COMPARISON OF MULTIVARIATE LOGISTIC REGRESSION AND CLASSIFICATION TREE TO ASSESS FACTORS INFLUENCING PREVALENCE OF ABORTION IN NIGERIAN CATTLE BREEDS

The aim of the study was to investigate the application of binary logistic regression and classification tree to assess the potential factors associated with the prevalence of abortion in two indigenous cattle breeds in Nigeria. Three thousand, six hundred and four calving records of cows from a total of 115 cattle herders within Nasarawa, Benue and Plateau, north central Nigeria from the year 2010-2013 were utilized in the study. The cattle which were of Bunaji and Sokoto Gudali genetic groups originated from different flocks and were reared under the traditional extensive system. The risk factors investigated were dam's breed-group, season and parity. The incidence of abortion was higher in Bunaji (16.9%) compared to Sokoto Gudali (13.8%) cattle. Abortion was also more in the rainy (19.6%) than the dry (13.0%) and harmattan (lower temperatures with winds)(13.5%) seasons respectively. Animals which have calved once and twice aborted (23.9% and 21.1%, respectively) more than their counterparts which have recorded up to three (9.8%) and four births (5.8%). However, the multivariate logistic model (chi(2) = 8.56, P = 0.197) and classification tree (risk value = 0.123 +/- 0.004) revealed that parity number and breed were the most important parameters affecting the prevalence of abortion in cattle. The present information may be exploited in management practices to attenuate the incidence of abortion, thereby increasing the productivity of the animals in Nigeria, sub-saharan Africa.