Some members of the life-extension fraternity, such as Ray Kurzweil – whom I enjoy reading – have been challenged head-on over the argument that taking vitamin supplements does any good in terms of enhancing overall health or warding off cancer. Here’s today’s lead story in the Daily Telegraph website:
Popular vitamin supplements taken by millions of people in the hope of improving their health may do no good and could increase the risk of a premature death, researchers report today.
They warn healthy people who take antioxidant supplements, including vitamins A and E, to try to keep diseases such as cancer at bay that they are interfering with their natural body defences and may be increasing their risk of an early death by up to 16 per cent.
Researchers at Copenhagen University carried out a review of 67 studies on 230,000 healthy people and found “no convincing evidence” that any of the antioxidants helped to prolong life expectancy. But some “increased mortality”.
The story concludes with the usual call for sale of vitamin supplements to be controlled, blah, blah. Even so, the supplement advocates have just been given a serious challenge. What I do not quite understand, however, is why ‘natural’ vitamins are okay but artificial ones are not. The article does not really explain this point.
Full disclosure: I take multi-vitamins occasionally if I feel under the weather and I have felt slightly better as a result. That is not, of course, proof that they are going to seriously add years to my life.
Well, I think the general argument deployed on Number Watch will apply here.
The “research” is almost certainly junk. It’s a meta-study (67 other studies surveyed), which are notoriously unreliable. A Trojan Number of 230,000 is deployed to impress us. The reported relative risks are all very low (1.16, 1.07, 1.04) and one can be very confident (95% confident in fact, ho ho) that this is just a data dredge and the figures are just statisical noise.
Of course there’s no difference between a “natural” and an “artificial” vitamin. They’re the same chemical.
Epidemiologists are a major enemy of lovers of freedom. They provide a great proportion of the grist to the statist mill.
It gets worse JP. Our government has also concluded that anti-asthma measures such as flash vacs and anti-dust sheets are pointless.
6.5m Brits have asthma. My brother does. My sister in law does. It can kill you. So this is a useful hint about pointless prophelaxia?
Not according to the Asthma Society. They took a “if it works for you” line which I found refreshing.
Source: BBC News 24 this morning.
Ian, Epidemiologists are just meant to study the science of the things, and as such they are the enemy only of quacks.
What the state does with that information, or some epidemiologists getting political, is the real problem.
Vitamin supplements are only of use if your diet is inadequate. And yes, it is possible to have too much of a good thing. You can die from overdoses of vitamins. You can die from drinking too much water.
Another good blog which talks about the science behind trials etc (it’s not libertarian though) is:
http://www.sciencebasedmedicine.org
Nick, the “if it works for you” line is meant to be comforting. Whether it’s useful is another matter entirely.
Gib, if something helps you stop an asthma attack, I would say it is more than just “comforting”.
Jonathan,
“Natural” vs. “artificial” is usually bollocks, but it does make sense in this context. It’s really difficult to take vitamins from ordinary food in the quantities necessary to do you serious harm. Not so if you buy concentrated supplements.
If you take vitamin supplements at usual doses and you have an ordinary balanced diet, then you are probably just wasting your money. If you take water-soluble ones, such as vitamin C, to vast excess you won’t do anything significant as far as anyone knows for sure. But using supplements it is fairly easy to get up to acute toxicity with A & D & some of the Bs, so we probably shouldn’t be surprised that long-term dosages aren’t good for you.
Given “no difference” I’m biased in favor of the natural vitamin, i.e. eating vegetables.
Maybe, but I don’t have a chemical laboratory, and I can’t prove it.
All these food suplements are 99% quackery and snake oil. Still people should be free to consume as much as they want, even if it causes some harm.
Yes, if it actually does. But thinking it does, and it actually helping are not the same, and that’s what studies are for. To separate the placebo or anecdote from the real effect.
Jacob, most vitamins are quite simple molecules, and it doesn’t matter whether they are synthesised in a factory or in a plant. You don’t need a lab to prove it, just take the word of a chemist, ie me.
Of course, eating the veg may give you other benefits, I wouldn’t dispute that at all.
Jonathan, you took vitamin supplements and “felt” better.
