Animal-level risk factors for Trypanosoma evansi infection in camels in eastern and central parts of Kenya.

Point prevalences and animal-level risk factors for Trypanosoma evansi infection were investigated in a cross-sectional study that involved 2227 camels from eastern and central parts of Kenya. The screening tests used were haematocrit centrifugation technique (HCT), mouse inoculation and latex agglutination (Suratex). All camels were screened with HCT, while 396 and 961 of them were, in addition, screened with mouse inoculation and Suratex tests, respectively. Parasitological and Suratex test results were used in parallel to determine the number of camels exposed to T. evansi infections. Statistical analyses were conducted using Statistical Analysis Systems. Parasitological and Suratex test results in parallel were dependent variables in multivariable logistic regression models that determined risk factors for T. evansi infection. Herd-level clustering was corrected with general estimation equations. The prevalences were 2.3% and 19.6%, using parasitological and Suratex tests, respectively, and 21.7% when both tests were used in parallel. There was a positive association between the screening tests (McNemar's test = 104.8, P = 0.001) although the strength of association was low (Kappa = 0.2; 95% CI: 0.1-0.3). Before accounting for herd-level clustering, dry season (OR = 1.5; 95% CI: 1.0, 2.1) and nomadic pastoralism (OR = 1.8; 95% CI: 1.1, 3.2) were associated with increased odds of a camel being exposed to T. evansi infection compared to wet season and ranching, respectively. Following this correction, only nomadic pastoralism was significantly associated (OR = 3.1; 95% CI = 1.0, 14.4) with T. evansi infection compared to ranching. It is concluded that camels managed under nomadic pastoralism had higher risk of being exposed to T. evansi infections than camels from ranching systems of management.