There was a time, not all that long ago, when recommendations regarding multi-nutrient supplements (multis for short, and the term I will use here to refer to the incorporation of a whole profile of vitamins and minerals into a single pill or capsule) were both simple and non-controversial.
That was then. This is now, and now is rather more complicated. At the very least, we have all heard that the simpleminded "at least they can't hurt" concept was flawed, as we should have surmised from the start. Anything with the power to influence human physiology for good has the power to ... influence human physiology, and that invokes the law of unintended consequences. Of course multi-nutrient supplements can do harm.
Now is also a time of ever more rigorous scientific scrutiny and a time of growing skepticism. And now is a time of endless reverberations in cyberspace, so that when research findings do capture the public attention, they echo on into a dense shroud of opinion-based distortions. Seeing the distinction between baby and bathwater in such a haze requires truly dedicated scrutiny; we are the beneficiaries of such an effort far too seldom.
Which brings us back to multi-nutrient supplements. I have written on this topic over recent years, more than once and more than twice, because of the evolving science and attendant opportunities for misinformation. We now have new entries in both those categories, courtesy of research and commentary recently published in the Annals of Internal Medicine, which obligate me to weigh in again.
In case you want the punch line early, here you go: I will continue to recommend multi-nutrient supplements selectively, as I have long done (there was a time I went with the prevailing flow and recommended them routinely, but that ship has long since sailed). I remain open-minded about their potential to do real good yet mindful about their potential to do some harm, particularly if used haphazardly.
When I do recommend multi-nutrients, I will favor those known to be of high quality and packaged in a sensible way to maximize convenience. I will also continue to point out the potential advantages of whole food concentrates, which may be ideal for plugging holes in less than stellar diets, while potentially avoiding the liabilities of conventionally formulated multis.
[Read: Top-Recommended Health Products.]
How can I justify such a temperate response to the recent research when the accompanying editorial in the Annals of Internal Medicine was entitled "Enough Is Enough: Stop Wasting Money on Vitamin and Mineral Supplements"? How can I be so noncommittal when the New York Times chose to run an opinion piece entitled "Skip the Supplements"? Because, on the basis of the actual science, such assertions are unjustified. They are, in fact, as biased and misleading – albeit in the opposite direction –as a sales pitch promising you your multi will save your life.
Here's why. The publications that spawned the recent round of unfounded pontification represented three studies. One examined the role of multis in preserving cognitive function among male physicians and found no clear benefit over placebo. Just under 6,000 participants were followed for roughly 12 years.
Pause for a moment to consider what this study actually tells us. Taking a multi-nutrient supplement did not, all other things being equal, make male physicians smarter – or keep them smarter longer. They did no discernible harm compared to placebo either, by the way.
Of course, if multis do make male physicians smarter but exert such a modest effect that you only see it one time in 10,000, or over a span of decades rather than years, the study would have been blind to such effects. But more importantly, did you ever think that taking a multi would make you smarter? Let's move on.
The second study in the already notorious issue of the Annals examined the effect of multis on recurrent heart disease over an average span of four and a half years in slightly fewer than 2,000 adults who had already had one heart attack. The rate of such events was, in fact, lower in the group assigned to multis but not significantly lower than in the placebo group. There was no evidence of any harmful effects of the multis.
And, finally, study three, a systematic review that aggregated prior evidence, concluded that we lack clear evidence that multis can prevent heart disease or cancer. The actual conclusion was that "limited evidence supports any benefit from vitamin and mineral supplementation for the prevention of cancer or CVD." There was no sweeping rhetoric about uselessness, proof of ineffectiveness or net harm.
In other words, what we have is just what we have long had: relative absence of the evidence we need to know things for sure, not clear evidence of absent effects. We conflate these two routinely and risk much in doing so.
Among the many supplement proponents unhappy with these studies and the hyperbolic indictments they have engendered are purists who point out that the doses and preparations of multis chosen for the trials were suspect. That may be so, but we needn't go nearly so far. We can actually indict the indictment on its own terms by using the very statistics on which research findings depend.
