Abstract
Background
Rabies is an acute and progressive viral zoonotic disease of the nervous system, which widely affects domestic animals in Latin America. Vampire bat-borne rabies virus (RABV) has significant negative impacts on the livestock industry via animal mortality. Nevertheless, the landscape level factors that facilitate or limit RABV transmission from vampire bats to livestock remain elusive.
Methods
To determine how abiotic and biotic factors modulate RABV spillover from vampire bats to livestock, we assessed the role of different landscape variables on the occurrence of RABV spillover from Desmodus rotundus to livestock in Colombia. Using ecological niche modeling as the theoretical and analytical framework, we analyzed ecological and epidemiological RABV data to reconstruct spillover transmission events.
Results
Anthropogenic variables including livestock and human density were consistently selected as predictors of RABV spillover from vampire bats to livestock. Cattle density had the highest average relative contribution to final ecological niche models (64.7%). We also found improvement of RABV spillover risk estimates when sampling bias in the form of cattle density was used in the modeling process. High risk for RABV spillover (0.75-0.98) was consistently predicted in the Caribbean region of Colombia. Nevertheless, more widespread moderate RABV spillover risk was predicted more broadly across the country when sampling bias was accounted for.
Conclusion
Our modelling effort revealed that variable selection and use of bias surface have tractable impacts on final projections of spillover risk. Our results also indicate that human activity drives RABV spillover risk to a greater extent than ecological or climatological factors. Results from this study provide important information about landscape conditions linked to RABV transmission risk, where livestock vaccination should be prioritized.