Doubts about anti-oxidant supplements

The new public health site, The Pump Handle (TPH), continues to produce top notch posts. The latest is by David Michaels, Professor and Associate Chairman in the Department of Environmental and Occupational Health, the George Washington University School of Public Health and Health Services (and former Assistant Secretary for Heath at the Department of Energy in the Clinton administration). It's about the dietary supplement industry, or as I prefer to call it, the quackery in a capsule biz. One of their favorite products is the anti-oxidant cancer preventative scam. There is quite a body of evidence that anti-oxidants don't prevent cancer and may even result in increased cancer risk.

More of that evidence came out this week in the Journal of the American Medical Association (JAMA; abstract):

The JAMA paper examined the results of 68 randomized trials of dietary supplements, involving 232,000 participants. The authors found that, in the trials with strongest methods (47 trials involving 180,938 subjects), antioxidants were associated with significantly increased mortality risk. This is consistent with the evidence to date. There have been several clinical trials that found antioxidants, especially beta-carotene, increased lung cancer risk; there is little human evidence that they reduce cancer risk. The JAMA paper is a systematic compilation and synthesis of data from previously published papers. (David Michaels, The Pump Handle)

The antioxidants are (beta carotene, vitamin A, and vitamin E). You need vitamins, but the megadoses in dietary supplements you don't need. Mostly you eliminate them in your urine, making American urine some of the most expensive in the world. But the data increasingly are showing they also increase mortality.

I'd say, "no sooner was this news released, when . . . " but I can't even say that. Before the article appeared the Council for Responsible Nutrition (1984 Newspeak for Irresponsible Health Information) had issued a press release asserting the study was "muddled" and scientifically unsound. The net effect, as Michaels observes, was to blunt the impact of the paper and submerge it in noise.

Michaels analyzes the tactic in his excellent post at TPH, a venue that becoming one of the best public health sites in the blogosphere. It's new, but it deserves greater recognition.

[Full disclosure: the Reveres and colleagues are TPH blogparents.]

Categories

More like this

being able to cross-post comments, now that would be web 3.0. I did post this at The Pump Handle (TPH). First post there, but I do think I did a good job of posting. So...

One of my teachers thought a great job would be a Epi guru (M.D. or not) who could respond to the newest health news with a spinner or temperature gauge behind them. They could give a quick summary and then turn to their gauge and say "This research ranks above coffee talk for soccer moms, but below ever mentioning to anyone else" or some such sillyness. The top of the gauge could be reserved for "freak out and call your doctor immediately" and the bottom "call your congressman and complain about why this was funded".

I'm sure like all things it would get perverted to the dollar or personal agenda, but at least it would be better than hearing another story about how they found the gene that causes farts or some other such sillyness.

Also, I give a nod to Dr. Shao and the diet supplement industry association for their own pseudo-type 3 error[sic]. JAMA does a phenomenal job of packaging their results with canned video interview and 'what does this mean' synopsis so the MSM will air their articles, for a fee of course. JAMA, imo, does deserve some noise for claiming that the 95% interval in 180,000 participants is above 1.0. No shit. Gimme a 180,000 participants and I'll show you a significant relationship between booger picking and all-cause mortality. No, they don't need noise for pointing out the obvious, but then again, is the noise from Dr. Shao et al. really going to stop the message from getting to his customers? No, they aren't going to listen anyways.

A former colleague who developed a positively prodigious green tea habit to prevent cancer died of cancer a few years ago.

One should always be suspicious of those who tell you you need to be taking some medication/supplement/diet when you are perfectly healthy. It's a sure sign of quackery.

Supplementation in the absence of a specific nutritional deficit or evidence of need (folic acid in pregnancy, B12 in elderly/vegans) is quackery. Anyone who promotes it is a quack. Healthy people getting adequate nutrition do not need 800x the daily RDA of vitamin E.

Also, notice the divide between the safety of the aqueous versus lipid molecules? Vitamin C and Selenium - safe but not particularly effective. The others? Not so much. Not too surprising there. That more than anything should be a clue here. If you can't just piss it out, you shouldn't be ODing on it.

Ok, so should one take no vitamin/mineral supplements, and then add them one at a time as advised/necessary? Seems expensive, and complicated, compared to taking a multi. (Not to mention, you're even more likely to get high doses in supplements with just 1-3 nutrients in them.)

