by Denkstrom
All storiesYale analysis: Lifestyle linked to lower epigenetic age

Yale analysis: Lifestyle linked to lower epigenetic age

An analysis of 51 intervention studies from Albert Higgins-Chen's laboratory at the Yale School of Medicine shows that lifestyle changes and certain medications were associated with a measurable reduction in epigenetic age, while the over-the-counter dietary supplements studied showed no effect on lowering it. Whether this also means healthier aging remains an open question, according to the researchers.

Raghav Sehgal, first author in the laboratory of Albert Higgins-Chen, an assistant professor of psychiatry at the Yale School of Medicine, compiled data from 51 intervention studies on longevity in a study published in Nature Medicine. Lifestyle changes consistently lowered epigenetic age estimated using DNA methylation; prescription medications showed the greatest reduction.

A common panel for 51 studies

For the analysis, the researchers harmonized study metadata and calculated a standardized panel of 16 epigenetic clocks for each study. They then used paired tests to compare biomarker values before and after the respective interventions. They assigned the interventions to four categories: lifestyle, including diet and exercise; medications such as metformin, semaglutide and anti-TNF therapies; dietary supplements; and medical procedures.

epigenetik

When effect sizes were compared across all clocks, only the medication and lifestyle categories significantly lowered epigenetic age. Whether these marker changes are directly caused by the respective intervention or whether other, unaccounted-for factors play a role remains unclear, according to ScienceDaily. According to the New Haven Register, Sehgal specifically identified medications for diabetes and weight management, as well as autoimmune and autoinflammatory diseases, as the drugs that had the strongest effect on epigenetic age. Two different Mediterranean diets lowered similar subsets of biomarkers in healthy groups, and anti-TNF therapies changed nearly all generation 2+ biomarkers to a similar degree in patients with arthritis or inflammatory bowel disease. According to Yale, newer clocks for determining biological age performed better than older models, and biomarkers changed more in people with illnesses than in healthy volunteers.

A comparison: Multivitamins and calorie restriction

Among participants undergoing calorie restriction in the CALERIE study, the pace of aging measured with DunedinPACE was on average 2 to 3 percent lower than in the control group, while changes in other clocks such as PhenoAge and GrimAge did not differ significantly between the groups. In the COSMOS study published in March 2026, 958 participants were assigned to four groups that received cocoa extract and a multivitamin, either of the two preparations, or two placebos daily. According to the study by Li and colleagues, compared with placebo, the multivitamin preparation reduced the annual increase in two epigenetic clocks by 0.113 years for PCGrimAge and 0.214 years for PCPhenoAge. The accompanying cocoa extract, by contrast, showed no effect on any of the five clocks examined. Both interventions thus changed individual clocks among those tested, while in the new review analysis, the medication category produced the greatest average reduction in epigenetic age across all 16 clocks. The COSMOS and DO-Health studies were not included in the Yale analysis dataset because of limited data availability, according to NutraIngredients.

altern

Where experts express doubts

A systematic review by Yi and colleagues, which evaluated 26 studies of geroscience interventions, found inconsistent agreement between changes in epigenetic clocks and clinical endpoints. They also rated the certainty of the evidence as very low throughout and recommend that such clocks currently be considered exploratory or supplementary biomarkers, rather than validated clinical surrogate endpoints. A reanalysis by Borrus and colleagues did not confirm several effects previously reported using a single clock when additional clock models were included and advised reassessing findings from individual clocks. Higgins-Chen, the analysis's senior author, himself urged caution: They had developed a systematic approach to comparing changes, but the actual significance of those changes still needed to be fully understood. Marco Di Antonio, an expert in biological aging at Imperial College London who was not involved in the multivitamin study, said the relationship between such clocks and practical aspects of aging remained unclear. Pilar Guallar Castillón, an academic at the Autonomous University of Madrid, pointed out that the clinical trial associated with the study had found no effect of multivitamin consumption on the main causes of mortality and disease. Also according to ScienceDaily, which draws on the same information from Yale, it has not yet been demonstrated whether lowering epigenetic age leads to meaningful improvements in long-term health.

Rules and criteria for a reliable finding

Sehgal envisions a future in which a blood sample taken in a clinic could indicate both overall biological age and the age of individual organs, allowing targeted interventions to be derived from the results. In his own words, such a testing option had not yet been available in August 2026. Sehgal and Higgins-Chen propose two rules for robust biomarker effects: Clocks of the same generation should respond to a similar degree and in the same direction within a study, and a second study of the same intervention should change the same biomarkers. Borrus and colleagues add three further criteria for greater confidence in a result: Reliable clocks based on principal components must respond, clocks trained on mortality or the pace of aging must respond, and multiple clocks must move in the same direction. According to the specialist website NMN.com, no single longevity intervention has yet been shown to slow aging in humans in general.