About the AIM Projects#
This page contains plain language summaries of the 7 Research projects that form the AI for Multiple Long-Term Conditions Research Consortium.
AIM-CISC#
The AIM-CISC team will use Artificial Intelligence to analyse and understand this information to attempt to establish what patterns of Multiple Long-Term Conditions (sometimes called “multimorbidity”) are most common, which most affect people’s lives, and help improve quality and safety of care
AIM-CISC project is based at the University of Edinburgh, with collaborators from The Roslin Institute, NHS Lothian, and University College London
You can find More information about AIM CISC on their website: https://www.ed.ac.uk/usher/aim-cisc
Background to AIM-CISC’s Project#
Long-term conditions are health issues which persist over years. Many people have more than one long-term condition (e.g. having both diabetes and asthma), which is usually referred to as having Multiple Long-Term Conditions, or “multimorbidity”. Multimorbidity often seriously affects how well people feel and what they are able to do.
However, health professionals like doctors or nurses often focus on one condition at a time. This can make it hard to ensure that care is properly joined-up and coordinated for people with Multiple Long-Term Conditions. Research also tends to focus on one condition at a time. This is partly because people with Multiple Long-Term Conditions, can have very complicated mixes of different conditions and treatments.
Artificial Intelligence (AI) uses powerful computers to analyse and understand this kind of complex information, which is carefully gathered by the NHS and others under strict rules to keep it safe and confidential. New AI methods can help us better understand what patterns of Multiple Long-Term Conditions are common and which most affect people’s lives, and help improve the quality and safety of care.
What AIM-CISC Aims to achieve#
We propose a programme of research which will use AI methods to:
Identify the most common combinations of long-term conditions that people have, and examine whether people inherit a tendency to get particular combinations of conditions from their parents. This will help us understand what causes Multiple Long-Term Conditions, and identify potential treatments.
Examine whether particular combinations of long-term conditions are more common in some areas or communities. This will help us understand how where people live affects their health (for example, through differences in air pollution, or in access to safe open-space to play/exercise, or in having local shops that sell affordable healthy food).
Develop new methods for identifying people who are likely to have unexpected health problems like falls or bleeding. This will help us to work out what changes to their treatment or care could prevent these problems. One example of a possible change is medicines review to make sure that people take the right medicines for them as an individual. Another is rapid, comprehensive assessment of people admitted to hospital to ensure that their care suits all of their needs rather than just the problem that brought them into hospital.
Our team combines expertise in AI methods with experience in delivering and researching healthcare for people with Multiple Long-Term Conditions. We will work closely with patients and members of the public to help us focus and interpret our research, and to help us make our findings widely known. We will also work closely with other research teams to share learning and methods, and with the NHS and government to ensure that the research turns into practical improvements in care.
AI-MULTIPLY#
The AI-MULTIPLY Project is using artificial intelligence (AI) to characterise the complex relationships between Multiple Long-Term Conditions and polypharmacy across diverse UK populations to inform and develop effective health care pathways.
AI-MULTIPLY is based at Newcastle University and Queen Mary University London.
Background#
Many people live with two or more ‘long-term health conditions’ which include lots of different illnesses, such as cancer, heart, and mental health problems. People living with multiple long-term conditions may progress to poor health and have a shorter life expectancy. Treating multiple health conditions is a balancing act. People are often prescribed many different medicines together (known as ‘polypharmacy’). Sometimes these medicines (and their side effects) can interact in unexpected ways, causing further problems.
Aims#
To better understand the relationships between multiple long-term conditions, polypharmacy, personal and social factors to optimise treatment for individual patients. We will:
Use existing health data from many sources (including from people living in East London and the North-East of England) to understand how long-term health conditions and polypharmacy change and interact over time.
Look for patterns of change in long-term conditions and polypharmacy and how these patterns relate to personal/social factors and future health outcomes.
Use knowledge of these differences to understand when best to intervene to reduce health inequalities between patient groups and to improve outcomes.
Methods#
Our group has experience in using new developments in computer technology, called ‘artificial intelligence’ (AI). We will study electronic healthcare data collected from GP practices and hospitals from different regions across the UK. This data is anonymised, large and complex. AI is very good at spotting patterns in this kind of data. We will use AI to look for relationships between long-term health conditions, prescribed medicines, and personal/social factors, including measures of deprivation. In the long term, our research will lead to strategies for improved management of multiple long-term conditions including targeted review of medicines.
