GrimAge Outperforms Other Epigenetic Clocks in the Prediction of Age-Related Clinical Phenotypes and All-Cause Mortality.
Level 3 - non-randomized controlled study
Prospective longitudinal cohort study (TILDA) with up to 10-year follow-up
PubMed 33211845 · doi:10.1093/gerona/glaa286
What was done
Assessed the predictive performance of four DNA methylation epigenetic clocks (Horvath, Hannum, PhenoAge, and GrimAge age acceleration [AA]) against nine clinical and cognitive outcomes: walking speed, grip strength, Fried frailty phenotype, polypharmacy, Mini-Mental State Examination (MMSE), Montreal Cognitive Assessment (MoCA), Sustained Attention Reaction Time, 2-choice reaction time, and all-cause mortality over up to 10 years of follow-up. The analysis evaluated 490 participants from The Irish Longitudinal Study on Ageing (TILDA) using minimally adjusted and fully adjusted (controlling for social and lifestyle factors) regression models.
What was found
The abstract reports direction and statistical significance of associations without quantitative effect sizes (e.g., hazard ratios or beta coefficients): - HorvathAA and HannumAA showed no significant predictive utility for the tested health outcomes. - PhenoAgeAA was associated with 4 of 9 outcomes (walking speed, frailty, MoCA, MMSE) in minimally adjusted models, but lost significance after adjustment for social and lifestyle factors. - GrimAgeAA was associated with 8 of 9 outcomes (all except grip strength) in minimally adjusted models, and retained significant predictive power for walking speed, polypharmacy, frailty, and all-cause mortality in fully adjusted models.
Why it matters
Among commonly used DNA methylation clocks, GrimAge demonstrated superior utility for capturing phenotypic physiological decline and long-term mortality risk beyond traditional lifestyle and social covariates.
Limits
The abstract provides no exact numerical effect estimates, confidence intervals, or p-values. The sample size is relatively small (n = 490) and restricted to a single demographic cohort (TILDA in Ireland). As an observational cohort study, residual confounding remains possible and findings do not establish a causal mechanism between DNA methylation changes and functional decline.
Cited by
- context Epigenetic clocks have a correlation of approximately 0.5 with one another after regressing out chronological age, sex, and other covariates.