Risk-Based Targeting of Animals for Ancillary Testing during a Bovine Tuberculosis Breakdown Is Associated with a Reduced Time to Test Failure: Indirect Evidence of Mycobacterium bovis Exposure?

Bovine tuberculosis (bTB) continues to have significant economic and veterinary health impacts on cattle herds where the disease remains endemic. The continual tailoring of policies to address such maintenance requires an in-depth analysis of national data, underpinning new control strategies. In Ireland, when outbreaks occur, ancillary testing of herd mates deemed to be at the highest risk of exposure to reactors is undertaken using the interferon gamma (GIF) test. This highest risk cohort was hypothesised to be of a higher future risk despite this ancillary testing. We used a dataset from Ireland to model bovine test failure to the comparative tuberculin skin test using a survival analysis (observations: 39,248). Our primary exposure of interest was whether an animal that tested negative had a GIF test after the disclosure of infection within a herd during a bTB breakdown. There was evidence that animals with a negative GIF test during a breakdown had an increased risk of failing a test relative to other animals from the same herds without this exposure. The time to failure was 48.8% (95%CI: 38.3-57.5%) shorter for the exposed group relative to the unexposed group during a two-year follow-up period (2019-2022; time ratio: 0.51; 95%CI: 0.43-0.62; p < 0.001). The results from this study suggest that animals who were GIF-tested, having been deemed to have a higher risk of exposure, subsequently had shorter time-to-test failure periods. The absolute numbers of failure are small (only 2.5% of animals go on to fail during 2-year follow-up). Importantly, however, a high proportion of these high-risk herds included in the dataset failed at least one test at the follow-up (21/54 herds), impacting breakdown duration or recurrence. Such risk-informed targeting of animals could be utilised in future control policies, though further research is warranted.