1072. The Role of an On-site Infectious Disease Specialist in Hospital-Based Antimicrobial Stewardship Programs

Abstract Background Antimicrobial stewardship programs (ASPs) are now a requirement for many hospitals, but a large proportion of US hospitals lack an on-site Infectious Disease (ID) specialist. We sought to compare the processes and outcomes of ASPs at Veterans Health Administration (VHA) hospitals with and without an on-site ID specialist. Methods This retrospective cohort included all acute-care patients in VHA hospitals admitted during 2016, or 2 years after a VHA mandate for hospital-based ASPs. Data from a mandatory nationwide survey were used to identify hospitals that self-reported the absence of an on-site ID specialist, including an ID physician or ID pharmacist, in 2016. Antimicrobial use was quantified at the hospital-level as days-of-therapy (DOTs) per 1,000 days present and categorized based on National Healthcare Safety Network definitions. A facility-level negative binomial regression model with risk adjustments made for aggregated case-mix and facility-level factors was used to determine the association between the presence of an on-site ID specialist and antimicrobial use. Results Eighteen of 122 (14.8%) hospitals lacked an on-site ID specialist. Non-ID hospitals had fewer admissions per month than ID sites (mean 107.3 vs. 425.4, P < 0.01). An ASP policy and an ASP pharmacy champion were present at ≥90% of hospitals with and without an ID specialist. Core ASP strategies were frequently used in both ID and non-ID sites, including prior authorization (90.4% vs. 83.3%, P = 0.41) and prospective audit-and-feedback (76.9% vs. 66.7%, P = 0.38). Broad-spectrum antibacterial use (263.9 vs. 317.6 DOTs per 1,000 days-present, P = 0.01) but not total antimicrobial use (600.8 vs. 634.3 DOTs per 1,000 days-present, P = 0.34) was lower at ID vs. non-ID hospitals. After facility-level risk-adjustment, broad-spectrum antibacterial use (OR = 0.81, 95% CI 0.69–0.94) but not total antimicrobial use (OR = 0.92, 95% CI 0.70–1.21) was lower at ID hospitals. Conclusion An on-site ID specialist was not associated with greater use of core ASP strategies, but the presence of an on-site ID specialist was associated with less frequent prescribing of broad-spectrum antibacterial agents. An on-site ID specialist may be an important part of an effective hospital-based ASP. Disclosures All authors: No reported disclosures.