The Supplement Paradox: Why are so many claims about supplements B.S.?
When you’re watching a magic trick, you know the effect is not real. You know that coins don’t really vanish into thin air and that women aren’t really sawn in half on stage. But the illusion can be compelling anyway, particularly when the source of the deception is hard to find. Some arguments are like this too; you might have good reason to believe the conclusion is false, and yet find yourself compelled by its logic and struggle to spot where it went wrong.
When it comes to dietary supplements, here are two arguments that work like magic tricks; empirical evidence (some of which is discussed below) does not support their conclusions, but many people nevertheless find them very compelling and struggle to identify anything wrong in their reasoning. What do you make of them?
Argument 1: Nature didn't always provide an abundance of the various vitamins, minerals, and so on that we need to survive and thrive. Our ancient ancestors didn’t have the easy access that we do today. Therefore, shouldn’t the average person now be able to add supplements to their diet that improve wellbeing and health, even without some specific health concern that they’re targeting?
Argument 2: Modern society is vastly different from the societies that we inhabited during our evolutionary development. It stands to reason that those differences could mean we're missing out on some important things we used to have. Including in our diets. Fortunately, dietary supplements can provide things our modern environment is missing or has less of and that we would have gotten before. Therefore, shouldn't the average person now be able to add supplements to their diet that improve wellbeing and health, even without some specific health concern that they’re targeting?
Indeed, arguments like these are all over the place and consumer behavior appears to endorse their conclusions. Roughly 60% of adults in the US report taking supplements, and a nationally representative US survey of 12,529 people found that:
37% of supplement takers reported using them to “improve overall health”,
35% reported using them to “maintain health”, and
and yet only 27% of the supplements people reported using were used on the advice of a health care professional.
Chart taken from survey results (source)
Hence, the truth about supplements matters because it potentially impacts your health and because the supplement industry is massive. If supplements don't work as claimed, that's an enormous misuse of resources. An editorial in the Journal of the American Medical Association claims that “people in the US spent an estimated almost $50 billion on dietary supplements” in 2021. There are many stores with whole walls of supplements. It’s easy to get swept up in all the marketing hype without deeply considering the question: Are they trick or treatment?
In this article, we argue that the balance of evidence does not support the idea that most generally healthy, non-pregnant adults in the US should be taking untargeted nutritional supplements (e.g., vitamins, minerals) to improve or maintain their health.
What we’re not saying
Before we dive into critiquing the common arguments, it’s important to point out that we are not saying supplements are useless to all people in all places at all times. We’re specifically looking at non-pregnant adults, and two other important kinds of exceptions should be noted:
For many nutritional deficiencies, supplements genuinely are incredibly helpful. If you go a long enough time without vitamin C, you will likely get scurvy; a lack of vitamin A can lead to Night Blindness, and a severe deficiency of Vitamin D can cause serious bone issues. It's absolutely critical that you get enough of each of the essential vitamins and minerals. Importantly, though, this is a different question than whether most people in the US will benefit from untargeted supplementation.
There is robust evidence in favor of some supplements having positive effects for specific medical conditions or in some contexts. We discuss this more below, but for example, folic acid supplementation can benefit people who are pregnant or planning to become pregnant in the next month or so, and creatine (when paired with resistance training) is genuinely useful if you're trying to improve muscle strength.
Our focus is the claim that most people should be on supplements because they produce positive effects in average adult people without deficiencies and who lack specific medical conditions that can be treated by those supplements. In other words, we’re criticizing the view that the general adult population should all be taking untargeted supplements to improve or maintain their health.
It’s important that we’re clear about this: if you're taking supplements and finding benefits, or your doctor has recommended you take them, we are not suggesting you should stop taking them.
The evidence of large trials
If you look at studies on supplements, you'll find that earlier trials tend to show a wide variety of different benefits of many supplements. But a common pattern of events goes like this: those early and promising-seeming studies (which may be smaller, target more specific demographics, or have otherwise-biased samples) motivate larger, more rigorous studies in the general population, but the larger trials don’t find the same results in healthy adults.
We don’t have the space to list out all the cases where this has happened, so let’s focus on a classic example: Vitamin D. You can find promising early trials on specific populations showing that vitamin D can help with many different things, including cancer, depression, falls, and more. But when really large trials were conducted, these effects were not found for generally healthy adults. Examples include this study of vitamin D’s effects on cancer, this one on depression, and this one on falls.
There is evidence that darker-skinned individuals, those who don't get much sun, and anyone with low levels of vitamin D in their diet (e.g., vegans) may have an elevated risk of deficiency, but that's still not grounds for a general population recommendation. And large-scale studies giving vitamin D supplements to the general population generally fail to find benefits for those without deficiency.
In some cases, large trials revealed that supplements that initially seemed promising were actually likely to cause harm rather than benefit. We’ve previously discussed the fact that early observational studies suggested that diets rich in beta-carotene might help prevent lung cancer, but then larger and more robust studies (and later meta-analyses) found that beta-carotene supplements apparently do the opposite: they appear to increase rates of lung cancer, especially for smokers and asbestos-exposed populations. The results were so concerning that one large trial was stopped 21 months early because of “substantial evidence of possible harm.”
