Classification: Optimum cut-off screen value for foot-and-mouth disease (FMD) immunity evaluation

Abstract

Circulating antibody levels are used in a dichotomous classification scheme for determining the foot-and-mouth disease (FMD) immune status of cattle. Such a classification scheme may result in 4 categories of animals based on the true state of nature of the animals: the truly protected; the truly unprotected; the falsely protected; the falsely unprotected. The 2 last categories are misclassifications, the frequency of which depend on the specific cut-off value selected for the screening procedure. A methodology for selecting, from all possible values, the optimal antibody-titer cut-off value is applied in such a way that the number of misclassifications is minimized. This methodology is extended for situations where false protection is considered as being more damaging than false unprotection. Bayes' theorem is used for the evaluation of the performance of the neutralization test, in screening the immune status of cattle. The sensitivity and specificity of the classification scheme, and its ability to predict protection and unprotection are used to evaluate the classification scheme. Methodology is provided for developing estimates of the true FMD protection rate and its variance in cattle populations.