Multivariate analysis for the prediction of dementia status and risk

 
   Lead Investigator:    David Llewellyn
   Institution      :    University of Exeter Medical School
   E-Mail           :    David.Llewellyn@exeter.ac.uk
   Proposal ID      :    895



   Proposal Description:
   This project will identify which predictors of dementia status and risk are most useful, in order to clarify which signs and symptoms of dementia may be most important clinically. A combination of advanced multivariate statistical techniques will be employed, such as bootstrap regression and Bayesian modelling. This is an extension of an ongoing project using NACC data to develop dementia disease signatures for differential diagnosis (title: ?Advanced computational modelling for the differential diagnosis of dementia?). Please note that we are collaborating with Professor Andrew Zhou  on this project.    The specific objectives of the project are:  ? To develop and validate multivariate predictive models of dementia status.  ? To extend this model by predicting conversion to dementia in patients free from dementia in their initial assessment.    In summary, this project aims to advance our understanding of the ?high risk? groups that should be targeted within the dementia identification pathway. In particular this will inform future clinical guidelines regarding the signs and symptoms that should trigger further investigations.