The idea that multivitamins do harm is only backed up by studying people who use them versus those who don’t. There is no mechanism defined. It’s just as likely that the reason is that people who feel the need to take multivitamins aren’t as healthy as those who don’t.
The only proposed mechanisms I am aware of is toxicity of fat soluble vitamins which can’t just be flushed out the system, and the idea that free radicals are sometimes beneficial. (For example, they can halt cancer before it starts.) But this still leaves a large number of vitamins and minerals unaccounted for.
And the good multivitamins don’t include an excess of the problematic nutrients.
This thread is getting perilously close to the “natural medicine is better” crowd. Think about it. Science is out there isolating the nutrients that cause certain effects, but people are saying that you’re better off sticking with what nature provides in food. How different is that from the issue of medicine, where scientists are out there isolating certain effects, and people saying that it’s better to go natural?
For the vast majority of people, multivitamins help make up any possible nutrient deficiencies or just don’t do anything. Only a small number of people have problems, problems that have still not been isolated to be caused by the multivitamin itself.
Some of the studies showing potential harm have been randomized controlled trials.
As in this study in which the multivitamin (antioxidants) group had a 2.8 times greater risk of death than those taking the placebo (with no benefit even though diets with those levels of vitamins are associated with large benefits). The mechanism is speculated to be an adverse effect of the vitamins on LDL and HDL particle size, but correct is not defiitively defined.
Agreed that the amounts in one standard issue multi is likely doing neither much harm nor good for any standard issue American. That said the standard issue comment of “it couldn’t hurt” is not justified. It could. It might. Not likely but at least as likely as it helping is.
Your point about going natural is in this case anti-scientific. Science has NOT isolated specific nutrients that supplemented above levels obtained by the typical American eating the typical American crap have specific effects (other than ones that have consequently already been fortified into foods, such as B vitamins to reduce neural tube defects). It has instead determined that the whole foods do it and that the extra specific nutrients in isolation do not. It has not figured out if it is because of other isolatable substances (e.g. phytochemicals, fiber) or something about the package that changes the delivery rate and or location that compounds are released or what.
One small illustration of what we do and do not know - We know that having several servings of whole grains foods per day is associated with a decreased rate of developing diabetes compared to the same number of more refined grain products or no grains. Certain fiber types seem to be part of that effect. Also certain starches are resistant to digestion when still contained in less processed whole grains and therefore more slowly absorbed, and also reaching the colon and feeding the microbiota to a greater degree. And doing something else too? We don’t know. We do also suspect that other substances, as of yet not clearly determined, within the whole grain also are part of the story. And maybe the physical structure impacts where and when they are released and the nature of their beneficial impact.
Reductionism is nice and so is an appreciation that how the parts are put together makes a difference. A car is not just a collection of parts; it is a specific selection of part in a particular balance put together in a particular way. The same parts in a random pile does not have the same functionality. Raining down a host of complete bus and train parts on a city without transport does not provide transportation and may cause harm. Even those are *all *the parts contained within buses and trains.
This likely was mentioned earlier, but the SELECT trial (results published in 2011) demonstrated an increased risk of prostate cancer in men who took the most commonly available vitamin E supplement - but no such effect has been observed in men who get their vitamin E from dietary sources.
It is common for promoters of supplements in general to hype the alleged value of their products while ignoring or denigrating foods that contain these compounds. Another example of this phenomenon is resveratrol, heavily promoted for supposed anti-aging, pro-cardiac health, anti-cancer effects etc. (though virtually all the positive findings thus far are limited to biochemical markers and not disease reduction, and have been demonstrated in tissue culture and rodent models, seldom in humans). You can obtain resveratrol through diet (red wine has been publicized, but resveratrol is also found in foods like grapes, blueberries, peanuts and so on). The supplement dealers would have you believe that food sources are no good and that you must buy their pills instead.
Too bad this self-serving view is not backed by good evidence.
