LAUSR

INTRODUCTION Health systems are increasingly implementing direct-to-consumer telemedicine for unscheduled acute care, however quality of care may be variable. Acute respiratory infection antibiotic prescribing rates in telemedicine visits performed by emergency physicians affiliated with medical centers has not been compared to care by unaffiliated, vendor-supplied physicians (a heterogeneous group). We hypothesized that, in virtual visits for acute respiratory infection, affiliated physicians would prescribe antibiotics at a lower rate than unaffiliated physicians. METHODS We performed a retrospective analysis of on-demand telemedicine visits available to health system employees and dependents at a large urban academic health system from March 2018 to July 2019. We performed multivariable logistic regression to determine the effect of physician affiliation on antibiotic prescribing patterns for acute respiratory infection, adjusting for patient age, visit weekday, and overnight visits. RESULTS Of 257 telemedicine encounters related to acute respiratory infection, affiliated physicians prescribed antibiotics in 18% of visits, compared to 37% of visits by vendor physicians. In multivariable analysis, patients seen by a vendor physician for acute respiratory infection had 2.3 higher odds (95%CI 1.1-4.5, p < 0.01) of being prescribed antibiotics, an average marginal effect of 15% (95%CI 2-29%). DISCUSSION In this study of virtual visits for unscheduled acute care in a single health system, vendor-supplied physicians were predicted to prescribe an antibiotic in 15% more acute respiratory infection visits compared to system-employed emergency physicians (35% vs 19%). Physician affiliation and familiarity with a health system, in addition to other factors, may be important in guideline adherence and antibiotic stewardship in direct-to-consumer telemedicine encounters.