Adaptation and validation of an antibiotic prescribing, peer comparison metric for respiratory tract diagnoses in walk-in clinics: a mixed-methods analysis

Abstract

Abstract Objective: Antibiotic overuse is common across walk-in clinics, but it is unclear which stewardship metrics are most effective for audit and feedback. In this study, we assessed the validity of a metric that captures antibiotic prescribing for respiratory tract diagnoses (RTDs). Design: We performed a mixed-methods study to evaluate an RTD metric, which quantified the frequency at which a provider prescribed antibiotics for RTD visits after excluding visits with complicating factors. Setting: Seven walk-in clinics across an integrated healthcare system. Participants: We included clinic visits during 2018–2022. We also conducted 17 semi-structured interviews with 10 unique providers to assess metric acceptability. Results: There were 331,496 visits; 120,937 (36.5%) met RTD criteria and 44,382 (36.7%) of these received an antibiotic. Factors associated with an increased odds of antibiotic use for RTDs included patient age ≥ 65 (OR = 1.40; 95% CI 1.30–1.51), age 0–17 (1.55, 95% CI 1.50–1.60), and ≥1 comorbidity (OR = 1.22; 95% CI = 1.15–1.29). After stratifying providers by their antibiotic-prescribing frequency for RTDs, patient case-mix was similar across tertiles. However, the highest tertile of prescribers more frequently coded suppurative otitis media and more frequently prescribed antibiotics for antibiotic-nonresponsive conditions (eg, viral infections). There was no correlation between antibiotic prescribing for RTDs and the frequency of return visits (r = 0.01, P = 0.96). Interviews with providers demonstrated the acceptability of the metric as an assessment tool. Conclusion: A provider-level metric that quantifies the frequency of antibiotic prescribing for all RTDs has both construct and face validity. Future studies should assess whether this type of metric is an effective feedback tool.