00:00
Hey everybody, it's Mikki here. You're listening to a mini mikkipedia on a Monday and today I want to take a deeper dive into multivitamins and particularly when someone might benefit. Now I will say upfront that I am a fan of supplementation because you cannot get everything you need from your diet and this isn't just me that I feel or I believe this.

00:30
We literally know this is true from the studies that are conducted showing insufficient intake amongst several groups in the population. And in particular, if you are in a calorie deficit, therefore you aren't eating the same number of calories as you would be if you were maintaining your weight, then you are limiting the opportunity to get nutrients from your food. So particularly in that instance, I am a fan of

00:59
someone taking a broad spectrum multivitamin. Now, I wouldn't make this a compulsory part of any plan that I sell, and I don't do that actually, because I don't want someone to sign up to Monday's Matter or the Accelerated Plan or my Flow or Man Plan, and then be given this list of supplements that they must take as an added cost. But I am very, I guess, upfront

01:24
by telling people what the research says and also what we know from people who are limiting their calories. So with that out of the way, I actually want to look at the literature around a multivitamin. And I want to talk you through two papers that sit within the same large randomized controlled trial, actually, the COSMOS study. So COSMOS stands for Cocoa Supplement and Multivitamin Outcomes Study. It's a big trial. It's well-designed.

01:53
And it does give us some of the best data we have on the question of a multivitamin, particularly as it relates to brain aging. And I came across this trial when I was listening to a couple of different podcasts that Rhonda Patrick was on recently, and she highlighted these trials. And I thought, is pretty interesting. So thought I'd go and have a look myself. Now, these aren't just observational studies where we just look at correlations between supplement use and outcomes.

02:21
These are randomized double-blind placebo-controlled trials, which ideally we would want to see this. And often this is the case with drugs, pharmacological trials where you just give someone a pill and it's funded by pharmaceutical industry and it's a very well controlled trial. This is done with a multivitamin. So I'm going to look particularly at the two studies that

02:49
come from the Cosmos Web Ancillary Studies. An ancillary study is a study that looks at an outcome outside of the primary outcomes. Now, the Cosmos study was set up to look at multivitamin use and cancer risk and also cardiovascular disease risk. And I will talk briefly about that, but we are going to do a deeper dive into the cognitive findings which were uh

03:18
pretty interesting actually. So first paper was published in 23 by Jung and colleagues and a second paper was published in 24 both in the American Journal of Clinical Nutrition, which is a high standard nutrition journal. And these studies look at cognitive outcomes and they look at them within the entire study and they give you a really thorough picture of a multivitamin as it relates to cognitive outcomes. So.

03:47
Let me give you the headline first. Across both of these papers, daily multivitamin supplementation, specifically Centrum Silver, easily accessible and affordable supplement, showed a statistically significant benefit for episodic memory in older adults. And when the three sub-studies were pulled together, the evidence for benefit on both global cognition and episodic memory was clear.

04:13
So the magnitude of the effect was equivalent to slowing cognitive aging by about two to three years. And it sounds impressive and it is. But as always, of course, the devil is in the detail. you know, why does this even matter? Cognitive decline is one of the biggest concerns for people as they age and we are an aging population. research consistently shows that maintaining cognitive abilities is a top health priority for older adults.

04:43
To be fair, it's a top health priority for anyone as they're heading into middle-aged. We have surprisingly few proven strategies to slow it down. We know that healthy diet patterns are associated with slower cognitive aging, and of course, exercises also associated with slower cognitive aging. And many older adults already take vitamins or supplements believing that they will help. Until recently, actual trial evidence for multivitamins and cognition has been pretty mixed.

05:11
So some earlier meta-analyses suggested a possible benefit of multivitamins for immediate verbal memory, but others showed null effects or no beneficial outcomes. The biggest trial before COSMOS was the physician's health study two, PHS2, which followed close 6,000 older men over about 10 years. And that trial found no benefit of a daily multivitamin supplement on cognition. But there was a critical design limitation.

05:40
cognitive testing didn't actually begin until about two and a half years after randomization. So any early benefits in the first couple of years in that trial would have been completely missed. you know, in fairness, the cognitive tests used in the PHS-2 trial, they were fairly blunt. So they were designed to detect clinical impairment and not the subtle changes in otherwise healthy people that we're interested in here. And this is where Cosmos comes in.

