Marc Joseph Nutrition - Blog

Over the course of the last couple of years, there have been numerous articles and reports in the mainstream press on the negative effects of taking supplements. The most recent subject of scorn are antioxdiant supplements, which, according to a study published in the Journal of the American Medical Association (JAMA), apparently increase our risk of death.

The analysis, which pooled data from 68 studies involving more than 232,000 people, found no evidence that taking beta carotene, Vitamin A or Vitamin E extends life span and, in fact, indicated that the supplements increase the likelihood of dying by about 5 percent. Vitamin C and selenium appeared to have no impact — either way — on longevity …

… “The message is: We shouldn’t be putting anything in our mouths until we know whether it works,” said Christian Gluud of the Copenhagen University Hospital in Denmark, who led the study. “It appears as if these substances may be harmful.”

The Whole Story?

As a meta-review analysis (essentially an analysis of the existing body of research in aggregate), a key factor determining this study’s outcome is the selection of the studies to include in the analysis. As the study’s abstract notes, the meta-review included all randomized studies examining antioxidants in primary and secondary prevention trials published prior to October 2005.

But the researchers didn’t stop there. They then …

… divided that group of studies into low- and high-bias studies, with the former group supposedly being more representative of the true effects of antioxidants. The abstract states the researchers made this distinction between low- and high-bias groups based on “randomization, blinding, and follow-up.”

This definition is the key to the study’s result. For, as noted in the abstract:

When all low- and high-bias risk trials of antioxidant supplements were pooled together there was no significant effect on mortality.

Isn’t that interesting?

So, if there was no significant effect on mortality when the low- and high-bias sub-groups were pooled, and there was a significant negative effect on mortality observed in the low-bias studies, that means that there was a significant positive effect on mortality (a lengthening of average lifespan) in the high-bias studies.

What would be really interesting to know is what good things were happening in those “high-bias” studies? These studies were randomized and worthy of initially being included in the analysis, but were later assigned to a “high-bias” sub-group that resulted in a significant negative result being found in the “low-bias” sub-group.

Hmmm?

Bottom-line, as a meta-review analysis, this study is subject to selection bias* and the interpretation of the researchers.

* Selection bias is the error of distorting a statistical analysis by pre- or post-selecting the samples. Typically this causes measures of statistical significance to appear much stronger than they are, but it is also possible to cause completely illusory artifacts.

Other Problems

In a rebuttal, The Council for Responsible Nutrition (CRN) also noted several other potential problems with the meta-review analysis:

• The studies varied widely in terms of dosage, duration, study population and nutrients tested.
• • For example, one included study looked at a single-day dose of 200,000 IU vitamin A, while other studies looked at vitamin A given for years. That’s like comparing apples and oranges.
  • Likewise, some studies included other nutrients (e.g., zinc, lutein) that would make it difficult to isolate potential effects.
• The vast majority of included studies examined secondary prevention (use of antioxidants in people with pre-existing conditions), rather than primary prevention (preventing disease in healthy populations). Yet, the results of the meta-analysis were portrayed by the researchers and in the press as being applicable to a healthy general population.
• Many of the included studies attempted to use antioxidants as sole treatments for pre-existing conditions. One can’t expect antioxidants alone to overcome serious diseases such as heart disease and cancer.

Conclusion

I have to agree with Andrew Shao, Ph.D. vice president, scientific and regulatory affairs for the CRN:

“There’s a large body of data that shows that antioxidant
supplementation is beneficial. The message to the average consumer is: Don’t pay attention to this. This doesn’t apply to you. You can go ahead and continue taking your antioxidant supplements in addition to the other things you do in your life to stay healthy.”

Antioxidant supplements aren’t a panacea. They’re part of an overall healthy approach that includes a balanced diet, broad-based supplementation, exercise, and stress management.

The fact that the mainstream media and traditional allopathic medicine primarily focus on negative nutrition studies, doesn’t mean positive ones don’t exist. It just means that they’re often conveniently ignored.

Empower yourselves …

• Leading journals:
• • American Journal of Clinical Nutrition
  • Journal of Nutrition
• The OHSU Linus Pauling Institute Micronutrient Center

And check out this related earlier post for more on the topic:

Questioning the Usefulness of Nutritional Supplementation

This entry was posted on Thursday, March 1st, 2007 at 10:58 pm and is filed under Antioxidants, Vitamins, Minerals. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.