Impact of an Emergency Medicine Pharmacist on Appropriate Empiric Antibiotic Prescribing for Community-Acquired Pneumonia and Intra-Abdominal Infections

BACKGROUND: Antibiotics are the second most common drug class prescribed in the Emergency Department (ED); therefore, it is critical to engage ED providers in antimicrobial stewardship programs (ASP). Emergency medicine pharmacists (EMP) play an important role in ASP by working with providers to choose the most appropriate antimicrobial agent, dose, and duration. This study aimed to determine the impact of an EMP on appropriate empiric antibiotic prescribing for community-acquired pneumonia (CAP) and community-acquired intra-abdominal infections (CA-IAI). METHODS: A retrospective cohort study was conducted evaluating adult patients admitted with a diagnosis of CAP or CA-IAI. The primary outcome of this study was to compare guideline-concordant empiric antibiotic prescribing when an EMP was present vs. absent. We also aimed to compare the impact of an EMP in a new ASP (2014) vs. established ASP (2016). Secondary outcomes included in-hospital mortality and hospital-acquired Clostridium difficileinfection (CDI). RESULTS: 320 patients were included in the study (EMP n = 185; no-EMP n = 135). Empiric antibiotic selection was more likely to be guideline-concordant when an EMP was present (78% vs. 61%, P = 0.001). Guideline-concordant empiric prescribing occurred more often when an EMP was present in the subgroup of CAP patients (95% vs. 79% P = 0.005) as well as in the subgroup of CA-IAI patients (62% vs. 44% P = 0.025). Overall guideline-concordant prescribing significantly increased between the new ASP and established ASP (60% vs. 82.5%, P < 0.001) and was more likely when an EMP was present (new ASP: 68.3% vs. 45.8%, P = 0.005; established ASP: 90.5% vs. 73.7%, P = 0.005). Patients receiving guideline-concordant antibiotics in the ED were continued on appropriate therapy on admission 82.5% of the time vs. 18.8% if the ED antibiotic was inappropriate (P < 0.001). The presence of an EMP did not impact hospital-acquired CDI (1.1% vs. 1.5%, P = 1.0) or in-hospital mortality (4.3% vs. 1.5%, P = 0.2). CONCLUSION: The presence of an EMP significantly improved guideline-concordant empiric antibiotic prescribing for CAP and CA-IAI. This impact was demonstrated in both a new and established ASP. Inpatient orders were more likely to be guideline-concordant if appropriate therapy was ordered in the ED. DISCLOSURES: All authors: No reported disclosures.