Another Negative Study of Vitamins

Perhaps one of the most common questions I receive from those who wish to utilize science-based medicine for their own health is what I think about vitamins. Even among hard-nosed skeptics, this question is often perplexing. On the one hand, vitamins themselves were discovered by medical and biological science, they play a vital role (by definition) in the healthy functioning of our bodies, and deficiencies of vitamins can cause disease. So they seem perfectly legitimate. On the other hand the market is full of exaggerated and even magical claims about the cure-all power of vitamins.

It’s difficult for people to come to a bottom-line conclusion – should they take vitamin supplements or not. Is it woo or not woo?

Well – it’s complicated. But there is large body of research to help inform our decisions about vitamins. Now, the largest study to date has been published (Multivitamin Use and Risk of Cancer and Cardiovascular Disease in the Women’s Health Initiative Cohorts) looking at 161,808 post-menopausal women over 8 years and finding no benefit for heart disease, cancer risk, or overall survival. This study comes on the heels of other recent studies showing no benefit from routine supplementation.

The recent study

There were two components of this new study. The first pooled data from several trials where multivitamin use was tracked, and also randomized patients to dietary modification, hormonal therapy, and calcium and vitamin D supplementation. The second component of the study was an observational study – which means that multivitamin use was tracked but subjects were not randomized.  The study found:

After a median follow-up of 8.0 and 7.9 years in the clinical trial and observational study cohorts, respectively, the Women’s Health Initiative study provided convincing evidence that multivitamin use has little or no influence on the risk of common cancers, CVD, or total mortality in postmenopausal women.

The strengths of this study are its size and duration. Essentially the study had enough power to detect even a modest effect if present. The weaknesses are that the observational part of the study was not controlled. This opens the door to multiple factors that could influence the results.

Some have also charged that 8 years is no sufficient, as benefit from supplementation may manifest only over decades. This is true, but is a generic criticism that can be leveled against any negative study. It doesn’t change the fact that there is still a lack of evidence for routine supplementation. Also, I think that 8 years is a substantial follow up, so if there were a measurable benefit from vitamins this study was probably powerful enough to see it.

On another point, the study’s lead author, Marian Neuhouser, is quoted as saying:

“Get nutrients from food. Whole foods are better than dietary supplements.”

This may be true, but this is not something that can be concluded from this study.  Advocates for the preventive benefits of nutrition often make this claim – that nutrition from food is better than nutrition from vitamins. This is partly based on other observational studies that link a diet rich in fruits and vegetables with lower risk of heart disease and certain cancers. However, another way to interpret this discrepancy is not that vitamins from food are better than vitamins from pills but rather that vitamins are not what is healthy about such diets.  For example, a diet rich in fruits and vegetables may be healthy because it is not full of animal fat. Or, people who take care to eat such diets may also generally take better care of themselves.

The weakness of observational studies cuts both ways.

Routine Supplementation

To back up a bit – let me say that I am a strong believer in good nutrition. From a basic science point of view, it just makes sense. We need specific nutrients for various biochemical reactions in our bodies.  Deficiencies are clearly related to specific diseases. Further, in areas where nutrition is poor overall development is also poor. The connection is clear.

The question of routine supplementation, however, goes further. It asks whether or not taking vitamin supplements are necessary or helpful for someone who has a reasonable diet and does not have any specific disease or condition that would impact their demand for specific vitamins – i.e. routine. In most wealthy industrialized nations lack of quality food is not a problem for most people. Also, modern methods of growing, preserving, and shipping food means that more people have access to a wider variety of foods all year round.

Supplement sellers have tried to perpetuate a myth that foods grown today lack key nutrients. There is just no evidence for this, however. The foods we have access to today have, if anything, superior nutrition to those of the past. Further, botique organically grown crops have only a minor advantage in nutritional content that is not likely to be meaningful.

In short, lack of nutrition does not appear to be a problem for the industrialized west. The modern nutritional scourge is, rather, too much of the wrong kinds of food. This gets back to the point about diets rich in fruits and vegetables – their benefits may be more in what they do not contain than what they do.

