Investigating a propagation of emerging carbapenemase-producing Enterobacteriaceae in Dutch broiler production pyramid through stochastic simulation

Simulating resistant bacteria transmission in livestock informs surveillance strategies for emerging threats like Carbapenem-resistant Enterobacteriaceae (CPE), aiding targeted surveillance and detecting CPE through active methods. We employed a simulation model to assess three potential scenarios for introducing CPE: 1) a single import of live animals, 2) the use of contaminated feed, and 3) multiple imports of live animals. Employing the SimInf package, we constructed a population model for broiler production, encompassing rearing farms, multiplier farms, hatcheries, and broiler farms. Subsequently, we introduced CPE and allowed it to spread throughout the population using the Susceptible-Colonized (Infectious)-Susceptible model. The model ran for 10 years with 100 runs. In the single import scenario, 1-2 rearing and multiplier farms saw major outbreaks in all 100 runs, while the broiler farm experienced major outbreaks in only 10 out of 100 runs; in the feed scenario, major outbreaks occurred in rearing farms in 32 runs and in multiplier farms in 26 runs, with major outbreaks in broiler farms observed in all 100 runs; in the multiple import scenario, outbreaks in rearing and multiplier farms happened in all 100 runs, with these major outbreaks reaching the broiler farm in 91 out of 100 runs. CPE transmission from imported or colonized broilers is rapid but short-lived within the broiler population, contrasting with the sporadic and prolonged emergence of CPE from contaminated feed, resulting in lower cumulative probabilities of detection from imported or colonized animals (0-0.50) compared to contaminated feed (0.9-0.97) over a 10-year period. Sensitivity analysis indicated that key outcomes such as farm outbreaks, chicken colonization, and outbreak duration are highly correlated with age-associated reductions in transmission (ψ).