P-208. Does an electronic alert for secondary C. difficile prophylaxis result in vancomycin prophylaxis and reduction in recurrent CDI?

Abstract Background Secondary vancomycin prophylaxis (VP) is one strategy to reduce recurrent C. difficile infection (CDI). Clinical decision support (CDS) was implemented in the electronic health record (Epic®) in 2020 to remind providers about eligibility for VP at time of prescription of broad-spectrum antibiotics in patients with a history of CDI. The effectiveness of this strategy was assessed. CDIFF BPA CDIFF Secondary Prophylaxis Best Practice Alert Methods Retrospective review of an interruptive best practice alert (BPA) for VP at 8 hospitals from 2021-2022. Patients were classified via manual chart review as immunocompetent (immune) or immunocompromised, and were eligible for 3 months and 12 months respectively after an episode of CDI. Demographic characteristics, immune status, receipt of VP, and outcomes (30-day recurrence CDI, death during hospitalization or within 30 days of discharge) were collected. Only the first BPA per patient was included in the demographic and death analysis. Only one alert per hospitalization was included in the VP use and recurrence analysis. Patients with CDI within 10 days prior to the BPA and on the same day as BPA alert were excluded. Descriptive statistics were used to analyze differences. Results Among 616 unique hospitalized patients for whom the BPA was triggered, median age was 64 years; 40% were immune. Approximately 50% were women, with 24% African American and 24% Hispanic. After 1562 BPAs, 13% of immunocompromised and 10% of immune patients were prescribed VP within 24 hours. Median days from previous CDI to BPA was 97 days for immunocompromised and 45 days for immune patients. For immunocompromised patients, frequency of CDI recurrence (11% versus 6%) and frequency of death (14% versus 18%) were not statistically different in the group that got VP and the group that didn’t. Similarly for immune patients, frequency of recurrent CDI (7% versus 8%) or death (16% versus 13%) were not statistically different in the group that got VP and the group that didn’t. Conclusion VP was infrequently initiated despite an interruptive BPA that identified patients with recent history of CDI at the time of prescription of broad-spectrum antibiotics. In this limited sample size, patients had similar outcomes regardless of VP. Further research is needed to understand the barriers for the low uptake, including CDS design and efficacy of VP. Disclosures All Authors: No reported disclosures