Right now the evidence supports the conclusion that for most people who have a reasonable diet there is no benefit from routine supplementation. Of course, what’s a “reasonable” diet? This opens the door for those who wish to take a multivitamin as “insurance” because they feel their diet is not very good. The evidence suggests that they would be better off having a healthy diet rather than living on steak and vitamin pills.

While lacking evidence, the “insurance” argument is not entirely unreasonable. The advice I give my own patients who ask my opinion is this. If you feel you need a supplement because your diet is lacking than go ahead. But – do not take megadoses of vitamins, do not go broke on fancy expensive supplements, and do not use vitamins as an excuse to otherwise have an unhealthy lifestyle. And, in the end your are probably better off working on having a better diet.

Targeted Supplementation

None of what I said above has anything to do with targeted supplementation. I prescribe vitamins almost every day – for those with conditions known to be improved or prevented by specific vitamins or co-factors, and for those with a documented deficiency. For example, many patients with certain neurological diseases have low folate or vitamin B12 levels. So I check these frequently and supplement them if the levels are low.

There is evidence that vitamin B2 may improve migraines in some patients – so that is an option I offer migraine patients. B6 may improve recovery from nerve damage (but only at proper doses, megadoses actually causes nerve damage). And folate has been shown to reduce the risk of spinal cord defects, and so should be taken by any woman who might become pregnant (once you find out you are pregnant it’s too late).

Also, there has been much evidence recently about the value of vitamin D in preventing certain cancers and neurological disorders. It may be that the recommended daily allowances of vitamin D were too low, and they are being increased as new evidence comes in.

Targeted supplementation is evidence-based and specific.


There is a clear trend in the evidence we have to date. The benefits of routine supplementation are unproven. High doses of vitamins probably cause more harm than good. If you are concerned about your nutrition then improve your diet.

Further, as with good science-based medicine in general, questions must be specific. We should not ask – do vitamins work. Rather we need to consider specific supplementation in specific situations and conditions.

What may surprise members of the public who have been exposed for years to the propaganda of a supplement industry eager to take their money, or CAM proponents eager to appear legitimate – is that vitamins have always been part of mainstream science-based medicine. There is a tremendous amount of quality research to inform our conclusions about vitamins, and they are part of everyday practice.

Posted in: Nutrition

Leave a Comment (27) ↓

27 thoughts on “Another Negative Study of Vitamins

  1. Dacks says:

    I think the representation of vitamins as “supplements” gives CAMers another inroad into legitimate science, allowing tincture of burdock to be sold alongside vitamin C pills.

    Every doctor I have seen has encouraged me to take a multivitamin, which I have never done. I consciously eat a diet with plenty of fruits, vegetables and whole grains and have always been healthy. I guess the vitamin recommendation is just a fallback position.

  2. adina says:

    I’m curious what you think about the research pertaining to omega-3 fatty acids. I know that they’re macromolecules, rather than vitamins, but I’m often confused about which claims about them are legit and which are bunk. Thanks.

  3. fetner says:

    Would you consider a healthy vegan diet one that doesn’t require supplements? I hear that B vitamins are a concern, as are D, since vegans miss out on getting that in milk.

  4. daedalus2u says:

    There is another explanation for why nutrients from food appear to produce better health than nutrients from supplements. Virtually all studies of the effect of diet on health are retrospective studies from a self-selected diet. If healthy people choose what we consider to be a healthy diet, our definition of what is a “healthy diet” may simply be due to that choice. Food choice may be the consequence of good health, not the cause.

    A major class of nutrients focused on is antioxidants, which destroy free radicals and limit their concentrations. A widely held (but simplistic) assumption is that free radicals are universally bad because they can damage proteins and DNA. Unfortunately that assumption is wrong (like most simplistic assumptions). Radicals are extremely important signaling molecules. They are ideal signaling molecules because the background is very low and they are extremely reactive. Free radicals are widely used as signaling molecules in every cell. For example mitochondria are regulated by the interplay between NO and superoxide (as are the cytochrome P450 enzymes).

