Bayesian estimation of the prevalence of antimicrobial resistance: a mathematical modelling study 

The study aimed to determine if a Bayesian approach could be a more reliable and resource-efficient method for estimating the prevalence of antimicrobial resistance (AMR) than traditional frequentist methods. It used retrospectively collected healthcare data and iterative random sampling and cross-validation to assess predictive accuracy and efficiency. Results showed that Bayesian estimation of AMR prevalence made fewer extreme estimation errors and required fewer observed antimicrobial susceptibility results per pathogen. The Bayesian approach was maximally effective and efficient for drug-pathogen combinations where the actual prevalence of resistance was not close to 0% or 100%.