Centrum Benefits and Side Effects: Does Multivitamin Use Extend Lifespan? (2026 Clinical Trial Evidence)

Originally written in Korean and translated for international readers.

One in three US adults takes a multivitamin^[1], and it's probably much the same wherever you live. Yet ask people why they take it, and most will say something like, "Well… because it's good for you?"

Is there a real benefit here, or are we just spending money on expensive urine? I finally decided to settle the question for myself. I went straight to the large clinical trials and meta-analyses published between 2024 and 2026, and what they add up to turned out to be more interesting than I'd expected.

A study that followed 390,000 people for 20 years: it won't make you live longer

The first thing I came across was a 2024 study from the NIH's National Cancer Institute, and the scale is staggering: it tracked 390,124 healthy adults across three large US cohorts for more than 20 years^[1].

The result? There was no meaningful difference in lifespan between the people who took a multivitamin every day and those who didn't^[1]. The odds of dying from heart disease, cancer, or cerebrovascular disease were much the same across the board. If anything, overall mortality came out 4% higher in the daily-use group (first half of follow-up HR 1.04, 95% CI 1.02–1.07; second half HR 1.04, 95% CI 0.99–1.08) — but it would be a mistake to read that number as "multivitamins are harmful." This was an observational study from the start, with no randomization and no placebo: people decide for themselves whether to take a multivitamin, which brings a well-known trap with it. People whose health has begun to slip tend to start taking a multivitamin, figuring "I should at least do this much," so the so-called "sick user effect" can manufacture a small positive association in observational data through reverse causation and residual confounding^[1]. Sure enough, in the second half of follow-up the HR for the largest cohort attenuated and its 95% confidence interval included 1, so the statistical significance vanished^[1]. The authors' own conclusion is "no evidence of lifespan extension," not "harmful."

In 2026 came a meta-analytic review that painted an even bigger picture. A health-longevity team at the National University of Singapore pulled together 19 meta-analyses published over the past 25 years; tally up the various health outcomes and the participants number more than 5.5 million. Looking at mortality on its own, it reached the same verdict — multivitamins have no effect on all-cause mortality^[2].

If you stopped reading here, you'd conclude there's no reason to bother. But the story doesn't end here.

It doesn't extend lifespan — but it slows the pace of aging?

In March 2026, a rather intriguing paper landed in Nature Medicine. From a team at Harvard Medical School and Mass General Brigham, it has an eye-catching premise: taking a multivitamin every day modestly slows the pace of biological aging^[3].

This is an ancillary study of a large clinical trial called COSMOS, so COSMOS itself is worth a quick word.

COSMOS (COcoa Supplement and Multivitamin Outcomes Study) is a randomized controlled trial (RCT) that tested a cocoa extract and a multivitamin at the same time in 21,442 US adults (women 65 and older, men 60 and older). In plain terms: half the participants got a real multivitamin and the other half an identical-looking dummy pill, and with no one knowing who was taking what, they were followed for more than three years. This is the design widely regarded as the gold standard of clinical trials.

The multivitamin in question was Centrum Silver; Pfizer (now Haleon) donated both the product and the placebo.

For 958 of the participants, researchers analyzed DNA methylation patterns in the blood. DNA methylation shifts in characteristic ways as we age, and tools called epigenetic clocks use those shifts to estimate "biological age" — a readout, separate from your chronological age, of how old your cells actually are.

The results were striking. The first-generation clocks (which track chronological age) showed no difference, but on the second-generation clocks — which more strongly predict future death and disease — the group that took Centrum Silver for two years aged at a statistically meaningfully slower rate.

More intriguing still, the effect was larger in people who were already biologically older at baseline — those whose cellular age was running ahead of their chronological age. In this group, PCGrimAge slowed by 0.24 years per year (95% CI −0.38 to −0.09), whereas in those who were already aging well there was essentially no difference (−0.01, 95% CI −0.13 to +0.10; P for interaction = 0.018). It fits the familiar pattern: the benefit is greatest when you're filling a gap in someone who's running short. Still, even though this subgroup (interaction) analysis was prespecified, the effect is small and needs further confirmation, so it's too early to call.

