Epidemiological modelling helps control endemic animal diseases

Context

Controlling endemic diseases is a major challenge for sustainable animal husbandry and competitive agri-food chains, but also for veterinary public health. These diseases have an impact on animal husbandry because they persist for a long time, generate heavy losses, degrade animal welfare, and sometimes lead to the use of antibiotics. Predicting their spread must take into account the specificities of animal husbandry populations: heterogeneous, managed, and interacting. Managing these diseases also requires different options to be arbitrated according to the losses due to the disease and the management costs. A better understanding of infection processes, resource allocation to management and farmer compliance is needed. However, these processes are rarely observed. Mechanistic modelling can be used to describe the spread of pathogens in a wide range of contexts and to explore the effectiveness of strategies at different scales (animal, farm, region, supply chain).

Results

A group of multi-institutional modellers discussed the benefits and challenges of mechanistic epidemiological modelling for endemic animal diseases (e.g. bovine viral diarrhoea, bovine paratuberculosis, porcine respiratory dysgenic syndrome). We addressed the issues of modelling the spread of pathogens at the farm and regional levels, in order to better assess the control of these diseases and propose targeted options. A focus was placed on intra-host models representing host immune response and heterogeneity between animals. Finally, the link between mechanistic models and observations was discussed. Our conclusions are :

- Multi-scale mechanistic models contribute to a better understanding of the spread of endemic pathogens;

- Interdisciplinarity is the key to building such models;

- Territorial anchoring, fed by observations, ensures realism and robustness of predictions;

- The complexity of the models must be compensated by affordable data;

- Economic and epidemiological models should be coupled to better guide the control of unregulated diseases.

Perspectives

Complementary skills must be mobilised in modelling, statistics, computer science, infectious diseases, immunology, epidemiology and economics to meet future challenges: (1) ensuring FAIR (Findable, Accessible, Interoperable, Reusable) data, as animal health data is heterogeneous and held by many actors; (2) capturing the complex behaviour of the system and the decisions of actors to manage it; (3) converting simulated results into usable knowledge for stakeholders; (4) ensuring realistic and practical results. Academic models are too complex to be transferred to health managers, while health managers are starting to use their results to plan their actions. Decision support tools based on academic models need to be developed in collaboration with users to prioritise management alternatives in a wide range of situations.

Valorisation

Article from a working group convened within the framework of the PIA ANR-10-BINF-07 (MIHMES), co-financed by the ERDF Pays de la Loire, coordinated by P. Ezanno (BIOEPAR, INRAE, Oniris).

Bibliographic references : 

Ezanno P., Andraud M., Beaunée G., Hoch T., Krebs S., Rault A., Touzeau S., Vergu E., Widgren S. 2020. How mechanistic modelling supports decision making for the control of enzootic infectious diseases. Epidemics 32:100398, https://doi.org/10.1016/j.epidem.2020.100398.

Epidemic infectious diseases (red) cause significant losses over short periods of time, while endemic infectious diseases (orange) persist and can result in significant cumulative incidence. Data collection is necessary to assess the impact of diseases and evaluate the effectiveness of control strategies.