Mediation Analysis in Understanding Mechanism of Alzheimer’s Disease Risk

Although studies often show reduced risk of dementia with late-life cognitive activity, concerns about residual confounding and reverse causation cast doubt on these findings. In Chapter 1 of this dissertation, we review epidemiologic studies of cognitive activity and incidence of Alzheimer’s disease (AD) and all-cause dementia, and conduct a bias analysis that indicates the observed inverse associations are likely robust to unmeasured confounding, and probably only partially explained by reverse causation. While pursuing enjoyable cognitive activities may reduce dementia risk, better characterization of the type, duration, and timing of activity associated with late-life cognitive benefit is needed to develop recommendations applicable over the lifecourse. The apolipoprotein episilon4 allele (APOE e4) is the most well established genetic risk factor for AD, and is also a risk factor for cerebrovascular disease (CVD). In Chapter 2, we use the counterfactual approach to mediation analysis to investigate the degree to which the negative effect on cognition of the e4 allele is attributable to its effects on CVD. Using neuroimaging and neuropsychological data from approximately 4,000 participants of the population-based Age, Gene/Environment Susceptibility–Reykjavik Study, we found that 9% of the e4 effect on cognition was jointly mediated by white matter lesion volume and cerebral microbleeds (CMBs). While our finding that the e4 effect largely operates through non-vascular pathways aligns with previous research and present understanding of the action of apoE in AD and CVD pathogenesis, our study is the first to show a small effect specifically via markers of CVD pathology. In Chapter 3, we investigate the role of CMBs further using the newly developed four-way decomposition approach. We found that when comparing e4 heterozygotes to e4 non-carriers, the e4 effect on memory was independent of CMBs. By contrast, when comparing e4 homozygotes to e4 heterozygotes, the e4 effect on memory was attributable to interaction between the effects of e4 alleles and CMBs, perhaps suggesting a greater vascular contribution for these individuals. Similar analyses in other population-based studies will be needed to confirm these findings and further elucidate the contributions of CMBs and CVD to the e4 effect on cognition.