Imagine, for instance, if multis actually do work – both to prevent cancer and heart attacks. But, of course, they exert only a modest effect because many other things matter, too, and some of them matter a lot more. Among the other factors are diet, activity level, genes, tobacco use, drug treatment and so on. What we are examining here is the isolated effect of a multi, all other things being equal.
So let's imagine they do work but that the modest effect in question translates to, say, one less heart attack and one less cancer per 100,000 people taking a multi for two years. If a randomized trial assigned 100,000 people to multis and a corresponding 100,000 to placebo and followed these 200,000 people for a full decade, there would be five fewer heart attacks and five fewer cancers in the multi group.
Such a small difference in so large a population would almost certainly fail to meet the standards of statistical significance, and the conclusion would be reached that multis don't work. It would take at least 20 or 30 fewer events in the multi group, and perhaps a bigger difference than that, to lend statistical support for their use. To see that effect would require following well over a million people for a decade or our 200,000 subjects for half a century.
And yet, if multis actually did prevent one heart attack and one cancer per 100,000 people every two years, it would be a big deal. For every million people taking them for a decade, there would be 50 fewer heart attacks and 50 fewer cancers. If 100 million of us took them for that long, multis would have prevented 5,000 heart attacks and 5,000 cancers. Try telling those 10,000 families that such an effect is "trivial."
[Read: Tips for Heart Health.]
I am not saying multis exert such an effect; I very much doubt it. But I am saying they could exert just such an effect, and all of the research done to date, collated together, would be insufficient to show it. That is the important difference between absence of evidence and the evidence of absence some seem inclined to pretend we have.
And finally, I have known a lot of people taking multis over the years. Very few if any were doing so to prevent a heart attack or cancer. In fact, I can't recall ever hearing that particular rationale. For the most part, people seem to take multis in the hope of feeling a bit better, having more energy, helping their body function as it should each day and maybe succumbing to the common cold just a bit less often. The entire batch of recent studies was blind to these outcomes.
Which leads to my conclusion that headlines telling us to skip supplements or stop wasting our money on them are based on conclusions the authors reached before ever the new research came along, not on any of the actual data. The actual data support no such pretensions.
What the data do indicate is that multis are not a panacea, and it would be a fool's errand to rely on them preferentially to prevent premature death or serious chronic disease. As noted, I don't know anyone who does that, but if you are one of those rare individuals, I suggest you desist. All the more so since lifestyle practices can reliably add years to life and life to years, and cut risk for all chronic disease by some 80 percent. At the very best, supplements are supplements to – not substitutes for – living well.
The data also show some potential for nutrient supplementation to do harm. But, in fact, multis have been impressively harm-free in most trials. So what's good for the goose should be good for the gander. On the basis of the same studies propagating indictments, multis may be vindicated of any appreciable harmful effects.
And finally, the study methods did not address the real reasons most people take multis – to feel and function a bit better every day. Maybe they exert this effect; maybe they don't. Basically, we still don't know what we didn't know before.
The new studies looked at multis for the mission of preventing serious chronic disease, not the mission most supplement users actually have in mind. They used methods limited in statistical power and blind to important real-world effects. The related headlines seem to reflect preconceived notions rather than science considered carefully and dispassionately.
No wonder, then, the message we got was badly muddled.
Hungry for more? Write to email@example.com with your questions, concerns, and feedback.
David L. Katz, MD, MPH, FACPM, FACP, is a specialist in internal medicine and preventive medicine, with particular expertise in nutrition, weight management, and chronic-disease prevention. He is the founding director of Yale University's Prevention Research Center, and principal inventor of the NuVal nutrition guidance system. Katz was named editor-in-chief of Childhood Obesity in 2011 and is president-elect of the American College of Lifestyle Medicine. He is the author of "Disease Proof: The Remarkable Truth About What Makes Us Well."