Vitamins and Good Sense

By Bernadine Healy M.D.

This story appears in the March 12, 2007 print edition of U.S. News & World Report.

Vitamin studies always seem to stir controversy, but certainly not visions of death. On that score, last week's report on antioxidant vitamins, published in the Journal of the American Medical Association, was a doozy. The researchers concluded that people taking the antioxidants vitamins A, its precursor beta carotene, and vitamin E, for whatever reason, at whatever dose, and for however long, may be putting their lives in jeopardy. But before you toss out your vitamin pills, let's examine this alarmist study a little bit closer.

Researchers from Copenhagen University Hospital set out to determine whether the antioxidant supplements lengthen one's life. That's difficult to answer, since most people taking vitamins are healthy. So the researchers identified antioxidant clinical trials large and small, as long as they reported at least one death. Any death counted, whether from heart disease or cancer, kidney failure or hip fractures, murders or suicides. Out of 747 antioxidant trials reviewed, 68 met the bill. Then, in what is called a meta-analysis, the 68 trials were combined into what is effectively one study.

PHOTO ILLUSTRATION BY JEFFREY MACMILLAN FOR USN&WR

Predictably, the trials that made the cut were skewed towards the elderly, who are closer to the Pearly Gates. In fact, the mean age in several trials was over 80 (one participant was 103). Diseases in the patients were legion: heart disease, cancer, cirrhosis of the liver, kidney failure, arthritis, diabetes, cataracts, and Alzheimer's and Parkinson's diseases. Trials of heavy smokers and those with premalignant conditions were included, as were those of people in nursing homes. In stark contrast, there were only a few studies here and there of health buffs.

Participants in the trials were not the only ones all over the map. So was their vitamin exposure. Some trials gave a single vitamin; others multiple ones. And close to 40 percent of the studies included beta carotene, a discredited supplement ever since it was found to increase, not decrease, a smoker's chance of lung cancer. (In high doses, beta carotene interacts with tobacco smoke and appears to change its stripes from being a friendly antioxidant to a hostile co-carcinogen.) Vitamin E in excess of 400 international units, or IU's, may also lose its antioxidant value. Among the 68 trials, dose varied from below the recommended daily intake to way beyond the tolerable upper limit. For example, vitamin E ranged from 10 IUs (22.5 is recommended) to 5,000 (1,5200 is considered toxic).

The duration of vitamin usage also plagues this analysis. Many of the participants barely tasted the supplements, as they were given for only a few days or a few months. The average was three years. And no surprise that individual studies varied in outcome. One study, of 13,017 middle-aged patients drawn from the general population, showed that nutritional levels of vitamins C and E and selenium supplements were associated with a 22 percent reduction in mortality after seven years. Compare this with the study of 18,314 patients, all heavy smokers averaging 58 years of age, who took high doses of vitamin A and beta carotene for four years. They had a 16 percent increase in mortality.

Just silly. Blenderizing these diverse trials into one giant 232,606-patient-strong study to come up with a seductively simple proclamation is just silly. When the researchers tallied up the mortality from the 68 trials, there was no difference based on vitamin intake. The headlines that these supplements significantly increase the risk of death by 5 percent overall came only when the researchers pulled out the 47 trials they deemed to have been the best executed. Actually, in the 21 randomized trials they peeled off, mortality was decreased by 9 percent among those taking the vitamins.

Sure, statistics can prove anything. But this study violates a cardinal rule of meta-analysis. Pooled studies must be compatible. That means combining apples and apples—or at the least, similar patients and comparable doses and duration of treatments. The first question to ask in evaluating any such study is whether the combination makes sense, both common sense and medical sense, in the first place. On both, the study flunks.

Where does this leave us? With the same advice that we had before: Don't smoke. Consume at least five fruits and vegetables daily; go easy on the saturated fats; enjoy fish and maybe even fish oil tablets; take supplemental calcium and vitamin D until studies say otherwise. And assuming your daily diet isn't always perfect, its just fine to add on a multivitamin. The whole package will help you live longer. And I don't need a meta-analysis to say so.

This story appears in the March 12, 2007 print edition of U.S. News & World Report.