444. SHEA FEATURED ORAL ABSTRACT: INSPIRE Trial: A 92-Hospital Cluster Randomized Trial of INtelligent Stewardship Prompts to Improve Real-time Empiric Antibiotic Selection versus Routine Antibiotic Selection Practices for Patients with Abdominal Inf

Abstract Background Up to 40% of hospitalized patients receive empiric extended-spectrum (ES) antibiotics despite low risk of multidrug-resistant organism (MDRO) infection, increasing the risk for adverse effects and future resistance. We evaluated whether computerized physician order entry (CPOE) prompts providing patient-specific MDRO risk estimates could reduce ES antibiotic use compared to routine stewardship in patients hospitalized with abdominal infections. Methods This 92-hospital cluster-randomized trial compared CPOE prompts providing patient-specific absolute risk estimates for MDRO abdominal infection and recommending standard-spectrum antibiotics for risk < 10% vs. routine stewardship. Trial population: adults treated with antibiotics for abdominal infection in non-ICUs in the first 3 days of admission (empiric period). Prompts were triggered if ES antibiotics were ordered. Trial periods: 12-month Baseline (Jan 2019-Dec 2019); 5-month Phase-in (Aug 2022–Dec 2022); 12-month Intervention (Jan 2023-Dec 2023). Primary outcome: ES antibiotic days of therapy (ES-DOT) per patient per empiric day; secondary outcomes were a) vancomycin and b) anti-pseudomonal DOT per empiric day. Unadjusted, as-randomized analyses used (1) generalized linear mixed effects models to assess differences in ES-DOT rates across intervention and baseline periods between groups clustering by patient, hospital, and period and (2) proportional hazards models to assess safety outcomes: days to ICU transfer and hospital LOS. Results We randomized 92 hospitals in 15 states. Across baseline and intervention periods there were 100,890 and 97,680 non-ICU patients with abdominal infection in the routine and CPOE prompt groups, respectively. The CPOE prompt group had a 35% reduction in ES-DOT compared to routine care (rate ratio 0.65 [95% CI 0.60-0.71], p< 0.001). Vancomycin and anti-pseudomonal DOT were reduced by 20% and 39%, respectively (Table, Figure 1) without significant differences in LOS or ICU transfers. Conclusion INSPIRE CPOE prompts providing patient-specific MDRO risk estimates recommending standard spectrum antibiotics in low risk patients significantly reduced empiric ES prescribing in adults admitted with abdominal infection. Disclosures Ken Kleinman, ScD, Xttrium Laboratories: Conducting studies in which participating hospital patients received contributed antiseptic products outside the submitted work Richard Platt, MD, MSc, GlaxoSmithKline: Contract to academic department|Janssen: Contract to academic department|Pfizer: Contract to academic department Susan Huang, MD, MPH, Xttrium Laboratories: Conducting studies in which participating hospital patients received contributed antiseptic products outside the submitted work