I don’t see it that way. All of the studies showing harms from vitamins involve megadoses, far beyond the US RDA. The linked study involved 1000mg of Vit C -17 times the US RDA- and 800mg of E, or 16 times the US RDA. Now it’s interesting they found some effect among those taking hormone replacement therapy. But it’s not especially applicable to the ordinary multi-V.
That said, I advise all taking multivitamins to let their physician know, especially if you are on other sorts of medication. There are occasional interactions.
If a study shows no effect, it could be because the effect was too small to measure or there were confounding factors. If a study shows some effect, it could be due to oversampling or confounding factors. The point is we have some evidence of benefit from multi-Vs for the general population and no evidence of harm. And some groups definitely need it.
Furthermore there is better but still mixed evidence for Vitamin D, B-6 and B-12. And calcium.
Still the bottom line for me is that a 4 cent pill is a tiny fraction of the food budget. If you don’t like profiteering hucksters, just buy supplements with guidance from Consumer Reports or hop on over to Amazon and buy the CostCo store brand, now selling for 3 cents per tablet (price drop!).
There’s also the concern that multi Vitamins might be substituting for good health practices. But it’s just as likely that they could be encouraging health consciousness: after all it’s a daily ritual. But I agree that taking a multi-V doesn’t do squat for your fiber intake or physical activity. (I confess I’m frankly dubious about fiber supplements, absent a solid biological explanation of fiber’s benefits, with all pathways detailed.) And there are a lot of vitamins unceremoniously dumped into prepared foods (along with heaps of sugar and saturated/trans fat) by our corporate masters.
It only took me a few minutes to find studies that showed harmful effects in doses found in multis. Based on that, it’s reasonable to assume there are more. Examples:
The Costco multi you linked to has over 100 mg of beta carotene (and plenty more forms of Vitamin A). The following study found more lung cancer in those taking just 20 mg of beta carotene than those taking a placebo.
The same multi above has 400mcg of folic acid. In the study below, taking that amount or more had a significant increase of breast cancer. No signiificant increase in breast cancer was found from those getting folic acid from food.
As far as BigT’s contention that those taking supplements may be less healthy:
I’m not claiming the above studies are conclusive or can’t be picked apart. That’s true for all studies. I’m merely showing that the claim that there no studies showing harm from doses found in multis is incorrect.
It is verycommon for vitamin/multivitamin pills to contain hundreds or thousands of times the RDA. And the customer may well figure that if one pill is good to handle “stress”, why not two or three?
I’m just not seeing a good health justification for mindlessly cramming down daily doses of vitamins, which at best are a waste of money for the great majority of people and at worst (see previously provided examples) could be causing real harm.
The Mayo Clinic seems to use a Vitamin A conversion rate of .30 mg per IU. So the linked multis have 3500IU * .30 * 29% Beta Carotene = 304.5 mg Beta Carotene. Not 100 by my calc. So it’s even worse than you indicated, according to my half-assed calculation.
The folic acid dosages checked out.
I was dubious about the last link though, as the study addressed “US women aged 55 to 69 found that those who used multivitamins and a range of other supplements”. Emphasis added. Yeah, megadosing can be possibly harmful, but it’s not clear whether multis or megas are driving the effect.
2 out of 3 isn’t bad though and you’ve demonstrated your point methinks.
Methinks your brush is too broad. And 8000% RDA, while ill-advised, is ~80x RDA, not 800 or 8000x RDA. I don’t doubt that vitamins can be abused, but so can paint thinner, automotive oil or any number of useful products.
X-Ray’s argument still holds though: we can’t rule out that multis are leading to actual (if probably modest) harm rather than modest benefit.
Given the large sample size (40,000 US women) ISTM that vitamin dosage was probably self-reported, not double blind. So while “…women who used multivitamins regularly had a 2.4 percentage point increased risk of dying during the follow-up…” it seems fair to question whether some of that subsample was taking other vitamin pills as well. The cite was a newspaper article, so it isn’t clear. Careful analysis could control for this, but it’s not clear whether the original authors exerted such effort.
