Spatial Targeting for Bovine Tuberculosis Control: Can the Locations of Infected Cattle Be Used to Find Infected Badgers?

• DOI: 10.1371/journal.pone.0142710
• Publication Date: 2015-11-13T19:02:25Z
• Authors: Catherine M. Smith; Sara H. Downs; Andy Mitchell; Andrew C. Hayward; Hannah Fry; Steven C. Le Comber
• Keywords: 0301 basic medicine; Science; Population Dynamics; Kaplan-Meier Estimate; Random Allocation; 03 medical and health sciences; Mustelidae; Animals; Cluster Analysis; Proportional Hazards Models; Spatial Analysis; Geography; Data Collection; Q; R; Mycobacterium bovis; 3. Good health; England; Communicable Disease Control; Geographic Information Systems; Medicine; Cattle; Animal Distribution; Tuberculosis, Bovine; Research Article

Bovine tuberculosis is a disease of historical importance to human health in the UK that remains a major animal health and economic issue. Control of the disease in cattle is complicated by the presence of a reservoir species, the Eurasian badger. In spite of uncertainty in the degree to which cattle disease results from transmission from badgers, and opposition from environmental groups, culling of badgers has been licenced in two large areas in England. Methods to limit culls to smaller areas that target badgers infected with TB whilst minimising the number of uninfected badgers culled is therefore of considerable interest. Here, we use historical data from a large-scale field trial of badger culling to assess two alternative hypothetical methods of targeting TB-infected badgers based on the distribution of cattle TB incidents: (i) a simple circular 'ring cull'; and (ii) geographic profiling, a novel technique for spatial targeting of infectious disease control that predicts the locations of sources of infection based on the distribution of linked cases. Our results showed that both methods required coverage of very large areas to ensure a substantial proportion of infected badgers were removed, and would result in many uninfected badgers being culled. Geographic profiling, which accounts for clustering of infections in badger and cattle populations, produced a small but non-significant increase in the proportion of setts with TB-infected compared to uninfected badgers included in a cull. It also provided no overall improvement at targeting setts with infected badgers compared to the ring cull. Cattle TB incidents in this study were therefore insufficiently clustered around TB-infected badger setts to design an efficient spatially targeted cull; and this analysis provided no evidence to support a move towards spatially targeted badger culling policies for bovine TB control.