A Daily Multivitamin for $0.11 Slowed Brain Aging by Two Years — But the Fine Print Matters

Three successive studies from one of the largest supplement trials ever conducted suggest that an over-the-counter multivitamin may protect the aging brain. The claims are specific, the data is real, and the price is negligible. But so are the effect sizes — and the conflicts of interest are not.

The Supplement: Centrum Silver at 11 Cents a Day

The supplement at the center of this research is Centrum Silver, a standard multivitamin-mineral (MVM) tablet formulated for adults over 50 and containing more than 20 micronutrients including vitamins D3, B12, E, zinc, and iron [1]. At major U.S. retailers, a bottle of 200 tablets sells for roughly $15–$22, placing the per-tablet cost at approximately $0.08 to $0.11 per day [2].

That price point stands in stark contrast to existing FDA-approved cognitive treatments. Generic donepezil (Aricept), the most commonly prescribed cholinesterase inhibitor for Alzheimer's disease, costs $8–$38 per month for a 30-day supply depending on dosage, or up to $170 without insurance [3]. Brand-name Aricept runs approximately $660 per month [3]. Memantine, the other major Alzheimer's drug, carries a similar price range for generic formulations. The multivitamin costs roughly $3.30 per month — less than 10% of the cheapest generic Alzheimer's prescription.

The Study: COSMOS and Its Three Cognitive Sub-Trials

The evidence comes from the COcoa Supplement and Multivitamin Outcomes Study (COSMOS), a randomized, double-blind, placebo-controlled trial run by researchers at Brigham and Women's Hospital (a Harvard Medical School affiliate) in collaboration with Columbia University and Wake Forest University [4]. The parent trial enrolled 21,442 participants — women aged 65 and older and men aged 60 and older — across the United States, with a median follow-up of 3.6 years [4].

Within COSMOS, three non-overlapping cognitive sub-studies tested whether a daily Centrum Silver tablet could slow cognitive decline:

• COSMOS-Mind (n = 2,158): telephone-based cognitive assessments [5]
• COSMOS-Web (n = 3,562): internet-based neuropsychological testing over three years [6]
• COSMOS-Clinic (n = 573): detailed in-person cognitive evaluations [7]

In the COSMOS-Web study, participants assigned to the multivitamin showed significantly better immediate recall on the Modified Rey Auditory Verbal Learning Test (ModRey) at one year — the primary endpoint — as well as across three years of follow-up on average. The researchers estimated the multivitamin improved memory performance above placebo by the equivalent of 3.1 years of age-related memory change [6].

A meta-analysis pooling all three sub-studies (over 5,000 participants) found statistically significant benefits for both global cognition (mean difference: 0.07 standard units; 95% CI: 0.03–0.11; P = 0.0009) and episodic memory (mean difference: 0.06 standard units; 95% CI: 0.03–0.10; P = 0.0007) [7]. The overall magnitude of effect on global cognition was estimated as equivalent to slowing cognitive aging by approximately two years [1].

The benefits appeared consistent across age, sex, race, ethnic group, weight, diet quality, and baseline cognitive ability [8].

Source: OpenAlex

Data as of Jan 1, 2026CSV

How Does This Compare to Alzheimer's Drugs?

The comparison with prescription drugs is instructive — but not straightforward. Donepezil, tested in people with diagnosed Alzheimer's disease, shows a mean improvement of 2.67 points on the ADAS-Cog scale (a 70-point test of cognitive function) and 1.05 points on the MMSE (a 30-point screening tool) after 26 weeks [9]. These are considered modest but clinically meaningful effects in a symptomatic population.

The COSMOS multivitamin results were measured in cognitively normal older adults using different scales — standardized composite scores rather than disease-specific instruments. The 0.07 standard-unit difference in global cognition is a small effect size by conventional standards (where 0.2 is considered "small" and 0.5 "medium" on Cohen's d). Direct comparison is difficult because the populations, endpoints, and goals differ: donepezil treats existing disease; the multivitamin appears to slow normal age-related decline [9][7].

Source: Multiple clinical trials (compiled)

Data as of Jan 1, 2025CSV

Who Paid for the Research

COSMOS was funded through a mixture of public and private sources. The National Institutes of Health provided grants, including NIA award R01AG050657 for the COSMOS-Mind sub-study [4]. But significant industry support also flowed in: Mars Edge, a nutrition research division of the candy and pet food company Mars Inc., provided an investigator-initiated grant that included infrastructure support and donation of study pills [10]. Pfizer Consumer Healthcare — the manufacturer of Centrum Silver — provided partial provision of study pills and packaging [10].

