Mathematical modelling of the environmental and ecological drivers of zoonotic disease with an application to Lassa fever

Due to the essential role of zoonotic hosts, zoonotic spillover results from a complex system of environmentally-driven ecological and epidemiological processes. Despite being a global public health concern, zoonotic disease epidemiology is rarely viewed through the lens of disease ecology, meaning that the ecological factors driving zoonotic disease risk are typically not quantified. In this thesis I develop mathematical models to understand the zoonotic disease system from a process-based perspective informed by ecology, dependent on environmental variables, and tested using human health data. I focus these methods on a case study of Lassa fever which has historically been a neglected zoonosis but now may have improved opportunities for disease mitigation and surveillance. I present an overview of the topic in Chapter 1, outlining the challenges in zoonotic disease modelling and management. In Chapter 2, I find evidence of a severity bias in Lassa fever case data and estimate that infection incidence is likely on a much greater scale than previously thought. To elucidate environmental and ecological drivers of the Lassa virus system, in Chapter 3 I quantify the climatic dependence of reservoir host demographic processes. Along with strong seasonality, I estimate that year-on-year changes in precipitation can lead to substantial changes in the reservoir host population. In Chapter 4, I extend this population model to include pathogen transmission dynamics. Applying this model to states in Nigeria and linking reservoir host virus dynamics to observed human cases, I find that patterns of Lassa fever are significantly and positively correlated with predicted prevalence of infectious reservoir hosts. Finally, in Chapter 5 I summarise the findings and discuss future directions for the management and mitigation of zoonotic disease, concluding that ecological process-based modelling – facilitated by increased integration of knowledge, methods, and data – is essential for understanding zoonotic disease systems.