| dc.description.abstract | Background: While cholera has been eradicated in many parts of the world thanks to improvements in water and sanitation, the water-borne disease remains globally prevalent and causes around 100,000 deaths each year, most of whom are young children. In the Eastern Indian city of Kolkata, cholera is endemic. Due to its close association with the local aquatic environment, cholera is widely considered to be closely linked to changes in climate, however the nature of this relationship varies widely between regions. In this thesis, I identify the particular relationship between cholera and climate within the context of Kolkata. Based on these findings, I then develop a mathematical model linking cholera and climate factors and apply it to both infer the likely effects of climate change on cholera and analyse the effectiveness of potential intervention types (e.g. provision of water and sanitation, flood mitigation and vaccination programmes).
Methods: Generalised Additive Modelling (GAM) was utilised to interpret the non-linear relationship between cholera cases and different climatic factors, including temperature, rainfall and sea surface temperature over a period of 21 years. A GIS-based approach was used to map the per capita burden of cholera by district, allowing for an exploration of the spatial distribution of the disease and its relationship with key demographic and geographical variables. The influence of larger climatic phenomena such as El Niño Southern Oscillation (ENSO) and the Indian Ocean Dipole (IOD) on cholera was explored using wavelet coherence analysis over historical and recent time intervals. Four mathematical models were developed which were then calibrated using Markov Chain Monte Carlo (MCMC) parameter estimation, the optimal model was selected by Watanabe-Akaike Information Criterion (WAIC). Finally, climate projections for 2080-2099 were bias-corrected and used as input to the selected model to simulate the effects of climate change and potential interventions (e.g. provision of safe water and sanitation, flood mitigation, and vaccination programmes).
Results: Results reveal a dual-peak cholera phenomenon in Kolkata, mediated by temperature and rainfall. Significant spatial heterogeneity in cholera distribution is also evident, with densely populated and lower elevation regions having higher risk. Analysis of ENSO and IOD demonstrates a general de-coupling between cholera and these variables in recent years. Application of the model suggest a median 81% - 150% increase in annual cholera infections depending on climate change assumptions. The most effective intervention type simulated was limiting contact between individuals and contaminated water bodies.
Implications: In clarifying the climate-cholera relationship in Kolkata, this thesis provides a framework for adapting to other regions and contributes to the broader goal of developing a general cholera-climate model. | en_GB |