06:08
It was designed differently and it did collect baseline cognitive data before randomization. So you could actually measure change from the start. So I want to explain the parent trial structure first because I do think understanding this helps make sense of both papers. So the COSMOS was a large two by two factorial randomized controlled trial. And that just means that every participant was randomly assigned to one of four groups. Active cocoa extract plus active multivitamin

06:38
Active cocoa extract plus a multivitamin placebo, cocoa extract placebo plus active multivitamin, or a double placebo. And this factorial design means you can independently evaluate the effects of each intervention. The total trial enrolled over 21,000 older adults from United States. So the women had to be at least 65 and the men at least 60. Participants were followed for a median of about

07:05
3.6 years, which is a long time in terms of nutrition research in this randomized controlled trials. The primary outcomes of the parent trial, as I indicated earlier, were cardiovascular events and cancer, not cognition. The cognitive findings I'm going to focus on today come from those ancillary studies, meaning secondary studies that were bolted onto that parent trial using the same participants and infrastructure, but with their own specific outcome measures.

07:35
And there were three cognitive ancillary studies. There was the Cosmos Mind, the Cosmos Web, and the Cosmos Clinic. So before we come to the cognitive results, it is worth knowing what happened with the outcomes the parent trial was actually designed to test. And that's both cancer and cardiovascular disease. For cancer, the multivitamin did not significantly reduce total invasive cancer. There were 518 cancer events among about 10,700 participants taking the multivitamin.

08:04
versus 535 in the placebo group. That's a very small difference and it wasn't statistically significant. And when individual cancer types were examined, there was no effect on breast, colorectal, prostate cancer, or melanoma. There's an interesting signal for lung cancer, about a 38 % reduction with multivitamin use, but that was a secondary outcome and it wasn't adjusted for multiple comparisons. So I think that would need to be interpreted quite cautiously.

08:34
In cardiovascular disease, the daily multivitamin had no effect on total major cardiovascular events or on any individual cardiovascular endpoints. So essentially the parent trial was nil for both of its primary outcomes. Multivitamin did, however, improve circulating levels of several nutritional biomarkers, including folate, vitamin B12, and 25-hydroxyvitamin D. And this is where the cognitive ancillary studies become particularly interesting because the

09:03
strongest findings from the entire study arguably came from the not from the primary endpoints but from these add-on cognitive studies and it is a good example of how ancillary studies can sometimes yield the most clinically relevant results from a large trial. Now first of all I do want to address the funding question head-on because I knew some of you will be wondering the multivitamin used for Centrum Silver as I mentioned and it was provided by Pfizer Consumer Health Care.

09:32
which is now actually called Halion. They supplied the study pills and the packaging. The cocoa extract arm was funded by Mars Edge and the trial also received substantial support from the National Institutes of Health. So both papers are explicit that no funding source had any role in the design or conduct of the study, the data collection, management, analysis or interpretation, the manuscript preparation, review or approval.

09:59
or the decision to submit for publication. The trial was investigator initiated and led by independent academic teams at Brigham and Women's Hospital, Harvard and Columbia University with an independent external data safety monitoring board. Now, some investigators did declare receiving investigator initiated grants of court from Mars Edge and one received honoraria from the Council for Responsible Nutrition and

10:29
uh These are standard disclosures and don't indicate that industry directed the research, but you should know the funding landscape. The bottom line was that the product was industry supplied, but the science was independently run. Now, on the supplement itself, and I do think this is really important for context and it's something the papers don't discuss, but I do think it matters if you're thinking about practical application.