Public and patient involvement (PPI)#
We collaborated with our patient advisory group to develop an innovative PPI structure, and to ensure the study aims and outcomes reflect patient and public priorities. During a development award phase when we prepared for this project, we engaged with 42 people with experiences of long-term health conditions, social inequalities, and from ethnic minority groups. Going forward, our PPI network will facilitate engagement with 33 of these individuals throughout the course of the project. Meetings are planned for key time-points during the project to ensure questions are relevant and findings interpretable. Training activities include jargon busting sessions on AI and healthcare data. PPI partners and researchers will be offered training to enable mutual understanding and effective engagement.
Dissemination#
We developed a tailored dissemination strategy, in collaboration with PPI members, to reach multiple audiences. As well as reaching academics and clinicians via journal articles and conferences, we will create transcribed and translated videos for the project website. We will work with our PPI networks and use targeted social media to ensure findings are accessible to patients and members of the public, particularly underserved groups.
DECODE#
The DECODE project aims to map the challenges and requirements for Data-driven, machinE learning aided stratification and management of multiple long-term COnditions in adults with intellectual DisabilitiEs (ID). This description is how the team came up with the project name DECODE.
The project is based in Leicestershire and has collaborators at the Leicestershire NHS Trust, Loughborough University, University of Leicester, and De Montfort University.
You can find more information about DECODE on their website: https://decode-project.org/
Background to the DECODE project#
About 1 in 100 people are identified as having Intellectual Disabilities (also known as learning disability in UK). Of these, about two-thirds have two or more long-term health problems, known as Multiple Long-Term Conditions (MLTCs). Sometimes these conditions, such as diabetes and heart problems, can be delayed or prevented through lifestyle changes. For other conditions like epilepsy, better management can improve the quality of life. We are a group of researchers, health care professionals, people with intellectual disabilities and carers. We are committed to valuing diversity and improving the lives of people with intellectual disabilities to manage and prevent MLTCs.
Aims of the research#
We will use computer programs, called machine learning, to learn more about MLTCs in people with intellectual disabilities. We will find out which conditions are more likely to occur together and what happens over time. This information will help us to develop a new way of providing the care for people with Intellectual disabilities and MLTCs which would be better co-ordinated, effective and timely.
Design and methods used#
Our research combines Healthcare Information (data) and what we hear from experts. The experts include Experts by Experience (people with intellectual disabilities and their carers) and Professional Experts.
Healthcare information (data) We will use healthcare data on people with intellectual disabilities from England and Wales to find out:
MLTCs that tend to occur together, and
What happens to some of these MLTCs over time and the because of factors such as people’s lifestyle habits, and their financial and social situations.
Input from Experts by Experience People with intellectual disabilities, their families, and carers will either act as research partners (as part of Public and Patient Involvement groups) or as research participants. We will recruit research participants for workshops or interviews where this is needed. They will help to:
Decide on the most important MLTCs affecting the lives of people with intellectual disabilities.
Make recommendations about the care of people with MLTCs.
Produce visual information such as graphs and infographics specially designed for people that will use them, including people with intellectual disabilities. When we have finished this work, we will propose a new model which aims to improve care and support for people with intellectual disabilities and MLTCs.
Patient and Public Involvement#
We have worked closely with people with intellectual disabilities and their carers to develop this project. We are currently training a group of people with intellectual disabilities to become research partners. Our work with these groups of people helped us understand the support they need to take part in the research.
Continued involvement of patients and members of the public will be supported by our carer co-applicant, charity partners and Sense College, a local college supporting people with intellectual disabilities
Dissemination#
As well as research publications, our findings will be shared via social media, our charity partners, intellectual disability conferences, and national committees that contribute to guidelines.
DynAIRx#
The DynAIRx research team looks at how Artificial Intelligence (AI) could be used to transform how medicines and prescriptions can be optimisation for people living with Multiple Long-Term Conditions.
The project is based at the University of Liverpool and has collaborators at the Uiversity of Leeds, University of Manchester, and the ARC North West Coast.
You can find more information about the DynAIRx project on their website: University of Liverpool | DynAIRx
Aims#
DynAIRx aims to develop new, easy to use, Artificial Intelligence (AI) tools that support GPs and pharmacists to find patients living with Multiple Long-Term Conditions (two or more long-term health conditions) who might be offered a better combination of medicines.
We will focus on three groups of people at high risk of rapidly worsening health from Multiple Long-Term Conditions:
Older people with frailty,
People with mental and physical health problems, and
Other people with four or more long-term health conditions who are taking ten or more drugs.
Background#
Multiple Long-Term Conditions is becoming more common as the population becomes older and people with long-term health conditions live longer. Currently, people with Multiple Long-Term Conditions are treated separately for each condition and prescribed different drugs, each to treat one condition. Taking multiple medicines, or polypharmacy, increases the likelihood of serious side effects. The NHS introduced Structured Medication Reviews by GPs and pharmacists to reduce the number of people taking potentially harmful combinations of drugs. However, there is no easy way of predicting which patients are most likely to benefit from a medication review and prioritising them. The review team is then faced with gathering and making sense of information from records held in different places and piecing the information together to see how the patient’s conditions and treatments changed over time.
