"Vitamins linked with higher death risk in older women"

[Panth]

Oh gosh, my appologies! I didn't mean to come off that way! *Must proofread things before posting them!* I'm actually up to my eyeballs reviewing other research articles atm, so I think I got my wires crossed somewhere. What I *meant* to say was that the report from the BBC was a bit vague, which tends to happen when paraphrasing, no matter what the original source. My deepest appologies to anyone who might've been offended by my previous post.

Panth, I'd love to see the research article in full! I currently can't PM due to newbie status, but that should change soon. Once I have access to PM'g priviledges, I'll send you my email addy. Thanks for offering to send that out to us!

Now that I realize I've made a glorious fool of myself, I'm going to go hide in a corner now until I've found time for a nap so I can think straight again!

Oh, goodness, please don't feel bad!

I wasn't being critical of *you*, rather simply the way you phrased that particular thing. Also, I wasn't trying to single you out - you were just the last comment that illustrated what I was talking about.

I can certainly understand being overwhelmed with journal articles and the like and having a brain-fritz. I'm currently kinda like that at the moment - start of term and lots to read!

[swords & roses]

Oh, goodness, please don't feel bad!

I wasn't being critical of *you*, rather simply the way you phrased that particular thing. Also, I wasn't trying to single you out - you were just the last comment that illustrated what I was talking about.

I can certainly understand being overwhelmed with journal articles and the like and having a brain-fritz. I'm currently kinda like that at the moment - start of term and lots to read!

Fair enough! I'm at the end of grad school. It's chaos! Exciting, but chaos! My brain hurts! Lol!

[heynormy]

I have a question, and someone please correct me if I'm wrong, but was this research based solely around the consumption of iron supplements in older women?

[Panth]

Fair enough! I'm at the end of grad school. It's chaos! Exciting, but chaos! My brain hurts! Lol!

Ah, I'm at the beginning of grad school. *grin* Good luck with managing the chaos!

[ZenGwen]

Correlation is not causation. I've been seeing this report all over the place and all I can do is slap my forehead at the amount of people taking it seriously.

I think it's like the diet soda effect. A lot of people, when they have a diet soda instead of a normal soda, think they can afford to give themselves a treat - perhaps a bar of chocolate, for example, which will contain more calories and fat than the original soda. Before long they're eating all kinds of crap and excusing it with diet soda, skim milk, or eating a piece of fruit or a vegetable to "make up for" treats.

I see it in sports a lot too. I go to the gym at least three times a week, and swordplay at least once a week. I see people at the gym guzzling all kinds of nutrition drinks and supplements and then doing crappy workouts - like 20 minutes of cardio, maybe a half-hearted attempt at some light weights on the machines - and thinking the supplements will make up for the lack of effort. Or the chubby geeks at swordplay who think that because they're hefting a bastard sword around for a couple hours a week, and sweating heavily under all the fencing kit, they're doing enough exercise. (I don't wish to sound mean - they're my friends, I love them, I used to be like that too, but it's really not enough. I am also a geek, but a fit geek!)

Same thing with vitamins. Take a multivitamin in the morning, think you've done something good, let your guard down about everything else you eat that day.

Sure, I take multivitamins, and drink protein shakes after workouts, and use skimmed milk. But I also don't drink any soda, I lift weights, I run regularly, I very rarely have sweets or high-sugar fruit juice, I get 6-8 servings of fruit and vegetables a day, lots of wholegrain carbs... people just need to do their research and actually take responsibility for their health, rather than relying on the idea that if they're doing some things which are healthy, their lifestyle is suddenly great for them.

Anyway. Yes. That's what I think is going on here. (And other people have mentioned it too, although I haven't taken the time to read through all the responses.) Taking a multivitamin will not make you keel over, using it as an excuse to live in an unhealthy way will.

[CherrySilver]

Correlation is not causation. I've been seeing this report all over the place and all I can do is slap my forehead at the amount of people taking it seriously.

No kidding. You'd be surprised at the number of people out there who are not capable of critical thinking. They believe what they want to believe. But, would they ever take the time to go find something out themselves? Noooooo. Unless Dr. Oz or some "authoritarian figure" tells something is good or bad, they won't buy it.

