One-third of American adults take a multivitamin, and roughly a quarter of kids and teens do too, according to the NIH Office of Dietary Supplements. That's not a niche habit. It's a default. And the newest attempt to settle what all that pill-popping actually buys you landed with a familiar verdict: not much, if what you're after is a longer life.
A rapid review published in Ageing Research Reviews pulled together 19 meta-analyses spanning more than 5.5 million participants — including over 333,000 pregnancies — to see what multivitamin and mineral (MVM) supplementation actually does across the research record (ScienceDirect). The headline finding tracks almost exactly with what I'd have predicted from a decade of prior trials: randomized controlled trials showed no benefit for all-cause mortality. None. The same null result held for COVID-19 outcomes and for most cognitive domains tested.
Where the Evidence Actually Points Somewhere
This isn't a review that found nothing, which is worth sitting with for a second, because "multivitamins don't do anything" is the lazy takeaway and it's not quite right. The RCT evidence in this new review did find real signals: improved global cognition, episodic memory, and immediate recall in older or cognitively intact adults, reduced psychological symptoms in healthy individuals, and lower systolic blood pressure in at-risk populations (ScienceDirect).
That's a genuinely mixed picture, and mixed is usually the honest answer in nutrition science. The problem is that none of those benefits are the ones people are implicitly buying when they reach for a bottle promising "complete daily nutrition." Nobody takes a multivitamin hoping for a few points of memory recall. They take it hoping it keeps them from dying of a heart attack or developing cancer, and on both of those specific questions, the review found nothing in the RCT data, and in fact reported a higher risk of age-related macular degeneration progression in some analyses.
The observational data tells a slightly different, messier story — associations with reduced colorectal cancer, coronary heart disease, cataracts, and fragility hip fractures — but observational associations about supplement users are notoriously confounded by the fact that people who take supplements also tend to eat better, exercise more, and see doctors more often. The review itself flags this inconsistency rather than papering over it, noting that MVM effectiveness varies substantially by population, age, and health status (ScienceDirect).
The Null Result Has Been Hiding in Plain Sight
What makes this new review useful isn't that it overturns anything — it's that it confirms, at a much larger scale, what individual landmark trials already showed. The Physicians' Health Study II followed nearly 15,000 male physicians for over 14 years and found no reduction in major cardiovascular events or cancer incidence between the multivitamin group and placebo (Health Habits). A 2018 Cochrane review reached a similarly unglamorous conclusion: routine multivitamin use doesn't convincingly extend life or prevent chronic disease in well-nourished populations (Health Habits).
Johns Hopkins researchers made the same point years earlier in an Annals of Internal Medicine editorial bluntly titled "Enough Is Enough," citing an analysis of 450,000 people that found multivitamins didn't reduce heart disease or cancer risk, a 12-year study of nearly 6,000 men showing no cognitive protection, and a trial of over 1,700 heart attack survivors where death rates were similar regardless of supplementation (Johns Hopkins Medicine). The new 2025 review doesn't contradict any of that. It just adds several million more data points to the same conclusion.
So Why Does a Third of the Country Still Take Them?
One recent PubMed-indexed commentary makes an argument worth taking seriously: maybe the trials have been measuring the wrong thing entirely. People don't report taking multivitamins to prevent a heart attack 15 years from now — they take them to feel less fatigued, more resilient, generally "covered." The paper argues that mortality and disease-incidence endpoints are methodologically misaligned with the quality-of-life outcomes consumers actually care about, and that future research should measure fatigue, cognition, and resilience directly rather than defaulting to hard endpoints (PubMed).
That's a fair critique of trial design, not a rescue of the mortality claim. If you're healthy, eating reasonably, and taking a multivitamin as insurance against an early death, the accumulated evidence — old and new — says that particular insurance policy doesn't pay out. The NIH itself doesn't recommend routine MVM use for healthy adults getting adequate nutrition from food (NIH). Whether a pill makes you feel sharper or steadier is still an open and legitimate question. Whether it extends your life is one of the more settled nulls in nutrition science — and it just got a few million more participants' worth of confirmation.