I'm sorry but chemo and radiation therapy, the only legally permitted cancer therapy in US hospitals, kills far more people than vitamin pills.

There is considerable evidence for the notion that our industrialized food and thus the American diet is seriously deficient in micronutrients, including many so called 'anti-oxidants'. Little has been done to evaluate the health effects of this situation. It would make sense to try supplementing some of these.

I really can't believe that a bunch of epidemiologists are going to let this hokey study in JAMA go by without criticism.

Another relevant link from LEF.org:

http://www.lef.org/magazine/mag2006/jun2006_cover_media_01.htm

And what about the dangers of pharmaceutical drugs?

"The recent explosion of FDA-approved killer drugs suggests that the agency's overzealous approach to cherry products might be better directed at pharmaceutical manufacturers whose products are one of the leading causes of death in America. Adverse drug reactions cause more than 100,000 fatalities each year and send a million and a half people to the hospital annually.30 Those are the documented cases; the actual number of people who become sick, hospitalized, or die from drugs is unknown.31"

That's from:

http://www.lef.org/magazine/mag2006/mar2006_cover_cherries_01.htm

I won't be convinced to give up my supplements until I hear a reasoned scientific argument debating the claims of the Life Extension Foundation - a very credible not-for-profit organization with many physician members.

By Jon Schultz (not verified) on 01 Mar 2007 #permalink

Ron, Jon: I don't think the argument that "they" are worse than "we" are (I am using shorthand here for conventional medicine and dietary supplements advocates) is a very good one, nor is the comparison of a desperate treatment for a terminal disease with an elective dietary choice. Ron, if you have a substantive critique of the JAMA meta-analysis, by all means let's hear it. So far all I've heard is you think it is hokey. Why? (other than you don't like the answer.) This isn't about the dangerous and often ineffective products of Big Pharma. We've done that here and many people have done it elsewhere (see the blog Health Care Renewal on our blogroll, for example). The dietary supplements industry is full of quackery, as is Big Pharma. Neither should be immune to criticism.

Jon: I don't know that the JAMA article is relevant to any claims made by the LEF. To the extent it is, then it seems the burden is on you/LEF to say why the JAMA article isn't a reasoned scientific argument. If the JAMA article isn't relevant to any LEF claims then you are off topic and using this thread to promote your own interests and I would ask you not to do this. I don't care if you give up your supplements or not. I wasn't trying to convince you to do one thing or another. I was pointing out something that was happening, the well worn tactic of manufacturing doubt via press release by a trade association. Big Pharma does the same thing as do chemical companies when a scientific result appears that is inconvenient for them.

I was responding to your "quackery in a capsule" statement. This may be a fair characterization in some cases--ultimately the diet supplement industry is in the same boat as Big Pharma--but it obscures the importance of the underlying issue.

I don't have access to the full JAMA article, but judging from the abstract, yes I would call the study 'hokey', meaning a questionable methodology is used to fabricate a predetermined answer.

I think it is highly questionable to seek statistical relationships across 385 studies, done for varying purposes, using highly varied assumptions and methodologies, doses and therapeutic environments. Calling all this representative of a a single population (or 'meta-population')and a single experiment is totally unjustifiable. And to test for effect on all causes of mortality--they may have just as well pointed to the truly significant trend in the data that death is very significantly correlated with birth! It would have about the same significance

Many of the studies themselves, to judge by similar ones I have had a chance to review, are flawed in the first place. They are designed to not-find effects. The effects of 'supplements' are usually tested with no information on what they are 'supplementing', ie the diet or nutritional status of the subjects. In this case it would appear that studies are thrown together of healthy people, sick people of different kinds, people undergoing different types of therapy, etc--and in all this noise we are supposed to find something?

It looks like they did find something though, because in order to report a non-significant result, they had to divided their 'sample' ad hoc into 'low risk bias' and 'high-risk bias'! I mean, really, as we say in Mexico, no se vale.

