Benchmarking Antimicrobial Use in Emergency Departments among Community Hospitals in the Duke Antimicrobial Stewardship Outreach Network

BACKGROUND: Antimicrobials are frequently administered in the Emergency Department (ED), and there are unique challenges to implementing antimicrobial stewardship in this setting. Benchmarking antimicrobial use (AU) in the ED may assist hospitals in identifying areas to target stewardship interventions. EDs are not currently included in the standardized antibiotic administration ratios calculated by the CDC Module. Therefore, we aimed to establish a method to compare AU among EDs across our network. METHODS: This descriptive cohort included hospitals participating in the Duke Antimicrobial Stewardship Outreach Network (DASON) with available electronic medication administration records from EDs for calendar year 2016. ED encounters were estimated using encounters data collected for LabID event reporting. Overall AU, AU by antimicrobial category, and AU by antimicrobial agent were calculated in days of therapy (DOT) per 1,000 ED encounters and then compared among hospitals using descriptive statistics. Outlying use for individual agents was defined as >95th percentile. RESULTS: Thirteen community hospitals in the southeastern United States provided AU data for over 724,627 encounters (median 57,199, range 29,063–86,574). Median overall AU was 206 DOT/1000 ED encounters and showed variation among hospitals (interquartile range (IQR) 180–231) (Figure 1). The majority of administered agents were antibacterials (96.9%), followed by antifungals (1.18%), antivirals (0.13%), and other (0.12%). Antimicrobial categories showing greatest variation among EDs included antifungals (coefficient of variation (CV) 57.47, median 2.04, IQR 1.39-3.36) and antipseudomonal β-lactams (CV, 44.8, median 19.3, IQR 14.07-21.76). Outlying use in individual EDs were identified for ertapenem, ceftaroline, daptomycin, and fluconazole (Figures 2 and 3). CONCLUSION: We observed large variability in AU in EDs among acute care community hospitals and identified several outliers for certain agents among hospital EDs using the metric DOT per 1000 ED encounters. These data will help inform future stewardship interventions at these hospitals. DISCLOSURES: All authors: No reported disclosures.