AI health prediction could help healthcare professionals identify future medical problems earlier for people with learning disabilities. Researchers from the DECODE project have developed a digital tool using artificial intelligence, statistical methods and health records from more than 33,000 adults across England and Wales. The system estimates which conditions an individual may be more likely to develop and how their health could change over time based on patterns observed in people with similar profiles.
The project is led by Loughborough University and Leicestershire Partnership NHS Trust and funded by the National Institute for Health and Care Research. Researchers hope the technology could ultimately become part of routine annual health checks for people with learning disabilities.
AI Health Prediction Targets Earlier Intervention
The DECODE tool was developed in response to significant health inequalities affecting people with learning disabilities.
Researchers report that people with learning disabilities die, on average, around 20 years earlier than the general population. About 39% of deaths among adults with learning disabilities have been classified as avoidable, nearly twice the rate seen in the wider population.
These disparities make earlier identification of health problems especially important.
Rather than waiting until symptoms become severe, the DECODE tool is designed to show clinicians possible future risks so preventive action can be considered sooner.
AI Health Prediction Uses Personal Health Factors
The technology uses information including a person’s age, sex and existing medical conditions.
Healthcare professionals can enter these characteristics and view possible health trajectories over a selected number of years. The predictions are based on patterns identified within population-level health records rather than determining with certainty what will happen to a specific patient.
This distinction is important. The tool is intended to support professional judgment and preventive planning, not provide a definitive diagnosis of future disease.
DECODE Analyzes Multiple Health Conditions
The project focuses heavily on multimorbidity, meaning the presence of several long-term conditions in the same person.
DECODE researchers found that 86% of the people with learning disabilities studied were living with at least two long-term conditions, with some individuals experiencing as many as 12. These included conditions such as diabetes, cardiovascular disease and mental health disorders.
Understanding these combinations is challenging because one medical condition can influence the likelihood, management or severity of another.
AI Health Prediction Maps Disease Pathways
Researchers used AI and statistical methods to analyze how conditions appeared together and developed over time.
The resulting digital tool produces visual maps showing possible relationships between existing and future conditions. The strength of connections between health problems can also be displayed, helping clinicians explore how an individual’s medical profile might evolve.
Researchers found that some combinations of conditions seen in people with learning disabilities differed from patterns typically observed in the general population.
This could make population-specific prediction tools particularly valuable because general healthcare models may not fully represent the risks faced by this group.
Annual Health Checks Could Use AI
The researchers ultimately want to integrate the tool into existing annual health checks.
In England, anyone aged 14 or older who is on their GP’s learning disability register can receive a free annual health check. These appointments are intended to identify health problems early, review medications, discuss existing conditions and create or update a health action plan.
Between April 2025 and March 2026, NHS data recorded 273,430 learning disability health checks, representing 79.8% of eligible people on the register aged 14 and older.
AI Health Prediction Could Strengthen Checks
Adding predictive information could broaden the purpose of these appointments.
A clinician would not only review conditions a person already has but could also see which health problems may warrant closer monitoring in future.
For example, an elevated predicted risk could lead to additional screening, lifestyle support, medication review or more frequent follow-up.
Dr Satheesh Gangadharan, co-lead of DECODE, said earlier visibility of potential conditions could enable clinicians to consider preventive interventions before those illnesses develop.
Communication Is Central to the Project
The DECODE initiative is not focused solely on algorithms.
People with learning disabilities and their carers have helped shape the research, discussing healthcare experiences, guiding research questions and advising on how the technology should be used.
Researchers noted that some people with learning disabilities can find it difficult to explain symptoms or understand physical changes. This can contribute to symptoms being missed or incorrectly attributed to the learning disability itself.
AI Health Prediction Includes Accessible Resources
The project has therefore developed accessible communication tools alongside the predictive technology.
These include simple color-coded icons representing different health conditions and an animated resource explaining how health can change over time.
The aim is to help patients participate more actively in conversations about their health rather than designing the system exclusively for clinicians.
This patient involvement may also improve trust in AI-supported care by making predictions easier to understand and discuss.
Further Testing Is Still Required
The DECODE tool is not yet ready for routine NHS deployment.
Researchers are seeking additional funding to refine the technology and test it with larger healthcare datasets. They estimate it could be ready for trials in real healthcare environments within approximately four years if further development progresses successfully.
The planned next phase will also examine integration into healthcare systems and apps while expanding accessible patient resources.
Researchers intend to produce policy recommendations focused on earlier intervention, preventive care and more coordinated services for people with learning disabilities.
AI Health Prediction Could Support Prevention
The potential significance of DECODE lies in shifting healthcare from reacting to illness toward anticipating risk.
For people living with multiple long-term conditions, identifying patterns earlier could help clinicians prioritize screenings, monitor warning signs and develop more individualized prevention strategies.
However, predictions must remain clinically validated, understandable and used alongside professional judgment. Researchers will also need to demonstrate that identifying additional risks leads to meaningful improvements in patient outcomes.
If further testing confirms its value, AI health prediction could become an additional tool within annual health checks, helping healthcare professionals move beyond documenting current conditions toward preventing future ones.
For a population that continues to experience substantial health inequalities and avoidable mortality, that transition could make predictive technology particularly valuable.
For more healthcare industry updates, insights and news, visit DistilINFO. Click here to subscribe to stay informed.
