Dr Isabella Centeleghe1,2, Kirsty Thomson2, Ashley Baring2, Ella Knox2, Professor Philip Russo1,2
1Monash University, Melbourne, Australia, 2Cabrini Health, Melbourne, Australia
Biography:
Isabella works as a Research Fellow for the Nursing Research department at Cabrini Health, as a joint appointment with Monash University. She moved from the UK to Melbourne to take on this opportunity. She currently has 6 published papers and presented overseas during her time as an early career researcher.
Abstract:
Introduction
Surgical site infections (SSI) are the most common healthcare associated infection (HAI) in Australia. Infection prevention and control (IPC) staff spend approximately 36% of their time undertaking routine surveillance. Emerging methodologies suggest this workload could be reduced by over 95% using automated approaches. This study evaluated, for the first time in the Australian private healthcare sector, a decision-based algorithm to identify SSI in patients undergoing hip or knee arthroplasty, or coronary artery bypass graft surgery.
Methods
A published algorithm was implemented at a private hospital in Melbourne and compared with traditional surveillance methods. A waiver of consent was obtained. Approximately one month after the procedure, data were collected via REDcap, including; number of microbiological cultures, length of stay, readmissions and duration of antibiotic exposure. Algorithm-identified outcomes were compared with those detected by the IPC team and reported to VICNISS.
Results
Over seven months, a total of 882 patients were prospectively screened for the probability of developing an SSI. The algorithm identified two patients as a “high probability of SSI”, compared to nine patients identified through traditional methods. Notably, the patients flagged as high probability by the algorithm were classified as “close calls” by the IPC team.
Conclusion
Discrepancies between the algorithm and traditional surveillance were mainly due to the use of paper-based medical records, which made it difficult to accurately track readmissions. Additionally, VICNISS guidelines apply a 90-day surveillance window for deep infections and a 30-day window for superficial infections, resulting in some patient cases not being picked up.