DNA Methylation Clocks Outperform Telomeres for Predicting Mortality Risk
Large US study reveals epigenetic aging markers, especially GrimAge, predict death better than telomere length across racial groups.
Summary
Researchers analyzed aging biomarkers in over 4,000 Americans from three major studies (NHANES, HRS, HANDLS) using advanced Bayesian network analysis. DNA methylation-based epigenetic clocks, particularly GrimAge acceleration, proved superior to telomere length for predicting mortality risk. The study revealed significant racial and sex differences: women showed slower biological aging, while Black Americans exhibited accelerated aging through specific markers. These findings suggest epigenetic clocks offer more precise tools for assessing biological age and mortality risk than traditional telomere measurements.
Detailed Summary
Understanding biological aging has become crucial as life expectancy gaps persist across demographic groups in the US. This comprehensive study examined whether DNA methylation-based epigenetic clocks or telomere length better predict mortality risk, using sophisticated Bayesian network analysis to map complex relationships between aging biomarkers.
Researchers analyzed data from 4,113 participants across three major US cohorts: NHANES (2,522 participants), Health and Retirement Study (1,029), and HANDLS (92-470). They measured multiple epigenetic aging markers including GrimAge, HannumAge, PhenoAge, and DunedinPACE, alongside telomere length, tracking mortality through the National Death Index.
Epigenetic clocks consistently outperformed telomere length as mortality predictors. GrimAge epigenetic age acceleration emerged as the strongest predictor, with hazard ratios of 1.61 in NHANES and 1.67 in HRS for each standard deviation increase. Women showed significantly slower biological aging across multiple markers, while Non-Hispanic Black adults exhibited accelerated aging particularly through DunedinPACE, partially explaining their higher mortality risk. Hispanic adults showed unique associations with PhenoAge acceleration.
The Bayesian network approach revealed complex interconnections between age, sex, race, and biological aging markers that traditional statistical methods might miss. These networks consistently identified GrimAge and chronological age as key mortality predictors across cohorts, while highlighting demographic-specific pathways to accelerated aging.
These findings have important implications for precision medicine and health equity. Epigenetic clocks may provide more accurate biological age assessments than telomere length, potentially enabling earlier intervention for at-risk individuals. However, the study's observational design limits causal inferences, and the predominantly older, White samples may not fully represent younger or more diverse populations.
Key Findings
- GrimAge epigenetic clock predicted mortality 61-67% better than telomere length
- Women showed significantly slower biological aging across multiple epigenetic markers
- Black Americans had accelerated aging via DunedinPACE, explaining higher mortality risk
- Hispanic adults showed unique PhenoAge acceleration patterns linked to mortality
- Bayesian networks revealed complex aging pathways missed by traditional statistics
Methodology
Cross-sectional analysis of 4,113 participants from three US cohorts using additive Bayesian networks, Cox proportional hazards models, and generalized structural equation modeling. Mortality tracked through National Death Index with up to 20-year follow-up.
Study Limitations
Observational design prevents causal conclusions. Samples skewed toward older, predominantly White participants may limit generalizability to younger or more diverse populations. Some cohorts had small sample sizes for subgroup analyses.
Enjoyed this summary?
Get the latest longevity research delivered to your inbox every week.