Multidiscipline Ecosystem - Lifecourse Determinants of Multimorbidity
Research type
Research Study
Full title
Developing a Multidisciplinary Ecosystem to study Lifecourse Determinants of Complex Mid-life Multimorbidity using Artificial Intelligence (MELD)
IRAS ID
293759
Contact name
Simon DS Fraser
Contact email
Sponsor organisation
University of Southampton
Duration of Study in the UK
0 years, 7 months, 31 days
Research summary
Summary of Research
Society faces many challenges related to the growing number of people living with multiple long-term health conditions. Throughout life many things influence the chances of developing such conditions, such as age and ethnicity, illnesses or accidents, smoking and diet. Perhaps even more important are broader influences such as the environment people grew up in, their education, work etc. People from more disadvantaged backgrounds are more likely to develop multiple conditions at an earlier age and there is evidence that the order of developing conditions influences subsequent conditions’ development. Understanding those broader influences and how they affect that order is vital to inform when and how we should intervene.
To achieve this, we need to study large numbers of people over their lifetime, but such datasets don’t exist. Large health datasets from NHS GPs are helpful but haven’t been running long enough to track from birth to later life. They include lots of information on long-term conditions but not much about broader issues. We have access to one dataset of about 700,000 people to identify health conditions and also to data from the ‘1970s birth cohort’ –about 17,000 people born in the same week of 1970 followed throughout their lives providing information about many broader issues every few years.
The aim here is to safely and ethically establish the necessary environment, systems and methods to allow artificial intelligence techniques to ‘connect’ birth cohort data with large GP datasets. This will allow us to ‘connect’ information on the broader, lifecourse issues with the GP long-term condition information to:1. Identify the kinds of people who develop combinations of long-term conditions by middle-age.
2. Understand the order of developing long-term conditions through life and which ones develop first.
3. Work out how broader issues affect that order and the resulting combination of conditions.Summary of Results
As with many countries we are facing challenges related to the growing number of people living with multiple long-term health conditions like diabetes, heart disease or dementia. All the way through peoples’ lives many things influence the chances of developing such conditions. This includes some things that are hard to research - broader issues throughout life such as the environment people grew up in, their education, work, and so on. Sadly, people from more socially and economically disadvantaged backgrounds are more likely to develop multiple conditions at an earlier age. There is also evidence that the order of developing conditions varies considerably and influences what then happens to people. This makes understanding these broader issues and how they affect that order vital to inform when and how we should intervene to prevent conditions developing.
To achieve this, we need to study large numbers of people over their whole lifetime, but such datasets do not exist. Very large health datasets collected from NHS GPs are helpful but haven’t been running long enough to track from birth to later life. They include lots of information on long-term conditions but not much about broader issues.
In our Development Award (called ‘MELD’) we had access to one such dataset of about 700,000 people, which we used to identify health conditions. We also accessed data from the ‘1970 British Cohort Study’ – a long-running research study called a ‘birth cohort’ - about 17,000 people born in the same week of 1970 followed throughout their lives. This provided detailed information about many broader issues every few years up to age 46 for about 8000 people from across the country. Our aim was to safely and ethically establish the necessary environment, systems and methods to allow artificial intelligence (‘AI’) techniques to ‘connect’ birth cohort data with large GP datasets. This would allow us to connect information on the broader, lifetime issues with GP information on long-term conditions. We explored these AI ‘transfer learning’ methods between datasets but technically this is difficult to do and we encountered some problems. Despite this we made some important ‘proof of principle’ achievements. We identified the kinds of people who developed certain combinations of burdensome long-term conditions by middle-age. We also identified the order of developing long-term conditions and which ones developed first using specific examples. We explored how broader issues influenced the resulting combination of conditions. Importantly, we built up our team by bringing in expertise that will be needed for the future Research Collaboration (called ‘MELD-B) and submitted our application to NIHR. In this future Research Collaboration we will apply everything we learned in MELD to three birth cohorts and two much larger routine datasets to estimate the risk at different life stages and identify key time points for targeted public health interventions. We will also have a Patient and Public Advisory Board overseeing the whole project and engaged with every step, and a whole section of the work dedicated to dissemination and engagement, so we make sure that our work will have impact.
REC name
London - Camden & Kings Cross Research Ethics Committee
REC reference
21/PR/0023
Date of REC Opinion
21 Jan 2021
REC opinion
Favourable Opinion