For twenty years the standard medical advice on multivitamins has been a polite shrug. Big reviews concluded they don't prevent heart attacks, cancer, or early death in well-nourished adults, and the phrase "expensive urine" became the internet's favorite verdict. So it caught a lot of people off guard this month when a rigorous new analysis reported that a plain daily multivitamin — the kind sold in every pharmacy — measurably slowed the body's biological aging clocks.
The finding is genuine, it comes out of one of the largest supplement trials ever run, and it deserves to be taken seriously. It also deserves to be read carefully, because the gap between "slowed a biological clock in a study" and "will keep you young" is exactly where longevity headlines usually go off the rails. Here is what the researchers actually measured, what they found, and how much it really means for the bottle in your cabinet.
The study that put a number on it
The data come from COSMOS — the COcoa Supplement and Multivitamin Outcomes Study — a large randomized trial that enrolled more than 21,000 older U.S. adults and split them, by chance, into groups taking a daily multivitamin, a cocoa flavanol extract, both, or matching placebos. COSMOS was originally designed to test whether these supplements prevent heart disease and cancer. But because the researchers had banked blood samples, they could go back and ask a more modern question: did either supplement slow aging at the molecular level?[1]
To answer it, a team led by researchers at Brigham and Women's Hospital and Harvard Medical School analyzed blood from 958 participants — 482 women and 476 men — who had samples taken at the start and again two years later. Everyone in this sub-study took either the multivitamin (the widely sold Centrum Silver) or a placebo, plus either cocoa extract or its placebo, so the trial could untangle each supplement's individual effect.[1] Crucially, this was a prespecified analysis — the scientists committed to the question before looking at the results, which guards against the fishing-expedition problem that plagues so much nutrition research.
What an "epigenetic clock" actually measures
To gauge aging, the researchers didn't weigh anyone or run a treadmill test. They read DNA methylation — chemical tags that sit on top of your DNA and switch genes on and off. As you age, the pattern of these tags shifts in predictable ways, and scientists have learned to read that pattern like the rings of a tree. The algorithms that do this are called epigenetic clocks, and they estimate your biological age — how worn your cells actually are — as opposed to the number of birthdays you've had.[3]
Not all clocks are equal. The COSMOS team looked at five. Two are "first-generation" clocks (PCHannum and PCHorvath) that were trained simply to predict chronological age. The other three are "second-generation" clocks — PCPhenoAge, PCGrimAge, and DunedinPACE — built instead to predict health outcomes like disease and death, which makes them the ones geroscientists care about most.[1] GrimAge, for instance, was designed specifically to forecast lifespan and healthspan, and in validation studies it strongly predicted time to death and time to heart disease.[4] DunedinPACE, developed from a New Zealand cohort tracked for two decades, measures the speed at which someone is aging right now.[3] When a supplement moves these particular clocks, it's at least pointing in a direction that matters.
What the multivitamin actually did
Over two years, the placebo group's clocks ticked forward, as everyone's do. The multivitamin group's ticked forward a little more slowly. Compared with placebo, the daily multivitamin reduced the yearly increase in PCGrimAge by 0.11 years (95% confidence interval −0.205 to −0.020; p = 0.017) and in PCPhenoAge by 0.21 years (95% CI −0.410 to −0.019; p = 0.032).[1]
In plain terms: for each calendar year that passed, people on the multivitamin aged roughly a tenth to a fifth of a year less on these two health-predicting clocks than people on the dummy pill. The two first-generation clocks and DunedinPACE didn't budge significantly — it was specifically the outcome-focused clocks that responded. That's a modest but coherent signal, and it lines up with the same trial's earlier finding that the multivitamin slightly improved memory and global cognition in older adults, an effect the researchers estimated was equivalent to shaving about two years off cognitive aging.[2]
The people who benefited most
The most interesting wrinkle isn't the average — it's who drove it. When the team split participants by how fast they were aging at baseline, the multivitamin's effect on GrimAge was concentrated almost entirely in people who were already aging faster than their years. In that accelerated-aging group, the slowdown was clear (−0.236 years per year); in people aging at a normal or slow pace, it was essentially zero (−0.013), and the difference between the two groups was statistically significant.[1]
That pattern makes intuitive sense. A multivitamin fills nutritional gaps, and the people most likely to have gaps — and therefore the most to gain from filling them — are often those whose bodies are under more strain to begin with. It also hints at why decades of multivitamin trials in generally well-nourished populations came up flat: if the benefit lands mainly on people who are depleted or aging quickly, an average across everyone will wash much of it out.