This pattern of insufficient evidence of benefit applies to many supplements and many health concerns. Another example comes from the US Preventive Services Task Force, which recently conducted a large review of the evidence and found that vitamin and mineral supplementation (including multivitamins) provided “little or no benefit” in preventing cardiovascular disease, cancer, or death in healthy adults. There was a slight reduction in cancer incidence with multivitamins, but the task force determined this was insufficient to recommend their general untargeted use. They also explicitly warn against taking beta-carotene supplements to prevent cardiovascular disease or cancer, because the harms outweigh the benefits.
There have been lots of studies of the effects of multivitamins (such as those in the review just mentioned, and others like this RCT and this RCT and this meta-analysis). The fact that such studies involve giving people lots of vitamins at once and yet they don’t find enough benefit to warrant recommending multivitamins for the general population of non-pregnant healthy adults suggests that (even though multivitamins have targeted uses for some people) the untargeted “spray and pray” approach to supplements is not effective.
Again, we’re not claiming that nutritional supplements (e.g., vitamins and minerals) do nothing. They really can help some people who have specific health concerns or contexts, and people with genuine severe deficiencies must change their diet or add supplements. If you’re experiencing benefits from supplements or have been recommended supplements by your doctor, please do not stop using them on the basis of anything in this article.
And, of course, there’s always a chance that there’s some benefit (or some harm) to generally healthy adults that hasn’t yet been discovered. We’re just arguing that the balance of evidence does not support the view that most average, healthy, non-pregnant adults should be taking supplements for untargeted improvements in or maintenance of their health.
What do the magic arguments get wrong?
What we’re calling “the supplement paradox” is the fact that there are compelling arguments to think that routine, untargeted dietary supplements could improve the health of average healthy adults (without nutritional deficiencies), and yet empirical research has not found the kinds of improvements that could justify such supplementation. We’re not going to solve the paradox by giving a deep diagnosis of all the failures of reasoning in those compelling arguments, but we do want to make a couple of observations about where they go wrong.
The first argument
Here’s a reminder of how the first argument goes:
Argument 1: Nature didn't always provide an abundance of the various vitamins, minerals, and so on that we need to survive and thrive. Our ancient ancestors didn’t have the easy access that we do today. Therefore, shouldn’t the average person now be able to add supplements to their diet that improve wellbeing and health, even without some specific health concern that they’re targeting?
An assumption that can drive this kind of argument is: if something is needed by your body, then more of it is likely to be beneficial. But that assumption fails to consider alternatives, such as the ones mentioned by the British physician, academic, and science writer Ben Goldacre:
“Perhaps an excess of [supplement] is simply excreted, or turned into something else. Perhaps it just sits there doing nothing, because it’s not needed. After all, half a tank of petrol will get you across town just as easily as a full tank. Or perhaps, if you have an unusually enormous amount of [it] lying around in your body doing nothing, it doesn’t just do nothing. Perhaps it does something actively harmful. That would be a turn-up for the books, wouldn’t it?”
The second argument
Here’s the second argument again:
Argument 2: Modern society is vastly different from the societies that we inhabited during our evolutionary development. It stands to reason that those differences could mean we're missing out on some important things we used to have. Including in our diets. Fortunately, dietary supplements can provide things our modern environment is missing or has less of and that we would have gotten before. Therefore, shouldn't the average person now be able to add supplements to their diet that improve wellbeing and health, even without some specific health concern that they’re targeting?
Notice that all this does is establish that we could be missing out on important components of nutrition. But plenty of things that could be true are actually false. In order for this to be a compelling argument for supplements, it would need to further establish that we actually are (or are likely to be) missing out, and that supplements can fill in the gap.
It’s also worth noting that natural selection favors traits that increase reproductive fitness rather than optimizing maximum lifelong health. This means that speculation about the diets of our ancient ancestors could give us signals for things to investigate, but it does not entail that deviations from the behaviors of our ancient ancestors will have negative consequences.
What should you do?
We can’t tell you what to do (and we’re not medical professionals), but we can tell you how we read the evidence. It seems to us that if:
Footnotes
1 The infographic simplifies things more than we would: For example, it states that this study “suggests that vitamin D3 may reduce mortality” (without specifying a specific population). Taken at face value, this might seem to suggest that vitamin D supplementation for the general population could be beneficial (contrary to what we’re saying the evidence shows). However, that study includes 26 trials using participants with vitamin D insufficiency and 19 trials using participants with adequate vitamin D. The authors of the study note that “Vitamin D had no statistically significant effect on mortality in the trials including participants with vitamin D adequacy” and “The effect of vitamin D3 on participants with adequate vitamin D status is unknown.”
2 Some of the cited studies have been superseded: For example, that same study of vitamin D supplementation included trials up to February 2012, but more recent evidence undermines the generalized claim the infographic makes. For example, a recent systematic review found that “there was no evidence to support empiric vitamin D supplementation [...] in healthy adults aged 19 to 75” (but it did find some benefits for people aged 75+ and those with high-risk prediabetes).
Note: The authors of this article are not medical professionals and this article is not intended to constitute medical advice.
