Cross-scale effect of microbiota dynamics on resistance

Antimicrobial resistance (AMR) is often carried asymptomatically in the human gut microbiota, yet the role of this carriage in AMR persistence and transmission remains poorly understood. To address this gap, we developed a cross-scale model that links within-host microbiota dynamics of extended-spectrum beta-lactamase (ESBL)-producing and non-ESBL-producing bacteria with transmission across hospital and community settings. The model incorporates plasmid exchange, antibiotic effects, and host heterogeneity. It integrates diverse data sources, including microbiota dynamics, community antibiotic consumption, and detailed surveillance data from a large university hospital, complemented by literature estimates. This model identifies mechanisms that enable coexistence of resistant and sensitive strains at both individual and population levels, revealing host heterogeneity as a critical driver of resistance persistence. Unlike classic epidemiological models, our analysis finds a continuum of colonization states within individuals rather than discrete resistant or sensitive categories, and highlights that realistic detection thresholds may substantially underestimate resistance prevalence. Finally, we use this framework to evaluate interventions such as hospital screening and isolation, showing that their impact strongly depends on local transmission dynamics and resistance prevalence. By bridging within-host and population processes, our model provides a novel tool for investigating AMR spread and optimizing prevention strategies in healthcare and community environments.