High- vs. Low-Intensity Prospective Audit and Feedback on Internal Medicine Wards and Impact on Antimicrobial Use at a Community Hospital

BACKGROUND: Antimicrobial stewardship program (ASP) interventions, such as prospective audit and feedback (PAF), have been shown to reduce antimicrobial use and improve patient outcomes. However, there is a lack of data comparing different PAF approaches. We examined the impact of a high-intensity interdisciplinary rounds-based PAF compared with low-intensity PAF on antimicrobial use on internal medicine wards in a 400-bed community hospital. METHODS: Prior to the intervention, low-intensity PAF was performed by ASP pharmacists with a focus on targeted antibiotics (fluoroquinolones, anti-pseudomonal penicilins, carbapenems, vancomycin, clindamycin, third-generation cephalosporins). Recommendations were made directly to the internist for each patient. High-intensity rounds-based PAF was introduced to 5 internal medicine wards sequentially. Rounds occurred twice weekly, reviewed internal medicine patients receiving any antimicrobial agent, and were interdisciplinary (ASP PharmD, internist, ward pharmacist, ASP MD). The primary outcome was antimicrobial use on internal medicine wards measured in defined daily doses (DDD) per 1000 patient-days (PD) 1–24 months prior compared with 1–24 months after the intervention. We performed interrupted time series analysis using linear regression to compare prescribing rates while accounting for autocorrelation within wards. Adjusted models included covariates to account for secular and seasonal trends. RESULTS: Following the intervention, there was a non-statistically significant drop in antimicrobial use from 469 to 435 DDD/1000 PD. See Table 1 and Figure 1 for analyses of antibiotic use. CONCLUSION: Although high-intensity PAF did not result in lower antibiotic use compared with low-intensity PAF overall, a delayed reduction (>12 months) in usage was seen. Prospective studies are needed to determine the optimal approach to PAF. DISCLOSURES: All authors: No reported disclosures.