Spatio-temporal modelling of localised health inequalities in Glasgow

Ismail, Riham Hamza (2024) Spatio-temporal modelling of localised health inequalities in Glasgow. PhD thesis, University of Glasgow.

Full text available as:
[thumbnail of 2024IsmailPhD.pdf] PDF
Download (8MB)


The main aim of this thesis is to develop a statistical clustering methodology for disease mapping. Disease mapping studies aim to understand a disease’s spatial pattern and identify areas with low or high disease risk. These studies play an essential role in epidemiology and public health by providing information on how disease exposures differ geographically and assisting in allocating resources for prevention or intervention strategies. Commonly, such studies are based on areal data, which partitions the study region into a set of non overlapping sub-regions. The standard clustering techniques for grouping data ignore the spatial dependencies between nearby areas in areal data. Therefore, the first model proposed in this thesis incorporates the spatial information within a Poisson finite mixture model for clustering areal data. The disease data are usually available over multiple timepoints, providing a valuable opportunity to carry out examinations of temporal trends and patterns. Thus, the two other methods proposed in this thesis are both forms of spatio-temporal generalised additive mixed model designed to capture trends and variations over both temporal and spatial dimensions. The first of these approaches estimates the disease risk over time and then identifies the high and low-risk clusters of spatio-temporal disease risk data. The final model in this thesis considers the potential clustering structure in spatial data over time and thereafter estimates the disease risk. These models are each used to assess the spatial and temporal trends of COVID-19 cases in the Greater Glasgow and Clyde Health Board areas. A key finding was that areas in different clusters often exhibited similar temporal trends but somewhat different means. The study also clearly identified several waves of COVID-19 cases during the study period, most notably an increase in COVID-19 cases in September 2021, potentially influenced by the UK’s rules in managing the COVID-19 epidemic.

Item Type: Thesis (PhD)
Qualification Level: Doctoral
Subjects: H Social Sciences > HA Statistics
Colleges/Schools: College of Science and Engineering > School of Mathematics and Statistics
Supervisor's Name: Anderson, Dr. Craig and Dean, Dr. Nema
Date of Award: 2024
Depositing User: Theses Team
Unique ID: glathesis:2024-84289
Copyright: Copyright of this thesis is held by the author.
Date Deposited: 02 May 2024 10:03
Last Modified: 02 May 2024 10:03
Thesis DOI: 10.5525/gla.thesis.84289

Actions (login required)

View Item View Item


Downloads per month over past year