Antibiotic review kit for hospitals (ARK-Hospital): a stepped-wedge cluster-randomised controlled trial.
Llewelyn MJ., Budgell EP., Laskawiec-Szkonter M., Cross ELA., Alexander R., Bond S., Coles P., Conlon-Bingham G., Dymond S., Evans M., Fok R., Frost KJ., Garcia-Arias V., Glass S., Gormley C., Gray K., Hamson C., Harvey D., Hills T., Iyer S., Johnson A., Jones N., Kang P., Kiapi G., Mack D., Makanga C., Mawer D., McCullagh B., Mirfenderesky M., McEwen R., Nag S., Nagar A., Northfield J., O'Driscoll J., Pegden A., Porter R., Powell N., Price D., Sheridan E., Slatter M., Stewart B., Watson C., Weichert I., Sivyer K., Wordsworth S., Quaddy J., Santillo M., Krusche A., Roope LSJ., Mowbray F., Hand KS., Dobson M., Crook DW., Vaughan L., Hopkins S., Yardley L., Peto TEA., Walker AS.
BACKGROUND: Strategies to reduce antibiotic overuse in hospitals depend on prescribers taking decisions to stop unnecessary antibiotic use. There is scarce evidence for how to support these decisions. We evaluated a multifaceted behaviour change intervention (ie, the antibiotic review kit) designed to reduce antibiotic use among adult acute general medical inpatients by increasing appropriate decisions to stop antibiotics at clinical review. METHODS: We performed a stepped-wedge, cluster (hospital)-randomised controlled trial using computer-generated sequence randomisation of eligible hospitals in seven calendar-time blocks in the UK. Hospitals were eligible for inclusion if they admitted adult non-elective general or medical inpatients, had a local representative to champion the intervention, and could provide the required study data. Hospital clusters were randomised to an implementation date occurring at 1-2 week intervals, and the date was concealed until 12 weeks before implementation, when local preparations were designed to start. The intervention effect was assessed using data from pseudonymised routine electronic health records, ward-level antibiotic dispensing, Clostridioides difficile tests, prescription audits, and an implementation process evaluation. Co-primary outcomes were monthly antibiotic defined daily doses per adult acute general medical admission (hospital-level, superiority) and all-cause mortality within 30 days of admission (patient level, non-inferiority margin of 5%). Outcomes were assessed in the modified intention-to-treat population (ie, excluding sites that withdrew before implementation). Intervention effects were assessed by use of interrupted time series analyses within each site, estimating overall effects through random-effects meta-analysis, with heterogeneity across prespecified potential modifiers assessed by use of meta-regression. This trial is completed and is registered with ISRCTN, ISRCTN12674243. FINDINGS: 58 hospital organisations expressed an interest in participating. Three pilot sites implemented the intervention between Sept 25 and Nov 20, 2017. 43 further sites were randomised to implement the intervention between Feb 12, 2018, and July 1, 2019, and seven sites withdrew before implementation. 39 sites were followed up for at least 14 months. Adjusted estimates showed reductions in total antibiotic defined daily doses per acute general medical admission (-4·8% per year, 95% CI -9·1 to -0·2) following the intervention. Among 7 160 421 acute general medical admissions, the ARK intervention was associated with an immediate change of -2·7% (95% CI -5·7 to 0·3) and sustained change of 3·0% (-0·1 to 6·2) in adjusted 30-day mortality. INTERPRETATION: The antibiotic review kit intervention resulted in sustained reductions in antibiotic use among adult acute general medical inpatients. The weak, inconsistent intervention effects on mortality are probably explained by the onset of the COVID-19 pandemic. Hospitals should use the antibiotic review kit to reduce antibiotic overuse. FUNDING: UK National Institute for Health and Care Research.