Antimicrobial resistance (AMR) is a growing European and global health problem in both humans and animals, leading to limited or poor treatment options for many diseases. Containment of AMR spread is part of the European Commission action plan against the rising threats from AMR. However, consensus estimates on actual exposure risk, sources, transmission routes and disease burden is scarce. A major reason for the lack of well-developed analyses might be the complexity of these approaches and their data requirements, and their interdisciplinary nature. Crucial to progression on this theme is the development of modelling methodology and the systematic integration of data. Currently much microbiological, molecular and epidemiological data is being gathered on AMR development, prevalence, spread, transmission, risks, etc. in the veterinary, food and human domain. However, assessment of the importance of different transmission routes and quantifying public health effects (i.e. disease burden) associated with AMR represent major knowledge gaps. In order to make full use of all available data, to check consistency, and to find consensus estimates, both problems (attribution and disease burden) require an integrated framework. In different WPs this project will generate genomic information to feed into transmission models (WP1), will establish estimates for the relation between antimicrobial use and the development of resistance (WP2), develop transmission models for AMR spread between and within animal and human population, establish risk assessment models to estimate the risk of spread through the food chain (WP3), and will construct a methodological framework for estimating the disease burden of AMR (WP5). The extreme diverse nature of the data available and generated in this project requires innovative methods to effectively integrate this data into final estimates for source attribution, risks and disease burden. To that end we will employ machine learning techniques (WP4) and Bayesian Evidence Synthesis (WP6). To this end innovative modelling techniques such as machine learning and Bayesian evidence synthesis will be employed. This project is expected to deliver harmonized modelling approaches and methodologies that are clearly beyond the state-of-the-art and will change the way of approaching the AMR problem to an integrative approach.

• dr. L. (Lapo) Mughini Gras

• Eelco Franz (RIVM)
• Tine Hald (DTU)
• Aline de Koeijer (WBVR)
• Thomas Sellhorst (BfR)
• Christine Muller-Graf (BfR)
• Annemarie Kaesbohrer (DfR)
• Luca Busani (ISS)
• Annalisa Pantosti (ISS)
• Robin Simons (APHA)
• Emma Snary (APHA)
• Madelaine Norstrom (NVI)
• Camilla Sekse (NVI)
• Michel-Yves Mistou (ANSES)
• Pascal Sanders (ANSES)
• Sophie Granier (ANSES)