Implementation of a Carbapenem Prescribing Algorithm at a Community Hospital

BACKGROUND: Carbapenem (CAR) antibiotics have broad-spectrum activity against both Gram-positive and Gram-negative bacteria. Unrestricted use can lead to limited susceptibility profiles. Imipenem-cilastatin was also identified as the antipseudomonal β-lactam with the lowest susceptibility to Pseudomonas aeruginosa (88%) at our 79-bed community hospital. The Antimicrobial Stewardship Committee (ASC) developed an initiative to decrease overall CAR usage by improving prescribing habits and promoting the appropriate use of these antibiotics. METHODS: A CAR Prescribing Algorithm (Figure 1) was developed as guidance to prescribers to promote appropriate CAR use; it was implemented in December 2015. Pharmacists utilized the prospective audit and feedback strategy for new orders that failed to meet the algorithm’s CAR use criteria and recommended alternative therapy when appropriate. A three-part Medication Use Evaluation was conducted to evaluate the impact of implementing the algorithm: Pre-intervention (P1: January to March 2015), Post-intervention (P2: January to March 2016), and six months Post-intervention (P3: July to September 2016). Frequency of appropriate prescribing, antimicrobial days of therapy per 1,000 patient-days (DOT) and cost savings utilizing pharmacy purchasing data were evaluated. RESULTS: In P1, 107 patients received CARs with only 21 (20%) of patients meeting algorithm use criteria. In P2, patients receiving CARs decreased to 31 with 11 (35%) of patients meeting algorithm use criteria. In P3, 27 patients received CARs with 19 (70%) meeting algorithm use criteria. A three-fold decrease in DOT was observed from P1 to P2, 131.8 to 40.2; DOT remained low in P3 at 42.9 (Figure 2). Using 2015 as a baseline, we calculated a $75,000 pharmacy cost saving in 2016 attributed to the CAR Algorithm. CONCLUSION: The implementation of a Pharmacy-driven CAR Prescribing Algorithm at a small community hospital improved prescribing habits and led to a three-fold reduction in overall use. Success with the algorithm continued 6 months after implementation and led to cost-savings for the hospital. DISCLOSURES: All authors: No reported disclosures.