Controlling multi-host pathogens within their natural wildlife reservoirs has inherent advantages over targeting humans or domestic animals that contribute little to long-term disease dynamics. Interventions such as vaccination and culling have the potential to reduce or eliminate pathogens, but it can be difficult to obtain sufficient population-level coverage to impact transmission, particularly for reclusive wildlife. Interventions that autonomously spread from treated individuals to susceptible hosts are a promising theoretical option, but have not been explored in real-world systems. We used field experiments with an innocuous biomarker to approximate the spread of oral vaccines or poisons (“vampiricide”) in wild populations of common vampire bats (Desmodus rotundus), the primary reservoir of rabies in Latin America. Mathematical models incorporated field-derived transfer rates to examine whether vaccination or culling was more effective at controlling rabies within single bat colonies. Finally, we built and validated a metapopulation model of rabies based on data from rabies invasions in southern Peru and utilized this model to identify optimal intervention strategies for rabies control across a broad geographical landscape.
Results/Conclusions
Our biomarker experiment tracking mock vaccine/vampiricide spread revealed that for each treated bat, 1.85 other bats ingested the biomarker through allogrooming. The within-colony stochastic model of rabies transmission indicated that outbreak sizes could be reduced by 50% through vaccinating as few as 20% of bats within a single location. We considered three epidemiological scenarios that represent different candidate management strategies: (1) a 'preventative' approach, where vaccine/vampiricide was applied to prevent viral invasion into historically rabies--free bat populations; (2) a 'proactive' approach, which represented an intervention in a rabies endemic area, but in a colony that was not currently infected; and (3) a 'reactive' approach where intervention followed an outbreak. In all three scenarios, vaccination was more effective at reducing the size, probability, and duration of rabies outbreaks than culling under realistic application levels. Additionally, results from our metapopulation model strongly supported vaccination rather than culling for long-term rabies prevention, control, and eventual eradication. This is the first evidence that strategic vaccination of bats may be significantly more effective than the current policy of culling to reduce the burden of rabies in Latin America.