Factors associated with BoDV seropositivity in SP breeding horses:
Variables considered potentially associated with exposure to BoDV in the group of SP breeding horses sampled from 2011 to 2013 were included in a statistical model. The results of the final model and the effect of statistically significant variables (p˂0.05) are presented in Table 2.
There was no clustering at the stud farm level, indicating that the between-stud variability was insufficient to warrant necessary incorporation of a stud farm-level random effect. The Hosmer–Lemeshow goodness-of-fit test indicated that the multivariable logistic regression model fitted the data adequately (p=0.85). The area under the ROC curve was 0.7, indicating that the model had adequate overall predictive power. The main risk factor associated with seropositivity to BoDV was the presence of a cold semi-arid climate (BSK climate from the Köppen-Geiger Climate Classification) (Rubel and Kottek, 2010) in the area where the horse resided (OR = 3.04, 95% CI 1.48 to 6.26). Factors associated with decreased odds of being BoDV seropositive included the age and the time of the year when the horses were sampled. Adult horses, 7-14 years old, were 3.2 times less likely to be BoDV seropositive than young horses, 1-6 years old (OR= 0.31, 95% CI 0.16 to 0.63). Horses sampled from February to June were 2.7 times less likely to be BoDV seropositive than horses sampled from July to January (OR= 0.37, 95% CI 0.19 to 0.73).