1235. A Survey of Surgical Site Infection (SSI) Surveillance Practices in US Hospitals, and their Association with SSI Rates

Abstract

Background

Current US hospital reimbursement models rely on self-reported SSI rates. The impact of variability in SSI surveillance on publicly reported SSI rates is unknown.

Methods

Cross-sectional survey to US hospitals administered during November 18 – 2/19 through the Association for Professionals in Infection Control. We assessed SSI surveillance practices, and asked for self-reported facility standardized infection ratios (SIR) for hysterectomy and colon surgeries. We performed bivariate analysis and used Kendall’s ranks correlation for trend analysis.

Results

Of the 2,851 hospitals surveyed, 491 (17.2%) responded. Table 1 shows facility descriptors. Critical Access Hospitals (OR 6.11 [3.12 – 11.750, P < 0.005) and Ambulatory Surgical Centers (OR 3.92 [1.68 – 8.64], P < 0.001) were more likely to have less than one full-time ICP. University Hospitals were more likely to have ≥4 ICPs (OR 12.15 [6.73 – 22.04, P < 0.001). The majority (83%) of the 477 respondents reported electronic software for SSI surveillance, with Epic (23%), Theradoc (22%), and Cerner (11%) as the most common packages used. Manual surveillance was more likely for Critical Access Hospitals (OR 2.80 [1.47 – 5.19], P < 0.001). University Hospitals were more likely to have higher rates in 2016 for colon surgery (P = 0.02) and hysterectomy (P = 0.002). Table 2 shows characteristics of SSI surveillance practices reported by study participants. Ambulatory Surgical Center ICPs were more likely to use reports from surgeons and/or surgical staff as the initial trigger for SSI surveillance. University Hospital ICPs were significantly more likely to spend increased time (mean hours/month 69.77 vs. 28.99, P < 0.001), and to use more data sources for SSI review (mean 4.58 vs. 3.99, P = 0.001). In our trend analyses, we found the number of data sources used for SSI surveillance to be positively associated with higher SSI rates: (K_T =0.14, P = 0.028 for colon SIR in 2017; K_T = 0.20, P = 0.009; K_T = 0.25, P = 0.001 for hysterectomy SIR in 2016 and 2017, respectively).

Conclusion

SSI surveillance practices across US hospitals vary significantly, and rigorous surveillance methods are associated with higher SSI rates. Standardizing SSI surveillance is necessary to accurately capture SSI burden of disease.

Disclosures

All authors: No reported disclosures.

Session: 147. HAI: Surgical Site Infections

Friday, October 4, 2019: 12:15 PM