[Panth]

I have a question, and someone please correct me if I'm wrong, but was this research based solely around the consumption of iron supplements in older women?

Nope.

They took 38772 women. The average age at the start of the study was 61.6 years. They then studied them between the start of the study (1986) and the end (2008 ). In 1983, 1997 and 2004 they asked the women to report fill out a standard survey (this particular survey is commonly used in this field of research). Some of the questions included vitamin usage. Between 1986 and 2008 they kept a record of every death.

They then divided the women up by which vitamins they'd taken (none, a multivitamin, vitamin A, beta-carotene, vitamin B6, folic acid, vitamin B complex, vitamin C, vitamin D, vitamin E, selenium, iron, magnesium, zinc, copper and calcium) and used statistics to determine whether taking a particular vitamin increased the likelihood of death during that timespan.

Additionally, there were some shorter-term studies of the same sort that looked only at calcium, iron or no supplements for 4, 6 or 10 years.

For each vitamin, the change in absolute risk of death (i.e. the increase/decrease in death risk as compared to the control, non-vitamin-taking women) was as follows:
- a multivitamin - 2.4% greater chance of death
- vitamin B6 - 4.1% greater chance of death
- folic acid - 5.9% greater chance of death
- iron - 3.9% greater chance of death
- magnesium - 3.6% greater chance of death
- zinc - 3.0% greater chance of death
- copper - 18.0% greater chance of death
- calcium - 3.8% less chance of death
- the rest - no significant change in chance of death

These results were replicated in the shorter trials.

The total baseline death rate for the duration of the study was 57.0% (that's my calculation, based on the no. at the start of the study and the no. quoted as being alive at the end). However, during their calculations, they excluded deaths from injury, accident or suicide as they considered these were highly unlikely to be related to supplement use.

As for confounding variables, the questionnaire also asked about age, height, educational level, place of residence (farm, rural area other than a farm, or city), diabetes, high blood pressure, weight, waist & hip circumferences, hormone replacement therapy, physical activity, and smoking. These were then taken into account when doing the statistics.

For those theorising that the issue is that supplement use was used to make up for a healthy lifestyle, note they say: "supplement users had a lower prevalence of diabetes mellitus, high blood pressure, and smoking status; a lower BMI and waist to hip ratio" ... "supplement users were more likely to have lower intake of energy, total fat, and monounsaturated fatty acids, saturated fatty acids and to have higher intake of protein, carbohydrates, polyunsaturated fatty acids, alcohol, whole grain products, fruits, and vegetables." Note, of course, that these variables were controlled for in the analyses.

The supplement they are most concerned about is iron, which they say was "strongly and dose dependently associated with increased total mortality risk" and "the association was consistent across shorter intervals, strengthened with multiple use reports and with increasing age at reported use".

They acknowledge the shortcomings of their study - e.g. that subjects might have changed their supplement routine if they were diagnosed with a disease during the study, or that some subjects genuinely needed their supplements due to a particular disease. They also emphasise that the study is only giving evidence about white women of age approx. 60-80 and the findings can't be extrapolated onto other groups. They say "Based on existing evidence,
we see little justification for the general and widespread use of dietary supplements. We recommend that they be used with strong medically based cause, such as symptomatic nutrient deficiency disease."

What you want to take from that is up to you. It seems to be a reasonably good study (as much as a non-specialist can tell, anyway) but it is only one study and they acknowledge that some of their results conflict with those of other studies.

[MsBubbles]

Thank you for the breakdown, Panth. That was extremely helpful.

Here's something I got via pseudo-junkmail, which I find interesting regarding that study:

http://drhyman.com/why-you-should-no...=Get+the+story

[Panth]

Thank you for the breakdown, Panth. That was extremely helpful.

Here's something I got via pseudo-junkmail, which I find interesting regarding that study:

http://drhyman.com/why-you-should-no...=Get+the+story

Hmm... to give another little breakdown, I personally think that report you link to is (partially) flawed. Basically, he is correct in saying that the study only showed correlation, that the population group studied was quite limited (i.e. only white, menopausal/post-menopausal American women) and that as an epidemiological observation-based study that it cannot be used to influence medical advice or treatments without further research.

