Vitamin Study Flawed

A study released a few weeks ago stated, “…dietary vitamin and mineral supplements may be associated with increased total mortality risk.” (Archives in Int. Med. Vol. 171. No. 18. Oct. 10, 2011). This article made the rounds in the media with headlines proclaiming, “Dietary Supplements Linked to Higher Mortality.” (Medicalnews.com).

Are supplements dangerous? If you believe the media, the answer is “yes.” Let me sift through the study for you so that you can make an educated decision.

The authors of the study looked at 38,772 older women in the Iowa Women’s Health Study. Their mean age at baseline, in 1986, was 62 years. The study participants self-reported their use of supplements three times over an 18 year period. The authors split the women into two groups; those that took dietary supplements and those that did not.

Let’s go through the results. However, this is where things get tricky. In Table 2, the scientists reported that women who took dietary supplements such as vitamin B complex, Vitamins C, D, and E and calcium had a slightly decreased death rate as compared to nonusers of supplements. When the researchers adjusted the data for various factors such as educational level, place of residence, body mass index and others, the benefits of the supplements disappeared (except for calcium which still showed a benefit).

In Table 3, comparing supplement users to non-users, the study found the risk of cancer mortality decreased in the vast majority of supplement users who used such items as a multivitamin, vitamins A, C, D, E an calcium. However, copper supplementation showed a higher cancer mortality rate. Most other nutrients studied showed a neutral effect. When the authors ‘adjusted’ the data for place of residence, diabetes, high blood pressure and other items, most of the beneficial effects disappeared.

Table 4 looked at the risk of mortality from the use of supplements across the three time periods where the subjects turned in their questionnaires. The only supplements that showed an increased mortality rate were folic acid and iron.

This was a very difficult study to read. The authors seem to have ‘adjusted’ the data to make supplement use appear to be problematic. However, even with their ‘adjustments’ I did not feel the study indicated that supplement use was detrimental. In fact, this paper found many different supplements (Vitamins C, D, E and calcium) actually decreased mortality rate. When the researchers began ‘adjusting’ the data, the positive numbers all began to look worse. However, the authors emphasized the negative results in the abstract and did not mention the positive results (except for calcium). This negative interpretation is what was picked up by the media.

It is interesting to look at Table 4 where multivitamin users had a decreased mortality rate as compared to nonusers. I wonder why the media did not comment on this finding.

This study can be faulted for many reasons. It looked at three surveys from 38,000 women over an 18-year time period. Think about that; these women were only surveyed three times in 18 years. No laboratory tests were ordered. Which supplements did the women take? Did they take them continually over the 18 years? Were the supplements doctor recommended? Did anyone check blood levels of these nutrients? No one knows. Data from surveys are notoriously problematic.

The negative findings of this study occurred when the authors ‘adjusted’ the data. Even most of the negative findings were not significant. There was only a small increase in mortality—about 1% from those taking a multivitamin. This is a very small effect and could be due to chance.

I say, “Forgetaboutit.” This study is a bunch of nonsense. If the authors had emphasized the positive aspects of nutritional supplementation found in this study it would never have been published in this journal.

There are hundreds of articles on nutritional supplements every month. Some are positive, some are negative. My experience has shown the judicious use of supplements has many positive benefits.