Well you would, wouldn’t you. That is how homeopathy works.
Whether you were better is quite another matter.
On the radio this morning they qualified that it was not the everyday multivitamin taking they were talking about but people who had cancer say and were taking massive doses to help ‘cure’ the cancer.
Feeling a tad under the weather, I asked my excellent doctor in Frankfurt…deutsche Grundlichkeit… if I should supplement my good, somewhat excessive diet with vitamins. He said that he had never met any reasonably healthy person to whom such additives would be advantageous. I immediately felt much better.
Regarding the artificial/natural thing, what I have read is that in order for vitamins and such like to be effectively used by the body, the body needs other substances which help the process along, and these substances are found in natural foods, but not in vitamin tablets. As such, it’s better to get vitamins from foods rather than tablets, because they can be used more efficiently.
I’m not a biologist or a chemist though, so whether that’s true or not, I am unsure. Perhaps someone here can enlighten us?
Sorry, but I can’t quite let this one pass:
The “research” is almost certainly junk. It’s a meta-study (67 other studies surveyed), which are notoriously unreliable.
Let’s see. It’s not a meta-study. It’s not even a meta-analysis, which is perhaps what you meant. It is a systematic review. Systematic reviews are the foundation of modern evidence-based medicine. Meta-analysis, while less useful than SRs, are still of course far more useful than individual studies, even individual high-quality studies like RCTs. To suggest that “meta-studies” are notoriously unreliable, and not provide evidence for that assertion is peculiar. Certainly you can produce a low quality SR and a low quality MA, but in general they are far more useful than individual studies.
It so happens that this systematic review is published by the Cochrane Collaboration, which is one of, if not the, most respected sources of high quality systematic reviews of medical research papers. This is likely to be about as high quality as evidence gets on this particular topic.
A Trojan Number of 230,000 is deployed to impress us.
What the hell is a trojan number? A number that looks friendly, like ’34’, but is actually filled with evil numbers like ’99’? It’s a number that’s not deployed to impress us. The Telegraph may have quoted this number to impress us, but it’s deployed to indicate the number of individuals in the combined trials, which is a fairly key number in any systematic review.
The reported relative risks are all very low (1.16, 1.07, 1.04) and one can be very confident (95% confident in fact, ho ho)
The relative risks are low, but the study does not say otherwise and does not conclude that vitamin pills are dangerous. If the Telegraph suggests otherwise, that’s its failing. Your attempt at humour fails, except to indicate you don’t know the difference between 95% confidence and a 95% confidence interval, the second being relevant to the study, and the former being of no relevance at all.
that this is just a data dredge and the figures are just statisical noise.
A systematic review differs from a ‘data dredge’ in that it declares openly the criteria for including a given study in its review, and then seeks to comprehensively include all and only studies meeting those criteria. Since you seem unaware of this I’d hazard a guess that you don’t know what you are talking about, and didn’t bother to read even the freely available ‘plain English’ summary of the original systematic review. But perhaps you’d like to now, it is here:
(Link)
Of course there’s no difference between a “natural” and an “artificial” vitamin. They’re the same chemical.
You are assuming the supplement pills contain only the labelled vitamins and inert substances. It is possible that the vitamins in supplement pills de-nature faster than thought. It is possible the inert fillers used have some clinical effect. It is possible that manufacturers don’t put in as much active ingredient as they claim. But, assuming the pills contain what they say they do there may be a difference between consuming vitamins in concentrated doses, and consuming them mixed with other food in lower doses. We know the effects of many drugs are changed depending on what food a patient is (or isn’t) eating with it.
Epidemiologists are a major enemy of lovers of freedom.
Interesting. I’m not aware of any well-designed studies into the political views of Epidemiologists. I would be interest in the evidence behind this assertion.
They provide a great proportion of the grist to the statist mill.
They may well do – certainly it’s fashionable for politicians to use medical research figures to support illiberal laws. Are you blaming the researchers for this? Should we simply stop doing clinical research for a decade or two until politicians improve?
So, J…as Eric Morcombe said, ‘There’s really no answer to that.’
To echo what Guy said, I imagine the study is reported as “applying only to synthetic supplements and not to vitamins that occur naturally in vegetables and fruit” because the former is much easier to isolate and study.
