Search

A major frontier in disease ecology is understanding the transmission dynamics of generalist pathogens, where multiple host species are involved in ongoing circulation of a single pathogen. These dynamics violate assumptions of the simplest epidemic models and lead to challenges in characterizing the determinants of transmission because of the interwoven contributions of different host species. Since many pathogens (including all zoonoses, by definition) can infect multiple hosts, learning about complex multi-host systems from partial evidence is a pervasive problem in disease ecology. Understanding transmission dynamics of a generalist pathogen in a community of wildlife host species faces the additional challenges of analyzing sparse wildlife data. Biological and logistical sampling constraints in wildlife systems can create temporally and spatially coarse data, which can be difficult to analyze by conventional statistical methods. Thus, analyzing and integrating these data streams requires the development of novel methods. The coastal California ecosystem provides an opportunity to explore the transmission dynamics of a generalist pathogen in a multi-host community in the presence of complex wildlife data challenges. My primary focus is on the reintroduced Channel Island fox (Urocyon littoralis) population on Santa Rosa Island off the coast of Southern California. After a population crash in the late 1990s and subsequent captive breeding program in the early 2000s, the Santa Rosa Island fox population was reintroduced to the wild gradually from 2003 to 2009. Shortly thereafter, an outbreak of Leptospira, a pathogenic bacterium known to infect most mammals, was detected in the reintroduced population. Within the broader coastal California ecosystem, Leptospira was known to circulate endemically in the California sea lion (Zalophus californianus) population since at least the 1980s. The context of the Leptospira outbreak on Santa Rosa Island prompted questions regarding its origin on the island and determinants of transmission in this wildlife community. In this dissertation, I present three studies analyzing the ample cross-sectional and longitudinal ecological data on the fox population, which has been monitored by the National Park Service for two decades, and pathogen genetic data isolated from four potential host species to understand transmission dynamics of a multi-host, generalist pathogen in this wildlife community. Ecological data offers only one perspective of Leptospira transmission in the coastal California ecosystem. In chapter one, I use bacterial genomic data from four host species to provide qualitatively different evidence addressing two particular questions: (i) What was the source of the pathogen introduction in the reintroduced island fox population? (ii) Did pathogen fadeout occur in California sea lions during a period when other evidence showed an unprecedented pause in Leptospira transmission? To address both questions, I construct Bayesian time-calibrated phylogenies and use the topology to infer epidemiological linkages between hosts. For the former question, I show that bacterial isolates from Santa Rosa Island form a distinct cluster in the tree, ruling out sea lions as the direct source of the pathogen introduction to the reintroduced island fox population. For the latter question, I show that isolates obtained after the suspected pathogen fadeout period are not descended from those isolated before, suggesting that pathogen fadeout did occur in the California sea lion population, and that the post-fadeout circulation must have arisen via pathogen introduction from an external reservoir. These results are consistent with and important corroboration for other lines of evidence. This work demonstrates the utility of whole genome sequencing as a component of a multi-disciplinary, multi-data source study to untangle the complexity of wildlife disease systems.In chapter two, I focus on the Leptospira outbreak in reintroduced island foxes to identify and quantify risk factors for infection while addressing two challenges that frequently arise in wildlife data: interval censoring and time-varying covariates. I first addressed the challenge of extensive interval censoring by bootstrapping a set of datasets with an imputed time of infection for individual foxes using a quantitative serology model. The resulting synthetic datasets lacked interval censoring, which allowed the use of a Cox proportional hazards model and enabled me to account for time-varying covariates through a counting process formulation. I find that higher 24-month cumulative precipitation increases the risk of infection and suggests that long-term fluctuations in the water table significantly influence transmission of Leptospira. I also show that risk of infection decreases with increasing fox abundance, contrary to conventional expectation, which may be due to the stabilization of fox social structure on the island as the population increased post-reintroduction. This chapter highlights the need for intensive and sustained data collection and new methodologies for analysis in wildlife systems and lays a foundation for future studies to investigate transmission risk to inform prevention and control strategies in wildlife populations.Collecting movement data is a resource-limited endeavor, and many studies monitor the locations of individuals at a coarser scale for purposes other than movement analyses. In chapter three, through the integration of a novel spatial data type, obtained through field notes and expert interpretation, and innovative methodology, I propose a method to construct wildlife movement trajectories from location data of varying resolution, again using the Santa Rosa Island fox population as a case study. I integrate unconventional expert-drawn polygons with traditional GPS data by resampling locations on the date of observation for every individual fox. I then fit smoothing splines through the coordinate directions to interpolate each fox’s location for every day in its observation window. By pairing the movement estimates with time of infection estimates estimated via a quantitative serology model, I reconstruct the spatiotemporal origin of the Leptospira outbreak and estimate transmission to most likely have begun in mid-to-late 2005 along the northern shore of the island. Our approach lays the groundwork to reap the full benefit of rich, long-term monitoring datasets, which could provide vital insights into a species’ movement ecology and better inform conservation and management.The studies presented in these three chapters apply a breadth of analytic methods to tackle challenges of wildlife data and illustrate how methodological developments and integration of different data streams can be utilized to describe the transmission dynamics of a multi-host generalist pathogen. This work lays the foundation to capitalize on ample lower quality data collected from long-term monitoring programs and to address fundamental challenges in studying wildlife disease ecology.

For many emerging or re-emerging pathogens, cases in humans arise from a mixture of introductions (via zoonotic spillover from animal reservoirs or geographic spillover from endemic regions) and secondary human-to-human transmission. Interventions aiming to reduce incidence of these infections can be focused on preventing spillover or reducing human-to-human transmission, or sometimes both at once, and typically are governed by resource constraints that require policymakers to make choices. Despite increasing emphasis on using mathematical models to inform disease control policies, little attention has been paid to guiding rational disease control at the animal-human interface. In this thesis, I introduce a modeling framework to analyze the impacts of different disease control policies, focusing on pathogens exhibiting subcritical transmission among humans. We quantify the relative effectiveness of measures to reduce spillover (e.g. reducing contact with animal hosts), human-to-human transmission (e.g. case isolation), or both at once (e.g. vaccination), across a range of epidemiological contexts. This analysis provides guidelines for choosing which mode of control to prioritize in different epidemiological scenarios and considering different levels of resource and relative costs. We contextualize our analysis with current zoonotic pathogens and other subcritical pathogens, such as post-elimination measles, and control policies that have been applied. This work provides a model-based, theoretical foundation to understand and guide policy for subcritical zoonoses, integrating across disciplinary and species boundaries in a manner consistent with One Health principles.