I have been collecting these studies that seem to be proliferating over the past few months, reporting negative results for effects of supplements and herbal medicines. They all tend to commit these methodological card tricks and they completely ignore previous research results that contradict their findings There seems to be a plot afoot, no doubt to legislate to restrict availability of supplements and herbal treatments to bring under the control of the pharmaceutical companies and the medical establishment. In other words, pseudo-scientific justification to eliminate the alternatives, one of the reason's we have the sickest health system in the world--corporate profits have priority over peoples' health, not to mention scientific truth

I think I will send this over to Good Math, Bad Math so he can have a look see.

Caia: no, you shouldn't be taking single supplements instead of your multi. You should be eating a balanced diet instead of taking your multi! How's that for a revolutionary idea?

By Mathematician (not verified) on 02 Mar 2007 #permalink

Ron: The meta analysis is of randomized clinical trials. Meta analysis is a well accepted technique and is just a formal version of what you are doing when you review the literature except its assumptions are explicitly displayed, allowing someone to evaluate them, unlike the usual "review" paper on the same topic. So I don't think you can say it is "hokey" without impugning your own method (unless you don't pay attention to trials at all, which for me would be a problem).

Of course you may send it to Mark C-C at Good Math, Bad Math, but that seems rather pointless as it isn't a math problem. The question isn't whether using some technique like inverse variance weighting of effect measures is appropriate or not (BTW, Mark C-C is not a statistician, he is a computer scientist so I don't know if he even has any opinions about the mathematics).

What you seem to be saying is that you don't believe the JAMA study. Period. Fair enough. Just tell me why. I'm sure there is room for disagreement among experts, which is normal in science.

BTW, the big dietary supplement outfits are owned by Big Pharma. They have found many ways to practice their quackery, whether it is supplements or "conventional" drugs.

Come on, revere. You were doing a lot more than "pointing out... the well worn tactic of manufacturing doubt..." You said, "It's about the dietary supplement industry, or as I prefer to call it, the quackery in a capsule biz." That clearly implies what I think is the ridiculous argument that all recommendations that people take dietary supplements is quackery. If you didn't mean that then you should retract that statement. You are correct, however, that my bringing up the alleged dangers of pharmaceutical drugs was irrelevant.

I won't argue that there is no fraud and quackery in the dietary supplement industry. I agree that there is, but I respect the Life Extension Foundation, and trust the quality of the particular products they distribute, and on their general recommendations I do take what the FDA and/or many doctors would categorize as megadoses not only of antioxidants but other categories of supplements as well (although they also recommend that people only take supplements under a physician's guidance - the problem is finding a local one who is aware of their work). I don't know that this is a wise course of action - it's simply my best guess based on the arguments I have seen. But those arguments, as presented in LE Magazine and in the LEF protocols - both available on their website and a sample of which I provided in the links I posted - are a lot more convincing than blanket statements such as yours and some others here which clearly imply that all "megadose" supplement-taking is foolishness.

If LEF publishes a rebuttal to the JAMA paper you cited - which probably wouldn't be for some time - I will post a link to it. They do publish such rebuttals, or occasionally mail them out to members such as one about the recent widely-reported negative study on DHEA.

If any event if you are going to indict the entire "dietary supplement industry" as quackery, as again you clearly implied, then that includes LEF and it is therefore not irrelevant for me to bring them up as an example of MDs and other people who recommend megadoses of various supplements but do not seem to be practicing bad medicine. If I am wrong about that then I would like to see a good article that convinces me of such, and not just a few flippant statements.

By Jon Schultz (not verified) on 02 Mar 2007 #permalink

Mathematician: That's fine if you can get a 'balanced diet' (i.e., quality food). Growing your own and buying fresh from local organic growers will probably let you do that. But if you are stuck with the normal American faire, (some 30% now from fast-food restaurants)then you are a long way from getting many essential micronutrients that a 'balanced diet' should give you. Besides, 'balanced diet' is a meaningless term: what we should be looking for is an 'optimal diet.'

Revere: Meta-analysis isn't so well viewed by all. I need to look at the paper but I am familiar with what passes for 'randomized trial' studies. Magic words that often cover mere hocus pocus. They give the illusion that things are 'controlled' when there are so many hidden assumptions built into the models that the the facade of 'scientific method' is really laughable. And tweaking the design will easily guarantee that 'no significant result' is inevitable. The main assumption is that all of the 'subjects' in the trial are equivalent, yet, usually, no examination is made of nutritional status or diet. Sometimes medical histories are taken and sometime note is made of ongoing therapies (though usually not of alternative therapies), but for the most part the group is considered as a 'representative sample' of a supposedly homogenous population (i.e. homogenous on all relavant parameters except the variables being measure) This is an erroneous assumption.

