Diagnosing disease with shopping data
Publication Date
July 19, 2024
Creator
Abstract
In an era of widespread digital data collection, this research addresses critical knowledge gaps related to the responsible and effective use of personal transactional data for health research. The thesis explores the potential of shopping data to yield valuable insights into chronic and infectious diseases, with a specific focus on respiratory conditions, including COVID-19, and ovarian cancer. Investigating this emerging field, the research aims to uncover practical applications, with a particular emphasison utilising loyalty card data.
The primary research objectives encompass assessing the predictive capabilities of shopping data for disease forecasting and disease risk evaluation, with a particular focus on harnessing the potential of machine learning methods in these contexts. The research also endeavours to understand the public’s willingness to share individual loyalty card data for health research, including developing appropriate protocols and evidence-based justifications, to enable effective data collection and analysis.
Comprising a collection of five interconnected studies, this research adopts a mixed-methods approach. The qualitative method of interviews with the public, and in-depth surveys to capture the lived experiences of individuals self-managing symptoms through their shopping behaviours. Additionally, quantitative research utilises both aggregated and individual sales data linked to health information.
The research findings underscore how shopping data could play an important role in health research, illustrating the public’s willingness to share data when certain conditions are met. They reveal useful findings into the self-management of symptoms of ovarian cancer through shopping patterns and the use of shopping data for respiratory disease surveillance. Furthermore, the collection of individual loyalty card data, along with the application of machine learning techniques, demonstrates the potential of shopping data in advancing early disease predictions. Additionally, the research offers actionable recommendations and considerations for future researchers in this evolving field.
Item Type
ethesis
Thesis Type
PhD
Supervisors
Keywords
Subjects (LC)
Associated Schools / Departments
Nottingham University Business School (UK)
eprints ID
77924
UoN Repository URI
Except where otherwise noted, this item's license is described as
File(s)![Thumbnail Image]()
Name
Dolan_Elizabeth_20219229_corrections.pdf
Type
Full-text
Description
Examined. Second submission of thesis with minor corrections
Size
14.05 MB
Format
Adobe PDF
Checksum (MD5)
5d855b260bffed601f6a765d4a4d9ec8