Estimating the effects of air pollution on human health in Greater Glasgow in space and time

Halbert, Greg Jonathan (2013) Estimating the effects of air pollution on human health in Greater Glasgow in space and time. MSc(R) thesis, University of Glasgow.

Abstract

It is well documented that air pollution has an adverse effect on human health. With the increased risk of global warming, there has been an international effort to decrease emissions and pollution concentrations throughout the globe over the past sixty years, and these values are monitored by many laws and acts of governments. This thesis is a long term study of the effects of air pollution on the health of a Scottish population, specifically the incidence of respiratory disease cases in the Greater Glasgow and Clyde National Health Service (NHS) health board. As this is a long term study, the main points of interest are what effects pollution concentrations have on the hospitalisation counts of patients with respiratory disease on a yearly basis, and what other covariates, if any, have an effect on disease incidences. Furthermore, as this is a study in space and time we need to take into account any spatial and/or temporal correlation that may exist within the data. The study region is split up into 271 small areas based on population size and we evaluate what effect two specific pollutants, Nitrogen Dioxide (NO2) and Particulate Matter (PM10) have on respiratory disease across these areas.

The rest of this thesis is structured as follows. Chapter One will present an introduction to the data and a literature review of the previous studies in this field. Chapter Two gives an outline of all of the statistical methods used throughout this study, including Poisson generalised linear models, diagnostic tests for overdispersion and spatial correlation, Bayesian models
and conditional autoregressive models. Chapter three gives a description of all the data in the study and how it was obtained, as well as some preiliminary tables and plots. Chapter Four gives the results of all the purely spatial models discussed in Chapter Two. Chapter Five gives the results of the spatial-temporal health models where the entire space-time data set is modelled. Finally, Chapter Six presents an overall conclusion to the thesis, a
discussion of any problems that occurred during this study, as well as what future work could be produced based on this study.

• Item Type: Thesis (MSc(R))
• Qualification Level: Masters
• Subjects: H Social Sciences > HA Statistics
• Colleges/Schools: College of Science and Engineering > School of Mathematics and Statistics
• Supervisor's Name: Lee, Dr. Duncan
• Date of Award: 2013
• Depositing User: Mr Greg J Halbert
• Unique ID: glathesis:2013-4336
• Copyright: Copyright of this thesis is held by the author.
• Date Deposited: 30 May 2013 13:36
• Last Modified: 30 May 2013 13:36
• URI: https://theses.gla.ac.uk/id/eprint/4336