The memory data is the most convincing of all

In fact, the biggest talking point to come out of COSMOS wasn't aging but cognition. That work was split across three ancillary studies, each asking the same question a different way: can a multivitamin slow the aging of the brain?

First, in 2022, came COSMOS-Mind (2,262 people, cognitive assessment by telephone, 3 years). The multivitamin group scored significantly better on cognition than placebo, and the researchers put the effect at roughly the equivalent of slowing cognitive aging by about 1.8 years^[4].

COSMOS-Web (3,562 people, online neuropsychological testing) followed in 2023. Here memory in particular improved significantly, with the effect estimated at about a 3.1-year delay in memory aging^[5].

Then, in 2024, COSMOS-Clinic (573 people, in-person neuropsychological assessment) rounded out the set, making a meta-analysis of all three possible. Pooling more than 5,000 non-overlapping participants, the multivitamin group came out significantly ahead on both overall cognition (p=0.0009) and memory (p=0.0007). The researchers' pooled estimate was a delay in cognitive aging of about 2 years^[6].

The crux is this: the same COSMOS trial was assessed three different ways — telephone, online, in-person — and consistently pointed the same direction. A finding from a single method could be a fluke; a finding that holds up when you change the method is a different proposition. In some analyses the effect was larger in participants with a history of cardiovascular disease (an exploratory result that didn't hold across all studies), one reading being that this is a group more likely to be nutritionally deficient.

So does it help prevent disease?

By the lights of the 2026 meta-analysis, the areas where multivitamins help and where they don't split fairly cleanly^[2].

On the "yes" side, RCTs confirmed better cognition in older adults, lower blood pressure in people with hypertension, and fewer infections in younger adults (under 65). Bring observational studies into the mix and you can add a lower risk of colorectal cancer and fewer cases of coronary heart disease.

On the "no" side, there was no effect on all-cause mortality, breast cancer, or prostate cancer — and no effect on blood pressure in people whose blood pressure was already normal.

A pattern comes into focus. Multivitamins help the person who's short on something, and make little difference to the person who already has enough. It sounds obvious, but it carries real weight to have it confirmed across data from millions of people.

OK, but what about people in their 30s and 40s? Here it gets murky

Every COSMOS study so far has been in adults 60 and older. "I'm 35 — will Centrum do anything for me?" Honestly, there are almost no large randomized controlled trials (RCTs) that can answer that.

Why not? The reason is practical. Healthy people in their 30s decline slowly and rarely get sick, so to pick up a multivitamin effect statistically you'd have to follow thousands of them for more than a decade. The cost is enormous — and if you're going to spend that kind of money, studying older adults is far more efficient.

There are, however, a handful of small RCTs in younger adults, all of them double-blind and placebo-controlled — properly designed, in other words.

Kennedy et al. (2010) gave 215 healthy male office workers aged 30–55 a high-dose B complex plus vitamin C and minerals (Berocca) for 33 days. Stress (PSS) fell, vigour (POMS vigour) rose, and general-health (GHQ-12) scores improved^[7].

Pipingas et al. (2013) followed 138 adults aged 20–50 for 16 weeks. The intriguing wrinkle here is that mood didn't change significantly when measured in the lab — yet when participants logged their mood at home on a smartphone, stress, physical fatigue, and anxiety all came down. The researchers themselves were left scratching their heads over the discrepancy: was the lab setting too artificial to register the effect, or was the at-home self-report colored by expectation bias?^[8]

White et al. (2015) ran 58 adults aged 18–40 for 4 weeks. Short as the window was, blood homocysteine dropped (p = 0.01) and the "depression-dejection" score improved (p=0.018). A blood marker actually moving points to a real physiological effect — but in this very study the homocysteine change and the mood change weren't directly linked, so it's hard to be sure the two are causally connected^[9].

There's also a meta-analysis that bundles these and other trials — 8 RCTs in healthy adults all told. The synthesis by Long & Benton (2013) runs like this: statistically meaningful improvement was confirmed across all five domains of stress, psychiatric symptoms (overall), anxiety, fatigue, and confusion. Depression, however, showed no effect^[10].

That said, don't read these results too generously

This is the point to step back and look at it all with a cooler eye.