    Like everything that is important in physiology, free radical signaling is regulated, but the details of that regulation are mostly unknown. If physiology has adjusted itself to promote signaling via superoxide, a diet rich in antioxidants would interfere with that signaling, and physiology would be forced to destroy those antioxidants via generation of still more superoxide (which physiology can produce in unlimited quantities). People in ill health may choose diets low in antioxidants to reduce the metabolic demand to destroy those antioxidants so as to maintain the antioxidant level that physiology is calling for.

    In “the wild”, organisms self-select a diet that allows them to survive and reproduce. Presumably their physiology causes them to self-select a diet that contains all essential nutrients in sufficient quantities to sustain life and reproduction. Diet choice pretty much has to be part of the control system that regulates levels of nutrients.

    I think the generation of excess superoxide to destroy excess antioxidants is the mechanism for the slight negative effects observed in many of the large placebo controlled studies of antioxidant supplementation. Non-physiological levels of virtually anything can only have non-physiological effects. Non-physiologically effects are most likely to be adverse but that depends on what is considered “adverse”. Our physiology didn’t evolve for us to live happy lives, it evolved because our ancestors survived and reproduced.

  5. harryc says:

    Here’s a fact: people have evolved over millions of years eating real food.

    Why not do what you are designed by nature to do?

    Here’s another: most people today, as this article points out, have access to an incredible variety of good food.

    It’s common sense and studies have shown (in case you need proof) that eating good food is good for you in many ways.

    Eat real food. Eat good food.

  6. Infinite Monkey says:

    Now, I’m just a regular consumer trying to keep myself informed. An informed consumer is a good consumer. So, if I’ve comprehended what you’re saying correctly, then taking a generic OTC mutlivitamin is ok, even if its effects are placebo. While there’s no evidence it does any actual good, there’s no evidence it does any actual harm. But, taking a multivitamin is not an excuse to eat nothing but steak and ice cream.

    Does that about cover it?

  7. Prometheus says:

    As HarryC pointed out, humans evolved to live on a diet of “real food”. More specifically, we evolved to be omnivores – eating animal and plant tissues.

    Another thing about our evolution – we also evolved in a situation – which still occurs “in the wild” – of “feast and famine”. Until very recently (evolutionarily), humans spent most of their time hungry and very a small fraction of their time with a full belly. We evolved to be able to make it through the “lean” times without vitamin supplements or a “precise balance of fatty acids”.

    Apparently, a new species has diverged from Homo sapiens sapiens, one which requires regular – even daily – administration of vitamins, “minerals”, the “right” fatty acids etc. to maintain “wellness”.

    I have tentatively named this new species Homo sapiens hypochondrians.

    Unlike its ancestral species, Homo sapiens hypochondrians cannot survive without dietary “supplements”, “organic” food, homeopathic remedies and regular chiropractic “adjustment”. In many ways, this new species resembles the auxotrophic mutants of many laboratory strains of bacteria, plants and animals.

    We need to consider – as a society – if we are willing to allow this new species of Homo to consume inordinate amounts of societal resources, which they will need to survive, given that they are clearly not able to survive even in the plush environment inhabited by Homo sapiens sapiens.

    Is this species worth preserving or whould we let selection pressures work unimpaired?


  8. tmac57 says:

    Michael Pollan says in his book ‘In Defense of Food’ that there are cultures of people that eat diets that range from almost exclusively animal protein to vegan, and thrive well on them , while we eating what he calls the “western” diet have many health problems and diseases that are nearly non-existent in those cultures. It seems like a fixation of food as “medicine” type of mentality that drives the various diet fads in the U.S. doesn’t have a lot of good science behind it. His advice: Eat food . Not too much. Mostly plants. But who knows? Maybe he’s wrong too.

