Antibiotics: easier to start than to stop? Predictors of antimicrobial stewardship recommendation acceptance

Objectives: Acceptance of prospective audit and feedback antimicrobial stewardship programme (ASP) recommendations has been shown to vary, but the drivers of recommendation acceptance are not well understood. We sought to identify the factors associated with recommendation acceptance at a large community teaching hospital. Methods: Data from an ASP recommendation registry were collected from 2010 to 2018. Variables included data about the infection, the prescriber, and the recommendation, categorized by whether they increase, decrease, or are neutral to antibiotic exposure. The primary outcome was acceptance of ASP recommendations. Adjusted odds ratios and 95% confidence intervals were estimated using logistic regression models with random intercepts in order to account for clustering by prescriber. Results: Over the 8-year period, a total of 11 014 evaluable recommendations were made to 146 prescribers, and 9058 (82.2%) were accepted. The most common recommendations were: reduce duration (n = 2796; 25%), stop antibiotics (n = 2184; 20%), de-escalate (n = 1876; 17%) and increase duration (n = 1176; 11%). Acceptance by service ranged from 70% (n = 843/1196) (surgery) to 86% (n = 6378/7444) (general medicine). In the multivariable analysis, compared to recommendations that have a neutral impact on antibiotic exposure, recommendations to decrease antibiotic exposure had lower odds of acceptance (aOR 0.73; 95%CI 0.64-0.84) while recommendations to increase exposure were associated with greater acceptance (aOR 2.00; 95%CI 1.62-2.45). Other factors associated with increased acceptance included the presence of the ASP physician during rounds and making the recommendation verbally. Conclusions: Recommendations to decrease antibiotic exposure had lower odds of acceptance than those to increase antibiotic exposure. This study presents important considerations for ASPs with prospective audit and feedback programmes aiming to evaluate and increase the impact of their recommendations. Crown Copyright (C) 2020 Published by Elsevier Ltd on behalf of European Society of Clinical Microbiology and Infectious Diseases. All rights reserved.