Everyone gets vitamins from vegetables and fruits, pretty much, so it’s very hard to control against.
It is, as always, utterly bollocks stupidity that the Telegraph references the study with no link to it; if it’s really not available online, they should at least say so outright. (And, honestly, science reporting is typically so bad I don’t trust it, on principle and in general.)
My strong suspicion is that the increased risk is associated with megadoses of the vitamins in question, not the levels found in, say, a daily multivitamin. But god forbid the mention of dosages occur at all, in a report on it!
(A little bit of digging suggests that the study in question is apparently a year old meta-study. If there’s some other Copenhagen study that’s a review of 67 studies, that isn’t a year old, that’s an amazing coincidence, isn’t it?
But if it is that study, why is the Telegraph’s “medical correspondent” reporting it as if it’s breaking news, and without even noting last year’s criticisms?
Incompetence?)
J,
What’s a “systematic review”, as opposed to a meta-study? A meta-study is a study of other studies, in which the results are statistically combined. I can see how you might review other studies without combining their results and make a distinction on that basis, but I’m not sure if this is what you’re doing. I haven’t seen the original paper either, so I can’t tell if that’s the case.
A trojan number is the claim of a large sample size in circumstances when the actual effective sample is much smaller. For example, if you track ten thousand people, and during the study interval ten of them die, and you look at the statistics of causes of death to derive your mortality figures, then the effective sample size is only 10. The large sample size number quoted giving the impression of statistical accuracy is known as a trojan number. I’m not sure the term applies in this case, since people who didn’t die still contribute to a simple mortality calculation, but it depends how it was done.
The point about low relative risks is for those cases where a small but statistically significant odds ratio has been used to claim something is dangerous. In this case, the researchers themselves say that no detrimental effect was found either, and draw no conclusion from the small mortality increase, but you’ll note the newspaper headline does.
A data dredge also openly declares the criteria for including a study in its review – the difference is that the criteria are generally defined after the result has been found. Calling something a ‘data dredge’ only implies that many possible hypotheses have been previously checked and rejected before the ones that show a correlation get published. The rejected hypotheses affect the significance of the correlations found, but are usually not reported or accounted for. Since the result reported was in fact negative, this is probably not an issue. This study would only be a data dredge if the authors considered multiple selection criteria for studies that all showed correlation before reporting on one that showed no result. There’s no evidence for that.
Interaction effects are all possible, but none of them constitute a difference between natural and artificial vitamins. Is vitamin C in apples different to the vitamin C in oranges? They’re clearly mixed with different chemicals, and could in principle have a different effect, but it has nothing to do with artificiality.
As noted above, the only real difference is dose.
I don’t think epidemiologists are enemies of freedom so much as common tools of the enemies of freedom. Bad epidemiologists have a tendency to seek out and emphasise evidence of unexpected harm as a way of getting more papers published, more media publicity and hence grants. (I have heard that it is most common with inexperienced researchers just after completing their PhD who use data dredging as a quick way to get a few easy papers under their belt, before they settle down to a more responsible approach when their career is established and their tenure more secure.) Apparent evidence that X is bad for you is then used by statists as an argument to ban it. It’s not unknown for the two to be the same, with the epidemiologists who have found the evidence being prominent campaigners for its ban.
But I think this particular set of epidemiologists are on the side of the angels. They’re saying that there’s no solid evidence that vitamin supplements have any effect at all, and it’s only the newspaper desperate for some more drama that is spinning it that way.
Even for simple molecules there can be important differences between those produced synthetically and those from natural sources due to the issue of chirality, vitamin E is an example.
I have been taking multi-vitamins for my entire life – almost 49 years. Whenever I run out, I get sick – usually a cold. I rarely get sick. There were many years where I did not miss one day of school. I am also a statistician and although my study has a trojan number of 1, I believe in vitamins.
Isn’t this the tail end of the anti-oxidant fad?
If I recall correctly there are all these nasty free-radicals sloshing round in our bodies causing all sorts of mayhem and the cure was taking lots of anti-oxidants to combat them.