Other common errors are not looking at the form of the supplement. Many Vit E studies evaluate only d-alpha tocopheral rather than a mixture of tocopherols and tocotrienols. Optimum Vit E levels also depend on levels of other kinds of nutrients in the diet such as fatty acids. Oprimal diet involves a relative set of nutrients that interact. You can't just evaluate the effect of 400 mg a day of d-alpha tocopherol without taking that fact into account. You are 'randomizing' some things while ignoring others.

The insistance of the research establishment on 'randomized trial' as the only scientifically valid evaluation of nutrient composition(as if food were an antibiotic or something) begs the issues and sets up a situation in which 'supplements', evaluated in isolation from whole diet and ecology, will never show 'significant effects'. So no, I don't believe the study, mainly because the studies it is 'meta-analyzing' contain hidden assumptions that predetermine the results.

I bet you a good bottle of wine that Markcc will have something critical to say about the statistical methods of the JAMA study.

Ron: nobody has to get 30% of their food from fast food joints. Anyone who can afford to do that and buy supplements can afford to eat real food, and it doesn't have to be home grown, organic, or expensive.

By Mathematician (not verified) on 02 Mar 2007 #permalink

30 might not be just the right number but its up there, I can't put my figure on the statistics I saw recently but I'll keep looking. To give you an idea though, Schlosser writes

"In 1970 Americans spent about $6 billion on fast food; in 2001 they spent more that $110 billion. Americans now spend more money on fast food than on higher education, personal computers, computer software or new cars. They spend more on fast food than on movies, books, magazines newspapers videos and recorded music---combined." He also says that half the money used to buy foodin America is spent in restaurants, mostly fast food restaurants. (from the Introduction to 'Fast Food Nation' by Eric Schlosser)

Ron: You do meta-analysis every time you review the literature. If you are talking about how one summarizes the effect measures across studies, that is something we can argue about. Like everything in science, people have different opinions. Please look up the definition of a randomized clinical trial and then we can talk about it.

Feel free to ask MarkCC if you wish. My guess is he will decline because he is not a statistician nor an epidemiologist. Just because it uses mathemaics doesn't make any mathematician conversant with the methods, which go beyond algorithms. But do as you see fit. Since it is a subject I know something about (I teach it) I am glad to discuss it with you as well if you have specific questions. Finding fault with a study isn't hard. I can do it as easily in the studies that show benefit from supplements, as well. That is hardly the point, however. We need to look at the evidence and make a judgment based on it and our own priors (speaking in a Bayesian mode). I would guess the evidence would have to be very overwhelming for you to change your mind, since you seem to have made it up without even seeing the JAMA study.

Vitamins are not harmless, especially fat soluble ones (A, D, E and K). It is quite plausible to me that they are not beneficial in megadoses since they are genuine biological compounds and we have evolved to consume them in physiologic quantities typical of a feasible human diet from natural foodstuffs. If you want to consume them in huge quantities, go ahead.

But as my father used to say about diets, everything in moderation. Still good advice as far as I am concerned.

Jon: Please read what I said carefully. The dietary supplement business (now much of it owned by Big Pharma) is largely quackery IMO, based on unsubstantiated claims and implied promises. The use of dietary supplements by individuals is not quackery. It is a personal choice. If done wisely it is usually harmless or possibly beneficial. If done thoughtlessly it can be harmful.

You have made your position clear. I have tried to make mine clear. Let's call it a day.

revere: I don't see why it plausible that vitamins are not beneficial in megadoses since they are "genuine biological compounds..." - as opposed to saying that anything we are exposed to in our highly technologically-advanced world, which people were not exposed to several hundred or several thousand years ago, is plausibly not beneficial because we evolved out of an environment which didn't include them. And if that is the case then we should of course go back to being cave(wo)men...

By Jon Schultz (not verified) on 02 Mar 2007 #permalink

revere: Thanks for clarifying your position on the "quackery" statement.