However, he does also make some mistakes and mis-judgements in his argument.

1) The paragraph under "How Vitamins Save Money and Save Lives"

- he says the original study flies in the face of previous research - now, this alone is not an indicator that the study is wrong! (Particularly since the two big studies he mentions were literature reviews - i.e. potentially have an even more tenuous link to the real situation than an observation, particularly since they pool multiple groups of data that then have to be fudged to be compared, since they will not have been studying the same population group or using precisely the same measures.)

- for the telomere study, well, again, there's a problem with that - measuring telomeres is an indirect measure of longevity and as such is to be taken with a (very small, but still present) pinch of salt

- for the controlled studies - this is the crucial part and the bit I think he's missing - the original study (unlike these controlled ones) looked at all supplement taking, not just doctor-prescribed vitamin taking, not just taking vitamins for a specific illness/condition - it is entirely likely that what is beneficial to an ill person is not beneficial in the entire population (indeed, IMO that could well be what the original study is actually telling us)

2) Under the heading "Why Most Vitamin Studies are Flawed"

- this doesn't really count for the original study as (as I said before) it was looking at total supplement consumption, including non-prescribed ones. Most individuals are not going to know about complex vitamin interactions and therefore could well take combinations that won't work - thus, the study reflects reality (as it must, as it was observational - i.e. did not tell anyone to change what they did, just measured certain parameters whilst they carried on as normal)

- however, this argument could stand for the previous studies he mentions ... and thus would weaken his arguments above that the original study must be wrong since it goes against previous ones

3) Under the heading "Obesity is linked to Malnutrition"

- this does indeed seem to be the case

- however, among other things the original study controlled for weight, and also for waist & hip circumferences - i.e. the two best (easy) parameters for obesity that we have

- in any case, if an obese person is malnourished they should go to a doctor, get tests for vitamin levels taken and then supplement based on known deficiencies - not, as in the original study, just take random supplements (like the individuals studied)

4) Under the heading "Flaws in the "Vitamins Kill You" Study:

1. This is a patent falsity. If you note my comment above, hormone replacement therapy *was* one of the questions in the survey and *was* taken into account.

2. This is true ... but iron was not *given* to the post-menopausal women in the study - they took it themselves regardless of the fact they were being studied. It is an observational study. So this cannot be a criticism of the study, rather it says again that people should not just randomly take supplements but should consult a doctor.

3. Patient background was not completely ignored (as I said in my previous comment, many parameters of health, e.g. weight, were monitored by survey). As for the whole 'people could have started/stopped taking supplements due to developing an illness', well - that is a valid criticism, but it is one that the authors made of their own study. It is a limitation of the original study but does not make the entire thing invalid, rather it calls for further research.

4. No, the population was not representative. The population rarely is in studies, even in the Phase I, II and III clinical trials that are needed to get a new drug licensed. Basically, it's 2 things - there are more white people in the rich countries that fund the majority of studies; and well-off white people are more likely to volunteer for studies (education probably influences this). So, yes, this is a valid criticism ... but if you discounted all studies that had this problem you'd never have another drug on the market - so, it's something that has to be worked around.

5. True, but really ... I personally think that is rather out of the scope of the original study. It's point was not to provide the be all and end all of all information about the subject -no study could ever do that, particularly not on its own- it was rather, to provide an initial investigation that could then be followed up in more detail.

6. Like no. 5 but so much more so. A study, being what it is, looks at a tightly proscribed part of life. It has to, otherwise the variables may be great enough to obscure the data (unless impossibly enormous samples are taken) and/or the amount of data generated would be completely unmanageable and impossible to analyse, drawn conclusions from or report on. Incorrect medication prescription and hospital deaths have nothing whatsoever to do with vitamins. Nothing. It's just a great big red herring. Yes, people may die from that. Maybe you could argue it's more important to study that. So what? Other people are studying that. That's not what the original study was about. It was about people taking supplements. *gah*

So, erm, yeah. So, basically, a few criticisms are valid. Most are complete red herrings.

....aaaand then there's the great big glaring fact that the authors of the original study had no conflictions of interest, yet Dr. Mark Hyman sells vitamins for a living. Suspicious, no?