Then they did a big trial feeding anti-oxidants/placebos to a load of cancer patients which they promply abandoned when the patients on the vitamins started keeling over at a greater rate than those on the placebo.
The, (off the cuff), theory behind this seeming anomaly was that by filling ourselves up with anti-oxidants we were somehow interfering with our bodies natural defences.
I’m patiently waiting for nanny to abandon her five-a-day propagand and apologise for misleading us all.
I heard the attack on B.B.C. Radio 4 this morning.
And then I went out and bought some vitamin supplements – something I have not done in quite some time.
J states: The relative risks are low, but the study does not say otherwise and does not conclude that vitamin pills are dangerous.
But in the link he provided, we find the following:
Overall, the antioxidant supplements had no significant effect on mortality in a random-effects meta-analysis (relative risk [RR] 1.02, 95% confidence interval [CI] 0.99 to 1.06), but significantly increased mortality in a fixed-effect model (RR 1.04, 95% CI 1.02 to 1.06).
and:
In the trials with a low risk of bias, the antioxidant supplements significantly increased mortality (RR 1.05, 95% CI 1.02 to 1.08). When the different antioxidants were assessed separately, analyses including trials with a low risk of bias and excluding selenium trials found significantly increased mortality by vitamin A (RR 1.16, 95% CI 1.10 to 1.24), beta-carotene (RR 1.07, 95% CI 1.02 to 1.11), and vitamin E (RR 1.04, 95% CI 1.01 to 1.07), but no significant detrimental effect of vitamin C (RR 1.06, 95% CI 0.94 to 1.20). Low-bias risk trials on selenium found no significant effect on mortality (RR 0.91, 95% CI 0.76 to 1.09).
Now, I reckon that the term “significantly increased mortality” would portray to most people the notion that if they take the substance under consideration they are much more likely to take an early celestial bath.
I am somewhat confused by some of the numbers. Antioxidants significantly increase mortality with an RR of 1.05, so to vitamin E, RR 1.04) but not vitamin C at an RR of 1.06 and selenium has no significant effect with an RR of 0.91.
Well, if it’s in the Daily Mail…
it must be bollocks.
There are supplements and then there are supplements. Some people need to eat more. Others, for whatever reason, need a way to get micronutrients without eating more. In my own case: I’ve been fighting my waistline for some time now. Spending most of the working day sitting in a car is not helpful. That means needing to eat perhaps not as much as I otherwise might. And the vitamins still need to come from somewhere.
And then, the weightlifting world is full of all manner of idiocy about supplements. Look inside “Muscled and Witless” some time: the ads, and the rest of the magazine (which are usually just paraphrases of the ads) make all sorts of claims about how the supplements are damn near magical.
Nope. The ones that do any good are basically food. Protein powder is a way to get the protein from a quart of milk without having to drink a quart of milk. Well, unless it’s derived from eggs (which are also a good protein) or soybeans (which are okay if you don’t mind mediocre absorbtion and surplus estrogen). Creatine is, again, food. One teaspoon will also be found in 6-8 ounces of tuna or beef. (Canning and cooking may destroy it, but sushi is probably good for you too) and so on.
The exceptions being the drugs that manipulate hormone levels. Either they are testosterone, or they mimic testosterone, or they claim to do so. The ones that actually work are all controlled substances in the US. The others all carry the “This product is not meant to diagnose or treat any disease or condition” disclaimed because they haven’t been tested and would turn out to be placebos if they were.
Disclaimer: I take multivitamins. I occasionally use protein powder. I used to use creatine, but the place where I bought it shut down and I couldn’t be bothered to find a new one.
One interesting (I thought) bit of folklore about vitamin toxicity: it’s common for books on hunting or wilderness survival to caution against eating organ meat (especially liver) from arctic mammals. I know that Polar Bears were mentioned by name and I think some sort of seals were as well. The livers contained deadly levels of vitamin A.
Gosh!….49 years without a proper meal!
I highly doubt the findings of this report. It can only do good if taken moderately. I would agree that abuse could be detrimental to your health
People should first of all eat properly, and then if they don’t take what they need, as vitamins and minerals, from their food, they should think about taking supplements. I believe that nature has everything we need.
_____________________
buy legal steroids