In the last few weeks, there have been two articles on the benefit of supplements, specifically multivitamins and flavonols (extracted from cocoa), on improving memory. The media mentioned the multivitamins; the flavanols garnered no attention. What did the studies report?
Let’s begin with the widely reported study in the American Journal of Clinical Nutrition on the role of a daily multivitamin in enhancing memory, specifically word recall. The research cohort in this study comes from a much larger study, COcoa Supplement and Multivitamin Outcomes Study (COSMOS), that enrolled roughly 22,000 men and women investigating,
“whether taking daily cocoa extract supplements containing 500 mg/day cocoa flavanols or a common multivitamin reduces the risk for developing heart disease, stroke, cancer, and other important health outcomes.”
Last year, a group of researchers published results of a cohort of this larger study, COSMOS-Mind, finding a cognitive improvement in those participants in the research arm taking a multivitamin. The improvement in cognitive function was primarily in episodic memory (remembering events in your past), with a much smaller gain in executive function (abilities to organize, abstract, multitask, social interactions, planning.) The criticism of the study at that time, as I reported, was that the finding of a cognitive improvement might well have been an effect of practice taking the test rather than from the multivitamin.
Using another smaller cohort rebranded as COSMOS-Web , the current study takes a “second bite of the apple,” explicitly looking at the impact of a daily multivitamin vs. placebo on episodic memory. The episodic memory testing was based on word recall using a ModRey Task where 20 words were presented for 3 seconds each. Word recall is tested immediately after seeing the words and 15 minutes after completing other tests.  The reported value in the study is the number of words correctly identified immediately.
- Participants taking multivitamins “improved from a mean of 7.10 words at baseline to 7.81 words after 1 year.” For the placebo group, the change was from 7.21 to 7.65 after I year.
- Improvement was seen in both groups in years 2 and 3 – part of the argument for a practice effect, the original criticism of that initial report last year. The researchers write that this improvement was “above the practice effect in a placebo group.”
- The overall mean difference in immediate word recall after three years was a statistically significant 0.15 in favor of the multivitamin group.
- Multivitamin supplements did not significantly improve memory retention, executive function, or novel object recognition.
The researchers note that the effect size is small,
“…and may not be noticeable to all individuals receiving multivitamin supplementation, even small effect sizes can result in large health benefits at the population level.”
We might reduce that sentence to, may help, can’t hurt. And that might be the end of the discussion; it certainly was for the mainstream media. But the researchers did mention an additional study using the same cohort, COSMOS-Web, but now looking at another treatment arm, the cocoa supplements, specifically flavanols. That study was reported in the Proceedings of the National Academy of Sciences, PNAS.
Before jumping to those results, we need to mention that the two treatment groups in the multivitamin study had a Healthy Eating Index (HEI) of 43 – and was similar to a bit better than the American average. 
Using the same cognitive function tests, the flavanol study found no difference in those receiving flavanol supplementation versus placebo. But they then went on to stratify patients by their HEI – the lowest group with a diet consistent with the US population average, and the medium and upper groups with a better diet (HEI 38 to 48, or >48). The low HEI group had a lower baseline ModRey score but was the only group to show improvement; flavanols had no impact on participants with a slightly better-than-average US diet. As in the other study, flavanols did not affect memory retention, executive function, or novel object recognition.
For those with an average or poorer diet, flavanols resulted in a statistically significant mean treatment difference of 0.28 – almost double the value from the multivitamin study. Again, a small effect, and again, the researchers note,
“While the effect size of the flavanol-associated increase in memory might be considered small, sustained and small effect sizes are often characteristic of randomized clinical trials that are based on diet and lifestyle modifications in generally healthy people and at a population level are often meaningful to health maintenance, dietary guidance, and public health.”
Publication – the coin of the realm
In academics, in addition to luck and who you know, advancement is based on publications – publish or perish is real and alive. It is understandable why the same group of researchers using the same dataset, would write not one but two papers. But in comparing the two, the multivitamin study never reported whether the improvements were related to the HEIs of the participants, even though the data was available and was used by the same researchers in the flavanol study. If the flavanol study is correct that there is a “depletion–repletion” underlying etiology, then why omit that analysis in the multivitamin research - isn’t it incumbent on the same researchers, using the same dataset, to at least speak to depletion-repletion when it comes to multivitamins?
You will find both studies if you are deeply interested in this topic. Parceling results across several papers may enhance your academic standing but does little for the general public’s understanding. The media is deeply interested in what gathers attention and, as a result, ignores the flavanol study – misinformation by omission.
 The four treatment arms were Multivitamin and Cocoa Flavanol, Multivitamin and Placebo, Cocoa Flavanol and Placebo, and Placebo and Placebo.
 Those other tests looked at the ability to identify novel images and measure executive function.
 The Healthy Eating Index (HEI) is a measure of diet quality developed by the USDA that assesses how closely an individual's or population's diet aligns with the recommended dietary guidelines. A score of 100 is perfect.
Sources: Multivitamin supplementation improves memory in older adults: a randomized clinical trial American Journal of Clinical Nutrition DOI: 10.1016/j.ajcnut.2023.05.011
Dietary flavanols restore hippocampal-dependent memory in older adults with lower diet quality and lower habitual flavanol consumption PNAS DOI: 10.1073/pnas.2216932120