DECODE – Using artificial intelligence to improve the health & wellbeing of people with learning disabilities

The Population Data Science group at Swansea University is to collaborate on a new study led by Loughborough University and the Leicestershire Partnership NHS Trust, that will use Artificial Intelligence (AI) to improve the health and wellbeing of people with learning disabilities utilising the Wales national Trusted Research Environment (TRE) The SAIL Databank.

About 1 in 100 people are identified as having a learning disability.

Of this population, over 65% have two or more long-term health problems, known as multiple long-term conditions (MLTCs), and a life expectancy that is 20 years lower than the UK average.

The team at Swansea University will work alongside partner organisations using anonymised population-scale data from across the UK to provide valuable, anonymised healthcare data for people with learning disabilities.

Often, the physical ill-health symptoms experienced by those with a learning disability are mistakenly attributed to a mental health/behavioural problem, or as being inherent to their disability.

This means they do not always receive the same level of care as those without a learning disability.

And as there is no easy way to understand and predict the complex interactions between MLTCs and the care needs of individuals, it is difficult to provide effective joined-up care between health and social services.

For the DECODE (Data-driven machinE-learning aided stratification and management of multiple long-term COnditions in adults with intellectual disabilitiEs) project, the team will use machine learning to better understand MLTCs in people with learning disabilities.

The researchers will analyse healthcare data on people with learning disabilities from England and Wales to find out what MLTCs are more likely to occur together, what happens to some of these MLTCs over time, and the role other factors, such as lifestyle choices, financial position, and social situations, play in their MLTCs.

By their very nature, SAIL Databank and other TREs employ strong governance protocols to ensure all data is safeguarded with robust access controls. As such, each TRE will be used independently to run each analysis before the results are carefully brought together outside of their respective TREs following disclosure control for a broader meta-analysis for greater insights.

To uncover the wider factors associated with the health and wellbeing of people with learning disabilities, the incredibly rich and varied data assets held securely within SAIL Databank can be linked together in the analysis. AI and machine learning tools can then be deployed across the data to reveal patterns and clues for MLTCs.

The team will also work directly with people with learning disabilities, their carers, and the professionals who support them.

This will help to identify the most important MLTCs affecting the lives of people with learning disabilities, make informed recommendations about the care of people with MLTCs, and produce visual information such as graphs and infographics that can be easily understood.

The end goal is to create a new joined-up model of care for people with learning disabilities, that brings together the multiple clinical guidelines relevant to the dominant MLTCs in this population, in a format accessible for all users. Ultimately this will enable the better management of MLTCs by health and social care providers, and in some cases, prevent them from developing.

Loughborough University’s Dr Thomas Jun, a Reader in Socio-technical System Design, is co-lead for the project, said, “We are very excited about this collaboration opportunity, working with clinicians and experts in data science, AI, medical informatics, human factors, design, ethics and qualitative research, as well as those with lived experience of learning disabilities. We will be able to demonstrate how AI can create safe, ethical and cost-effective improvement to the quality of life for thousands of people with learning disabilities.”

SAIL Databank’s Chief Technical Officer, Professor Simon Thompson, said, “We’re very proud of SAIL Databank’s involvement in this type of cutting-edge data science, working alongside other leaders in the field. The SAIL Databank TRE platform, that’s powered by our sophisticated data technology, SeRP – the Secure eResearch Platform, will ensure the research team have all the tools and capabilities needed to deliver this important work that can help realise huge benefits for society.”

Swansea University’s Senior Research Manager and Data Scientist, Ashley Akbari, said, “We have a strong record of multi-organisational and multi-disciplinary collaborations like DECODE. It’s exciting to see TREs and research teams from across the UK coming together to better understand the rich data landscape that the UK can provide, working with academics, policymakers, health care professionals and members of the public to ensure the research we deliver provides impact and value to people and services. The potential for this project to utilise and accelerate data science techniques is exciting, and we are proud to be collaborating in this Team Science project”.

Co-lead Dr Satheesh Gangadharan, a Consultant Psychiatrist with the Leicestershire Partnership NHS Trust, added, “Moving forward we hope our research will shape how people with a learning disability and long-term conditions are supported in the UK and beyond. The links we have with the National Learning Disability Professional Senate, Royal Colleges, Health Education England, Public Health Wales, NHS England and NHS Wales will enable us to make a real impact and improve the care.”

The DECODE project is being funded by the National Institute for Health Research (NIHR), the research partner of the NHS, public health and social care, and is due to start in April. The other academic project partners include the University of Leicester, Swansea University, King’s College London, University of Plymouth, the University of Nottingham, and De Montfort University.

Adapted from an article first published by Loughborough University on 6th May 2022.