Supplementary Material for: Exploring the impacts of age at menarche on cognitive aging in late adulthood: Evidence from a Mendelian randomization study on the Taiwanese population

INTRODUCTION: The potential influence of age at menarche (AM) on cognitive aging remains inadequate, partly because of the difficulties presented by multiple confounders. To address this issue, the Mendelian randomization (MR) analysis was used to explore the causal impacts of AM on cognitive aging.

METHODS: Using the publicly accessible Taiwan Biobank, we identified single nucleotide polymorphisms (SNPs) significantly associated with AM as instrumental variables to estimate the effects of instruments on cognitive function assessed with the Mini-Mental State Examination (MMSE). We employed several MR methods, including two-stage least squares, inverse variance weighting (IVW), MR-Egger, weighted median, weighted mode, and constrained maximum-likelihood (cML) MR methods, to ensure the stability and reliability of the results.

RESULTS: MR analyses indicated no significant causal relationship between genetically determined AMs and total and subdomain MMSE scores, except the G5 subdomain (βIVW = 0.156, 95% confidence interval [CI]: 0.005, 0.307; βcML = 0.161, 95% CI: 0.014, 0.309). However, in a leave-one-out sensitivity analysis, we found a significant relationship between AM and cognitive aging after eliminating rs157863 and rs6758290, thus demonstrating the potential pleiotropic effects of these two SNPs. After these two SNPs were eliminated, a significant causal relationship between AM and overall MMSE scores (βIVW = 0.425, 95% CI 0.011, 0.839), though.

CONCLUSION: Evidence from the present MR study did not fully support a causal relationship between AM and cognitive function decline in later life. Potential pleiotropic effects of the genes underlying these two traits are worth of further investigation.