COSMOS Study: Do Multivitamins Really Slow Ageing?

A major name, a respected specialist journal and plenty of media hype – yet the final picture is far more complicated than the headlines suggest.

In recent days, international media have circulated a claim that daily multivitamins can measurably slow ageing, at least in people aged over 70. It is based on the COSMOS project, a study whose latest findings were published in Nature Medicine. A closer look makes the story less sensational – but all the more interesting when it comes to the question: what do dietary supplements really achieve in later life?

What lies behind the COSMOS study

The research project is called COSMOS, short for “Cocoa Supplement and Multivitamin Outcomes Study”. It was led by a team at Brigham and Women’s Hospital in Boston, a highly regarded hospital with a strong focus on preventive medicine. The analysis set out to examine whether multivitamins or cocoa extracts could slow biological ageing processes.

The sub-study currently being discussed involved just under 1,000 people, split roughly evenly between women and men. All participants were at least 70 years old. Researchers randomly allocated them to four groups, each of which received different capsules daily for two years:

• a commercial multivitamin preparation plus 500 mg of cocoa powder (containing 50 mg of epicatechin, an antioxidant)
• cocoa alone plus a placebo
• a multivitamin alone plus a placebo
• a pure placebo with no active ingredient

None of the participants knew which group they had been assigned to. This gives the study a sound basic design: it was randomised, controlled and double-blind, as would be expected in drug trials.

What was measured: “epigenetic age”

Rather than looking at wrinkles, fitness or blood pressure, the researchers examined something much more refined: so-called epigenetic clocks. These are specific DNA methylation patterns – tiny chemical markers on genetic material that change over the course of life and are statistically linked quite closely with life expectancy.

The study examined five different epigenetic clocks in blood that are regarded as biological markers of ageing.

Blood samples were taken from every participant at the outset, after 12 months and again after 24 months. Using five predefined epigenetic clocks, the team assessed whether expected “biological ageing” had slowed or accelerated.

COSMOS multivitamin results in figures

After two years, a pattern emerged that prompted the headlines:

• People taking a daily multivitamin showed signs of slightly slower ageing across all five epigenetic clocks.
• Two clocks that are particularly strongly associated with mortality produced the clearest response.
• The effect was especially apparent in people who initially appeared biologically “older” than their chronological age.
• When translated into “time”, the pattern equates to an advantage of roughly four months after two years compared with the placebo group.
• This analysis found no detectable benefit from the cocoa component.

A biological age difference of four months after two years is, viewed soberly, more of a small nudge than a rejuvenation treatment. The study authors themselves make precisely this point.

Why experts urge caution

Both the Nature article and the original paper take a far more restrained tone than many media reports. Although the researchers describe the effect as statistically significant, they clearly classify it as limited.

The data suggest a possible small benefit, but they do not answer whether this results in noticeable health effects.

Put simply, the epigenetic clocks do run minimally more slowly with a multivitamin. However, this sub-study does not show whether participants will have heart attacks later, remain independent for longer or genuinely live longer.

The research team calls for further studies to clarify several issues:

• How closely are epigenetic clocks actually linked to specific illnesses in old age?
• Can changes in these markers brought about by interventions such as vitamins translate over the long term into fewer illnesses or a longer life?
• What part do lifestyle factors such as diet, exercise and smoking play compared with taking a pill?

Multivitamins versus a healthy diet – the more compelling comparison

Howard Sesso, the study lead, stresses in the accompanying Nature article that multivitamins should not be considered in isolation. His primary interest is how supplementation compares with improving someone’s overall diet.

One realistic scenario would be to have one group of older people take multivitamins, while another receives intensive support to follow a Mediterranean-style diet rich in vegetables, wholegrains, pulses and high-quality oils. Whether multivitamins would still offer any identifiable additional benefit in that setting remains entirely unclear.

The limitations of epigenetic clocks

The authors also openly address another issue: epigenetic clocks are considered promising markers, but they have not yet been conclusively validated by science. They correlate with age and mortality risk, yet the underlying mechanisms are understood only in fragments.

Among the gaps that currently remain are:

• clear thresholds for determining when a change is genuinely medically relevant
• long-term datasets showing how interventions affecting these markers influence the course of disease
• comparisons with established measures such as fitness, muscle strength or cognitive performance

Anyone reading their “epigenetic age” on a laboratory report is therefore not holding a health score equivalent to an MOT certificate, but rather a risk indicator that is still being developed.

Who pays has a voice: sponsorship behind the COSMOS study

As is often the case, a sensitive aspect appears towards the bottom of the publication: the funders. Alongside public bodies such as the National Institutes of Health, several industry-linked organisations are listed.

The researchers state that they worked independently and did not permit sponsors to influence the study. Formally, the research is therefore considered free from conflicts of interest. Even so, an uncomfortable question remains: would the result have looked different if only neutral public funders had stood behind it?

The Council for Responsible Nutrition in particular has repeatedly faced criticism in the past over indirect influence, including funded travel, fees or other benefits for researchers. Such links do not automatically make a study worthless, but they do require especially critical reading – particularly when a commercially attractive product appears to come out ahead.

What does this mean for older people?

Anyone aged over 70 who takes several medicines and sees advertising promises about “anti-ageing vitamins” will understandably ask: is it worthwhile? The COSMOS findings do not provide a straightforward answer.

Nevertheless, some practical points can be drawn from them:

• Multivitamins can help to address specific micronutrient gaps, for example in people with a very limited diet or poor appetite.
• The observed slowing of epigenetic clocks is small and cannot currently be directly translated into specific health benefits.
• A balanced, fibre-rich diet, physical activity and stopping smoking have demonstrably much stronger effects on illness and life expectancy.
• Anyone with existing conditions or taking numerous medicines should discuss every supplement with their doctor because of possible interactions.

Multivitamins are more of a supplement than a miracle cure – they do not replace a good diet or an active lifestyle.

Risks and misconceptions

Many people assume vitamins are harmless in any quantity because they can be bought without a prescription. That is not the case. Fat-soluble vitamins, especially A, D, E and K, can build up in the body. Depending on existing health conditions, high-dose products may create risks such as kidney stones or liver problems.

There is also a psychological effect: people who rely on a pill may postpone other, more demanding changes. An evening walk, drinking less alcohol or eating one portion of vegetables each day does considerably more for blood vessels, the brain and metabolism than reaching for a box of tablets.

How to assess health stories more effectively

The COSMOS research is a clear example of how complex data can be turned too quickly into simple promises. Several questions can help when evaluating health news more generally:

• Does it concern hard outcomes such as heart attacks and deaths, or surrogate markers such as blood values or epigenetic clocks?
• How large is the observed effect in real-world terms, such as months or percentages, rather than solely statistically?
• Who funded the study, and does the funder have a commercial interest in its outcome?
• How do the authors themselves describe the findings – cautiously or in promotional language?

In the case of multivitamins, the honest answer at present is this: there are interesting signs of small laboratory effects. Whether these give older people a measurably longer period of good-quality life in everyday life remains open. Anyone wanting to improve their chances of healthy ageing is better served by the established basics: good nutrition, exercise and social contact – as well as specifically prescribed products where needed, rather than blindly choosing high-dose combination supplements.