by Marion Nestle
Oct 13 2011

Alas, vitamin supplements

Two studies released this week provide additional evidence that vitamin supplements are potentially harmful and, at the very least, do no good.

This depressing news comes from the Iowa Women’s Health Study.  Older women in the study who took supplements ranging from multivitamins to high doses of single nutrients had a greater risk of dying than those who did not.

Equally depressing are the results of a trial of high-dose vitamin E and selenium versus prostate cancer.  It found higher rates of the cancer among men taking vitamin E (selenium was somewhat protective).   In this trial, it was so obvious that the supplements did not protect against prostate cancer that the investigators ended it before its scheduled date of completion.

USA Today interviewed me and Dr. Jeffrey Blumberg (Tufts University) about our interpretations of these trials.

I think that the main conclusion to be drawn from this research is that supplements do not make healthy people healthier.   They may not cause harm at high doses, but they appear not to do good.

I don’t take them and I don’t recommend them—except to people who have diagnosed nutrient deficiencies or other problems handling nutrients.

Dr. Blumberg, in contrast, thinks multivitamins constitute a useful nutrition insurance policy and everybody should be taking them.

Supplements are a good example of how scientists can interpret research in different ways, depending on point of view.  I illustrate this point in Food Politics in a table in which I compare what I call “belief-based” (for lack of a better term) and “science-based” approaches to deciding whether supplements are needed, effective, or safe (see table 29, page 232).

For example, on the need for supplements, a belief-based approach rests on:

  • Diets do not always follow dietary recommendations.
  • Foods grown on depleted soils lack essential nutrients.
  • Pollution and stressful living conditions increase nutrient requirements.
  • Cooking destroys essential nutrients.
  • Nutrient-related physiological functions decline with age.

A science-based approach considers:

  • Food is sufficient to meet nutrient needs.
  • Foods provide nutrients and other valuable substances not present in supplements.
  • People who take supplements are better educated and wealthier: they are healthier whether or not they take supplements.

The statements in both approaches are true.

This is why point of view is such an important consideration in interpretation of nutrition research.

  • Thanks for your view on this. I have been wondering about these new studies. My husband and I represent both views (I don’t think I need a multivitamin, he thinks of it as an “insurance policy”.) Personally, I think the vitamin industry has influenced his view more than science has — but that’s just me.

  • Ecobabe

    If we understand a “belief” to be “confidence in the truth or existence of something not immediately susceptible to rigorous proof”, then I am confused about why you are terming the one viewpoint “belief-based”.

    The tenets that the “belief-based” viewpoint rest on appear to be empirical and testable — so unless you are implying that they are false observations that people adhere to despite evidence to the contrary or lack of evidence, they’re just as “scientific” as those of the “science-based” viewpoint, are they not?

  • Anthro

    It was Food Politics that first woke me science based nutrition and I’ve never looked back. The belief based points may be technically true, but only to a minor degree, no? Could you speak to this Marion as it does come up in conversations I have?

    I don’t think there’s any problem with taking a cheap, over-th-counter multi vitamin if a person doesn’t eat well all the time, but beyond that, where is the evidence for the belief based points?

  • RB

    One issue with supplements is that they are not regulated. They may not contain the level of nutrient stated on the label. They also my contain adulterants. They may not be safe because of adulterants or just poor quality control in manufacturing. We just don’t know.

  • Ellen

    Well, but the selenium supplement was protective against prostate cancer. How does a study showing one kind of supplement may reduce disease risk and another kind may increase disease risk lead to the conclusion that all supplements do not make people healthier? If the study was reliable enough use as evidence against high-dose vitamin E supplementation, isn’t it reliable enough to use as evidence FOR selenium supplementation?

  • Alexia

    @ Anthro — if the “belief-based points” are “technically true” then, well, they’re not just beliefs. They’re facts. Science deals with facts. Religion deals with beliefs.

    Also, I don’t think they’re only true “to a minor degree”. It seems pretty safe to say that most people in the US aren’t eating diets that are nutritionally sound. Would they be better served by eating healthful food rather than dietary supplements? Of course. Should we promote healthful diets? Of course. Does that mean no one will benefit from taking supplements? I doubt it.

    Likewise, there seems to be ample evidence that foods today are less nutritious than they once were. See this article: This particular study cites the variety of plants being grown rather than the health of the soil, but the outcome of less nutritious food is the same.

    For the record, I take very minimal amounts of dietary supplements myself. I don’t really care if people take them or not. But as a scientist myself, it irks me to see the term “belief-based” used here when we’re talking about measurable, empirical observations which the author states are TRUE. It’s understandable to make the case that your interpretation of the data is the correct one — all scientists do. But it confuses the issue of what’s science and what’s not to basically say both scientific interpretations are true but only one is science-based.

  • During undergrad, I was taught (1) you want MANY studies from recent years to purport something, and (2) every study has flaws.

    I’m never a fan of the media highlighting a new study or two on suggestive evidence. What needs to be done is more research and, in regards to these studies, start (or at least consider) regulation of the supplement industry.

