Evening chronotype and incidence of all-cause dementia and subtypes after accounting for other sleep traits: A prospective cohort study

Background Chronotype may affect neurophysiology, but whether its association with dementia persists independent of other sleep traits remains uncertain. Objective To examine associations of chronotype with incident dementia subtypes and potential modification by metabolic and social factors. Methods In 443,906 dementia-free participants, Cox models estimated dementia risk after adjustment for sociodemographic, lifestyle, clinical, apolipoprotein E ( APOE ) genotype, and sleep-related factors. We tested baseline interactions and robustness using multiple imputation, competing-risk models, and landmark analyses. Results Over a mean follow-up of 12.5 years, 6355 incident dementia cases were documented. Evening chronotype was associated with a small increased risk of all-cause dementia compared with morning types (HR, 1.07; 95% CI, 1.01–1.14). This association was observed primarily in “other dementias” (HR, 1.12; 1.04–1.21), with no significant association with Alzheimer's disease or vascular dementia (VaD). Exploratory subgroup analyses identified two interactions that remained significant after false discovery rate (FDR) correction: social visit frequency for all-cause dementia and napping for VaD (FDR-adjusted p for interaction is 0.024 for both). Evening chronotype was associated with higher all-cause dementia risk among participants reporting infrequent social visits and with higher VaD risk among those reporting usual napping. Conclusions Evening chronotype was associated with a small increase in the hazard of all-cause dementia after multivariable adjustment. Given the modest magnitude and observational design, evening chronotype may be considered a potential risk marker rather than an independent or causal risk factor. These findings require independent replication.