The classification of regions into high or low risk areas with respect to the trypanosomiasis-tsetse problem is essential to planning disease control activities. Discriminant analysis was performed on a multivariate data set obtained from Southwest Ethiopia, to obtain a mathematical function that can be used for classification into high or low risks using 24 explanatory variables. The prevalence of trypanosomiasis, the dependent variable, was evaluated at either 5% or 10%. For the classification rule based on a critical prevalence demarcation of 10%, the presence of forested areas (X7), human population density per cultivated hectare (X24), and fusca group tsetse (X29) were important; therefore, control of the disease should emphasize these factors. Various misclassification probabilities and an interpretation of the above functions in light of the epidemiology of the disease were presented. The overall significance of discriminant analysis as an epidemiological tool was discussed.
ASJC Scopus subject areas
- Animal Science and Zoology