  • Jeff

    The Iowa Women’s Health Study was a weak observational study. The participants were asked only three times over 18 years what supplements they remembered taking. The result: Those taking suplements had a 1% increase in overall mortality, a barely significant difference. Dr. Robert G. Smith has posted an analysis of the study;

    The SELECT Trial used a low-quality, synthetic form of vitamin E. Most vitamin E supplements use natural alpha-tocopherol, which has a much higher level of biological activity. This article points out some of the problems with the SELECT cancer study:

  • If media actually read the report they’d find; supplement users were significantly more likely than non-users to be much healthier. But that’s neither here nor there.

    The bottom line is that the researchers have no idea how much vitamins and minerals, and in what combinations, these women actually consumed or why they died. We don’t even know if these elderly women even took vitamins, they simply answered a survey 3 times over 22 years. The researchers didn’t even bother to back up their finding with any medical investigation or biological plausibility.

    To condemn multivitamins with such shoddy statistical inference is absurd.

    Additionally, this study is less than useless: it is dangerous, because it is being used by the media and the mainstream medical establishment to blacken the eye of nutritional supplements using poor data, bad analysis, and specious conclusions—otherwise known as junk science.

  • jwlucasnc

    Fascinating. I have always been borderline anemic. Various tests over the years identified no specific problem, other than I don’t absorb enough iron in my diet, which is appropriately varied and inherently vitamin rich. As such, I was put on daily iron suplement (Bifera), which seems to help. Judging by today’s post, I can’t help but wonder if that’s coincidence or wishful thinking. Thoughts?

  • cl

    What about nutrient deficiencies that are very common? These may be individually diagnosable, but testing for nutrient levels can be expensive, especially in states that require a doctor’s Rx for any test. For example, it appears that relatively few people have sufficient levels of vitamin D and magnesium.

  • Pingback: Healthy Eating Deserves a Food Fight – The Buzz Bin()

  • Velez

    I typically take a vitamin D supplement in the winter months and a B12 supplement year-round, as I am a vegan. I get everything else I need from my food.

  • Joe

    Is it not interesting to observe that among the general population (those who supposedly maintain a poor diet subsusting on junk foods) there is little vitamin deficiency. If it is a fact that we as a nation generally eat a nutrient sparse diet then it would seem rational to see vitamin and mineral deficiency on a large scale yet such is not the case. I guess then that the idea that we eat poorly is a “belief” and not a fact.

  • As a mother who is nursing her newborn, I take a multivitamin as directed by my physician. I also take herbal supplements based on what works for me. For example, I take a cranberry pill to ward off urinary tract infections. I suffered with recurrent UTIs for seven years (including one hospitalization and several ER visits). Since I began taking the supplement daily five years ago, I haven’t had a single UTI.

    I do not, however, take any supplement in large doses.

    I love reading your blog! The politics of our food supply always interests me and it is great to read the news which isn’t being reported about the food industry. This week I have featured your blog in my “Follow Friday.”

  • The general omission in all this is that fish oils are proven to be beneficial, especially for those that don’t eat enough fish and shell seafood.

    Many people derive huge benefits from taking fish oil supplements and they should continue to take them or start taking them (no interests to declare).

    Not all supplements are created equal, so don’t throw out the jewels with the rest of the trash.

  • CS

    I am pregnant choose not to take prenatal vitamins because the iron in them constipate me so much! Last pregnancy, when I took prenatals, I relied on OTC stool softeners and laxatives. I didn’t like that I was near-dependent on them. This pregnancy, from pre-conception, I have been more committed to eating a healthy diet. I eat plenty very healthy foods (including foods rich in iron). My fruits, veggies, eggs and beef are all grown locally and organically. I eat a lot of beans and nuts, and my milk and cheese are organic. At my 28 week blood draw, I am happy to say that I am NOT anemic. Last pregnancy, I was anemic at the 28 week point, and my doctor told me to take an additional iron supplement!!!
    I am a Registered Dietitian who does counseling for patients with Gestational Diabetes. I recommend that my patients take prenatals because I do not trust that they have the desire/education/time/money to eat as well as I do.

  • Vitamin D levels are relatively inexpensive to check and I have found that among the women that I see it is relatively common. In sunny southern California, one would think that it would be fairly rare. With Vitamin D, I believe that one should check the level prior to supplementing.

  • Pingback: Sweat Science » Beliefs vs. science for vitamin supplements()

  • Marion,

    I listened to you on On Point last week, and I was a little horrified that none of the nutritionists mentioned the incredibly important role of supplements in correcting deficiencies other than a very brief mention of iron and anemia by Walt Willet. Having been repeatedly diagnosed with nutrient deficiencies despite a nutrient dense diet, I do now regularly take supplements although my philosophy leans toward the varied diet full of whole foods end of the spectrum.

    On the other hand, I do wonder whether someone might have caught celiac a little earlier if they had more actively pursued the reason for the nutrient deficiencies!


  • Pingback: Food Links, 02.11.2011 « Tangerine and Cinnamon()

  • Pingback: Liquid Vitamin Supplements()

  • Pingback: Market AmericaGuest Post: More Facts on Fats | Market America()

  • Pingback: On supplements… | Medical Minute()