Data Use Application 007
Lead applicant organisation name
Name of the legal entity that signs the contract to access the data.
University of Birmingham (UoB). Applicant: Thomas Jackson, Clinician Scientist in Geriatric Medicine
Project title
The title of the project/research study request that the applicant is investigating through the use of health data.
Optimising therapies and disease trajectories for patients living with complex multimorbidity using detailed information about eye health as predictors and outcomes (OPTIMAL).
Lay summary
A concise and clear description of the project. This should outline the problem, objective and the expected outcomes in language that is understandable to the general public.
Conventional healthcare tends to treat each disease from which a person is suffering as
separate, prescribing different medicines for each condition. This approach might not help those with multiple long-term health problems because a drug prescribed for one disease may affect another disease. There is little information on how a drug for one condition impacts another condition, and some conditions have many treatment options. However, by grouping people with multiple health problems based on their diseases, we can study how a specific drug affects each disease combination.
New artificial intelligence (AI) methods can process large amounts of health data quickly. If an AI model could guide doctors in prescribing medicines, it might reduce the number of drugs patients need and prevent harmful drug interactions.
This study aims to (a) identify how combinations of diseases and drug treatments interact over time to affect a patient's health; (b) predict the other conditions people might develop, focusing on eye diseases; (c) identify drugs that help more than one disease, particularly those that benefit eye disease.
The study uses patient health records, including diagnoses, drugs, blood tests, and data from eye images, with AI to model how different disease combinations develop over time, with a focus on eye diseases. The model will identify drugs that cause or prevent new eye diseases and predict the risk of future eye diseases. The model may also determine if eye imaging data can predict the risk of other diseases, like heart disease. This approach could optimise medication for patients with multiple health conditions, improving their quality of life and benefiting patients, the NHS, and society.
Public benefit statement
A description in plain English of the anticipated outcomes, or impact of project on the general public.
Data from INSIGHT will allow categorisation of patients by the severity of their eye diseases. For example, images can measure central retinal thickness, which is a more accurate indicator of diabetic retinopathy severity than primary care diagnoses. This data can be modelled to explore whether certain medications improve or worsen eye diseases. Data from eye images may also help predict the risk of future conditions like heart disease.
The development of algorithms through the study could predict the next disease a patient may develop and identify medications that may be of benefit. The deployment of INSIGHT data could lead to optimised treatment for patients with multiple long-term conditions. This might involve removing unnecessary medications from a patient’s treatment regime, switching to less harmful ones, and adding beneficial ones — an approach that can potentially reduce the risk of developing other conditions and improve patients' quality of life.
Latest approval date
The last date the data access request for this project was approved by a data custodian.
17 July 2023
Dataset(s) name
The name of the dataset(s) being accessed.
Bespoke dataset from University Hospitals Birmingham ocular diseases and comorbidities data
Access type
Determines whether the data will be accessed with an Trusted Research Environment (TRE) or via data release.
Data Licence Agreement. Data released to a secure and controlled environment under contract with the research applicant.
Data sensitivity level
The level of identifiability of the data being accessed, as defined by Understanding Patient Data.
Anonymised