An extensive re-evaluation of evidence and analyses of the Randomised Badger Culling Trial II: In neighbouring areas

In the second investigation in a pair of analyses which re-evaluates the Randomised Badger Culling Trial (RBCT), we estimate the effects of proactive badger culling on the incidence of tuberculosis (TB) in cattle populations in unculled neighbouring areas. Throughout peer-reviewed analyses of the RBCT, proactive culling was estimated to have detrimental effects on the incidence of herd breakdowns (i.e. TB incidents) in neighbouring areas. Using previously published, publicly available data, we appraise a variety of frequentist and Bayesian models as we estimate the effects of proactive culling on confirmed herd breakdowns in unculled neighbouring areas. For the during trial period from the initial culls until 4 September 2005, we estimate consistently high probabilities that proactive culling had adverse effects on confirmed herd breakdowns in unculled neighbouring areas, thus supporting the theory of heightened risk of TB for the neighbouring cattle populations. Negligible culling effects are estimated in the post-trial period across the statistical approaches and imply unsustained long-term effects for unculled neighbouring areas. Therefore, when considered alongside estimated beneficial effects within proactive culling areas, these conflicting adverse side effects render proactive culling complex, and thus, decision making regarding potential culling strategies should include (i) ecological, geographical and scientific considerations and (ii) cost-benefit analyses.