Improving care pathways for people with multimorbidity

  • Research type

    Research Study

  • Full title

    Artificial Intelligence and Multimorbidity: Clustering in Individuals, Space and Clinical Context (AIM-CISC)

  • IRAS ID

    304406

  • Contact name

    Bruce Guthrie

  • Contact email

    bruce.guthrie@ed.ac.uk

  • Sponsor organisation

    University of Edinburgh

  • Duration of Study in the UK

    2 years, 7 months, 31 days

  • Research summary

    People in the UK are living longer, which means that more of us now live with multiple long-term health conditions (multimorbidity). People with multimorbidity are more likely to experience serious adverse events such as falls, fractures, confusion, bleeding, and kidney problems. These can be caused by the conditions a person has, but also by the medicines and other treatments they receive. They can happen in the community (where they often cause hospital admission), or can happen in hospital when someone is admitted for another reason. This study is part of a larger project examining whether we can reliably predict who will get a serious adverse event, or who will have complicated care pathways. This study aims to understand professional, patient and carer experiences of health and social care, to identify opportunities for improvement, and to design a complex intervention for people at the highest risk of serious adverse events.

    Results Summary
    Background and aim. Long-term conditions are health issues such as heart disease or dementia. Many people have multiple long-term conditions, which affect how well they feel and what they are able to do. However, doctors and nurses often only focus on one condition at a time, and research is needed to better understand who gets multiple conditions and how to better care for them.
    What we did. Patients and carers were involved in all the work. (1) We examined many different methods for identifying the different patterns of conditions affecting individuals. (2) Using several datasets in Scotland and Wales, we examined which types of neighbourhoods had the highest rates of multiple conditions. (3) We worked with patients, carers and professionals to redesign care for people with multiple conditions, including developing artificial intelligence tools to predict which people have worse outcomes.
    What we found. (1) We did not find any method for identifying patterns of conditions which produced results which were easy to understand or which seemed useful to clinicians or patients. (2) We found that people are developing multiple conditions at younger ages than in the past. Multiple conditions were more common in people living in poorer neighbourhoods, and in neighbourhoods where residents feel less part of the community or are lonely. (3) We successfully developed new tools to extract important information from medical records and to predict bad outcomes in patients. Working with patients, carers and professionals, we designed new ways of caring for people with multiple conditions.
    What happens next? We are developing all three areas of work. In particular, we continue to work with the NHS and with patients and carers to improve care for people with multiple conditions.

  • REC name

    North West - Greater Manchester West Research Ethics Committee

  • REC reference

    22/NW/0313

  • Date of REC Opinion

    16 Nov 2022

  • REC opinion

    Further Information Favourable Opinion