Artificial Intelligence and Multimorbidity: Clustering in Individuals, Space and Clinical Context

Objective 3

Examining clustering of complex multimorbidity in communities and places.

It is well known that there are stark inequalities in health outcomes across neighbourhoods in the UK, but little is known about how specific combinations of health conditions vary as a result of the characteristics of neighbourhoods, households and individuals. Objective 3 will investigate the complex ways in which an individual's risk of developing combinations of health issues is shaped by their own individual and household circumstances and as a result of the kind of place in which they live. We will link detailed data on people including health outcomes and their social circumstances to separate data on the characteristics of the places that they live in. The data on people will come from the Scottish Longitudinal Survey, which uses census records linked to health data, as well as the Dataloch and SAIL data repositories on use of health services in Lothian and Wales respectively. We will link the data on people to a databank on the characteristics of neighbourhoods across the UK including features such as neighbourhood deprivation, green space, pollution and the nature of local food outlets. We will create new and innovative indicators of neighbourhood characteristics, not captured in official statistics, such as community cohesion using analysis of themes that emerge in news papers and social media data about particular places . The overall aim is to produce detailed findings on the extent to which particular combinations of health conditions are concentrated in specific places and groups of people, and to provide explanations of why this may happen.

 

Geography

Clinical 

Social Science

Natural Language Processing