The Controversy Over Multi-Vitamins

An editorial was recently published in the Annals of Internal Medicine in which the authors exhorted the public to “stop wasting money on vitamin and mineral supplements.” They based this opinion on three studies printed in the same issue that showed an apparent lack of evidence of beneficial effect upon the diseases and disorders being studied.

Let’s take a look at these studies and evaluate the bases for the author’s opinions:

The Cognitive Function Study – The objective was to determine whether long-term multi-vitamin supplementation affects cognitive health in later life. The finding was that there was no difference in mean cognitive change over time or mean level of cognition between the multi-vitamin and placebo groups. Here are the flaws:

Poor Adherence Criteria – This study accepted subjects who adhered to their multi-vitamin regimen a minimum of 2/3 of the time. If they skipped their vitamins four months of the year, their data was still included in the results.
Weak Data Collection – Subjects reported verbally on their adherence once per year. This method is based on recollection and trust versus a quantifiable measure of compliance, such as returning unused product. There seems to be a great deal of room for error and bias in this method of assessment.
Low Potency – The multi-vitamins used in this study contained very low levels of nutrients. High quality multi-vitamin products contain nutrient concentrations four to five times higher than the products used in this study. Even the researchers allowed that “the doses of vitamins may be too low…”

Interestingly, despite the low potency of the vitamins used, some improvement in cognitive function was seen in the study group versus the control, but the change was not statistically significant and, since the data was inherently untrustworthy, it can’t be applied to either conclusion – good or bad.

The Cardiovascular Event Study – This was intended to assess whether oral multi-vitamins reduce cardiovascular events and are safe. The researchers concluded that “high dose oral multi-vitamins and multi-minerals did not statistically significantly reduce cardiovascular events in patients after MI who received standard medications.” That’s a very specific conclusion. Too bad the researchers were not so rigorous in their methodology.

Poor Adherence Standards – 46% of the study subjects discontinued their multi-vitamin regimen during the study. The fact that the researchers felt justified in analyzing efficacy of a treatment to which almost half of the subjects failed to adhere is ridiculous. That a respected journal would publish this flawed study is incomprehensible. At least the researchers were honest enough to cite this limitation in their findings.
Arbitrary Efficacy Standards – The researchers set a threshold of 25% reduction of cardiovascular risk for the minimum effect necessary to indicate benefit. By establishing that standard the researchers were able to downplay positive results of 11% and 18% risk reductions in two separate endpoints of the study.
Poor Subject Selection – Subjects in the multi-vitamin group had a higher rate of diabetes than the control group. As diabetes is a major contributor to cardiovascular risk, it is reasonable to believe that this discrepancy could have biased the results.
Low Nutrient Quality – Despite describing the multi-vitamins used as “high-potency” these products had very low levels of two critical nutrients for cardiovascular health – vitamins D and B12. The concentrations were about 25% of those seen in high quality multi-vitamins.

The U.S. Preventive Task Force Study – This was performed to review evidence for the benefit and harms of vitamin and mineral supplements in the prevention of cardiovascular disease and cancer. The study found limited evidence to support any benefit of vitamin and mineral supplementation for the prevention of cancer or CVD. It did mention two trials that showed borderline significant benefits on cancer in men.

Meta-Analysis – This was not a double-blind, placebo controlled study. It was an analysis of the findings of many studies. By their very nature, meta-analyses don’t provide answers. They raise questions and show correlations. But correlations do not prove causation and questions should not be mistaken for conclusions. Another point to consider is that meta-analyses don’t evaluate every study ever conducted, just the ones the researchers choose. Certainly this method opens up room for error and/or bias.

After having read the studies and seen the methodological flaws it is surprising to read the strong opinions written by the authors of the Annals of Internal Medicine editorial. The evidence simply does not support the assertion and that leads me to believe that either incompetence or bias is behind this article. These are smart people; I doubt that incompetence is the problem.

What the heck, on the off chance that there is still an “open neuron or two” left in their collective mind, here are a few studies they might refer to that demonstrated beneficial effects from multi-vitamin supplementation:

Physician’s Health Study II (PHS II):

8% reduction in overall cancer incidence
12% reduction in cancer death
39% reduction in fatal heart attack risk

Supplementation in Vitamins and Mineral Antioxidants Study (SU.VI.MAX)

31% reduction in total cancer incidence in men

Vitamins and Lifestyle Study (VITAL)

Inverse relationship between dietary selenium and pancreatic cancer risk

This selection is just three of many studies that support the use of nutritional supplementation. If the authors are interested in seeing more, I am sure they can find them on their own.

The one assertion the authors made with which I will agree, is that a lifestyle that depends upon multi-vitamins for health is a poor choice. The key word is supplementation. First and foremost, a healthy life is dependent on a diet of fresh whole foods cooked at home, adequate hydration and regular, vigorous exercise. The addition of supplemental nutrition should be based on conditional need.

I will confidently ignore the advice of the authors and will continue to recommend nutritional supplementation to my patients; and I will use it for my wife, my son and myself. Make your own choice – but if you are using a high quality multi-vitamin, I think you’ll be okay.