Lots of my patients take vitamin supplements. I don’t recommend them. My patients don’t ask me about it, they just tell me. In the absence of a specific vitamin deficiency or medical condition, there is absolutely no evidence that any vitamin improves any health outcome. I’ve always assumed that vitamins are generally ineffective, but harmless. Some patients are very eager to feel like they’re doing everything they can to be healthy, and I usually decide not to say anything and instead save my credibility for convincing them to exercise or to take their prescription medications as directed. I figure that building trust is better than arguing against their harmless placebo. But maybe I’ve been wrong.

Enthusiasm for vitamins is apparently as deep as the evidence that they don’t help. About twenty years ago antioxidants, including vitamin E, became all the rage, not because we knew them to be helpful, but because what we thought we understood about cell biology suggested that they should be helpful. Since then every rigorous trial has shown vitamin E ineffective in whichever condition it was supposed to help, most recently in preventing heart attack and stroke.

A large trial was started in 2001 to test if vitamin E or selenium might help prevent prostate cancer. Again, there were sound biological reasons to suppose that this might be the case. About 35,000 men were randomized to four groups. One group took 400 units of vitamin E daily. The second took 200 mcg of selenium daily. The third group took both vitamin E and selenium. The last group took placebo. The trial was stopped in 2008 as it became clear that neither selenium nor vitamin E decreased prostate cancer incidence.

But a new study published last week in the Journal of the American Medical Association followed these same men for a few more years and found that the men receiving vitamin E developed prostate cancer significantly more frequently than those receiving placebo. After the longer follow up there were 529 cases of prostate cancer in the placebo group and 620 in the vitamin E group. That means that for about every 100 men taking vitamin E there was one additional case of prostate cancer.

Why? We have no idea. We don’t understand either prostate cancer or vitamin E nearly well enough to understand this effect. But this shows the danger of predicting effects based on our limited understanding of biology. We thought estrogen would prevent heart attacks. It doesn’t. We thought vitamin E would prevent heart attacks. It doesn’t. We thought vitamin E would prevent prostate cancer. It actually slightly increases the risk of prostate cancer.

It’s a good reminder that the only reliable way to have any confidence about the effects of any substance on people is a randomized trial.

So eat a healthy diet. If you’re taking a vitamin for a specific deficiency or medical condition, keep taking it. But if you’re taking it for general health, stop it. There’s no reason to believe it helps, and increasing reason to believe it may hurt. And now when we’re reviewing your medications at your annual exam and you casually mention all the vitamins you take, I might have to gently suggest that you’re doing something risky.

Learn more:

More Evidence Against Vitamin Use (NY Times health blog)

Vitamin E Is Linked to Prostate Cancer (Wall Street Journal)

Vitamin E and the Risk of Prostate Cancer (Journal of the American Medical Association)