Adding to a growing scientific consensus, a large Danish study released Tuesday found that vitamin E and other antioxidant supplements provided no health benefits and might even produce a small increase in the incidence of death.
The report in the Journal of the American Medical Assn. was immediately criticized by vitamin makers and some researchers, but its findings are similar to other studies suggesting that supplements are not as quick and easy a way to improve health as their promoters claim.
Supplements "have great biological plausibility, and we all wish that they would work," but they do not, said Dr. Edgar R. Miller of Johns Hopkins University, who was not involved in the new study.
"We need to concentrate on lifestyle modifications -- stop smoking and lose weight -- and not be deceived into thinking that taking a supplement will lower the risk of mortality," he said.
Dr. Goran Bjelakovic of Copenhagen University Hospital and his colleagues performed what is known as a meta-analysis, combining the results from 68 previously published clinical trials involving nearly a quarter of a million patients.
Using all of the studies, they concluded that the supplements, which varied in dosages, were not associated with either an increased or decreased incidence of mortality. But when they used only the 47 trials that they considered the best-conducted, they found a 4% increase in incidence of death with vitamin E, a 16% increase with vitamin A and a 7% increase with beta carotene.
Vitamin C showed no benefit or risk in either analysis, while selenium showed a trend toward increased risk, they said. Their study was funded by the authors' institutions.
Though the percentages of increased incidence of death are low, an estimated 25% of the U.S. and European population are thought to take one or more of the supplements, so potentially large numbers of people would be at risk.
The most common objection to the report regarded the authors' selection of what they considered the best-conducted studies.
It was only after they did this, "using their own criteria, that they observed a statistically significant effect on mortality," said Andrew Shao, vice president of scientific and regulatory affairs for the Council for Responsible Nutrition, a trade association of supplement makers.
"This meta-analysis appears to be a predetermined conclusion in search of a method to support it," he said.
Miller, however, said that "there are good clinical trials and bad ones.... That's a fine way to do it."
Biochemist Balz Frei, director of the Linus Pauling Institute at Oregon State University, noted that all of the original studies used the supplements to try to produce an improvement in patients who already had a disease.
"That may work with a drug, but it is never going to work with a dietary supplement," he said. "You can't undo the damage of decades by giving a dietary supplement for a couple of years."
But Miller noted that the results from the Danish study were virtually identical to those from the Women's Health Study, which involved nearly 40,000 healthy women, showing that the supplements don't work even with subjects who do not have a disease.