Many (re)emerging infectious diseases in humans arise from pathogen spillover from wildlife or livestock, and accurately predicting pathogen spillover is an important public health goal. In the Americas, yellow fever in humans primarily occurs following spillover from non-human primates via mosquitoes. Predicting yellow fever spillover can improve public health responses through vector control and mass vaccination. Here, we develop and test a mechanistic model of pathogen spillover to predict human risk for yellow fever in Brazil. This environmental risk model, based on the ecology of mosquito vectors and non-human primate hosts, distinguished municipality-months with yellow fever spillover from 2001 to 2016 with high accuracy (AUC = 0.72). Incorporating hypothesized cyclical dynamics of infected primates improved accuracy (AUC = 0.79). Using boosted regression trees to identify gaps in the mechanistic model, we found that important predictors include current and one-month lagged environmental risk, vaccine coverage, population density, temperature and precipitation. More broadly, we show that for a widespread human viral pathogen, the ecological interactions between environment, vectors, reservoir hosts and humans can predict spillover with surprising accuracy, suggesting the potential to improve preventive action to reduce yellow fever spillover and avert onward epidemics in humans. This article is part of the theme issue ‘Dynamic and integrative approaches to understanding pathogen spillover’.

Many (re)emerging infectious diseases in humans arise from pathogen spillover from wildlife or livestock, and accurately predicting pathogen spillover is an important public health goal. In the Americas, yellow fever in humans primarily occurs following spillover from non-human primates via mosquitoes. Predicting yellow fever spillover can improve public health responses through vector control and mass vaccination. Here, we develop and test a mechanistic model of pathogen spillover to predict human risk for yellow fever in Brazil. Our environmental risk model, based on the ecology of mosquito vectors and non-human primate hosts, distinguished municipality-months with yellow fever spillover from 2001 to 2016 with high accuracy (AUC = 0.71). Incorporating hypothesized cyclical dynamics of infected primates improved accuracy (AUC = 0.79). Using boosted regression trees to identify gaps in the mechanistic model, we found that important predictors include current and lagged (mechanistic) environmental risk, vaccine coverage, population density, temperature, and precipitation. More broadly, we show that for a widespread human viral pathogen, the ecological interactions between environment, vectors, reservoir hosts, and humans can predict spillover with surprising accuracy, suggesting the potential to improve preventative action to reduce yellow fever spillover and prevent onward epidemics in humans. Many important (re)emerging infectious diseases in humans-including Ebola, sudden acute respiratory syndrome (SARS), influenza, Plasmodium knowlesi and other primate malarias, yellow fever, and leptospirosis-arise from spillover of pathogens from wildlife or livestock into human populations (1,2). While spillover is an important mechanism of human disease emergence, the drivers and dynamics of spillover are poorly understood and difficult to predict (3). Pathogen spillover requires favorable conditions to align in the reservoir (non-human animal), human, and pathogen populations and in the environment (3-5). Because these conditions interact, nonlinear relationships among the environment, host populations, and spillover probability are likely to emerge. Moreover, spillover is a probabilistic process that does not always occur, even when suitable conditions align. Despite these challenges, it is critical to predict pathogen spillover to enhance public health preparedness. Predicting spillover also provides an opportunity to test ecological approaches to solving globally important human health problems.Most previous attempts to predict pathogen spillover have used statistical models (6-8). These models may be locally accurate for within-sample prediction, but may struggle to detect multidimensional, nonlinear, and stochastic relationships among host populations, pathogens, the environment, and spillover. In contrast, mechanistic models can test our understanding of transmission ecology, reproduce the complex, nonlinear interactions emerging in disease systems, and potentially improve our ability to predict spillover. In parti...

No abstract

No abstract