JAMA Netw Open. 2026 Jan 2;9(1):e2549655. doi: 10.1001/jamanetworkopen.2025.49655.
ABSTRACT
IMPORTANCE: Antibiotics prescribed at hospital discharge are frequently unnecessary or suboptimal. Strategies to improve prescribing are not well defined.
OBJECTIVE: To evaluate whether a discharge-focused prospective audit and feedback process decreases antibiotic overuse at hospital discharge.
DESIGN, SETTING, AND PARTICIPANTS: This stepped-wedge cluster-randomized clinical trial was conducted across participating units at 10 hospitals with antibiotic stewardship (AS) teams and supporting staff from December 5, 2022, to November 17, 2023. After a 24-week baseline period, 1 hospital crossed into the intervention arm every 2 weeks.
INTERVENTION: The intervention consisted of disseminating institutional guidelines for oral antibiotic step-down to frontline prescribers and conducting a prospective audit and feedback process for inpatients receiving antibiotics with an anticipated discharge date in the next 48 hours.
MAIN OUTCOMES AND MEASURES: The primary outcome was postdischarge antibiotic use. Secondary outcomes included inpatient antibiotic use, length of hospital stay, and readmission. Manual electronic health record reviews were performed in 434 cases to assess optimal antibiotic prescribing at discharge for patients who met specific criteria. Analysis was performed on a per-protocol basis.
RESULTS: There were 21 842 patient admissions (baseline, 14 288; intervention, 7554) across 10 hospitals. The median (IQR) age was 66 (53-75) years, with 13 380 (61.3%) males. At the hospital level, the mean (SD) number of patients audited by the AS team per week was 19.9 (5.8); approximately one-quarter of these audits (mean [SD], 5.0 [2.6]) resulted in feedback to the frontline prescribers. There were 3133 patients (21.9%) prescribed postdischarge antibiotics at baseline compared with 1645 patients (21.8%) during the intervention (odds ratio, 0.94 [95% CI, 0.84-1.05]). The mean (SD) postdischarge antibiotic duration was 7.1 (5.2) days at baseline compared with 7.6 (5.6) days during the intervention (mean difference, 0.02 [95% CI, -0.50 to 0.53] days). There were no statistical differences during the intervention compared with baseline for inpatient antibiotic duration (mean [SD], 4.4 (3.6) vs 4.2 [3.5] days; mean difference, 0.04 [95% CI, -0.20 to 0.27] days), length of hospital stay (mean [SD], 5.4 [4.8] vs 5.4 [5.0] days; mean difference 0.11 [95% CI, -0.12 to 0.33] days), or hospital readmission within 30 days (odds ratio, 1.02 [95% CI, 0.88-1.18]). Optimal antibiotic prescribing was more common during the intervention (122 of 264 cases [46.2%] vs 100 of 170 cases [58.8%]; odds ratio, 1.61 [95% CI, 1.08-2.40]). A total of 112 inpatient frontline prescribers were sent a postintervention survey; 40 (35.7%) responded, and 34 of 36 (94.4%) believed that the initiative improved antibiotic prescribing at hospital discharge.
CONCLUSIONS AND RELEVANCE: In this stepped-wedged cluster-randomized clinical trial conducted across 10 hospitals, discharge-focused prospective audit and feedback did not decrease antibiotic use at hospital discharge but did improve optimal antibiotic prescribing for common and uncomplicated diagnoses. Other AS strategies are needed to decrease unnecessary antibiotic prescribing at this transition of care.
TRIAL REGISTRATION: ClincialTrials.gov Identifier: NCT05471726.
PMID:41511774 | DOI:10.1001/jamanetworkopen.2025.49655
