Geostatistical methods for modelling zoonotic spillover : characterising Leptospira transmission pathways through rattiness

Eyre, Max and Giorgi, Emanuele and Diggle, Peter and Costa, Frederico (2022) Geostatistical methods for modelling zoonotic spillover : characterising Leptospira transmission pathways through rattiness. PhD thesis, Lancaster University.

Text (2022eyrephd)
2022eyrephd.pdf - Published Version
Restricted to Repository staff only until 9 June 2025.
Available under License Creative Commons Attribution.
Download (45MB)

Abstract

Increasingly, zoonotic spillover research follows a One Health approach, which ac- knowledges the interdependence of the environment, animal health and human health. Commonly this means that geolocated data are collected for human participants, animal disease reservoirs and environmental factors within a single study. In this thesis we explore how geostatistical approaches can be used to study zoonotic spillover through the analysis of fine-scale spatial information collected at the animal-human-environment interface. We use the developed methods to characterise the mechanisms that drive Leptospira spillover transmission in urban settings. The thesis comprises three papers that describe the progression of our work. First, we develop the ‘rattiness framework’, a multivariate geostatistical framework for the joint analysis of geolocated animal abundance data that has been obtained using multiple imperfect indices of abundance. We apply this to rat abundance data to identify rattiness hotspots within a low-income urban community in Brazil where the Norway rat is the primary disease reservoir for Leptospira. We then introduce the ‘rattiness-infection’ framework for joint spatial modelling of disease reservoir abundance and human infection data. We use it to examine the role of the Norway rat in spillover transmission of Leptospira to humans in an eco- epidemiological study, comprising a community-based cohort and concurrent rat ecology study. We find evidence that transmission mechanisms vary with elevation. Finally, we examine the complex spatiotemporal interactions between the rat reservoir, environment and rainfall that determine the risk of spillover infection to humans. We do this by applying a spatiotemporal extension of the ‘rattiness- infection’ framework to data collected in a seven-year prospective community- based cohort study and a three-year rat ecology study. Our findings provide evidence for three Leptospira transmission pathways, suggesting different mechanisms for all-year-round infections and large outbreaks in urban settings. Together, the studies presented in this thesis demonstrate the usefulness of geo-statistical methods in delineating complex animal-environment-human interactions during zoonotic spillover and identifying opportunities for public health intervention. Specifically, they provide insights into the mechanisms that drive Leptospira spillover transmission.