Association Between Patient Antibiotic Stewardship Alert and Antibiotic Prescription Rates in Virtual Care Setting

Corresponding Author Ugur B. Balkanci, MD, Division of Infectious Diseases, Children’s National Hospital, Washington DC, USA, 111 Michigan Ave NW, Washington DC, 20010, (612) 422-0742, ubalkanci@cnmc.org Funding This work had no funding source outside the authors’ employment. Conflict(s) of Interest Ms. Marshall and Ms. Kleinschmidt are employed by Included Health. Included Health is the national virtual care practice whose data was the source for this manuscript. Drs. Balkanci and Hamdy have no conflicts of interest to report. Background Upper respiratory tract infections (URTI) are among the most common causes of virtual visits with medical providers. Most of these infections are viral in nature and would not benefit from an antibiotic prescription. Yet, antibiotics are frequently prescribed. One explanation is the provider’s perception of patient expectations for an antibiotic prescription. There have been a few studies showing a public display of clinician attestation to antibiotic stewardship within patients’ view has resulted in lower antibiotic prescription rates in in-person settings. No studies have shown if a similar approach is effective in nudging the providers in the virtual care setting. Methods This was a retrospective cohort study of 721 025 virtual visits with URTI-related patient-reported intake reasons (cough, sore throat, cold, influenza, coronavirus, fever, or nasal congestion) between June 2022 and May 2024 at a national virtual practice. Beginning in June 2023, patients with URTI-related intake reasons started seeing an antibiotic stewardship alert, stating the commitment of their providers to prescribe antibiotics only when indicated, as the main intervention of the study. Unadjusted odds ratios with 95% confidence intervals were calculated. Results Approximately 33.2% of all visits between June 2022 and May 2024 had an antibiotic prescription. The odds of receiving an antibiotic prescription were 1.095 [1.084; 1.11] times higher after the intervention than before the intervention. When limited to visits with bronchitis or URI diagnosis code (n = 198 873), the rate of antibiotic prescription was 9.1%. Within this subgroup, the odds of an antibiotic prescription were 0.88 [0.86; 0.91] times lower after the intervention than before the intervention. Conclusion Receiving an antibiotic stewardship alert when setting up a virtual visit did not decrease the odds of being prescribed an antibiotic at the visit. However, when limited to visits with a bronchitis or URI diagnosis, when antibiotics are never indicated, the antibiotic stewardship alert was associated with a decrease in the odds of clinicians prescribing an antibiotic. Study limitations include inability to account for potential coding bias and differences in patient clinical and demographic characteristics. Antibiotic stewardship alerts seen by patients before their virtual visits may be an effective tool for decreasing inappropriate antibiotic use for a select group of diagnoses.