  9. SDR says:

    tmac57, you have a good point in general, but the claim that people with western diets have health problems non-existent in cultures who eat either primarily animal protein or vegan diets does not necessarily mean that those diets are more healthy. Just as we have health problems those people don’t have, they also have health problems that we don’t have. This doesn’t necessarily follow that their diets are therefore overall healthier than ours.

  10. The Blind Watchmaker says:

    When I see new patients, we tell them to bring in all of their medication bottles in a bag. When I go through the bottles, there are usually 2 or 3 (sometimes many more) large bottles of vitamins.

    They hem and haw about how much their blood pressure pills cost, but have spent small fortunes at GMC on vitamins.

    When I suggest that they are not all required (with some exceptions, as Dr. Novella described), they act a bit shocked. “Well which ones should I be taking?”, they ask.

    I reply, “Three meals a day, don’t skip breakfast. Get fruits and veges with each meal. Try to get whole grains and other complex carbohydrates. Get your fats from non-animal sources (oils from vegetables, fish and legumes). Use low fat milk or other healthy sources of calcium and vitamin D. Avoid fried foods. Minimize processed carbs like sugars, soda, and white bread. And…..get 2 1/2 hours of exercise a week.”

    “But Doc, what about CoQ-10?”


  11. CarolynS says:

    Michael Pollan’s prescription sounds plausible but also has never been tested, plus is rather hard to define. Most people believe that what they are eating is “food” after all!

    It’s probably prudent to eat a variety of foods and to avoid diets that are too extreme. Other than that, a lot of the stuff about “lifestyle” really verges on woo anyway. “Lifestyle” gets treated as a cheap effective solution to health, disease, cancer, obesity, you name it, despite almost no real evidence and few interventions. Yes, lifestyle interventions reduced the risk of progrression to frank diabetes over 3 years in a subset of people who already had impaired glucose tolerance. But lifestyle didn’t eliminate diabetes in this group! The effects of a healthy diet and exercise are probably good but they should not be exaggerated. I think that some of the beliefs about lifestyle are just as evidence-free as some of the beliefs about other CAM-type modalities. And they encourage the idea that conventional medicine is unnecessary or excessive.

  12. tmac57 says:

    Pollan gives detailed explanations in his book about each of his recommendations . Those were just the points he boiled them down to . Briefly by “food”, he means eating foods that your great great grand parents might recognize. Not the highly processed crap like Pop Tarts or Captain Crunch etc. He thinks food fads like oat bran and Omega 3’s and anti-oxidants are wrong headed trends, and that eating a wide variety on fruits, vegetables and smaller amounts of meat are a good general guideline.

  13. SDR says:

    tmac57, criticizing “processed” foods is not helping your argument. There is nothing my nature about “processed” (a word with no real distinct definition when talking about food) food that by nature makes it any less healthy than any other. Some is, some processed food is actually very “healthy” (another vague term). Food is food. It’s just as ridiculous so think that out great grand parents food was any better than ours. We are, despite any problems with out food such as a lot of fried things, still much healthier altogether than they ever were. I just don’t understand what you’re attempting to argue here.

  14. tmac57 says:

    SDR ,
    Actually I would have to leave it to Pollan to justify his arguments, I read his book about 2 months ago and do not have a copy to refer to. I think that the processed foods (added sugars,highly refined grains etc.) are being implicated in type 2 diabetes and colon cancer. By “food that your great grand parents would recognize”, he is not talking about anything new, but that today we eat a lot of prepared foods from restaurants and grocery stores that contain too much and the worst kind of fats, sugars, too much highly refined grains,foods fried in hydrogenated oils,and so on. He would recommend eating more vegetables, whole grains,fruits and small portions of meat .Nothing fancy or unique about it,just basic nutrition.
    If you read the book it would be a lot clearer, I can’t do justice to it here, but I thought it made a lot of common sense.

