Interventions for improving antibiotic prescribing in the community: systematic review and meta-analysis

BackgroundAntibiotic resistance is a public health priority and antibiotic use in humans is a major contributing factor to its development. Interventions to improve antibiotic prescribing in the community, where most antibiotics are prescribed, are widely implemented with varying effect. The aim was to systematically review and meta-analyse evidence on effectiveness of different types of antibiotic prescribing interventions in the community. Methods and FindingsMedline, Embase, and the Cochrane Central Register of Controlled Trials, were searched from database inception to 16 August 2021 to identify randomised controlled trials comparing antibiotic stewardship interventions versus usual care in community settings. Two reviewers screened studies, extracted data, and assessed risk of bias. Studies were grouped according to type of intervention. Meta-analyses employed random effects models. The outcome for meta-analyses was change in total antibiotic prescribing rates attributable to the intervention, compared to usual care, calculated as percentage differences. Other measures of change in antibiotic prescribing were included in narrative description. Of 26,130 studies screened, 74 were included, with 53 comparisons from 45 studies meta- analysed. 50% of included studies had high risk of bias. Single interventions with statistically significant reductions in total antibiotic prescribing were point of care tests for antigen detection (-28.0% reduction, 95%CI-38.2 to-17.8); educational materials (-17.0%, -31.0 to - 3.0); printed decision-support systems (-10.8%,-15.7 to -6.0), educational workshops (- 8.7%,-12.8 to -4.7), and; educational outreach (-6.0%,-10.6 to-1.4). Multifaceted interventions were not more effective than single interventions (education + audit and feedback -9.9%, -12.8 to -7.0; other multifaceted -9.4%, -17.2 to -1.6). Effect sizes in sensitivity analyses excluding trials at high risk of bias were similar or larger. ConclusionsCommunity antibiotic stewardship interventions were effective but with considerable variation in effect size. The most effective trial interventions may be more difficult to implement in practice, a key challenge for antibiotic stewardship. Systematic review registration PROSPERO CRD42014010160