Examining the causal mediating role of cardiovascular disease on the relationship between subclinical cardiovascular disease and cognitive impairment: The Cardiovascular Health Cognition Study

Abstract

AbstractBackgroundSubclinical cardiovascular disease (sCVD) is associated with an increased risk of incident Mild Cognitive Impairment (MCI) or Alzheimer’s disease (AD). However, relatively little is known about the mechanisms by which sCVD leads to MCI or AD. One hypothesis is that cardiovascular disease (CVD) mediates the relationship between sCVD and MCI or AD.MethodUsing data from the Cardiovascular Health Cognition Study (N=3,602) and the R package medflex, we fit natural effects causal mediation models to test the hypothesis that CVD is a causal mediator of the effect of sCVD on the risk of MCI or AD within 10 years. All models were adjusted for age, sex, race, education, diabetes status, smoking status, and body mass index, and we excluded those with baseline CVD. We also stratified our analyses by apolipoprotein (APOE)‐4 carrier status and conducted several sensitivity analyses to assess the robustness of our results to violations of our assumptions.ResultAmong those in our analytic sample (N=2,377), sCVD significantly increased the odds of MCI or AD by 35% (OR: 1.35, 95% CI: (1.14, 1.56)) in total. Both the natural direct effect (OR = 1.31, 95% CI: (1.10,1.52)) and the natural indirect effect (OR=1.02, 95% CI: (1.01, 1.05)) were significant as well. CVD mediated approximately 9.5% of the total effect. Among APOE‐4 allele noncarriers, we also found a significant total effect (OR = 1.50, 95% CI: (1.25, 1.76)), natural direct effect (OR=1.44, 95% CI: (1.17, 1.70)), and natural indirect effect (OR=1.04, 95% CI: (1.01, 1.08)) of sCVD on MCI or AD risk through CVD, with approximately 11% of the total effect being mediated. Among APOE‐4 carriers, however, we failed to see any significant effects. Multiple sensitivity analyses showed our results were robust.ConclusionWe found that CVD causally mediates the relationship between sCVD and the risk of MCI or AD within 10 years for APOE‐4 noncarriers, provided that our analytic and causal assumptions hold. Interestingly, the proportion mediated by CVD was only 9‐11% in our analyses, indicating that there are other, possibly substantial, etiological pathways by which sCVD increases the risk of MCI or AD; future research is needed.