The lead investigators, Drs. Howard Sesso and JoAnn Manson at Brigham and Women's Hospital, reported receiving investigator-initiated grants from Mars Edge and Pfizer during the study. One investigator also received honoraria from the Council for Responsible Nutrition, a supplement industry trade group [10].

Marion Nestle, a prominent food policy researcher at New York University, noted the funding structure with cautious interest. She observed that the cardiovascular and cancer findings from COSMOS showed no benefit from either cocoa extract or multivitamins — results that contradicted industry hopes — which she viewed as a sign of scientific integrity despite the funding entanglements [11]. An independent Data and Safety Monitoring Board reviewed health outcomes annually [12].

Still, no group outside the COSMOS investigators has independently replicated the cognitive findings. All three positive sub-studies come from the same parent trial, share the same randomization, and involve the same research team.

The Proposed Biological Mechanism

The researchers have not identified a single mechanism driving the cognitive benefit, and this remains one of the study's open questions [8]. Several plausible pathways exist in the scientific literature:

Micronutrient deficiencies become increasingly common with age due to reduced dietary intake, impaired nutrient absorption, and medication interactions [13]. The brain depends on active transport systems — including the sodium-dependent multivitamin transporter (SMVT) — to move nutrients across the blood-brain barrier [14]. B vitamins (particularly B12, B6, and folate) are involved in homocysteine metabolism; elevated homocysteine is associated with accelerated brain atrophy. Vitamin D receptors are present throughout the brain. Zinc and vitamin E play roles in antioxidant defense against oxidative stress, a contributor to neurodegeneration [13].

A broad-spectrum multivitamin may work not through any single pathway but by correcting subclinical deficiencies across multiple systems simultaneously — reducing neuroinflammation, supporting mitochondrial function, and maintaining neurotransmitter synthesis [13][14]. This hypothesis is consistent with the finding that cocoa flavanols alone (tested in the same COSMOS trial at 500 mg/day) did not improve cognition overall, though they did show benefit in participants with lower baseline dietary flavanol intake [4].

The mechanism remains speculative. As Harvard's Dr. Anthony Komaroff noted, researchers still do not know "which of the over 20 vitamins and minerals in the pill are responsible" for the observed effect [8].

The Case for Skepticism

The history of supplement-brain research is littered with promising early results that collapsed under rigorous follow-up. Every major candidate has failed:

Ginkgo biloba, once the most popular brain health supplement in the world, was tested in the Ginkgo Evaluation of Memory (GEM) trial — a randomized, placebo-controlled study of more than 3,000 older adults. It showed no effect on reducing dementia incidence [15].

Vitamin E was examined in a 2017 Cochrane systematic review that found no evidence it prevented progression from mild cognitive impairment to dementia [15].

Omega-3 fatty acids have been tested in multiple high-quality trials, including the ASCEND trial in diabetic patients. Several systematic reviews found no convincing evidence that omega-3 supplements benefit cognitive function [15].

Dr. Pieter Cohen, a Harvard-affiliated physician specializing in supplement safety, has stated bluntly: "Nothing legally contained in supplements has been proven to improve your thinking or prevent memory loss" [16].

Specific methodological concerns apply to COSMOS:

• Effect sizes are small. A 0.07 standard-unit difference, while statistically significant with large sample sizes, may not translate to noticeable real-world cognitive improvement for individual patients.
• Follow-up is limited. Two to three years is short for a condition that develops over decades. Whether the effect persists, grows, or disappears beyond year three is unknown.
• Surrogate endpoints. The studies measured scores on cognitive tests, not clinical outcomes like dementia diagnosis, need for assisted living, or functional independence.
• Self-selected population. COSMOS participants who volunteered for cognitive sub-studies may differ systematically from the general population of older adults — the "healthy user" bias that has inflated supplement benefits in prior observational studies.
• No independent replication. All positive results come from one trial infrastructure, one randomization, and one research group [7][8].

The Prebiotic Fiber Story: A Parallel Finding

The multivitamin is not the only cheap supplement generating cognitive headlines. A 2024 study published in Nature Communications tested inulin and fructooligosaccharide (FOS) — inexpensive prebiotic plant fibers costing pennies per day — in 36 pairs of twins over age 60 [17]. Over 12 weeks, the fiber group scored better on the Paired Associates Learning test (an early marker associated with Alzheimer's disease) and on measures of reaction time and processing speed [17].

