Bluetongue virus (BTV), is an economically important, midge-borne Orbivirus (family: Reoviridae) that causes disease in sheep and cattle. In south India, bluetongue disease (BT) occurs annually, whilst sporadic outbreaks occur elsewhere in the country. Within this study, the mean annual number of outbreaks in each district (1992-2009) was examined in relation to land-cover, host availability and climate predictors using a Bayesian generalised linear mixed model with Poisson errors and Conditional AutoRegressive error structure. A rigorous, hierarchical variable selection process was utilised based on information criteria. Land-cover predictors and host predictors outperformed climate predictors as explanatory variables describing patterns in outbreaks. More outbreaks occurred on average in districts with higher coverage of irrigated croplands or standing water following flooding. Number of outbreaks also increased with densities of exotic and local livestock breeds, which may be more susceptible toBTV, and with buffalo density. The value of the resulting cross-India prediction maps of BT risk, to state and district veterinary officers and epidemiologists engaged in surveillance, is discussed.
展开▼
