Vitamin supplementation leads to an earlier death?

A study released last week from the Archives of Internal Medicine indicated that vitamin supplementation can lead to an earlier death. This was based on observing Iowa women.

The current study sought to evaluate the link between supplement use and total mortality rate, using data from the Iowa Women's Health Study. A total of 38,772 older women were included in the analysis. Women were aged between 55 to 69 years, with an average of 61.6 years at the beginning of the study in 1986. Self-reported data on vitamin supplement use were collected in 1986, 1997, and 2004.

Vitamin B6, folic acid, iron, magnesium, and zinc were associated with about a 3% to 6% increased risk for death, whereas copper was associated with an 18.0% increased risk for total mortality when compared with corresponding non use.

In contrast, use of calcium was inversely related to risk for death (hazard ratio, 0.91; 95% confidence interval, 0.88 - 0.94; absolute risk reduction, 3.8%).

The researchers assessed the findings for iron and calcium in more detailed analyses conducted during shorter periods (10-year, 6-year, and 4-year follow-up) and found results similar to those for the analyses conducted during the entire time.

As a nutritionist, naturopathic physician and a confirmed supplement user, I was curious to dig into the study to see if made sense to me.

Folks, this is junk science at it's worst!

Here are some salient points:

The data were observational: women in Iowa were asked what supplements they were taking three times over eighteen years,that is every six years. Who remembers what they have taken over six years?

In addition, it was all anecdotal: you didn't have to say what you were taking specifically, just vague terms like multivitamin. Were the vitamins synthetic or natural? How much did they take? Did they really take it, and for how long? Did they take it to stay healthy or because they had become very ill, perhaps with cancer? No one knows.

Dr. Robert Verkerk, the scientific director for Biotics Research Corporation, weighed in. His analysis reveals, among many other interesting points, that all of the data was adjusted by the authors using methods of their own choice. If you look at the study itself, the first thing you see is an adjustment for age and energy, whatever energy means in this case. After this adjustment, vitamins C, B complex, E, D, as well as calcium, magnesium, selenium, and zinc all appear to add to years lived.

This evidently wasn't an acceptable conclusion. So two more adjustments were made. First, if you had a healthy lifestyle and took vitamin C and lived longer, the longer life was attributed largely to the healthy lifestyle and not to the vitamin C. That put everything except B complex and calcium into neutral or negative territory.

Still the authors were not satisfied. They adjusted again, this time for healthy eating, with the result that every supplement except calcium, B complex, and vitamin D became a contributor to an earlier death, according to this undocumented and completely loony math, and only calcium actually lengthened life. Not surprisingly, almost none of this except possibly for the the use of copper supplements taken by 24 women at the end of the study could be claimed to be statistically significant, even using the authors own methods.

The only accurate conclusion that can be drawn from this data is that supplement users are generally healthier people. The why and how and whether it is meaningful is really unknown.

The authors of the study admitted they started out with a hypothesis that supplements wouldn't add to life. It appears, although it is not revealed, that the supplement users actually lived longer than the non-supplement users. But the authors just manipulated the data until they got what they wanted and more: Supplements not only didn't help they were killers! And the lazy, biased, or naive major media took it from there.

Life Extension Foundation also did its own scientific analysis of the Archives of Internal Medicine study. Among other things, it pointed out that copper and iron are pro-oxidants, so their overuse should be expected to lead to earlier mortality. It also noted that many people start taking supplements only after they become ill, which is not controlled for in any way, and that a sizeable minority of the supplements users were also taking drugs that have since been proved to be highly dangerous patented hormones in particular, although no attempt whatever was made to control for drug use.

To pretend to control for so many factors but not to control for drug use and to get through peer review this way is a sad commentary on the state of medical research today. Could this be related to the overwhelming influence of drug companies on medical research in general?

In addition to covering what it referred to as the study's statistical voodoo, it also reminded us that the Archives of Internal Medicine receives millions of dollars in advertising from drug companies, part of the $400 million that goes from drug companies to medical journals, and that the major media trumpeting the study in scary headlines also stay afloat from the $4.7 billion spent in Pharma-to-consumer ads (all of this data is from 2008, and is actually higher now).

Bottom line: look into research before making conclusions and always follow the money.


Dr. T