Desai, Roopal; (2025) What can Mendelian randomisation studies tell us about the relationships between modifiable lifestyle risk factors and dementia? Doctoral thesis (Ph.D), UCL (University College London).

Text Desai_10209283_thesis_id_removed.pdf Access restricted to UCL open access staff until 1 July 2026. Download (2MB)

Abstract

Dementia is a growing global health issue. Progress towards creating disease modifying pharmaceuticals has been slow and the current first generation of medications have limited impact. Therefore, prevention remains crucial. The Lancet Commission in 2020 identified twelve modifiable risk factors that explain up to 40% of dementia cases. This thesis used Mendelian randomisation (MR) to assess the causality of these associations and explored the genetic evidence for links between modifiable risk factors and different dementia subtypes. The first study used two-sample MR analysis to investigate ten of the twelve modifiable risk factors and their associations with Alzheimer’s disease (AD), frontotemporal dementia, and dementia with Lewy bodies. The findings suggested that smoking, obesity, and high blood pressure appear protective against AD. However, post-hoc analyses indicated these results may be influenced by survivor bias and pleiotropic effects with coronary artery disease. This study concluded that evidence from causal inference studies should be considered alongside epidemiological evidence and incorporated into reviews. The second study is a systematic literature review of MR studies examining the twelve risk factors. Taking evidence from 47 studies comprising of 160 analyses across ten risk factors the evidence was evaluated into one of four levels (robust, probable, suggestive, insufficient). Over half of the analyses were graded as providing ‘insufficient’ evidence for causal links with dementia. There was ‘probable’ evidence for smoking as protective which was potentially an artefact of survivor bias. Diabetes related metabolic dysfunction and blood pressure demonstrated the relatively strongest genetic evidence for causal links with dementia. Overall, the genetic evidence was inconclusive for most risk factors. The third study focused on hypertension, a major modifiable risk factor for dementia. Using observational and one-sample MR analysis in the UK Biobank cohort, this study found that genetically proxied systolic blood pressure was significantly associated with risk of VaD but not AD. The results emphasise the importance of dementia subtype-specific analyses when assessing the role of hypertension. Overall, this thesis highlights the need to critically evaluate genetic evidence and incorporate causal inference alongside epidemiological findings when assessing modifiable risk factors for dementia. The findings also highlight the need for more subtype specific research in the future. However, evidence from MR studies must be evaluated within the context of the limitations inherent to MR analyses such as survivor bias and construct validity. Further research is essential to clarify understanding of how lifestyle factors contribute to dementia risk and to guide prevention efforts.

Download activity - last month

Export as CSVXMLJSON

Download activity - last 12 months

Export as XMLCSVJSON

Downloads by country - last 12 months

Export as XMLCSVJSON

Archive Staff Only

View Item