No, it is not an RCT, but it is pretty good data. Not perfect, but pretty good.
The tools used to assess diet and supplents used seem, on the whole, to pretty well based. (Decide for yourself if this bothers you: “Although the dietary supplement portion of the Food Frequency Questionnaire used in the study was not validated separately,19 an evaluation20 with similar instruments has reported validity correlations of approximately 0.8.”)
Indeed those who chose supplements were more likely at baseline to be healthier, not only, as already pointed out, better educated and with better exercise and nutrition habits, but "… a lower prevalence of diabetes mellitus, high blood pressure, and smoking status; a lower BMI and waist to hip ratio … " This is a group who should be having substantially lower morbidity and mortality in the follow up period.
From the actual paper:
Limitations to conclusive interpretations are duly noted by the authors; this does not prove conclusively that multis cause harm. But there is reason to believe they might.
I am providing a link to the actual study that was discussed in the newpaper article cited so that readers can evaluate the strength of the data for themselves. As noted, it was not a randomized cotrolled study (RCT); it was bit more however than “a survey.” It was a prospective longitudinal cohort study to be precise.
So to summarize our discussion so far:
-we need vitamins
-we absorb supplements when manufacturers add them
-people on a limited budget don’t eat as well as those who aren’t
-it’s harmful to take certain vitamins above recommended amounts.
I would say that some people on a limited budget don’t eat as well as some and some people on a high budget do not. Statistically the odds of poor eating go up with a low budget but healthy nutrition can be had on one. Moreover the “not eating well” is more the lack of vegetables, fruits, and whole grain, the excess of refined carbs and both trans and saturated fats, and the excess of sodium, than a deficiency of vitamins.
And the risk of harm seems to vary with the vitamin, the level, and the form. Causing harm from vitamins above recommended amounts in food sources does not happen with any standard American diet.
The idea that healthy could be had for less discounts the reality that food choice is one of the few things a person on a budget has any control over. Not everybody wants to start or end their day with an apple and a handful of kale.
Ok, one of X-ray’s studies has turned out to be about megas, one still applies and then we have the survey. Thanks for providing the original.
To the extent that multi-V users have better education etc., that should be picked up mostly or at least partly by the control variables.
My concern remains. If I am reading Table 4 properly, the authors ran ~9 regressions, each with a single vitamin pill and various control variables. But surely multi-V use is correlated with megadosing with other vitamins. So the reported multi-V correlation may be spuriously reflecting higher megadosage use. Or visa versa I guess. Then again, I may be misinterpreting the paper.
Against that, we have a study saying multi-Vs reduce cancer risk among men - which admittedly might suffer from an analogous problem.
The next step would be to find a good survey of the literature on multi-Vs. Given the preceding though, I’m inclined to agree with** DSeid** and** x-ray vision**: multivitamins might do modest harm: we can’t rule that out. I’m hypothesizing that if they were a true menace (or panacea) that we’d know about it by now.
4a. If mega-Vs were a true menace, I’d think they would be associated with acute liver failure. (Maybe they are and I didn’t google hard enough.) Of course, “It doesn’t kill you fast,” isn’t the strongest recommendation.
I ordered a bottle of the stuff at Amazon about an hour before I read X-Ray’s 3-study post. Curses, foiled again! I had tried to increase my multi-V intake to 1 per day (as opposed to “When I feel like”) after reading Kevin Drum’s post last year. Now I’m thinking of scaling back to my previous usage. Until the bottle runs out.
Apples are fine snacks though the Red Delicious variety is best avoided. Unless you are fond of wooden blocks. Bananas are cheap.
I’d avoid gobbling supplements from any company with the word “Nature” or “Natural” in it. Just on general principles.
-true vitamin deficiencies manifesting as disease are rare in the U.S.
-a multivitamin is not a substitute for good nutrition
-megadoses of some vitamins may be harmful
-beware of unproven/false health claims for supplements
-eating well is feasible on limited budgets