Geographical Analysis for Detecting High-Risk Areas for Bovine/Human Rabies Transmitted by the Common Hematophagous Bat in the Amazon Region, Brazil

BACKGROUND The common hematophagous bat, Desmodus rotundus, is one of the main wild reservoirs of rabies virus in several regions in Latin America. New production practices and changed land use have provided environmental features that have been very favorable for D. rotundus bat populations, making this species the main transmitter of rabies in the cycle that involves humans and herbivores. In the Amazon region, these features include a mosaic of environmental, social, and economic components, which together creates areas with different levels of risk for human and bovine infections, as presented in this work in the eastern Brazilian Amazon. METHODOLOGY We geo-referenced a total of 175 cases of rabies, of which 88% occurred in bovines and 12% in humans, respectively, and related these cases to a number of different geographical and biological variables. The spatial distribution was analyzed using the Kernel function, while the association with independent variables was assessed using a multi-criterion Analytical Hierarchy Process (AHP) technique. FINDINGS The spatiotemporal analysis of the occurrence of rabies in bovines and humans found reduction in the number of cases in the eastern state of Pará, where no more cases were recorded in humans, whereas high infection rates were recorded in bovines in the northeastern part of the state, and low rates in the southeast. The areas of highest risk for bovine rabies are found in the proximity of rivers and highways. In the case of human rabies, the highest concentration of high-risk areas was found where the highway network coincides with high densities of rural and indigenous populations. CONCLUSION The high-risk areas for human and bovine rabies are patchily distributed, and related to extensive deforested areas, large herds of cattle, and the presence of highways. These findings provide an important database for the generation of epidemiological models that could support the development of effective prevention measures and controls.