By Jon Schultz (not verified) on 02 Mar 2007 #permalink

Ron: Thanks for your input, I'm in agreement with you. Wanted to post yesterday but I'm not the medical intellect that this blog attracts and most would have foo-fooed me away as nonsense.

revere: While my respect for you has grown I'd ask you to put as much fervor into really studying reliable benefits that supplements have produced. For me, after dozens of disapointing visits to medical doctors I turned to a Naturopathic Doctor. This simple act turned my health around after five appointments. Not being able to continue paying the fees for his services, and having a fairly balanced idea of the direction to go in, I launched off on self study spanning over eight years now. Through trial and error, paying attention to how the body reacted and a load of reading, my efforts have paid off in helping me attain better health. That and better eating habits.

Mathematician and revere: This is why caia should keep taking supplements, "industrialized food and thus the American diet is seriously deficient in micronutrients". If I were to guess I'd say probably 90% of the population is malnourished in the U.S.

caia: Keep taking your supplements, please.

While speaking about industrialized food, please be kind and bear with me on this, it is one small example of nutrition's missing in our foods. Drifting off to a study done on the benefits of flavanols, an antioxidant found in cocoa: Hagen Schroeter of Mars Inc., the candy company, paid for part of this research done by Norman Hollenberg of Harvard Medical School, and he said he found health benefits in the Cuna Indian tribe in Panama because they drank cocoa exclusively.

He said he found that when tribe members move to cities, their blood pressure rises. A major difference is the consumption of their own prepared cocoa, which is high in flavanols. In native areas, that is all they drink; in cities they adopt the local diet. In addition to having low blood pressure, Hollenberg said, there are no reports of dementia among the native Cuna.

You all can argue that moving to a city would change anyone's blood pressure, that's not the point I'm striving to make. The point is drinking their own prepared cocoa, not filled with crap from being processed.

And please don't launch off on the Mars candy paying for part of it, I only included it so you'd have a clearer picture of the article.

I don't know anyone who recommends 'megadoses' of the fat soluble vitamins. They accumulate and are toxic, especially vitamin A. As I said, we are looking for optimal levels, and that seems unlikely to mean megadoses in most cases.

OK Revere, I will do my homework and we can analyze this. This is actually a very important topic (validity of randomized clinical trials) that goes far beyond the supplement issue.

Mathematician:

I suppose I should have seen that comment coming, since I was foolish enough to ask a question about nutrition on the tubez. I do, in fact, eat rather well. However, I have at least three different reasons I require some nutritional supplementation. The simplest is that, as a woman of childbearing age, when I don't take iron (or eat fortified cereals, which basically contain a pill in flake form), I show symptoms of irom deficiency. My other reasons aren't something I feel like sharing at the moment, but you'll have to trust me that they are sound reasons, not whims plucked from the air.

However, there is another reason that applies to everyone, namely, the nutritional content of food has been declining. Obviously records of foods vitamin and mineral content don't go back that far, and are spotty and not entirely reliable. But where we have records, the same food now appears to have markedly less of certain nutrients than it did 25-50 or more years ago. Some blame hybrid crops; I suspect soil depletion and lack of crop rotation has more to do with it. Either way, a balanced and healthy diet may well be insufficient for gaining sufficient nutrition.

caia: The nutritional content of food has been declining for 50-60 years. I am convinced that the preservatives in our foods of 50-60 years ago, is one of the reasons my health has not been right. That and genetics.

So, you can add the 50-60 years calculation into current physical aliments people are experiencing today.

revere: What benefit, or not, is the American Institute for Cancer Research, (AICR)?

Lea: As far as I know, they support good cancer research.

Re. the science, I'm with Ron on this one. Too many implicit assumptions, some factors controlled and others not, etc. etc. And too many vested interests that can compromise the methodology in subtle ways that are not evident from the write-up.

Really now, let's be consistent. Whenever a meta-analysis of ESP research is issued, immediately up go the cries of "Meta-analysis is not valid! You're compounding the significance of a bunch of studies that were barely getting p less than .05, that's no good!" and blah blah blah. But when out comes a meta-analysis that supports the prevailing biases (i.e. "quackery in a capsule") the only sound I hear is the mooing of sacred cows. Moo!

