Does a daily multivitamin help protect memory in older age?
A daily multivitamin appears to be able to protect memory in older people who are at risk of deficiencies or have a history of cardiovascular disease, but for well-nourished older adults the benefit has not been convincingly demonstrated.
Three large randomised trials, summarised in a meta-analysis of more than 5,200 participants, show that daily multivitamin supplementation improves episodic memory and overall cognition in older adults compared with placebo. The effect on episodic memory is roughly equivalent to two to three years less memory decline. That is modest but measurable.
Not everyone benefits equally. In the COSMOS-Mind study, the benefit was greatest in people with a history of cardiovascular disease. A separate, long-running study in older male physicians (followed for up to 12 years) found no effect whatsoever on memory or cognition. The researchers attributed this to the specific population: well-nourished, highly educated, and probably without any meaningful deficiencies. People who already get sufficient vitamins through their diet likely have less to gain.
A smaller study of B vitamins (B6, B12 and folic acid) in older adults with mild memory complaints and elevated homocysteine levels showed positive signals: memory improved, homocysteine fell and depressive symptoms decreased after 12 weeks. Because that study was not randomised and included only 48 participants, no firm conclusion can be drawn from it, but it does give reason to think that people with nutritional deficiencies stand to benefit more.
A randomised trial in Scotland with 910 participants aged 65 and older found no effect on memory or verbal fluency in the overall group after 12 months. There was a weak, non-significant signal of benefit in the oldest participants (75+) and in those at elevated risk of deficiencies. Finally, cocoa extract (flavanols) was found to have no effect on cognition in the COSMOS studies.
In summary, a daily multivitamin appears to be most beneficial for older people who are at risk of nutritional deficiencies or who have had cardiovascular problems. For well-nourished, healthy older adults, the benefits have not yet been demonstrated in large-scale research.
The answer rests primarily on three randomised trials from the COSMOS programme, summarised in a meta-analysis of more than 5,200 participants. Those trials were conducted independently and show consistently positive effects, which increases confidence in the findings. Set against this is an equally large, long-running randomised trial (nearly 6,000 male physicians, followed for up to 12 years) that found no effect at all. The discrepancy between these studies makes it clear that population characteristics, particularly nutritional status, play a major role. The smaller studies (Scotland, Korea) add context but are less well suited to establishing cause and effect.