Summary
Results of testing of 19,731 samples from a serologic survey of cattle with bluetongue virus (btv) infections in Australia were analyzed for association between age, species, or sex and test result. Bivariate analysis indicated that all 3 host factors were associated with test result. After adjusting for confounding caused by the location of each animal in the study (high, moderate, and low btv prevalence regions), cattle ≥ 4 years old had an odds ratio of 4.33 (95% confidence interval, 3.99, 4.71) for a positive test result, compared with that for cattle < 2 years old. Cattle 2 to 4 years had an odds ratio of 2.28 (2.14, 2.54), compared with cattle < 2 years old. Bos taurus cattle had an odds ratio of 1.76 (1.63, 2.05) of a positive test result, compared with crossbred cattle, and B indicus cattle had an odds ratio of 1.20 (1.09, 1.33), compared with crossbred cattle. Sexually intact (+) male cattle were found to have an odds ratio of 3.13 (2.66, 3.49) for a positive test result, compared with castrated male (−) cattle, and female cattle were found to have an odds ratio of 1.38 (1.29, 1.48), compared with male (−) cattle.
Multivariate analysis of btv testing results was performed, using stepwise logistic regression. The most parsimonious model selected included age, species, and sex factors, and first-order interaction terms between these factors. This model was only able to be fit to data from cattle restricted to the high (> 25%) btv prevalence region. Odds ratios were found to increase with age for male (−) cattle of all species. Odds ratios were found to be greatest at 2 to 4 years of age for female cattle of all species and for B taurus and crossbred male (+) cattle.
