Bridging hosts: Domestic horse density and Hendra virus spillover risk in a changing landscape.

Abstract

BACKGROUND: Anthropogenic climatic and landscape change can drive behavioural shifts in wildlife and thus lead to increased risk of pathogen exposure for humans and domestic animals. While spillover research often focuses on the reservoir hosts or ongoing transmission in humans, livestock and companion animals can play important roles as bridging and amplifying hosts, facilitating the emergence of highly pathogenic diseases. OBJECTIVES: To investigate the distribution and density of domestic horses in the context of their role as bridge hosts for Hendra virus and build models to study zoonotic emergence. STUDY DESIGN: Cross sectional. METHODS: Government horse datasets (2011-2024) were analysed, and field surveys conducted in southeast Queensland and northeast New South Wales, Australia, to estimate domestic horse distributions and density. Zero-inflated negative binomial models were used to examine spatial correlations between horse population density, flying fox foraging areas and Hendra virus spillover events across different landcover types. Finally, random forest models were used to predict property-level horse densities based on 209 landscape and socioeconomic covariates. RESULTS: Horse populations were widespread across the study area, though field observations confirmed under-reporting in government datasets. Property size was the strongest predictor of horse density. A positive relationship in agricultural areas between Hendra virus spillover events and both locality-level horse density (p&#x2009;=&#x2009;0.001) and cumulative winter occupancy of flying fox roosts (p&#x2009;<&#x2009;0.001) was identified. These relationships were specific to agricultural landscapes, with negative associations in urban and forested areas. CONCLUSION: A previously undetected association between horse density and spillover was revealed, highlighting the importance of this integrated approach. Current limitations in horse population data present challenges for biosecurity and disease risk assessments in existing risk areas. Targeted surveillance and predictive modelling will be essential to mitigate future spillover risks and protect both animal and human health.