Antibiotic Stewardship Strategies and Their Association With Antibiotic Overuse After Hospital Discharge: An Analysis of the Reducing Overuse of Antibiotics at Discharge (Road) Home Framework

Abstract

Background: Strategies to optimize antibiotic prescribing at discharge are not well understood.

Methods: In fall 2019, we surveyed 39 Michigan hospitals on their antibiotic stewardship strategies. The association of reported strategies with discharge antibiotic overuse (unnecessary, excess, suboptimal fluoroquinolones) for community-acquired pneumonia (CAP) and urinary tract infection (UTI) was evaluated in 2 ways: (1) all strategies assumed equal weight and (2) strategies were weighted based on the ROAD (Reducing Overuse of Antibiotics at Discharge) Home Framework (ie, Tier 1-Critical infrastructure, Tier 2-Broad inpatient interventions, Tier 3-Discharge-specific strategies) with Tier 3 strategies receiving the highest weight.

Results: Between 1 July 2017 and 30 July 2019, 39 hospitals with 20 444 patients (56.5% CAP; 43.5% UTI) were included. Survey response was 100%. Hospitals reported a median (interquartile range [IQR]) 12 (9-14) of 34 possible stewardship strategies. On analyses of individual stewardship strategies, the Tier 3 intervention, review of antibiotics prior to discharge, was the only strategy consistently associated with lower antibiotic overuse at discharge (adjusted incident rate ratio [aIRR] 0.543, 95% confidence interval [CI]: .335-.878). On multivariable analysis, weighting by ROAD Home tier predicted antibiotic overuse at discharge for both CAP and UTI. For diseases combined, having more weighted strategies was associated with lower antibiotic overuse at discharge (aIRR 0.957, 95% CI: .927-.987, per weighted intervention); discharge-specific stewardship strategies were associated with a 12.4% relative decrease in antibiotic overuse days at discharge.

Conclusions: The more stewardship strategies a hospital reported, the lower its antibiotic overuse at discharge. However, Tier 3, or discharge-specific strategies, appeared to have the largest effect on antibiotic prescribing at discharge.

Keywords: antibiotic stewardship; pneumonia; quality of care; transitions of care; urinary tract infection.

Conflict of interest statement

Potential conflicts of interest. S. F. reports personal fees from Wiley Publishing. A. H. reports receiving funding from the Centers for Disease Control and Prevention (CDC), Agency for Healthcare Research and Quality (AHRQ), and National Institutes of Health (NIH) outside of the submitted work; participation on a Data Safety Monitoring Board or Advisory Board for National Institute of Allergy and Infectious Diseases (NIAID) and leadership or fiduciary roles for the Pediatric Infectious Diseases Society (PIDS) and Infectious Diseases Society of America (IDSA). V. V. reports a related program grant from an unrelated Diagnostic Error grant from Betty and Gordon Moore Foundation, and an unrelated National Heart, Lung, and Blood Institute (NHLBI) loan repayment program and speaking fees for lecture on coronavirus disease (COVID) from Thermo Fisher Scientific. All other authors report no potential conflicts. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.

© The Author(s) 2022. Published by Oxford University Press for the Infectious Diseases Society of America.

Figures

Figure 1.

Antibiotic overuse after discharge in patients treated for pneumonia or urinary tract infection, by hospital (N = 39 hospitals). Variation in antibiotic overuse at discharge across hospitals is shown. Each bar represents a single hospital. Antibiotic overuse is divided by 4 types of overuse: (a) unnecessary antibiotics (asymptomatic bacteriuria or patients treated for pneumonia who did not have pneumonia), (b) excess duration, (c) suboptimal use of fluoroquinolones, and (d) suboptimal use of fluoroquinolones and excess duration. Hospitals marked with a star are those reporting at least 1 Tier 3 or discharge-specific stewardship strategy. Hospitals 1, 2, and 3 are those described in detail in Table 3.

Figure 2.

A, Association of stewardship strategies with antibiotic overuse at discharge (stewardship strategies unweighted). Bars shown represent the adjusted (with 95% CI) number of days of antibiotic overuse at discharge in hospitals with the specific number of unweighted strategies (eg, hospitals with 2 unweighted strategies had, on average, 2.7 days of antibiotic overuse at discharge per patient). Yellow line (model) shows the expected number of antibiotic overuse days at discharge based on the number of unweighted antibiotic stewardship strategies. Thus, values represent the adjusted overuse duration for an average patient treated at an average hospital where the number of unweighted antibiotic stewardship strategies varies from 1 to 8. Although 12 disease-agnostic strategies are possible, the maximum reported by included hospitals was 8 and thus data are not reported for values >8. B, Association of stewardship strategies with antibiotic overuse at discharge (stewardship strategies weight by tier). Bars shown represent the adjusted (red) number of days of antibiotic overuse at discharge in hospitals with the specified number of weighted strategies (e.g., hospitals with 8 weighted strategies had, on average, 2.4 days of antibiotic overuse at discharge per patient). Yellow line (model) shows the expected number of antibiotic overuse days at discharge based on the number of weighted antibiotic stewardship strategies (with Tier 3 strategies weighted 3 times Tier 1 strategies). Thus, the values represent the adjusted overuse duration for an average patient treated at an average hospital where the number of weighted antibiotic stewardship strategies is then varied from 1 to 16. Although 25 disease-agnostic weighted strategies are possible, the maximum reported by included hospitals was 16, and thus data are not reported for values >16. No hospitals reported 6, 12, or 15 weighted strategies. Abbreviation: CI, confidence interval.

Figure 3.

A, Association of tiered stewardship strategies with antibiotic overuse at discharge, pneumonia. Bars shown represent the adjusted (red) number of days of antibiotic overuse at discharge for patients with CAP in hospitals with the specified number of weighted strategies (eg, hospitals with 8 weighted strategies had, on average, 1.9 days of antibiotic overuse at discharge per CAP patient). Yellow line (model) shows the expected number of antibiotic overuse days at discharge based on the number of weighted antibiotic stewardship strategies (with Tier 3 strategies weighted 3 times Tier 1 strategies). Thus, values represent the adjusted overuse duration for an average patient treated at an average hospital where the number of weighted antibiotic stewardship strategies is then varied from 1 to 20. Although 33 CAP-related weighted strategies are possible, the maximum reported by included hospitals was 20, and thus data are not reported for values >20. No hospitals reported 6 or 18 weighted strategies. B, Association of tiered stewardship strategies with antibiotic overuse at discharge, urinary tract infection. Bars shown represent the adjusted (red) number of days of antibiotic overuse at discharge for patients with a UTI in hospitals with the specified number of weighted strategies (eg, hospitals with 8 weighted strategies had, on average, 2.8 days of antibiotic overuse at discharge per UTI patient). Yellow line (model) shows the expected number of antibiotic overuse days at discharge based on the number of weighted antibiotic stewardship strategies (with Tier 3 strategies weighted 3 times Tier 1 strategies). Thus, values represent the adjusted overuse duration for an average patient treated for a UTI at an average hospital where the number of weighted antibiotic stewardship strategies is then varied from 1 to 32. Although 57 UTI-related weighted strategies are possible, the maximum reported by included hospitals was 32, and thus data are not reported for values >32. No hospitals reported <6, 7, 10, 16, or 27–31 weighted strategies. Abbreviations: CAP, community-acquired pneumonia; UTI, urinary tract infection.