An epidemiological investigation into the reasons for high bovine tuberculosis incidence in cattle herds of the Burren, Ireland, prior to 2020

TL;DR

This study investigates why bovine tuberculosis rates were high in the Burren region of Ireland before 2020, finding that higher rainfall and herd size may be contributing factors.

Contribution

The study identifies rainfall and herd size as key factors associated with bovine tuberculosis incidence in the Burren using spatial and statistical analysis.

Findings

Bovine tuberculosis hotspots in the Burren correlated with higher rainfall levels.

Rainfall was the strongest predictor of bTB breakdowns in a logistic regression model.

Higher herd sizes in high-prevalence areas suggest increased contact and wildlife interaction may contribute to bTB spread.

Abstract

Herd-level bovine tuberculosis (bTB) incidence was examined in the Burren, an area in the west of Ireland where herd owners practice distinctive transhumance practices, with upland winter grazing. Prior to the initiation of our study in 2020, bTB incidence had for many years been unusually high in the Burren in comparison with the rest of the country, although the most recent figures have come down to being closer to the national average. Using data from the period prior to 2020, we mapped bTB infection in Burren herds alongside a range of indicators thought to have an association with it - herd size, herd density, herd type, cattle movement, and badger (Meles meles) population and control data, as well as rainfall and altitude. We also looked at how summary statistics for these variables differed when Burren herds with a history of bTB were compared to other Burren herds, as well as bTB positive and negative herds from outside the Burren. We found that for many indicators Burren herds would be expected to be low risk when compared to other herds in Ireland. An exception to this was for rainfall: hot spot areas for bTB in the Burren were found in areas of higher rainfall, on average herds in the Burren experienced more rainfall than those outside it, and bTB herds in the Burren experienced higher rainfall than non-bTB herds. Separately, for Burren herds only, a logistic regression model was developed to explain bTB breakdown occurrence using a matched case-control approach. Cases were herds which had experienced a new bTB breakdown between 2015 and 2019 (n = 260) and these were matched on herd type and herd size with the same number of herds not experiencing a breakdown during this period. This showed that, of a range of exogenous variables, rainfall was the most strongly associated with herd-level bTB incidence. These results suggest that high levels of exposure to inclement weather, and/or better environmental survival of Mycobacterium bovis in the environment, may contribute to high bTB rates in the Burren. However, as rainfall showed a highly aggregated distribution, this relationship may be due to an unmeasured factor correlated with it. Mapping and graphical output suggested that, although herd sizes in the Burren were on average lower than nationally, within the Burren they were higher in areas of higher prevalence, suggesting that mechanisms associated with herd size, such as increased contacts between and within herd, and with wildlife, may also play a role.

Figures