A case-control study of dementia in people taking three commonly prescribed medications: rizatriptan, beclometasone and allopurinol

Study type
Date of Approval
Study reference ID
Lay Summary

As our population ages, an increasing number of people are becoming debilitated through failing memory and decision-making, with over 100 million being affected world-wide by the middle of this century. Apart from the personal impact on individuals, this puts a huge strain on the UK healthcare system, costing £1.7 billion annually. Unfortunately, extensive efforts to develop new drugs have so far failed and increasingly scientists have turned their attention to existing drugs that were developed for other conditions and testing them for their potential benefits in people experiencing problems with memory and decision-making. This approach is promising because it allows for faster progress since the safety of the drugs is already known. In recent tests on individuals aged 50 and older, our research group found that certain medications consistently improve performance in a panel of mental skills tasks. To determine if these drugs truly benefit people with memory and decision-making difficulties, we would like to know whether people on these medications are less likely to develop these problems. By analyzing prescription data, we will be able to answer this important question. It is expected that drugs showing a clear association with a decreased incidence in memory and decision-making problems can be readily prescribed for people showing mild symptoms of the condition, with the hope of slowing down disease progression.

Technical Summary

Our analysis of the the cognitive trajectories in the over 50s through the PROTECT[1] study showed that some medications are associated with favourable cognition profiles. This proposal aims to see whether these benefits are present in a wider cohort through an analysis of prescription data available on the CPRD platform. Our hypothesis is that candidate drug use is associated with a decreased incidence of dementia. To test the hypothesis we propose carrying out a case-control study comparing candidate drug prescription histories of those with dementia relative to those without. Cases will be individuals with a dementia diagnosis between 2012 and 2022. The control cohort will be randomly sampled from those with no dementia and matched for sex, GP practice and age in a 4:1 ratio. The prescription history of the candidate drugs will be assessed prior to the index date defined as the first dementia diagnosis date for the case and matched controls. The three distinct drug sets to be assayed are detailed in the Specific Aims section below. Positive outcomes from this retrospective investigation will enable us to apply for funding to conduct a trial in patients using online monitoring of cognition through our PROTECT platform (30,000 participants with 10% cognitively impaired). We envisage conducting such a trail rapidly and cost effectively given that the candidates have established safety profiles, adopting the protocol developed for our current VitaMIND trial of vitamin D (www.protectstudy.org.uk). Dementia therapy is an urgent unmet need, with UK cases predicted to exceed 1.6 million by 2050. Our investigation has the potential to fast track safe drugs for the symptomatic relief of dementia.

Health Outcomes to be Measured

The main outcome measure of the study is dementia diagnosis. Dementia diagnosis will be according to the Read medical codes.


Gareth Williams - Chief Investigator - King's College London (KCL)
Gareth Williams - Corresponding Applicant - King's College London (KCL)
clive Ballard - Collaborator - University of Exeter
Richard Killick - Collaborator - King's College London (KCL)