Development of an Ethnically Stratified Model for Causal Risk Prediction of Dementia
Dementia is a progressive condition that affects nearly 10 million new people worldwide every year, with around 900,000 cases in the UK alone. Despite the significant public health burden, existing predictive models are only moderately accurate due to their reliance on a limited number of associated factors, many of which may not be causal. The aim of the project is to develop an approach that effectively explores all potential risk factors for dementia and leave out those that are simply associated with the condition. This can then be built into a refined causal risk model for use in interventions. The objectives are, for each ethnic group, 1) to construct a causal network of dementia; 2) to develop a causal risk prediction model of dementia where the key causal risk factors are informed by 1); and 3) to develop free software to share with the community.
University of Manchester
Advisor: Hui Guo