  15. CarolynS says:

    Regarading Pollan, his recommendations are certainly pretty conventional and unlikely to be harmful, but that of course doesn’t mean they are correct or that they do any good or that anyone really has much evidence one way or the other. Of course your great-grandparents had much lower life expectancies anyway and probably higher rates of cardiovascular disease than people today. Do “prepared foods” have some specific harmful ingredient or ingredients? Are there any actual data? Or just kind of the usual set of prejudices about the evils of modern life and McDonalds? I am not saying his recommendations are wrong. I am just saying that they aren’t supported by evidence, and that maybe the same evidentiary standards should be applied, or at least there should be some acknowledgment that the evidence is weak or lacking and these are more “precautionary” recommendations than evidence-based recommendations.

    I am also trying to say that the whole “lifestyle” set of beliefs comes pretty close to verging on the same problems as CAM beliefs and in fact may encourage CAM-type beliefs. Even though you think the “lifestyle” beliefs are benign and the CAM beliefs are less so, they both deserve to be treated critically, not just accepted uncritically.

  16. Fifi says:

    My great grandmother and great great aunts all lived until their mid to late 90s, I’m not sure why people assume that everyone used to die young a couple of generations ago. Sure you could get killed during childhood and over the course of your life by various diseases we can now cure but if they didn’t kill you there was no reason good genes and moderation couldn’t carry you into a ripe old age.

  17. DavidEmerson says:

    The study that you refer to is hardly “evidenced-based.”

    As a 15 year survivor of multiple myeloma, the last 10 in complete remission without any conventional therapies whatsoever, it is clear to me that daily supplementation of vitamins, antioxidants, etc is helping me remain cancer free.

    Antioxidants during therapy will help many survivors undergoing therapy-

    David Emerson

  18. upnorth says:

    Do prenatal vitamins count as “targeted supplementation?” I’m highly skeptical of supplements in general, but the nurse practitioner I see suggested taking prenatals while trying to concieve. I have a healthy, very balanced diet… jeez, do I really need this stuff? I find that the level of woo surrounding conception and pregnancy is outrageous.

  19. pmoran says:

    David Emerson’s story is interesting, but typical of the testimonial upon which”alternative” cancer methods rely there is a lot of irrelevant material and critical lack of detail where it matters. We learn, in a third party account by a journalist, how awful the doctors were, and in even more exhaustive detail about the difficulties of getting Burzynski’s outrageously expensive and still highly dubious treatment paid for by other people.

    Yet there is an absence of detail where it matters for the question as to what was responsible for the favourable outcome. We are expected to infer that there was uncontrollable cancer without any evidence being supplied that this was the case or that any of the treatments claimed affected the indicators of active disease that were being employed.

    It’s not good enough for such important questions. His may be a remarkable case, but we cannot tell from the supplied material.

  20. Calli Arcale says:

    upnorth: a lot of the woo relating to pregnancy and conception revolves around the fact that it is very difficult to ethically study this stuff in humans, and so most researchers won’t touch it with a ten-foot pole. There’s woo, and there’s also quite a bit of cargo-cult science and old wives’ tales that are relied upon in the absence of anything more substantive. Combine that with the fact that nobody in the world wants to risk making things worse for a developing baby and you get quite reasonable, evidence-based folks recommending things on little more than a hunch. I mean, more vitamins aren’t gonna *hurt*, so why not?

    There are a few vitamins that are absolutely vital to pregnancy, and which really are a bit difficult to get. Folic acid is the most famous; deficiencies that cause no problems for the mother can result in severe neural tube defects in the child. Spina bifida, for instance. There was a vigorous campaign to stamp out spina bifida years ago. The two big things that came out of it were: 1) a recommendation that pregnant women take lots of folate supplements during the earliest part of the pregnancy (when the neural tube is forming), and 2) routine fortification of grains with folic acid.

    Do you get enough nutrients in your diet to sustain a baby as well as yourself? Probably. But given how delicate the early parts of a pregnancy are, it seems wise to recommend supplementation anyway, just in case. So far, unforseen consequences haven’t reared their ugly head. But I don’t think routine supplementation makes much sense in any other situation, unless the person has some sort of a known deficiency. And vitamins can actually be harmful in excess.

Comments are closed.