Supervisors: Peter Alexander, Glenn Marion, Dominic Moran
Project Description:
The project will combine epidemiological modelling of antimicrobials resistance (AMR) prevalence and response to selection pressures under different treatment regimes, with agent-based model (ABM) simulation of prescription and administration of antimicrobials and alternatives. AMR is recognised as one of the most important global (One Health) issues for human and animal health in terms of societal impact, with an estimated 10 million lives a year and a cumulative 100 trillion USD of economic output at risk due to the rise of drug resistance . AMR develops through a complex interaction of microbial processes in the environment driven by human actions, but there are key gaps in understanding the role of the environmental and the social dimensions.
Research questions addressed in this project include:
• How do the microbiological, environmental and social aspects interact within the agricultural sector to influence the rate of AMR development and spread?
• What sector interventions or policies are effective in reducing the spread of AMR, and do these actions create trade-offs with other sector objectives (e.g. food security)?
The project will develop AMR models within stylised regions linking multiple farms and veterinarian practices, based on an existing ABM framework of agricultural systems. ABMs allow the spatial and dynamic behaviour of complex systems to be investigated, including representing the process of learning and behavioural evolution, as well as cross-scale environmental interactions and heterogeneity in behaviours and agents . ABMs have been applied in studying the spread of infections in social systems, including behavioural responses, social networks and spatial aspects. There has been limited use for studying AMR within hospitals , and none in an agricultural settings.
Vets and farmers will be represented by a heterogeneous set of behavioural agents operating within environments containing mixed farm types. Levels of environmental, animal and human AMR will be characterised according to underlying models of microbial community response to agent decisions concerning treatment regime/system management. Human and microbiological behaviours will be based on, and parameterised using, information/methods from previously published research and via links to ongoing research, e.g. NE/N020162/1. Sensitivity analysis will take a probabilistic approach to allow uncertainty to be assessed using Monte Carlo methods. The resulting epidemiological systems models for AMR risk will account for a range of feedbacks influencing stakeholder behaviours; agents e.g. farmers and vets will adapt based their own experiences including observation of local microbial responses, as well as being influenced by others, behavioural cues, policies, and other socio-economic factors. Emergent systems-level outcomes in terms of both behavioural profile and AMR outcomes will be compared under alternative scenarios. These scenarios will be refined through interactions with related projects where AMR livestock scenarios are being developed, e.g. ES/P008194/1. This project will provide training in a range of highly transferable skills including in epidemiological and social-environmental ABM modelling, as well as related aspects of data science and analytics, and interdisciplinary working.
If you wish to apply for this project, please check this link and send your application to this email.
