Diet score linked to slower aging on a DNA-based measure
Higher diet scores were associated with 4.17 years slower aging on a DNA-based measure in 1,711 participants and 3.77 years in 2,311 others.
Cohort study, independent replication
Limit: Correlated socioeconomic advantages could partly explain the diet–aging association
Edited by Boris Djordjevic · AgingResearch.News
In an observational study, researchers developed a food-based dietary score in 1,711 participants and tested it in an independent sample of 2,311 participants. They designed the score to predict slower epigenetic aging, estimated using chemical marks on DNA.
Higher scores were associated with slower aging on this measure in both samples. In two US cohorts, higher scores were also associated with lower mortality. Across US and UK aging cohorts, they were linked to lower risks of aging-related diseases, limitations in physical functioning and having multiple diseases. Analyses in a large UK population-based cohort also found associations with more favorable measures relating to blood vessels, inflammation, metabolism and brain structure.
Why it matters
The index may provide a framework for studying dietary strategies relevant to healthy aging.
Caveats
The observational design cannot show that the dietary pattern caused slower aging or better health. The abstract did not specify the score comparison behind the reported aging differences.
From the paper
- Participants
The current analyses included 7398 participants from the US Health and Retirement Study (HRS, mean age of 68.7 years, SD 11.0; 41% male) as the development cohort, and 23,830 from US NHANES as the external validation cohort (mean age of 62.7 years, SD 11.2; 48% male).
- Result
Conversely, legumes (β = −0.076, 95% CI −0.147, −0.005), tea (β = −0.084, 95% CI −0.149, −0.018), whole grains (β = −0.120, 95% CI −0.207, −0.032), and fruits (β = −0.126, 95% CI −0.218, −0.035) were associated with lower GrimAge2Accel.
- Limitation
Although we adjusted for a wide range of demographic, socioeconomic, lifestyle, and clinical covariates, residual confounding by unmeasured or imperfectly measured health behaviours and correlated socioeconomic advantages cannot be fully excluded.
Quoted word for word from the full text on nature.com.
The paper
First Affiliated Hospital Zhejiang University
Nature Communications, Sep 25, 2026, Open access
