Abstract

Bovine tuberculosis is a notifiable infectious disease of cattle caused by Mycobacterium bovis. Although the potential for zoonotic transmission exists, pasteurization of milk has effectively eliminated this risk within Great Britain. The impetus for control is therefore largely economic. The current test-and-slaughter policy cost £108 million in 2008-2009. Recent research on bTB has focused on understanding the relative roles played by cattle-movements and the wildlife reservoir of infection in badgers. Despite intensive research, the role of badgers in the natural history of infection of cattle herds is still poorly characterised. However, the control of bTB is based on a diagnostic test that has both poor sensitivity and specificity leading to the possibly that recurrence of bTB on individual farms may be a function simply of leaving hidden infection on a premise when it is declared “bTB free”. We set out to develop within-herd transmission models of bTB, parameterised from measures of the within-herd persistence of bTB. We use these empirical targets in an Approximate Bayesian Computation (ABC) framework to obtain parameter estimates for our model which are consistent with the natural history of transmission of bTB in Great Britain. From our parameterised model we hope to explore how the efficiency of the current testing regime can be improved to reduce the burden of bTB testing in GB.