Are herders protected by their herds? An experimental analysis of zooprophylaxis against the malaria vector Anopheles arabiensis

Background: The number of Anopheles arabiensis (Diptera: Culicidae) and Anopheles pharoensis caught by human and cattle baits was investigated experimentally in the Arba Minch district of southern Ethiopia to determine if attraction to humans, indoors or outdoors, was affected by the presence or absence of cattle. Methods: Field studies were made of the effect of a surrounding ring (10 m radius) of 20 cattle on the numbers of mosquitoes collected by human-baited sampling methods (i) inside or (ii) outside a hut. Results: The numbers of An. arabiensis caught outdoors by a human landing catch (HLC) with or without a ring of cattle were not significantly different (2 x 2 Latin square comparisons: means = 24.8 and 37.2 mosquitoes/night, respectively; n = 12, P > 0.22, Tukey HSD), whereas, the numbers of An. pharoensis caught were significantly reduced (44%) by a ring of cattle (4.9 vs. 8.7; n = 12, P < 0.05). The catch of An. arabiensis in human-baited traps (HBT) was 25 times greater than in cattle-baited traps (CBT) (34.0 vs. 1.3, n = 24; P < 0.001) whereas, for An. pharoensis there was no significant difference. Furthermore, HBT and CBT catches were unaffected by a ring of cattle (4 x 4 Latin square comparison) for either An. arabiensis (n = 48; P > 0.999) or An. pharoensis (n = 48, P > 0.870). The HLC catches indoors vs. outdoors were not significantly different for either An. arabiensis or An. pharoensis (n = 12, P > 0.969), but for An. arabiensis only, the indoor catch was reduced significantly by 49% when the hut was surrounded by cattle (Tukey HSD, n = 12, P > 0.01). Conclusions: Outdoors, a preponderance of cattle (20: 1, cattle: humans) does not provide any material zooprophylactic effect against biting by An. arabiensis. For a human indoors, the presence of cattle outdoors nearly halved the catch. Unfortunately, this level of reduction would not have an appreciable impact on malaria incidence in an area with typically > 1 infective bite/person/night. For An. pharoensis, cattle significantly reduced the human catch indoors and outdoors, but still only by about half. These results suggest that even for traditional pastoralist communities of East Africa, the presence of large numbers of cattle does not confer effective zooprophylaxis against malaria transmitted by An. arabiensis or An. pharoensis.