Multivitamin Biological Clock Study: 4-Month Slowdown in Cellular Aging Prompts Cautious Optimism
A new multivitamin biological clock study embedded in the COSMOS randomized trial found that daily multivitamin use over two years modestly slowed measured biological ageing in older adults. The trial—using five DNA methylation “epigenetic clocks” on blood samples taken at baseline, one year and two years—reported small but statistically significant reductions on two second‑generation clocks that estimate mortality risk, while a cocoa extract arm showed no effect.
Multivitamin Biological Clock Study — background and methods
The analysis was a prespecified ancillary evaluation of the COcoa Supplement and Multivitamin Outcomes Study (COSMOS). Researchers randomized 958 generally healthy older adults (average age about 70) into four groups: daily multivitamin plus cocoa extract; multivitamin plus cocoa placebo; cocoa extract plus multivitamin placebo; or two placebos. Blood samples were collected at the start of the trial and after one and two years and were analyzed for five DNA methylation measures of biological aging: PCHannum, PCHorvath, PCPhenoAge, PCGrimAge and DunedinPACE. The multivitamin tested was a multivitamin–multimineral formulation used in the trial, and cocoa extract tested separately did not affect any of the five clocks.
Deep analysis: what the epigenetic clocks revealed
Results showed that compared with placebo, daily multivitamin supplementation modestly reduced the rate of increase of two second‑generation epigenetic clocks. The between‑group difference in yearly change was −0. 113 years for PCGrimAge (95% confidence interval −0. 205 to −0. 020; P = 0. 017) and −0. 214 years for PCPhenoAge (95% confidence interval −0. 410 to −0. 019; P = 0. 032). Over the two‑year follow‑up, investigators translated those changes to roughly four months less biological ageing in the multivitamin group than in the placebo group. The study team also found that the multivitamin effect on PCGrimAge was stronger among participants who had accelerated biological aging at baseline (between‑group change −0. 236 years) than among those with normal or decelerated baseline aging (between‑group change −0. 013 years). These patterning results suggest nutritional status or preexisting deficits may partly explain who experienced larger epigenetic shifts.
Expert perspectives and implications
Howard Sesso, associate director of preventive medicine at Brigham and Women’s Hospital and Harvard Medical School and the study’s senior author, cautioned that the findings do not imply all older adults must take a daily multivitamin. He noted that “there are no known risks for taking a multivitamin in our two large clinical trials. At the same time, we do not know for sure who benefits, and how, ” and emphasized that “ultimately, it is critical to determine the clinical relevance of our findings. “
Daniel Belsky, associate professor of epidemiology at the Robert N. Butler Columbia Aging Center, offered a methodological caveat already evident in the trial: “There is no gold standard measurement of aging, ” reflecting ongoing uncertainty about how epigenetic clock changes map onto tangible health outcomes. The trial authors likewise highlighted that while the multivitamin signal on epigenetic clocks is statistically significant but small, additional work is required to establish whether such shifts translate into meaningful reductions in age‑related disease or functional decline.
This multivitamin biological clock study sits alongside prior large randomized work with mixed outcomes. The trial team acknowledged earlier large-scale research that found daily multivitamins did not extend lifespan and in some analyses appeared associated with increased early mortality risk, underscoring the need to weigh disparate findings when considering public health guidance. Funding and material support for the COSMOS trial included contributions tied to the tested products; investigators noted commercial provision of study supplements in the trial context.
Clinically, the result is modest: only two of five clocks moved in a favorable direction, the cocoa extract arm produced no benefit on methylation measures, and the observed effect size—roughly four months over two years—is small. However, the larger effect among participants with accelerated baseline biological aging points to possible heterogeneity of benefit and the potential role of baseline nutritional status in mediating epigenetic responses.
Given those uncertainties, the multivitamin biological clock study raises a central question for geroscience and preventive medicine: can small molecular shifts measured by epigenetic clocks be linked to reduced incidence of age‑related disease or improved years of healthy life, or are they transient biomarkers without clinical translation?
