1205. Antibiotic-prescribing for all respiratory tract diagnoses across a network of walk-in clinics: clinician-level variation and correlation with other performance metrics

Abstract Background It is unclear which stewardship metrics are most effective for audit and feedback to outpatient clinicians. In this study, we explored a metric that captures antibiotic-prescribing for all respiratory tract diagnoses to determine if it could serve as an appropriate metric for feedback. Methods We performed a retrospective cohort study of in-person visits to the 7 walk-in clinics, including 3 Urgent Care locations, within University of Iowa Health Care during 2018-2022. Visits were categorized as respiratory visits (RVs) if a respiratory tract diagnosis was coded and the patient lacked complicating factors, such as a concomitant non-respiratory infection or certain comorbidities. We built a hierarchical logistic regression model that adjusted for antibiotic appropriateness tiers (1-3) to identify factors associated with antibiotic-prescribing for RVs. Using Spearman’s correlation, we compared the frequency at which clinicians prescribed antibiotics for RVs and other types of visits. Results There were 331,496 visits, and 96,546 (29.1%) led to an antibiotic prescription; 44,498 (46.1%) of these were for RVs (Tables 1 and 2). At the clinician-level (n=89), the mean frequency of antibiotic-prescribing for RVs was 36.4% (standard deviation 11.6). Factors at the visit-level associated with an increased odds of antibiotic use for RVs included age ≥ 65 (OR=1.5; 95% CI=1.4-1.6) and having at least one comorbidity (OR=1.2; 95% CI=1.1-1.2). The frequency at which a clinician prescribed antibiotics for RVs was strongly correlated with antibiotic use for all visits (r=0.77, p< 0.001). Clinicians who more often prescribed antibiotics for RVs also more often prescribed antibiotics for respiratory infections that are always viral (r=0.73, p< 0.001), sinusitis (r=0.49, p< 0.001), and non-streptococcal pharyngitis (r=0.69, p< 0.001). There was no correlation between antibiotic-prescribing for RVs and the frequency of return visits within 30 days (r=0.11, p=0.29). Conclusion A metric that quantifies the frequency of antibiotic-prescribing for all respiratory tract diagnoses correlated strongly with clinicians who over-use antibiotics for walk-in visits. Future studies should assess whether this type of metric is acceptable to clinicians and an effective tool for feedback. Disclosures Daniel J. Livorsi, MD, Merck: Grant/Research Support Kelly M. Percival, PharmD, Gilead Sciences Inc: Advisor/Consultant