Why cocoa flunked — and why that matters
Here's the detail that should raise your confidence in the multivitamin result: the cocoa flavanol extract, tested in the very same participants with the very same methods, did nothing to any of the five clocks.[1] That's an important internal control. If the multivitamin's effect were just noise, or an artifact of how the blood was processed, you'd expect the cocoa arm to throw off random-looking "effects" too. Instead one intervention moved specific, pre-chosen clocks and the other moved none — which is what a real, if small, biological signal tends to look like.
How big is "a little," really?
Now the honesty check, because it's the whole point. The effect sizes here are small. A tenth of a year of GrimAge slowing per year is not going to be visible in the mirror, and the researchers were careful to say so: they described the effects as "statistically significant but small" and stressed that it's still unknown whether nudging an epigenetic clock in a trial translates into actually living longer or avoiding disease.[1]
That caveat isn't boilerplate. Epigenetic clocks are powerful predictors in large populations, but no one has yet proven that deliberately slowing one with a pill changes a person's real-world fate. It's plausible — that's the entire premise of the field — but it's unproven, and a two-year change in a blood biomarker is a long way from a longer life. The study also enrolled older adults (COSMOS participants were 60 and up) who were mostly well-nourished volunteers, so the results may not carry over to younger people or to those with very different diets.
What the trial does not say is just as important as what it does. It doesn't say multivitamins reverse aging, that everyone should start taking one, or that more is better — megadoses of isolated vitamins have their own risks, and this was a single standard-dose formula. It's one well-run study measuring a surrogate marker, not a verdict.
So should you take one?
For most healthy adults eating a reasonably varied diet, a multivitamin remains what it's long been: low-cost, low-risk, and no substitute for food, sleep, or movement. What this study fairly adds is a modest thumb on the scale — a signal that a daily multivitamin may do a little more than nothing at the cellular level, especially for people who are older, nutritionally depleted, or aging faster than their age suggests. If you already take one, this is quiet reassurance rather than a reason to celebrate. If you don't, it isn't a mandate to start, but it's a more evidence-backed option than most of the pricier bottles on the longevity supplement shelf.
The bigger takeaway is about how we read longevity news at all. This is a good result precisely because it's small, controlled, and honestly reported — a real signal with real limits, not a miracle. Compared with the handful of other everyday interventions that have moved a biological clock in a proper trial, a boring daily multivitamin has quietly earned a seat at the table. Just don't expect it to turn back time.
Common questions
Do multivitamins actually slow aging?
In the COSMOS trial, a daily multivitamin modestly slowed two DNA-based "second-generation" aging clocks over two years — about 0.11 fewer years per year of PCGrimAge and 0.21 fewer years of PCPhenoAge versus placebo. The effect is real and statistically significant but small, and researchers say it's too early to know whether it translates into living longer or healthier.
Which multivitamin was used in the study?
Participants took Centrum Silver, a standard over-the-counter daily multivitamin-multimineral, once a day for two years. A separate cocoa flavanol extract tested in the same trial had no effect on any of the five aging clocks.
Who benefited most from the multivitamin?
The slowdown was concentrated in people who were already aging faster than their chronological age at the start. Those with accelerated biological aging saw a clear GrimAge benefit, while people aging at a normal or slow pace saw essentially none.