Spread of hepatitis E virus from pens to the national industry: complex network analysis and modelling

Hepatitis E is a zoonotic disease that is widespread in pigs and potentially severe in humans. Transmission between pigs occurs via the faecal-oral route, either by direct contact or via the environment. In addition, the main factor of propagation between farms is the transport of live animals. The pig production chain consists of many interconnected actors who form a complex network in space and time. This study aimed to test control scenarios at a national scale by considering the many transmission routes and movements of live animals. Pig movements from 2017-2020 provided by the French national pig traceability database (BDporc) were analysed. Exponential random-graph models were fitted for three subsets defined by the type of animals transported (sows, piglets or growing pigs) to estimate probabilities of contact between farms. The results were validated using cross validation of goodness-of-fit analysis and maps. The estimated contact probabilities and general characteristics of farms were used to build a demographic model that simulated pig movements at three levels − between pens, between production sectors and between farms − according to a deterministic schedule, but with random destinations weighted by the contact probabilities. Finally, the simulated dynamics were used to feed a multi-level stochastic epidemiological model that simulated within- and between-farm transmission. The prevalence of hepatitis E virus -positive pigs shipped to the slaughterhouse was estimated for different control scenarios to evaluate the effectiveness of biosecurity measures implemented at the farm level.