10:54
Centrum silver uses the cheaper, less bioavailable forms of several key nutrients. It contains folic acid rather than methylfolate, also known as 5MTHF. It contains cyanocobalamin rather than methylcobalamin or hydroxocobalamin for B12. And it uses peroxidine hydrochloride rather than pyridoxal 5-phosphate for B6. The vitamin D is cholecalciferol, D3, which

11:22
is actually the preferred form, so that one's fine, but the B vitamins forms are relevant here, given that the cognitive benefits may be partly driven by B vitamin status. So let's explain why the form matters. Folic acid is synthetic and requires enzymatic conversion through several steps to become the metabolically active form 5-methyltetrahydrofolate, which is the form that actually participates in that one carbon metabolism

11:51
homocysteine recycling and methylation reactions. Key enzyme in that conversion is MTHFR and somewhere between 30 and 40 % of the population carry polymorphisms in that MTHFR gene, the C677T variant being the most well studied that reduce the conversion efficiency. And so for those individuals, folic acid is a poor substrate. They may accumulate unmetabolized folic acid

12:19
in the bloodstream or still not generating adequate methylfolate for downstream processes. Similarly, cyanocobalamin requires conversion to the active coenzyme forms, methylcobalamin and adenosylcobalamin. That conversion isn't equally efficient in everyone either. Methylcobalamin is the form directly involved in methionine synthesis activity, links back to homocysteine metabolism and methylation. So the interesting implication is this.

12:49
If centrum silver produced a measurable cognitive benefit using these less than ideal nutrient forms, what might you see with a better formulated product using methylfolate, methylcobalamin, and activated B6? You could argue the results represent a flaw rather than a ceiling for what a well-designed multivitamin could achieve. On the other hand, it also means these findings can't be automatically extrapolated to other multivitamin products, which the authors themselves acknowledge.

13:18
The specific formulation matters because it isn't just the B vitamins, know, it's every sort of nutrient together that makes up a supplement. And ironically, the one tested isn't what most practitioners would choose if they were designing a supplement from scratch, which is super interesting. So the Vias paper also notes that the Centrum Silver formulation used in Cosmos did differ from the one used in the earlier PHS2 trial that I mentioned.

13:46
the addition of both lutein and lycopene and that these formulation differences may partially explain the different results between the two trials. The one other thing I will say before talking about the results, there are a lot of big trials now that show benefit from taking particular supplements and multivitamin supplements on the brain. And Professor Julia Rutledge talks a lot about the mental health benefits from these multivitamin supplements. And in her research, she doesn't use top of the range either.

14:16
she uses really cost-effective and easily accessible nutrients. if you are working one-on-one with me, I'm probably going to mention something that has the better quality ingredients in it. um But I think it's also quite good to know that you might not necessarily have to have top of the range in order to see some benefits. So I mentioned earlier, if I'm going to go look at the results, I'm going to look first at the COSMOS web trial.

14:45
One of the things I mentioned in the PHS trial was the blunt instruments used to assess clinical impairment. So one of the really innovative aspects of the Jung paper, the first one in 23, was the cognitive battery. Rather than using traditional in-person neuropsychological tests, the researchers developed the Cosmos Web Battery, which was a set of computerized tasks that participants completed at home on their computers. So was pretty easy.

15:13
So the primary outcome was something called the moderate immediate recall task. So participants were shown 20 words, one at a time, three seconds each, and then immediately asked to type in as many as they could remember. So this is a test of episodic memory. This is the kind of memory that depends on your hippocampus. And it's your ability to encode and retrieve specific events and experiences and the type of memory that allows you to recall what you had for dinner last night.

15:42
or what someone said to you in a conversation. It's distinct from something like procedural memory, how to ride a bike, for example, or semantic memory, knowing that Paris is the capital of France. Episodic memory is particularly dependent on the hippocampus, and it's one of the first cognitive domains to decline with normal aging. The second outcome included the moderate retention ratio.

16:07
which is the proportion of words you remember after a 15 minute delay relative to what you recalled immediately. And they also included the mod-bent task, which is a novel object recognition test related and a flanker task, which measures executive function and is mediated by the prefrontal cortex. So the reason for including these different tasks was to test specificity. The hypothesis was that if multivitamins affect the hippocampus specifically,

16:36
which is the brain region most vulnerable to normal aging, then you'd expect to see effects on hippocampus dependent tasks like immediate recall, but not necessarily on prefrontal dependent tasks like the flanker. And one more thing worth mentioning is that the computerized delivery removed variability from different examiners, different testing sites, and different environmental conditions. So that is um absolutely worth mentioning. Okay, so what were the results?

17:04
At one year, participants in the multivitamin group showed significant greater improvement in moderate immediate recall compared with placebo. Specifically, the multivitamin group improved from an average of 7.1 words at baseline to 7.8 years at year one. Placebo group went from 7.2 to 7.6. So both groups improved, which is expected due to practice effects from repeated testing.

