Prediction of the spread of highly pathogenic avian influenza using a multifactor network: Part 2 – Comprehensive network analysis with direct/indirect infection route

•HPAI could be spread by personnel/vehicles visiting farms, aerial influence.•HPAI spread network was constructed using information about fomites.•Farms that are vulnerable to infection could be identified from the network through centrality analyses.•The network was verified by comparing the centrality of infected and suspected farms.•This study explained the infection routes of HPAI using the network theory.

Epidemic diseases of domestic animals, such as high pathogenic avian influenza, can be directly spread by poultry-related business personnel/vehicles visiting farms and can be indirectly caused by aerial spread. However, most of comprehensive analysis studies have considered only direct or indirect causes. The purpose of this study is to construct a direct HPAI spread network based on the relationships between farms using poultry-related business data from 39 farms in the Kimje district of South Korea and an indirect HPAI spread network using the aerial spread from each farm during the HPAI outbreak in 2008. Direct/indirect HPAI spread networks were also analysed using centrality analyses to identify highly vulnerable farms for infection and highly influential farms on other farms so the initial spread can be prevented. This study proposes an optimal prevention solution through a comprehensive spread simulation and compares the centrality of infected/suspected farms by tracing the infection routes for network validation.