Modelling health related behaviours using geodemographics: applications in social marketing and preventative health.
Doctoral thesis, UCL (University College London).
The increased incidence of lifestyle related diseases, such as obesity and diabetes, across the western world is now an established fact, and presents many challenges to researchers trying to understand the determinants of poor health. Measurement of health needs and health outcomes is a fundamental component of evidence-based policy, strategy and delivery of health care services and interventions at scales from the local to the national. A central contention of this thesis is that health outcome indicators should be cognisant of factors such as personal behaviour, lifestyles, community influences, living and working conditions, accessibility to services and educational attainment which all impact upon the health of the individual and the wider community. It is therefore sensible to explore these differences by understanding both the social space comprising of different population sub-groups and the geographical within which they live. Good quality data underlie the functioning of evidence-based decisions. Data provide the building blocks for understanding the nature and composition of neighbourhoods, together with the expected health outcomes of their residents. But within the health arena there are many complicated data issues. Existing operational health data sets are often incomplete or not up-to-date and accessibility is often limited by data protection and medical confidentiality policies. They are derived from disparate sources: GP registers, Hospital Episode statistics (people who are admitted to hospital), Child Registry and Accident and Emergency records, all adhering to different data collection and storage standards and systems that vary between organisations. Cross-referencing between these datasets is technically difficult because of these issues. Frequent quality issues of operational health data limit the extent of analysis that can be carried out with confidence. Furthermore, health survey data are released at coarse geographical scales where the ecological fallacy limits the potential for exploring local variability. Given these limiting factors, the theme of this research is to extend the health inequalities research and its associated data framework to explore variability in the spatial and social domain. This enables the identification of social facts relating to health harming lifestyle choices and behaviour that contribute to 'diseases of comfort'. This is carried out by developing and exploring the usefulness of geodemographics for analysing health inequalities, thereby adding the social and spatial context to our undertanding of causes of health inequalities. This thesis presents a more straightforward yet effective alternative to exploring the measurement of health impacting behaviours and predicting health outcomes using operational health data, national health surveys and a geodemographic classification. Geodemographic analysis of health outcomes can capture different lifestyle behaviours, and has already proven useful not only in improving customer segmentation in the commercial sector, but also to better target public services (Harris et al., 2005). By applying geodemographic classifications to national health surveys and NHS operational datasets at postcode level, interesting conclusions can be drawn in terms of different health harming lifestyle behaviours at very fine scales. Furthermore it is common practice that academic research projects occur in isolation, and exploitation of research findings and best practices in local government sectors is often beset by many obstacles. Consequently, within local government the adoption of new innovative technique and tools may often be slow. An inner London Primary Care Trust (PCT) is used as a test bed for disseminating and evaluating the geodemographic framework and indicators. The concluding sections of the thesis discuss the practicalities of embedding geodemographics in particular and geography in general into a professional environment where these technologies are new and innovative.
|Title:||Modelling health related behaviours using geodemographics: applications in social marketing and preventative health|
|Additional information:||Permission for digitisation not received|
|UCL classification:||UCL > School of Arts and Social Sciences > Faculty of Social and Historical Sciences > Geography|
Archive Staff Only