Epidemiological risk factors associated with a diagnosis of clinical cyathostomiasis in the horse.

Multiple logistic regression was used to assess epidemiological risk factors associated with the diagnosis of cyathostomiasis in 87 cases of chronic diarrhoea in the horse. Age, season and the period since last receiving anthelmintics were identified as important risk factors using chi-square and two-sample t test analyses, whereas access to grazing, shared grazing with other horses and recurrence of signs were only weakly associated with a diagnosis of cyathostomiasis. Multivariate analysis of the parameters using logistic regression was performed. The final model included age, season and time since last deworming. At a predicted probability of cyathostomiasis of 0.5, the model had a specificity of 86.0%, sensitivity of 66.7%, overall correct classification of 79.3%, a positive predictive value of 71.4% and a negative predictive value of 83.1%. The results of this study indicated that the specified variables and factors may be useful in the differentiation of clinical cyathostomiasis from other causes of chronic diarrhoea, based on case history alone.