Design and methods#
DynAIRx will develop tools to combine information from electronic health and social care records, clinical guidelines, and risk-prediction models to ensure that clinicians and patients have the best information to prioritise and support Structured Medication Reviews. We will develop AI tools that combine information from multiple records and guidelines and calculate risks of hospital admissions and other adverse outcomes for our three multimorbidity groups. To ensure this information is easily understandable we will develop visual summaries of patients’ journeys, showing how health conditions, treatments and risks of future adverse outcomes are changing over time. These visual summaries will be tested in general practices across northern England and improved based on feedback from clinicians and patients. Patient and public involvement Members of the public, patients, carers, and clinicians will help develop and test these tools – focusing on what matters most to people whose health and care is being considered. We will also seek public views on using AIs in healthcare to ensure our approach is acceptable.
Dissemination#
We have built a large network of collaborators working on related projects across northern England, and internationally, over the past fifteen years. This network has produced the data, and the patient, public and practitioner involvement needed for building reliable healthcare AI tools. We will also use these connections to discuss and share our tools, ensuring they benefit a wide and varied population.
OPTIMAL#
The OPTIMAL project is about OPTIMising therapies, discovering therapeutic targets and AI assisted clinical management for patients Living with complex multiple long-term conditions. This description is how the team came up with the name OPTIMAL.
The OPTIMAL project is based at the University of Birmingham and University of Oxford, with collaborators in the University Hospitals Birmingham NHS Trust, NHS Greater Glasgow & Clyde, and the University of St Andrews.
What’s the research problem OPTIMAL is addressing?#
We treat each disease separately. This means we prescribe a different drug for each condition, which may not help people with four or more long-term health problems. (We say these people have complex multimorbidity or “cMM”.) A drug for one disease can make another disease worse or better. But there’s no information on the effect of one drug on a second disease. This means we are not sure which drug to prescribe in these people. We can group such people based on their mixes of disease. Then we can study the effects of a drug on each disease mix. This should help doctors prescribe better and reduce the number of drugs patients need. That would benefit almost all people with cMM.
The OPTIMAL project aims to#
Find the mix of diseases and drug treatments that interact over time to worsen or improve a patient’s health.
Predict the next disease that people might develop.
Find drugs that help more than one disease.
With input from lay people and health professionals, produce computer programs to help doctors improve the choice of drugs in patients with cMM.
How will OTIPMAL do this?#
Link large, detailed health records of patients who attend GP services and hospitals that have been collected since 1999. These include all diagnoses, disease severity, drugs, blood tests, readings such as blood pressure, scans, and specialist tests. Using artificial intelligence (AI) methods, model how the different mixes of diseases arise over time. The models will tell us what drugs cause or prevent a new disease. This will show if a drug helps improve symptoms of a disease or make them worse. The model will also help predict who may get another disease.
Ask lay people and doctors about their knowledge and views of AI-guided clinical choice making. This will help us make these models better to use in practice.
Use AI computer techniques to combine data, and together with the input from lay people and doctors, develop a computer program. This will predict which drug we should give and when we should give it to someone with cMM to reduce the risk of harm and bring about maximum benefit. It will also tell us what disease people may get next.
Examine the best way to present information in the computer program to lay people and doctors by asking them about what is important to them and what options they prefer.
The OPTIMAL team includes patients, public members, and world leading experts from three universities with expertise in biology, AI, medicine, health service research, public health, and general practice. All will improve the care of people with cMM. We will make all outputs including software and data models openly available.
You can read more information about the OPTIMAL project on their website: University of Birmingham Applied Health Research: OPTIMAL
MELD-B#
The MELD-B team uses an Artificial Intelligence-enhanced analysis of birth cohort data and electronic health records to identify lifecourse time points and targets for the prevention of early-onset, burdensome Multiple Long-Term Conditions.
The MELD-B project is based at the University of Southampton, with collaborators at University of Glasgow, Swansea University, Southampton City Council, the University of Aberdeen, and King’s College London.
Background to MELD-B’s Research#
A growing number of people are living with several long-term health conditions like diabetes, heart disease, depression or dementia. We call this Multiple Long-Term Conditions, which can be shortened to MLTC or “multimorbidity”. Many things throughout a person’s life influence the chances of developing health conditions. This includes their biology (e.g. age, ethnicity), things that happen to them (e.g. infections, accidents), behaviours (e.g. smoking, diet) and broader experiences (e.g. the environment people grew up in, their education, work, income). People from more disadvantaged backgrounds and/or certain ethnicities are more likely to develop MLTC and to develop it earlier. The impact (or ‘burden’) of MLTC, and the order that people develop conditions, also varies. We don’t know what all the possible opportunities are to prevent burdensome MLTC.
