Close
Ben Wheeler, University of Bristol
(Project no. 10338)
This study forms the central piece of research for my PhD, which started in October 1999. It is based on my own research proposal, which was drawn up in conjunction with the departments of Geography and Social Medicine at Bristol University, and is funded by the Medical Research Council.
An overwhelming body of evidence illustrates the presence and growth of socio-economic and geographic health inequalities in this country. The processes by which these inequalities arise are still not entirely understood. One potential mechanism is the unequal distribution of exposure to environmental health risks through social and physical space.
Issues of environmental inequality have received little attention in the UK, relative to a 20-year history of research in this area in the USA. Geographically referenced sets of environmental pollution data are now available at a national scale and offer opportunities for investigating relationships between environmental and social factors and health outcomes.
Effects of acute exposure to environmental pollutants have received much attention in the literature. However, effects of chronic exposure to 'every-day' levels of pollutants are little understood, and it is these that have the greatest potential for public health impact. Since this study is probably looking for a fairly small effect as a result of chronic exposure, the large sample size and longitudinal nature of the LS make it ideal for this purpose.
The aims of this project are to use the LS as a central dataset for an investigation of the potential health effects of unequal exposure to environmental risks. The underlying methodology will be epidemiological, and the study is set in the context of environmental risk management and public health.
Use of the LS data will allow testing of specific hypotheses under the more general hypothesis: "Socioeconomic and geographical variations in morbidity/mortality can be explained, to some extent, by variation in exposure to physical environmental hazards". Occupation and industry data will be used to control for potential confounding.
Current work with datasets from the Environment Agency and the National Air Quality Archive will generate small-area environmental statistics. The output from this work will be data for areas such as wards, local authorities or parliamentary constituencies. This data would be exported to the LS and attributed to individuals based on the geographic identifiers available. Any additional data required (such as ward-constituency look-up tables) would be supplied by myself. Statistical description and analysis will then be carried out, such as multivariable regression modelling of health outcomes with socio-economic and environmental explanatory variables. I am also interested in following up the possibility of applying multi-level modelling techniques to this study.