Lifestyle factors play a critical role in healthy aging, yet their relationships with aging biomarkers remain insufficiently characterized, particularly in Asian populations. This study aimed to examine the cross-sectional and longitudinal associations between 15 modifiable lifestyle factors and two DNA methylation (DNAm) clocks (GrimAge acceleration [AgeDev] and DunedinPACE) in a cohort of older Asian adults.

We conducted a cross-sectional analysis of 631 participants (median age 70.0 years; 72.6% female) and a longitudinal analysis of 114 participants (mean follow-up 3.96 years) from the Singapore Diet and Healthy Aging (DaHA) cohort. Lifestyle exposures were assessed using validated self-administered questionnaires. Peripheral blood DNAm profiles were generated using the Illumina MethylationEPIC array. Multivariable linear regression models were applied to evaluate associations between lifestyle factors and DNAm clocks, adjusting for sociodemographic covariates, health status, and immune cell-type proportions.

In cross-sectional analyses, smoking history showed robust positive associations with accelerated epigenetic aging (GrimAge AgeDev: β = 1.45, 95% CI 1.13–1.77, p < 0.0001; DunedinPACE: β = 0.63, 95% CI 0.22–1.05, p = 0.003). Conversely, weekly physical activity was associated with slower aging (GrimAge AgeDev: β = –0.22, 95% CI –0.40 to –0.04, p = 0.02), as was daily engagement in cognitively stimulating activities (GrimAge AgeDev: β = –0.16, 95% CI –0.31 to –0.01, p = 0.04). Weekly feelings of stress were initially associated with greater GrimAge AgeDev, but this relationship was attenuated after full adjustment. No significant longitudinal associations were detected, which may reflect limited statistical power and the stability of long-standing lifestyle behaviors over the follow-up period.

These findings highlight significant cross-sectional associations between key modifiable lifestyle factors, particularly smoking, physical activity, and cognitive engagement, and epigenetic aging in an older Asian cohort. The results suggest that interventions targeting these behaviors may modulate the pace of biological aging. The absence of significant longitudinal associations underscores the need for larger prospective studies with longer follow-up and continued validation of epigenetic clocks in diverse populations to confirm these relationships over time.