Bovine TB: Research project summary
Project SE3004: Multivariate analysis of the risk factors affecting tuberculosis incidents in cattle herds.
Project duration: 5 years 3 months
A retrospective case control study was undertaken of farms in 1997-98 to provide clarification of key variables contributing to the risk of a herd becoming infected with bovine TB. The study utilised data sets provided by telephone questionnaire, TB49 epidemiological investigation forms, farm level badger density estimates, estimates of landscape structures and climate variables.
TB49 forms - Analysis indicated that a minimum quality of completing the TB49 questionnaire was obtained that was suitable for it’s original purpose, to assist in the management of breakdowns. However, the quality in terms of individual variables was highly varied and consequently data in the TB49 forms could not be used reliably in any risk factor investigation.
Badger density estimates - Predictions were successfully made for all farms for main sett presence, total sett number and activity score. However, farm level badger abundance estimates provided by statistical modelling validated against field survey could not be used in case-control data analysis. This was because although they successfully captured general trends, their ability to predict badger sett density ranking was poor.
Climate variables - Estimates of climatic variables provided crude estimates of spring time farm level climatic variables.
Landscape structures - Predictions were successfully extracted for four landscape variables (areas of broadleaf woodland & pasture, common length of broadleaf woodland to field boundaries and heterogeneity of broadleaf woodland).
A combined analysis of all elements - Much of the data was excluded from multivariate analysis because of the large uncertainty surrounding some of the estimates. Therefore this aspect could not be undertaken. This research identifies the difficulties in obtaining suitable high resolution, badger-abundance and ecological variable data for risk analysis studies.
Back to Research projects
Page last modified:
July 7, 2008