17:32
but the multivitamin group improved significantly more. And there was a p-value there of 0.025. And this remained significant when averaged across the full three years of follow-up with a p-value of 0.01, suggesting the effect was sustained. Now, I do want to put that effect size in context. The researchers calculated that each additional year of age was associated with a loss of 0.074 words on the mod-ray at baseline.

18:01
So if you divide the treatment difference of 0.23 words by 0.074 words, that gives you roughly 3.1 years. So in other words, the multivitamin benefit was equivalent to reversing about 3.1 years of age-related memory decline. And interestingly, the secondary outcomes, retention, novel object recognition, and executive function did not show significant treatment differences, which

18:30
is consistent with their hypothesis that the effect is specific to hippocampus-dependent immediate recall, not a global across-the-board improvement, and hippocampus memory is specifically associated with aging. So compliance in the trial, it is worth noting, is also high. So over 90 % of participants reported not missing more than eight days of the pills per month at the 12-month mark. And this remained above 77%, even at

18:59
36 months or three years. Follow-up completion for the mod-ray was 95 % at year one, 87 % at year two, and 83 % at year three, which is pretty good. Solid numbers for a remote self-administered study. Now the second paper, the Cosmos Clinic and Meta-Analysis from Bias et al., brings in a sub-study. It's the smallest but arguably the most rigorous of the three cognitive ancillaries. It involved

19:27
573 participants from the parent COSMOS trial who lived close enough to the Brigham and Women's Hospital in Boston to come in for in-person assessments. So unlike the web trial, which used remote computerized testing, COSMOS clinic involved detailed face-to-face neuropsychological evaluations lasting about 45 minutes. And these were conducted at baseline and again at two years. The battery of tests included

19:55
11 cognitive tests spanning multiple domains and I'll include the papers in the show notes for you to get a real eye on the type of tests that they conducted. From these 11 tests they created three composite cores. The global cognition composite averaged z scores across all 11 tests. The episodic memory composite averaged the four memory tests and the executive function and attention composite averaged the

20:24
five remaining tests. Now a quick note for you unfamiliar with the z score. This is getting a bit technical but you can handle it. The z score standardizes each test to the same scale. A z score of zero means you performed at the sample average. Positive values mean better than average. Negative values means worse. This lets you meaningfully combine scores from tests that have completely different score scales. Now the sample of this little subset of

20:53
people were 49 % female, the mean age of about 69.6 years. Nearly 38 % reported pre-randomization use of a multivitamin. Baseline characteristics were well balanced between multivitamin and placebo groups, and adherence was excellent. 95 % of both groups reported taking at least 75 % of their pills at year one, and about 90 % at year two.

21:19
Over the two years, participants randomized to the multivitamin showed a modest benefit in global recognition compared with the placebo. A small, but the confidence interval did cross zero. So it's worth noting. However, the finding for episodic memory was stronger. The mean difference was 0.12 standard deviation units, and the confidence interval did not cross zero, and it was statistically significant. So to put in clinical context,

21:48
With the numbers that you can read in the papers, the researchers estimated this effect was equivalent to about 4.8 fewer years of cognitive aging and episodic memory, which is pretty stunning. Executive function and attention showed no significant difference between the groups. Therefore, the pattern here in this smaller trial does mirror what Jung and colleagues found in the Cosmos Web. A benefit for episodic and no benefit for executive function.

22:15
consistency across two completely different testing methods and largely non-overlapping populations is pretty meaningful. So within these papers, one last thing I want to mention is there is a meta-analysis pulling all three sub-studies too. So this is where the Viya's paper really adds value. The authors conducted this meta-analysis pulling these non-overlapping participants from all three of the cognitive ancillary studies. Cosmos Clinic with 573 participants,

22:44
Cosmos Mind with 2,158 participants and Cosmos Web with 2,472 participants, giving a total of over 5,000 participants. So they prioritized the clinic study first and then Cosmos Mind and then the web and this conservative approach was good because it does favor the smaller but more detailed study. So when you combine all of the results,

23:10
What they were able to determine was that the results were consistent across the three different testing modalities, three largely independent samples, and multiple research teams. this gives you lot of confidence in the findings of the research. Now, neither of the papers defines really a mechanism, but there are some reasonable hypotheses. First of all, the selectivity of the effect, benefiting that hippocampus-dependent memory,

23:40
suggests that the mechanism may involve the hippocampus specifically, obviously, and it is uniquely vulnerable to normal aging. It does also contain a high density of these micronutrient receptors, including vitamin D receptors. It's plausible that a broad-spectrum multivitamin supports hippocampal function through multiple converging pathways. Interestingly, I didn't mention this earlier, but subgroup analyses consistently showed larger benefits in individuals with cardiovascular disease.

