AI-based modelling of EHRs of patients with learning disabilities and multiple long-term health conditions

There are approximately 1.1 million UK adults aged 18+ living with a learning disability (LD). Individuals with LDs often experience poorer physical and mental health, as well as higher comorbidity rates and avoidable mortality than those without LDs. This talk presents our work on the NIHR-funded project entitled DECODE: using machine learning to Identify Clusters and trajectories of multiple long-term conditions (MLTCs) in Individuals with LDs. This talk will cover: 

1) Methodology for constructing an electronic health record dataset from the SAIL databank 

2) Development of an equitable machine learning model to predict hospital length of stay

3) Modelling longitudinal trajectories for this population.

The work of the DECODE team was explored as a feature article in the January Issue of VOLUME, an online magazine published by he University of Loughborough. The article explore how DECODE is supporting people with learning disabilities to play a significant role in co-creating new ways to access care that better meet and suit their needs.