated in NNIS risk indexation, surgeon procedure volume and differences in SSI prevention policy and practice, need further evaluation. The policy implications of the association between operative volume and SSI rate merits further discussion in Australia.