Modelling the Risk of Highly Pathogenic Avian Influenza H5N1 in Wild Birds and Poultry of China.

This paper applied an integrated spatial regression model to explore the associations between ten environmental variables and the highly pathogenic avian influenza (HPAI). A subtype H5N1 cases in wild birds and poultry in China, and to predict the spatial distribution of HPAI H5N1 relative risk. Here a generalized linear mixed model (GLMM) incorporated with a variogram model through its random effects item, used as the spatial regression model. Four environmental variables were found to have significant effects, including annual mean temperature, poultry density, distance to lakes and wetlands, and distance to bird migration routes. The Root Mean Square Error of arbitrary 15 sample data was 11.56. Further, the high predicted relative risk areas of HPAI H5N1 were mainly in the Northwest, Middle, Southwest and Southeast part of China. With its simple structure and good prediction ability, this spatial regression model was very promising for predicting the risk of other disease.