Libertarian rant department:

Hey everyone, ask yourself this: Who owns your mouth? Do you own your mouth, or does government or some corporation own your mouth? Who gets to tell you what you can put into your mouth? You, or the government, or some corporation?

If you start giving away control over what you can put into your mouth, next thing you know you'll be giving away control over what comes out of your mouth. Don't go there.

Now speaking as a professional geek, anyone here who disapproves of nutritional supplements is welcome to unplug their computer, pack it up in a box, and go back to using a typewriter. Ditto your office telephone with all those cool blinking lights. The entire geek industry, by which I mean the engineers who make your telephone/ computer/ online universe happen, is infested with people who gobble nutritional supplements and smart pills as if they were jelly beans. We do it to keep our brains in good shape. The difference it makes to us is typically in the range of about 10 to 20 IQ points. The difference it makes to you is that your telephone and computer keep working.

If you take away our vitamins, one of these days you're going to pick up your phone and get no dial tone, and one of these days you're going to turn on your computer and get the little Haiku that says the internet connection is down. And there won't be anyone around to fix either of them.

Mark my words.

"I don't know anyone who recommends 'megadoses' of the fat soluble vitamins."

I guess that depends on how you define "megadoses," Ron. In the first article I linked to above, William Faloon states that "the diets of more than 90% of Americans supply less than the 12 milligrams a day of vitamin E the government proclaims to be adequate..." But some doctors recommend 400 mg a day of alpha-tocopherol E (or more, if I'm not mistaken - plus other-tocopherol E) based on the results of some studies. If the FDA is saying that 12 mg is adequate - and most people don't get that much in their diet - then I imagine they probably consider 400 mg to be a "megadose." But my point, again, was that I don't see the logic in saying that such doses are plausibly not beneficial because vitamins are "genuine biological compounds." It seems more reasonable to think that because vitamins are genuine biological compounds, that there would tend to be a greater likelihood that high doses of them could possibly be beneficial as opposed to things, i.e. pharmaceutical drugs, which are not natural to our bodies.

By Jon Schultz (not verified) on 03 Mar 2007 #permalink

To add on the debate on vitamins and other (more questionable) food supplements, check out skeptalchemist.blogspot.com

There is more than vitamins out there being touted to be the ultimate anti-oxidant and immunesystem booster. Some of these products have been publicly objected to by Nobel Prize winners who investigated some of the concepts behind the creation of such proprietary formulations.

May I cite Ambrotose AO (tm)? Watch your sugars.. :)

Speaking of megadoses.

Back in the late 70s, the US RDA for vitamin C was 60 mg/day. At that time some idiot intoduced a bill that would require a prescription for any nutritional supplement that provided more than 1.5x the US RDA. That would have made a glass of orange juice, with its typical 90 mg. of vitamin C, a prescription drug.

Know what? It's all about the money.

Everyone wants to get their hands on your money.

The supplement industry does it with questionable advertising, but at least what they are doing requires voluntary consent on your part.

Big Medicine, Big Pharma, Big Insurance, and all the other Big Brothers and their bastard brethren, do it by legislation, regulation, distortion of markets, and other measures that are about as voluntary & consenting as date rape.

The role of FDA should be limited to issues of purity, potency, overt toxicity, and providing unbiased listings/links for peer-reviewed findings. If someone wants to market a pill called Active Placebo for ten bucks a dose, fine, as long as they list the ingredients, don't make overt medical claims, and the pills are exactly as described in the ingredient disclosure.

The idea that we are all children or borderline cognitively disabled and need to be protected from ourselves, is absurd and pernicious. It fosters a culture that is weak, brittle, stupefied, and incapable of dealing with significant challenges.

Anyone who laments the lack of avian flu preparedness ought to take that warning to heart.

Revere, thans for pointing this out.
This misinformation campaign was brilliantly successful and fooled me. I read the New Scientist news article (in the print edition) and thought the claim that "the trials included people who were already sick" meant they weren't double blind, and therefore the report was useless. In fact all the studies used for the report's conclusions were well blinded, so the supplement manufacturers claim is deceptive. In fact many sick people start taking supplements - eg vitamin A or carrot juice on being diagosed with breast cancer - so including the effects ontheir health is relevant.

http://www.newscientist.com/channel/health/mg19325934.000-can-vitamin-s…