24:08
These individuals tended to have worse micronutrient status, therefore you are able to correct these subclinical deficiencies. This may be a key driver of cognitive benefits also. And the parent COSMOS trial confirmed, as I mentioned, that the multivitamin increased circulating levels of folate, vitamin B12, and 25-hydroxyvitamin D. Also, thirdly, the multivitamin is a combination of over 20 nutrients.

24:34
Previous trials testing single nutrients or small combinations have generally shown no results. It may be that the benefit of a multivitamin lies in addressing multiple simultaneous micronutrient insufficiencies rather than any one nutrient isolation. Low levels of B12, folate and vitamin D have all independently associated with cognitive decline and dementia risk in observational studies. But as always, there are going to be strengths and limitations.

25:01
So the randomized double-blind placebo-controlled design obviously is a gold standard for a study like this. They signed cognitive testing before randomization means you can actually measure change. They had sample sizes that were pretty large. They had high compliance and they used three independent cognitive sub-studies with different testing methods and a meta-analysis showed no between-study heterogeneity, which strengthens the overall conclusion as I mentioned.

25:29
But of course there are important caveats. The first is generalizability. I mentioned the study group, the Cosmos Web sample, 93 % white, highly educated, computer literate, living independently. The Cosmos Clinic sample based in the Boston area was similarly limited in diversity. So these findings may not extend to more diverse populations, people with lower socioeconomic status, or those with existing cognitive impairment. Second, the effect sizes are small.

25:57
So a treatment difference of 0.23 words on a 20-word recall test may not be something an individual would necessarily notice. So the authors rightly note that even small effect sizes can be meaningful to population level for an individual considering whether to take a multivitamin or the brain health, the expected personal benefit may be modest. And I already discussed earlier that the supplement used was Centrum Silver. So we don't know whether a different formulation

26:25
would produce the same results, nor can we assume they wouldn't do better. And of course, fourth limitation is that we don't really know which specific nutrients are driving the benefit. You just cannot tell that from a multivitamin or the data. And then the translation of treatment effects into years of cognitive aging is a simplification, and everyone is well aware of that.

26:47
It's based on cross-sectional age associations, not longitudinal tracking, and assumes a linear relationship between age and cognitive decline, which isn't always the case. And the follow-up period was two to three years. So it wasn't short in terms of nutrition research, but it's certainly not decades, right? So we don't know whether these benefits persist, plateau, or diminish over a longer time. So to summarize, the COSMOS trial is

27:14
the largest and most rigorous examination of daily multivitamin supplementation and cognitive function in older adults that we have. The parent trial found no significant benefit of the multivitamin for its primary endpoints of cancer and cardiovascular disease, but the three cognitive ancillary studies tell a different story. Daily centrum silver consistently improved episodic memory compared with placebo, with an effect equivalent to maybe two to three years of reduced cognitive aging.

27:42
And while the trial was industry supplied, it was independently run by academic investigators with no funded involvement in the science. And the supplement tested, while effective, it used less bioavailable nutrient forms than what many practitioners would recommend, which may mean a higher quality formulation could do even better. So the effect is real and it was replicated and it's statistically robust, but it is small. And we do need to keep that in perspective. However, for a cheap, safe,

28:12
widely available supplement even a small benefit is worth knowing about. As always though, a supplement is just that, it is not a substitute for fundamentals of good nutrition, physical activity and sleep. Alright team, if you found this helpful share it with someone you know. I'll pop the papers in the show notes. Let me know your thoughts. over on Instagram threads in X @mikkiwilliden, Facebook @mikkiwillidenNutrition or head to my website @mikkiwilliden.com

28:42
a one-on-one call with me. Alright team, you have the best week. See you later.