Estimating risk factors for farm-level transmission of disease: foot and mouth disease during the 2001 epidemic in Great Britain.

Controlling an epidemic would be aided by establishing whether particular individuals in infected populations are more likely to transmit infection. However, few analyses have characterised such individuals. Such analyses require both data on who infected whom and on the likely determinants of transmission; data that are available at the farm level for the 2001 Foot and Mouth Disease epidemic in Great Britain. Using these data a putative number of daughter infected premises (IPs) resulting from each IP was calculated where these daughters were within 3km of the IP. A set of possible epidemiological, demographic, spatial and temporal risk factors were analysed, with the final multivariate generalised linear model (Poisson error term) having 6 statistically significant (p<0.05) main effects including geographic area, local cattle and sheep densities, and the number of non-IP culls. This model demonstrates that farms are heterogeneous in their propensity to transmit infection to other farms and, importantly, that it may be possible to identify holdings that are at high risk of spreading disease a priori. Such information could be used to help prioritise the response to an epidemic.