Periodic shifts in viral load increase risk of Hendra virus spillover from Pteropus bats.

Prediction and management of zoonotic spillover requires an understanding of infection dynamics within reservoir host populations. Spillover risk is commonly inferred from infection prevalence based on detection of viral genomic material, yet detection alone does not indicate the presence of infectious virus or a sufficient dose for transmission. We undertook a comprehensive investigation of Hendra virus shedding in its primary reservoir, Pteropus bats, analyzing quantitative PCR with reverse transcription (RT-qPCR) data from 6151 pooled urine samples collected across five sites over 3 years. We assessed longitudinal associations between viral prevalence (proportion of positive pooled urine samples), viral load proxies, and equine spillover, using generalized additive models and a permutation analysis. Peak prevalence periods associated with spillover events (N = 5) had a higher proportion of samples with high viral loads than periods without spillover. Prolonged periods of low viral load and low prevalence likely reflect noninfectious RNA or doses insufficient for cross-species transmission. Incorporating viral load metrics alongside prevalence can improve prediction of spillover risk.