Zou · Clinical epigenetics 2025 · prospective cohort study · n=1966

Associations between five indicators of epigenetic age acceleration and all-cause and cause-specific mortality among US adults aged 50 years and older.

Cited 5 times in the scientific literature.

Level 3 - non-randomized controlled study

Prospective cohort study using linked survey and mortality registry data

PubMed 40301953 · doi:10.1186/s13148-025-01872-6 · record verified 2026-08-30

What was done

Researchers analyzed 1,966 US adults aged 50 and older from the National Health and Nutrition Examination Survey (NHANES 1999–2002) linked to the National Death Index. They assessed five DNA methylation age acceleration (EAA) metrics (HorvathAge, HannumAge, PhenoAge, GrimAge, and GrimAge2). Weighted Cox proportional hazards models adjusted for age, smoking status, chronic conditions, and other covariates were used to evaluate all-cause, cardiovascular, and non-cardiovascular mortality, with restricted cubic splines to test for non-linear relationships.

What was found

All five epigenetic clocks were significantly associated with mortality risk, with GrimAge and GrimAge2 demonstrating the strongest linear associations. Each 5-year increase in GrimAge acceleration was associated with a 44% higher risk of all-cause mortality, a 33% higher risk of cardiovascular mortality, and a 54% higher risk of non-cardiovascular mortality. Each 5-year increase in GrimAge2 acceleration was associated with a 40% higher risk of all-cause mortality, a 33% higher risk of cardiovascular mortality, and a 47% higher risk of non-cardiovascular mortality. In contrast, HorvathAge, HannumAge, and PhenoAge showed J-shaped associations with all-cause and non-cardiovascular mortality, with inflection points for all-cause mortality occurring at accelerations of 2.29, 3.07, and -7.65 years, respectively.

Why it matters

This study directly compares five prominent epigenetic clocks in a nationally representative older population, confirming that second-generation mortality-trained clocks (GrimAge and GrimAge2) exhibit superior, linear predictive performance for long-term mortality compared to earlier clocks.

Limits

The sample is restricted to US adults aged 50 and older from a 1999–2002 baseline survey. DNA methylation was measured at a single time point, precluding evaluation of longitudinal changes in epigenetic age over time. Exact confidence intervals and median follow-up duration were not reported in the abstract.

Cited by