They were some of the most promising medicines of the 1990s -- wonder pills that appeared to fight cancer, heart disease, stroke and other ailments.
Laboratory tests and initial studies in people suggested that lowly vitamins could play a crucial role in preventing some of the most intractable illnesses, especially in an aging population. The National Institutes of Health gave them the same treatment as top-notch pharmaceutical drugs, investing hundreds of millions of dollars in elaborate clinical trials designed to quantify their disease-fighting abilities.
This month, two long-term trials with more than 50,000 participants offered fresh evidence that vitamin C, vitamin E and selenium supplements don't reduce the risk of prostate, colorectal, lung, bladder or pancreatic cancer. Other recent studies have found that over-the-counter vitamins and minerals offer no help in fighting other cancers, stroke or cardiovascular disease.
Research has even suggested that in some circumstances, the supplements can be unsafe.
Some physicians now advise their patients not to bother with the pills, and to rely instead on a healthy diet to provide needed vitamins and minerals.
"These things are ineffective, and in high doses they can cause harm," said Dr. Edgar R. Miller, a professor of medicine and epidemiology at Johns Hopkins University School of Medicine in Baltimore. "People are unhappy with their diets, they're stressed out, and they think it will help. It's just wishful thinking."
Yet faith in vitamins runs deep. The Council for Responsible Nutrition, a trade group in Washington, estimates that 64% of American adults take vitamin and mineral supplements.
Despite the steady drumbeat of reports questioning their efficacy, sales have risen steadily, from $5 billion in 1995 to $10 billion this year, according to the Nutrition Business Journal.
Scientists remain convinced that vitamins are essential to health. But they have puzzled over how their obvious benefits could be so elusive in randomized controlled trials, the gold standard of medical research.
"You really do need vitamin E. You really do need vitamin C. You really do need selenium," said Jeffrey Blumberg, director of the U.S. Department of Agriculture's Antioxidants Research Laboratory at Tufts University in Boston. "Without them, you die."
Blumberg and others now believe that a combination of factors -- including the versions of vitamins that were tested and the populations they were tested in -- probably doomed the studies from the start.
"In retrospect, maybe the expectations were a little bit unrealistic," said Blumberg, whose research has been funded in part by supplement makers.
Unlike observational studies, which look backward at groups of people to identify factors that are associated with a particular disease, a forward-looking randomized controlled trial has the power to show whether a particular factor can prevent the disease.
The studies rely on thousands of participants who are randomly assigned to receive a medicine or a placebo. Researchers carefully track the participants so they can account for other factors that might affect health outcomes, such as age, diet and amount of physical activity. After several years, the researchers compare the number of people in each group who developed a particular disease and use statistical tools to see whether the medicine made a difference.
Vitamin A and beta carotene were the first supplements to be tested that way. In observational studies from the 1980s, researchers noticed that people who ate lots of vegetables rich in beta carotene, a form of vitamin A, were less likely to be diagnosed with lung cancer and other cancers compared with people who didn't. Thus, they reasoned, the vitamin supplements probably reduced the risk of lung cancer.
The thinking was, "Let's get a bunch of people who smoked for 30 years and see if it's going to be a magic bullet for them," said nutritionist Andrew Shao, a vice president for scientific and regulatory affairs for the Council for Responsible Nutrition, a trade association for the supplement industry. "We were all guilty of wanting to go for that home-run-style randomized trial that would validate everything once and for all."
One trial that followed 29,000 male smokers in Finland for an average of six years found that men who took beta carotene were 18% more likely to be diagnosed with lung cancer and 8% more likely to die during the study than men who didn't.