Adaptation and implementation of the ARK (Antibiotic Review Kit) intervention to safely and substantially reduce antibiotic use in hospitals: a feasibility study

Background: Antimicrobial stewardship initiatives in secondary care depend on clinicians undertaking antibiotic prescription reviews but decisions to limit antibiotic treatment at review are complex. Aim: To assess the feasibility and acceptability of implementing ARK (Antibiotic Review Kit), a behaviour change intervention made up of four components (brief online tool, prescribing decision aid, regular data collection and feedback process, and patient leaflet) to support stopping antibiotic treatment when it is safe to do so among hospitalized patients; before definitive evaluation through a stepped-wedge cluster-randomized controlled trial. Methods: Acceptability of the different intervention elements was assessed for a period of 12 weeks by uptake of the online tool, adoption of the decision aid into prescribing practice, and rates of decisions to stop antibiotics at review (assessed through repeated point-prevalence surveys). Patient perceptions of the information leaflet were assessed through a brief questionnaire. Findings: All elements of the intervention were successfully introduced into practice. A total of 132 staff encompassing a broad range of prescribers and non-prescribers completed the online tool (19.4 per 100 acute beds), including 97% (32/33) of the pre-specified essential clinical staff. Among 588 prescription charts evaluated in seven point-prevalence surveys over the 12-week implementation period, 82% overall (76-90% at each survey) used the decision aid. The median antibiotic stop rate post implementation was 36% (range: 29-40% at each survey) compared with 9% pre implementation (P < 0.001). Conclusion: ARK provides a feasible and acceptable mechanism to support stopping antibiotics safely at post-prescription reviews in an acute hospital setting. (C) 2019 The Healthcare Infection Society. Published by Elsevier Ltd. All rights reserved.