The proposed mechanism runs through the gut-brain axis: the prebiotics increased populations of Bifidobacterium, a genus of gut bacteria that mouse studies suggest can reduce cognitive deficits by modulating gut-brain signaling [17]. The study was small — 72 participants total — and the twin design, while elegantly controlling for genetics, limits statistical power. It has not been replicated.

The Target Population and Public Health Stakes

The demographics are enormous. In the United States, approximately 10% of adults aged 65 and older have dementia, and another 22% have mild cognitive impairment (MCI), according to a 2022 study published in JAMA Neurology [18]. Globally, the prevalence of MCI in adults over 60 ranges from 6.7% to 25.2%, with rates increasing sharply with age — from 3% of people aged 65–69 to 35% of those over 90 [19].

The global population aged 60 and older is projected to reach 2.1 billion by 2050 [19]. Even in the United States alone, more than 55 million adults are over 65.

Source: World Bank Open Data

Data as of Dec 31, 2023CSV

If even a fraction of this population adopted a daily multivitamin based on the COSMOS findings, the public health arithmetic is striking. At $3.30 per month per person, 10 million adopters would cost $396 million annually — a rounding error compared to the estimated $345 billion the U.S. spends yearly on Alzheimer's and dementia care [18]. If the two-year cognitive aging delay is real, deferred institutionalization and reduced caregiver burden could yield substantial savings.

But potential harms exist. The supplement industry is largely unregulated for quality. The FDA does not test supplements before they reach shelves, and ingredient accuracy is not guaranteed [20]. Drug interactions are a concern for older adults taking multiple medications: vitamin K in multivitamins can interfere with warfarin; iron can reduce absorption of thyroid medications and certain antibiotics; calcium can interact with bisphosphonates [13]. Mass adoption without medical guidance could produce adverse effects that offset any cognitive benefit.

Regulatory and Commercial Stakes

Centrum Silver is already sold over the counter at every major U.S. pharmacy and retailer. Following the COSMOS publications, the brand's marketing now prominently features the study: Centrum's website describes itself as "clinically shown to support cognitive health" [2]. This is a structure/function claim — the only type the FDA permits for supplements — and must be accompanied by the disclaimer: "This statement has not been evaluated by the FDA. This product is not intended to diagnose, treat, cure, or prevent any disease" [20].

For Centrum's parent company (now Haleon, spun off from GSK's consumer health division, which acquired Pfizer's consumer health business), the commercial incentive is obvious. The global brain health supplement market exceeds $8 billion annually.

An FDA-approved health claim — which would allow Centrum to state that multivitamins reduce the risk of cognitive decline — would require a much higher evidentiary bar: "significant scientific agreement" among qualified experts, based on the totality of publicly available evidence [20]. The current COSMOS data, while the strongest evidence to date for any supplement's cognitive benefit, falls short: the effect sizes are small, the follow-up is limited, the endpoints are surrogate, and independent replication is absent.

In December 2025, the FDA considered relaxing label disclaimer requirements for supplement structure/function claims. Dr. Pieter Cohen called this "a very important step in the wrong direction," arguing it would make it harder for consumers to distinguish between evidence-backed and evidence-free products [16].

Source: OpenAlex

Data as of Jan 1, 2026CSV

What the Evidence Actually Supports

The COSMOS findings are not nothing. Three pre-registered, randomized, placebo-controlled sub-studies with over 5,000 participants, conducted by credible researchers at major academic institutions, found a consistent small benefit from a daily multivitamin on cognitive test scores over two to three years. The meta-analytic P-values are well below conventional significance thresholds [7].

But "statistically significant" and "clinically meaningful" are different things. A 0.07 standard-unit improvement in global cognition is real in a population average but invisible to any individual. The equivalent-of-two-years framing, while accurate as a statistical rescaling, risks overpromising: it does not mean a 75-year-old will think like a 73-year-old.

The most honest summary: for older adults who are not already taking a multivitamin, Centrum Silver at $0.11 per day is cheap, appears safe over two to three years, and may provide a small cognitive hedge — particularly for those with suboptimal diets. It is not a treatment for cognitive impairment, not a substitute for exercise and diet, and not yet proven to prevent dementia. The research is promising, industry-entangled, and unreplicated. That combination warrants attention, not enthusiasm.