University of Edinburgh

DataLab

DL_2023_027

More people are living with dementia. This can be a devastating and life-limiting condition that is of clear importance for the public, NHS and social care services. Having an idea of which people are at higher risk of developing dementia could help address known risk factors earlier in life, like smoking and obesity. We want to see how effective routinely entered health data from GP and hospital visits can be used to predict the future risk of developing dementia. This might include information like blood pressure measurements, other health diagnoses and blood test results. We will also see if other tools developed elsewhere can be used successfully in a Scottish population. If our approach is successful, we would provide some evidence for the use of dementia prediction tools to find people at higher risk. Future work could then test interventions in this group.

Dementia is a devastating and frequently life-limiting condition. As our population is getting older, more people are affected – over 90,000 people are living with dementia in Scotland. While there are no cures available, recent scientific breakthroughs suggest the potential to further slow dementia progression, but these medications are likely to be needed very early in the condition. We want to understand the potential for routinely collected health information to help predict those at higher risk of dementia before symptoms develop. If successful, this could be used to target interventions to better control known risk factors for dementia (e.g. blood pressure, smoking), as well as identify high-risk people in the future for newer medications. We are not the first team to think this is important, but many published prediction models use specialised data (e.g. brain scan images) which are only done when concerns about memory start. There is value in understanding if any of these tools could be applied to a Scottish population, and if so, how effective they are in predicting dementia. But we also want to test the effectiveness of routinely entered data in our population. While this would be unlikely to provide perfect individual risk prediction given the complexity of dementia, even identifying a group of people at substantially higher than average risk could provide value to the public and the NHS.

Health Services & Delivery

13/12/2023

De-Personalised

TRE