MELD-B’s Research Aims#
We aim to understand more about what things influence the way people develop early MLTC over their lifetime and the burden this has for them. By ‘early’ we mean before age 65. This will identify key time points in a person’s life where prevention efforts should be targeted or strengthened to reduce the risk of that person developing burdensome MLTC.
The Research Design and methods used#
To understand what ‘burdensome’ means for people with MLTC and how it could be prevented or reduced, we will ask patients, carers and MLTC experts their views and carefully summarise previous research on this issue.
Ideally, we would study very large numbers of people from birth to death. However, not enough of this type of data is available. Therefore, we will use Artificial Intelligence (AI) methods (using computers to learn patterns from data) to help us connect information and knowledge from two very big General Practitioner datasets with information from three ‘birth cohorts’ – research studies of people all born in the same year (e.g. 1970) and followed throughout their lives.
We will also use AI methods to help us understand the order in which people develop conditions and how they group together to become ‘burdensome’.
We will have experts and members of the public in a ‘People, Policy and Impact’ group identify opportunities to prevent burdensome MLTC.
How MELD-B will involve patients and memebrs of the public#
A diverse ten-person Patient and Public Advisory Board will oversee the project. Members of the Board will actively engage with each element of the project and be asked to discuss results as they emerge. This will ensure that the experiences and priorities of people living with MLTC will always be considered. Their input will ensure all decisions consider the experiences and priorities of people with MLTC.
CoMPuTE#
The Complex Multiple long-term conditions Phenotypes, Trends, and Endpoints (CoMPuTE) project uses Artificial Intelligence to help predict who is more likely to develop multiple long-term conditions. The project is funded by the National Institute for Health and Care Research (NIHR) under its Programme Grants for Applied Research Programme (NIHR204406). It is based on the University of Oxford, University of Leeds and University College London.
Background to the CoMPuTE project#
More than a quarter of adults in England have more than one health condition. By 2035 this is expected to increase by 10-17%. Having more than one condition is called ‘multiple long-term conditions’ (MLTC). The more conditions someone has, the more disabling the effects.
MLTC are difficult for both patients and carers: taking more medicines (with possible problems caused by conflicting or simply too many medications); the cost and wasted time of attending too many healthcare appointments; and the day-to-day challenges of living with multiple conditions.
This study hopes to predict who will suffer from MLTC and how MLTC will progress over a person’s lifetime. Previous research has focused on looking at causes of MLTC, however much is still unknown about why certain conditions appear together, how they relate to normal ageing, prevention, and appropriate care. Also, although the NHS currently invests significant amounts of money in trying to prevent specific health conditions (e.g. heart disease, cancer), many people do not engage. This is a missed opportunity to prevent future ill health.
The CoMPuTE Research Aims#
This project looks at whether using artificial intelligence (AI) can help us predict those more likely to develop MLTC – to get help sooner to those who need it and prevent people from developing MLTC in the first place.
Regular computer models are already used for research on electronic health records. We want to use AI techniques to process this information faster and more accurately. The data will be ‘anonymised’ so it cannot be traced to individuals. Because many people have concerns about how their data are used, members of the public have been involved in this work from the beginning and will be involved throughout. A public member leads one section of work. Other public members work on an equal level with the academic researchers.
This study hopes to see whether it is possible to predict who will suffer from MLTC and how MLTC will progress over a person’s lifetime.
It will investigate inequalities and the health and financial burden of MLTC.
It brings in the public perspective on ethical and social questions about the use of AI in healthcare. Members of often-excluded communities will be actively involved in discussion groups, the development of study materials and the writing of papers. This is important to ensure that plans to help people with MLTC address everyone’s health and care needs.
How CoMPuTE will involve patients and members of the public#
One of the three CoMPuTE Themes (Ethics, Patients and the Public’) is entirely public-led and aims to adopt best practices and break new ground in public and patient involvement in directing the research. Our public stakeholder group brings a wide range of competencies and experience, including personal, geographical, ethnic, socio-economic and age diversity and lived experience of the immigrant experience and living with MLTC, and professional competencies in engineering, randomised control trials, medical communication, project management and health care delivery, to name a few.
Public members are linked with the data and epidemiology Themes (Themes 1 and 2) and are increasingly embedded in that work. Public members have already helped shift the direction of research. Public members already work with and will be presenting to researchers within and beyond the programme; we are planning a series of public-led mini-webinar; our public members will be co-authoring papers, and they will be forging the broader dissemination strategy.
AI-MIXED Cluster#
Information coming soon.