Twenty-Four-Hour Rest-Activity Rhythms and Epigenetic Age Acceleration in Middle-Aged and Older Adults

Circadian rest-activity rhythms (RARs) are increasingly viewed as aging-relevant behavioral phenotypes, but links to DNA methylation-based epigenetic age acceleration (EAA) are not well defined.

To evaluate associations between multidimensional actigraphy-derived 24-hour RAR metrics (strength and robustness, regularity and fragmentation, and timing) and EAA across 4 epigenetic clocks.

This cross-sectional study used data from the Baltimore Epidemiologic Catchment Area Study, wave 5 (2017-2022). The cohort included community-dwelling adults in eastern Baltimore, Maryland, with 7-day wrist actigraphy and blood DNA methylation data. Analyses were conducted from January to December 2025.

Seven-day wrist actigraphy-derived parametric and nonparametric RAR metrics, including rhythm strength and robustness (amplitude, mesor, relative amplitude, and mean activity during the least active continuous 5-hour period [L5] and most active continuous 10-hour period [M10]), regularity and fragmentation (interdaily stability [IS] and intradaily variability [IV]), timing (acrophase, L5 and M10 start times, and rest and sleep midpoints), and day-to-day timing variability (rest midpoint variability [RMV] and sleep midpoint variability [SMV]).

Associations between EAA for 4 clocks (Horvath, Hannum, PhenoAge, and GrimAge) and each RAR metric, evaluated using linear regression models.

A total of 207 participants were included (mean [SD] age, 68.42 [7.48] years; 135 [65.2%] female). In fully adjusted models, each 1-SD increase in amplitude, relative amplitude, and IS was associated with lower GrimAge acceleration (amplitude: β, -0.67 [95% CI, -1.24 to -0.09]; P = .02; relative amplitude: β, -0.77 [95% CI, -1.31 to -0.23]; P = .005; IS: β, -0.65 [95% CI, -1.20 to -0.09]; P = .02). Each 1-SD increase in amplitude and IS was also associated with lower PhenoAge acceleration (amplitude: β, -1.20 [95% CI, -2.26 to -0.15]; P = .03; IS: -1.35 [95% CI, -2.36 to -0.34]; P = .009). Higher L5 activity and IV were associated with higher GrimAge acceleration (L5: β, 0.60 [95% CI, 0.06-1.13]; P = .03; IV: β, 0.72 [95% CI, 0.17-1.28]; P = .01). There were nonsignificant but directionally consistent outcomes for the Horvath and Hannum clocks. RAR-EAA effect sizes were generally larger in women than in men (amplitude, mesor, M10, RMV, and SMV; P < .05 for interaction) and in White participants compared with those in other race and ethnicity categories (mesor, IV, and M10; P < .05 for interaction); associations for rhythm-timing metrics varied with age; and Shapley additive explanations analyses identified IS, RMV, and amplitude as 3 of the most influential RAR factors in epigenetic age acceleration.

In this exploratory cross-sectional study, higher strength and stability of circadian RARs were associated with lower acceleration of epigenetic age (particularly GrimAge and PhenoAge), whereas greater irregularity and more variable activity timing were associated with greater acceleration, supporting the interpretation of RARs as markers of biological aging.