Know more

Context and issues

The control of infectious livestock diseases spread through the animal trade is a major challenge, particularly to ensure sustainable agriculture. This requires improved control strategies, particularly for unregulated diseases where decisions are left to individual farmers' initiatives or to collective regional initiatives. In this context, epidemiological models provide a detailed description of the spread of pathogens and can be used to assess the effectiveness of control measures. However, taking into account the individual decisions of farmers in such models remains a challenge.

Results

In a study published in Scientific Reports, researchers from the MaIAGE unit (INRAE, Université Paris-Saclay) in Jouy-en-Josas and the BIOEPAR unit (INRAE, Oniris) in Nantes addressed this question. They proposed a new integrative model of the spread of a livestock disease between farms via a trade network, by integrating the adaptive decisions of farmers regarding the adoption of a control measure in their herd. The modelling of the decision mechanism is based, at each moment of decision and for each farmer, on the economic implication of the previous individual decision as well as those of the trading partners. This formalisation takes into account certain real phenomena that can intervene in human decision-making: stochastic behaviour, learning and the emergence of imitation and free-riding. The two components of the model - network-based demo-epidemic dynamics and decision making - are linked by a feedback loop. A control measure based on vaccination was considered and the model was evaluated numerically by means of intensive simulations and sensitivity analyses. For a given epidemiological situation, the type of interaction between actors in the vaccination decision making was found to have a strong impact on the epidemic dynamics. Sensitivity analyses confirmed these results and validated the fact that the model is influenced by all types of parameters: epidemiological, economic and decision-related. In particular, the frequency of decision making and the effectiveness of the vaccine showed a very strong influence on the model's outputs, given fixed epidemiological parameters.

Perspectives

This study contributes to the understanding of the interaction between epidemic dynamics in farm animals and human behaviour regarding the voluntary adoption of measures to control them. The model developed can be used for other epidemiological model structures (theoretical or for specific diseases) or different types of interventions, by adapting its components.

Valorisation

This work is part of the ANR CADENCE project, led by E. Vergu of the MaIAGE unit. The model code is publicly available: https://github.com/CristanchoLina/IntegrativeEpiDecisionModel.

Bibliographical references

Cristancho Fajardo, L., Ezanno, P. & Vergu, E. Accounting for farmers’ control decisions in a model of pathogen spread through animal trade. Sci Rep 11, 9581 (2021). https://doi.org/10.1038/s41598-021-88471-6

(A) Schematic of the stochastic integrative model at the farm level, including the epidemiological and decision-making components, as well as a cost function evaluation step and a decision implementation. (B) Temporal dynamics of vaccination decisions for two different values (upper and lower panels) of the parameter reflecting the sensitivity of farmers to their own costs and used in the decision component. Results corresponding to a realisation of the stochastic integrative model schematised in (A). NV (non-vaccination), V (vaccination). Each colour represents a different vaccination scheme, defined by the sequence of vaccination decisions at each of the six decision instants. For example, the pattern [NV1, NV2, V3, V4, V5, V6] refers to farms that do not vaccinate their animals at the first two decision points and always vaccinate afterwards. Each vertical line represents a decision point in time and the width of the flows between two such points in time is proportional to the frequency in the population of farmers who have chosen a given vaccination scheme.