A comparison of an algorithm, and coding data, with traditional surveillance to identify surgical site infections in Australia: A retrospective multicentred cohort study

Abstract

Surveillance of healthcare associated infections (HAIs) in Australia is disparate, resource intensive, unsustainable and provides limited information. Traditional HAI surveillance is time intensive and agreement levels between clinicians has been shown to be variable. The aim was to compare two methods, a semi-automated algorithm, and coding data, against traditional surgical site infections (SSI) surveillance methods. This retrospective multi-centre cohort study included all patients undergoing a hip (HPRO) or knee (KPRO) joint replacements and coronary artery bypass graft (CBGB) surgery over 2 years at 2 large metropolitan hospitals. Routine SSI data were obtained via the infection prevention team, a previously developed algorithm was applied to all patient records, and the ICD-10-AM data were searched for those categorised as having a SSI. Overall, 1447, 1416 and 1026 patients who underwent HPRO, KPRO and CBGB respectively were included. The highest Se values were generated by the algorithm: HPRO D/O 0.87(95%CI:0.66-0.96), CBGB 0.86(95%CI:0.64-0.96) and HPRO all SSI 0.77(95%CI:0.57-89), the lowest Se was Code CBGB D/O 0.03(95%CI:0.00-0.21). The highest PPV values were generated by the algorithm: HPRO D/O 0.97(95%CI:0.77-0.99), CBGB D/O 0.97(95%CI:0.76-0.99) and the Code HPRO D/O 0.9(95%CI:0.66-0.99). Both the algorithm and coding data resulted in a substantial reduction in the number of medical records required to review. The application of algorithms to enhance SSI surveillance demonstrates high accuracy in identifying patient records that require review by infection prevention teams to determine the presence of an SSI. Coding data alone should not be used to identify SSI's.