Sustained Reductions in Overall and Unnecessary Antibiotic Prescribing at Primary Care Clinics in a Veterans Affairs Healthcare System Following a Multifaceted Stewardship Intervention

Background: Most antibiotic prescribing is in outpatient settings. However, antibiotic stewardship has focused overwhelmingly on hospitalized patients. In a few studies, behavioral interventions decreased unnecessary outpatient prescribing against acute respiratory infections, but data are conflicting on sustained benefits after intervention discontinuation.

Methods: We conducted a prospective, observational study in 7 primary care clinics, in which an intervention comprised of clinician education, peer comparisons, and computer decision support order sets was directed against all antibiotic prescribing. After 6 months, peer comparisons were discontinued. Antibiotic prescribing was compared in the baseline (January-June 2016), intervention (January-June 2017), and postintervention (January-June 2018) periods.

Results: Mean antibiotic prescriptions significantly decreased from 76.9 (baseline) to 49.5 (intervention) and 56.3 (postintervention) per 1000 visits (35.6% and 26.8% reductions, respectively; P values < .001). The rate of unnecessary antibiotic prescribing (ie, antibiotic not indicated) decreased from 58.8% (baseline) to 37.8% (intervention) and 44.3% (postintervention) (35.7% and 24.7% decreases, respectively; P = .001 and P = .01). Overall, 19.9% (27/136), 36.6% (66/180), and 34.9% (67/192) of antibiotics were prescribed optimally (ie, antibiotics were indicated, and a guideline-concordant agent was prescribed for guideline-concordant duration) during the baseline, intervention, and postintervention periods, respectively (baseline vs intervention and postintervention, P = .001 and P = .003, respectively). Differences between intervention and postintervention periods in overall, unnecessary, or optimal antibiotic prescribing were not significant.

Conclusions: A multifaceted outpatient stewardship intervention achieved reductions in overall, unnecessary, and suboptimal antibiotic prescription rates, which were sustained for a year after components of the intervention were discontinued. There is opportunity for further improvement, as inappropriate and suboptimal prescribing remained common.