To begin with, the effects are small. The SMDs (standardized mean differences) from the meta-analysis sit squarely in the "small effect" bracket. Put plainly, it comes down to agreeing a little more strongly with "my stress went down" on a questionnaire. It may not even rival the mood lift of a 30-minute walk.

The sample sizes are a problem too. 58 people, 138, 215 — nothing next to COSMOS's 21,442. The fewer the participants, the higher the odds of a positive result turning up by chance; statisticians call this a lack of statistical power.

And as the Pipingas study laid bare, an effect that shows up only in at-home self-report and never in the lab is tricky to interpret. The design is double-blind, so participants aren't supposed to know what they're taking — but if the vitamin B2 turns your urine fluorescent yellow, it's not hard to guess: "Ah, I really am on the vitamin." You can't fully rule out that this fed an expectation bias.

Above all, some studies found almost no cognitive benefit in healthy people whose nutritional status was already good. Which is to say a multivitamin doesn't push brain function "above normal" so much as it makes up a shortfall and brings it back "to normal."

And, decisively — the long-term data that could answer "If I start Centrum in my 30s, will I be healthier in my 60s?" simply doesn't exist yet, anywhere.

So — should you take it or not?

Having read through the papers, here's where I land.

A multivitamin is not an "all-purpose health pill," and it won't make you live longer. But it's hard to call it "useless," either. In older adults especially, the protection of cognitive function and the slowing of biological aging rest on fairly solid RCT evidence^[3–6] [3] Li S, et al. (2026). Effects of daily multivitamin–multimineral and cocoa extract supplementation on epigenetic aging clocks in the COSMOS randomized clinical trial. Nature Medicine.[4] Baker LD, et al. (2022). Effects of cocoa extract and a multivitamin on cognitive function: A randomized clinical trial. Alzheimer's & Dementia.[5] Yeung LK, et al. (2023). Multivitamin Supplementation Improves Memory in Older Adults: A Randomized Clinical Trial. American Journal of Clinical Nutrition.[6] Vyas CM, et al. (2024). Effect of multivitamin-mineral supplementation versus placebo on cognitive function (COSMOS Clinic subcohort and meta-analysis). American Journal of Clinical Nutrition..

Broken down by age, it shakes out like this.

If you're 65 or older, the case for a multivitamin is clearest. In COSMOS, Centrum Silver delivered a roughly 2-year delay in cognitive aging and a confirmed slowing of aging on second-generation epigenetic clocks. For this age group, Centrum Silver is a multivitamin backed by an unusually rich body of cognitive and aging evidence, validated head-on by a large RCT (COSMOS). For vegetarians and anyone who's been on metformin or acid-suppressing drugs long-term, the risk of B12 deficiency is higher, which makes the case stronger still.

If you're in your 30s or 40s, the evidence to say "you must take it" honestly isn't there. But if your diet is irregular, you're dieting, or you're under heavy work stress, you might see a modest easing of stress and fatigue. Think of it as a kind of "nutrition insurance." Even so, a single pill can never stand in for regular meals, 30 minutes of exercise, and 7 hours of sleep.

If you do decide to take it, take it with a meal — or right after one. The fat-soluble vitamins (A, D, E, K) are absorbed well alongside food, and taking it consistently at the same time each day matters more than which time you pick. And if you're on metformin, an acid-suppressing drug, or Parkinson's medication (levodopa), check with your doctor first.

One last thing. The fantasy of a "pill that lets you live to 100" has no data behind it. But the opposite cynicism — that it's all just expensive urine — is too sweeping in the face of the cognitive-protection and aging-slowing data in people over 60. A multivitamin sits somewhere in between, doing a small job for particular ages and particular circumstances. That, for now, is the whole of what the science tells us.

Cautions / Who can safely start

• Pregnant or breastfeeding women, children, and adolescents: use a dedicated prenatal or pediatric product rather than a general multivitamin.
• People with reduced kidney function: vitamin A, vitamin D, and minerals can accumulate, so check with your physician first.
• People taking warfarin: be sure to check the vitamin K content.
• People taking metformin, acid-suppressing drugs (PPIs), or levodopa: possible effects on B12 and iron absorption, plus drug interactions — decide in consultation with your doctor.

This article is general information, not individual medical advice. Talk to a specialist about your own health and the medications you take.