Antibiotics | The Impact of Pen-Level Sample Testing on Antimicrobial Selection for Bovine Respiratory Disease Treatment

This study employs a hybrid agent-based and discrete-event simulation (ABM-DES) model to investigate how different pen-level sampling and antimicrobial resistance (AMR) detection strategies affect antimicrobial use and resistance management in bovine respiratory disease (BRD) treatment within western Canadian feedlots, providing methodological support for future AMR risk intervention experiments.

 

Literature Overview
This study, titled 'The Potential for Sample Testing at the Pen Level to Inform Prudent Antimicrobial Selection for Bovine Respiratory Disease Treatment: Investigations Using a Feedlot Simulation Tool' published in Antibiotics, reviews and summarizes the impact of pen-level sampling and laboratory testing on antimicrobial selection for bovine respiratory disease (BRD) treatment, as well as the effects of sampling timing and detection methods on AMR prevalence and economic outcomes. Through simulation experiments, the article emphasizes the potential value of sampling and detection strategies in antimicrobial management while highlighting their dependence on multiple factors, including detection accuracy, AMR prevalence, and treatment switch thresholds.

Background Knowledge
Antimicrobial resistance (AMR) poses significant threats to global public health and livestock management, primarily driven by antimicrobial misuse and overuse. Bovine respiratory disease (BRD) is the most prevalent illness in North American feedlots and a major driver of antimicrobial use. Current BRD treatment strategies predominantly rely on clinical symptoms rather than laboratory testing, increasing risks of resistant pathogen spread and treatment failure. The research team previously developed a hybrid agent-based and discrete-event simulation (ABM-DES) model to simulate AMR dynamics in high-risk cattle populations and explore sampling strategies for antimicrobial management. This study aims to validate whether pen-level sampling and detection can optimize antimicrobial selection, reduce resistance development, and assess its applicability under diverse conditions, offering theoretical support for future AMR control strategies.

 

 

Research Methods and Experiments
The study builds upon a previously developed stochastic, continuous-time ABM model, expanding it into a hybrid ABM-DES framework to simulate pen-level sampling and detection workflows. The model incorporates a Testing agent to account for time delays in sampling, transportation, processing, and detection. Sampling timepoints include feedlot arrival (0 days on feed [DOF]) and 13 days on feed (DOF), with detection methods encompassing antimicrobial susceptibility testing (AST) and long-read metagenomic sequencing (MS). A 25% AMR prevalence threshold was established as a treatment switch criterion to evaluate its impact on BRD recurrence, antimicrobial use, and AMR prevalence.

Key Conclusions and Perspectives

• Under baseline epidemiological and treatment conditions, testing-informed (TI) strategies demonstrated no significant impact on antimicrobial management or economic outcomes.
• When assuming higher AMR prevalence thresholds, antimicrobial categories and usage patterns changed, though resistance or disease incidence remained unaffected.
• Sampling timing and detection accuracy significantly influence AMR positivity rates and related outputs, particularly under extreme BRD treatment scenarios where all cases receive 15-membered macrolide antimicrobials.
• Model simulations indicate implementation effectiveness of pen-level sampling and detection strategies depends heavily on BRD treatment protocols, AMR prevalence, detection accuracy, and treatment switch thresholds.
• This research validates the hybrid ABM-DES model's flexibility in simulating diverse intervention strategies, providing methodological support for future AMR control studies.

Research Significance and Prospects
This study establishes theoretical foundations for pen-level sampling and detection applications in BRD treatment. Findings suggest this strategy may reduce inappropriate antimicrobial use and control resistance development under specific conditions. Future research should optimize detection methods, sampling frequencies, and processing delays to enhance model practicality. Additionally, the model can be extended to other livestock systems for evaluating AMR control strategies, offering global antimicrobial management with data support and policy recommendations.

 

 

Conclusion
This study utilizing a hybrid ABM-DES model investigated pen-level sampling and detection strategies for antimicrobial selection in BRD treatment. Results indicate that while TI strategies showed limited improvement under baseline conditions, they demonstrated measurable impacts on antimicrobial use and resistance control under high AMR prevalence or extreme treatment scenarios. The model confirms complex interdependencies among sampling timing, detection accuracy, AMR thresholds, and treatment protocols, emphasizing their critical roles in AMR mitigation. These findings provide experimental validation and modeling support for future detection-based AMR intervention strategies, contributing to optimized antimicrobial management practices that enhance livestock sustainability and public health safety.

 

Dana E Ramsay, Wade McDonald, Sheryl P Gow, Nathaniel D Osgood, and Cheryl L Waldner. The Potential for Sample Testing at the Pen Level to Inform Prudent Antimicrobial Selection for Bovine Respiratory Disease Treatment: Investigations Using a Feedlot Simulation Tool. Antibiotics.