Introduction/Objectives

Surveillance of surgical site infections (SSIs) is essential; analysis of SSI time series can provide insights into the effects of ecological intervention measures and changes in SSI definitions. In 2017, the Centers for Disease Control and Prevention (CDC) reduced post-implant SSI surveillance from 1 year to 3 months. This study evaluated long-term SSI trends after arthroplasty and the impact of individual feedback.

Methods

Monthly SSI rates (Jan/2010–Dec/2025) following primary total hip arthroplasty (THA) and total knee arthroplasty (TKA) were analyzed. Definitions followed CDC criteria during each period (1-year surveillance through 2016; 3-month surveillance from 2017 onward), forming a hybrid time series. Interrupted time series analysis (ITSA) using segmented Poisson regression, was used to estimate immediate changes in SSI rates (level changes) and changes in temporal trends (slope changes) following each intervention. Interventions: (1) in-person feedback for knee surgeons (lecture and simultaneous printed individual reports); (2) electronic feedback for hip surgeons (lecture and emailed individual rates at different times). The definition change in 2017 was treated as a structural break point.

Results

The SSI definition change did not produce level changes on rates (TKA: RR 1.32; 95% CI 0.67–2.63; THA: RR 1.16; 95% CI 0.67–2.02). In-person feedback led to sustained SSI reduction after TKA (RR 0.95/month; 95% CI 0.92–0.98; p<0.001 – slope chage), but no level change. The first electronic feedback reduced SSIs after THA initially (RR 0.63; 95% CI 0.40–1.00; p=0.049 – level change), but effects attenuated over time (RR 1.05; 95% CI 1.01–1.08; p=0.006 – slope change). The second eletronic feedback did not produce immediate effect, but was associated with downward trend (RR 0.90; 95% CI 0.81–0.97 – slope change), which may have been influenced by concurrent prevention measures, including a decolonization protocol introduced in April 2025.

Conclusions

Synchronous in-person feedback was associated with a sustained reduction in SSI trends, whereas the long-term effect of asynchronous electronic feedback remained inconclusive. The surveillance-definition change was not associated with an immediate change in SSI rates.

As an ecological analysis, the study does not account for concurrent baseline risk changes or unmeasured factors. Feedback format and its social context may influence SSI prevention effectiveness.

Keywords:

Surgical site infection

Arthroplasty

Infection control

References

1) Ivers N  et al. Cochrane Database Syst Rev. 2025 Mar 25;3(3):CD000259.

2) Bernal JL et al.Int J Epidemiol. 2017;46(1):348–355.

3) Keely BK et al. Curr Rev Musculoskelet Med. 2018;11(3):357–369.