Studies of Dietary Supplements Come Under Growing Scrutiny
hen a California judge handed down a $12.5 million false-advertising judgment against the maker of an ephedra-based weight-loss pill late last month, he also issued what amounted to a bill of reproach against the science of dietary supplements.
The company, Cytodyne Technologies, maker of Xenadrine RFA-1, the supplement implicated in the death of a Baltimore Orioles pitcher, had not just exaggerated the findings of clinical trials it commissioned, Superior Court Judge Ronald L. Styn said in ruling on a class-action suit, but had also cajoled some researchers into fudging results in published scientific articles.
The evidence, Judge Styn said, had left him no alternative but to conclude that the researchers had set out to create a study that "justified the money being spent" by Cytodyne and would ensure that they received further work from the company.
The Cytodyne case is part of a swelling tide of litigation that is raising serious questions about the way makers of ephedra and other dietary supplements use and often misuse the promise of scientific proof to market their products.
In the last eight months, three leading manufacturers of weight-loss pills have been hit with false-advertising verdicts in the millions of dollars. A fourth has been rebuked by a federal judge for hiding evidence. The Missouri attorney general and a group of district attorneys in California have also brought false-advertising suits against manufacturers, and Congress has demanded Cytodyne's research records.
Under pressure in the courts and from regulators, a number of manufacturers, including Cytodyne, have refashioned some of their weight-loss products without ephedra. Just this month, legislative leaders in Albany agreed to make New York the second state to ban ephedra, after Illinois, and the federal government is considering its own ban.
But experts say a switch in ingredients does nothing to alter the industry's reliance on questionable science. In fact, they say, they are especially concerned because there is so little rigorous research on the new ephedra substitutes, primarily a substance called synephrine.
The truth is, supplement makers are not obliged to do any research. A 1994 law exempts them from having to prove, as drug manufacturers do, that their products are harmless and effective. But at a time of concern about the safety and effectiveness of its products, the $18 billion-a-year supplement industry has embraced research because, when carefully aimed, it can be worth money.
Precisely because the industry is not regulated, though, its research is sometimes less than strictly scientific, experts say.
"There will be 250 to 300 clinical trials on nutraceuticals this year," said Anthony Almada, a consultant and founder of EAS, the biggest sports nutrition company, who advocates scientific research on products but has become a critic of the way supplement makers conduct it. "The rigor applied in these studies on the average is somewhat notably less than that of a drug study."
Often relying on as few as a dozen subjects, these studies are scaled-down versions of the double-blind, placebo-controlled clinical trials required before drugs can be approved. Some are published in abbreviated form at meetings of scientific organizations, or in obscure journals, providing a basis for marketing claims like "clinically proven."
An industry spokesman, Steven Dentali, vice president for science and technical affairs at the American Herbal Products Association, acknowledged that "whenever there's a desired outcome, you've got the potential for bias." At the same time, he argued that supplement science is no worse than that done for pharmaceuticals.
"There's an interest in making sure you do good science and representing it properly because there's a spotlight on the industry," Mr. Dentali said.
Even the best available science about ephedra is ambiguous. The Food and Drug Administration has collected reports linking it to more than 100 deaths. But by and large, the studies done have been too small and too limited using only healthy subjects to assess the actual danger. In March, the RAND Corporation issued an analysis combining the results of the many small studies; the analysis found evidence that ephedra works for weight loss in the short term, but can produce numerous cardiac side effects.
Documents from the lawsuits, including e-mail correspondence and detailed experiment records, provide an inside look at how, for the supplement makers, the demands of marketing have sometimes compromised those of science. They show how companies and researchers suppressed negative data; removed product names from abstracts when the conclusions showed no positive effects; changed statistical methods to see if results improved; and left out subjects who complained of troublesome side effects like heart palpitations and high blood pressure.
For example, an e-mail message in the Cytodyne lawsuit shows how the company tried to influence one researcher.
The researcher, Jeffrey Armstrong of Eastern Michigan University, had refused to rewrite a journal abstract the way the company wanted, according to testimony, and Tim Ziegenfuss, the consultant who commissioned the study on behalf of the company, said he would try to change Dr. Armstrong's mind.
"As far as rewriting the abstract, since I am not recognized as a co-author on the study I am not allowed to do it," Dr. Ziegenfuss said on November 2000 in an e-mail message to a company official. He added, "In this case the best I can do is try to carefully nudge his interpretation/writing in Cytodyne's favor."
Cytodyne was unable to budge Dr. Armstrong "I will not be intimidated," he said in an e-mail message after being told that the company was threatening to sue him. But in another message, Dr. Ziegenfuss suggested that the study could still be marketed.
"Let Jeff do his hum-drum 'science' thing," he said.
"This will portray Cytodyne in an objective, favorable manner to the scientific community," he added. "This is particularly important now considering the recent bad press on ephedrine. And then since Jeff has no control over the use of data in ads, use percentage changes there to impress consumers."
The company did, turning a modest 3.1-pound average weight loss over several weeks into exclamatory copy saying that Xenadrine users lost 759 percent more weight than nonusers.
Cytodyne's lawyer, Brian Molloy, said the company had not threatened to sue Dr. Armstrong. E-mail messages to that effect, he said, were a "poor effort to get Armstrong's attention."
Mr. Molloy said that he feared court action would discourage future research on supplements. "Having more studies on dietary supplements is generally a good thing," he added.
Before and After
One company, a Canadian concern called Muscletech Research and Development, commissioned a number of studies in 2001, when the ephedra industry was beginning to come under pressure from regulators. Muscletech, maker of an ephedra-and-caffeine weight-loss product called Hydroxycut, was then in the midst of a major advertising campaign. The splashy advertisements in magazines like Men's Health and Self needed a flow of copy. Like Cytodyne's, those advertisements featured personal testimonials and before-and-after photographs and, often in big type, the results of trials.
Muscletech documents from a federal lawsuit in Oklahoma show how studies would be massaged for advertising copy. One study, for instance, found that Hydroxycut users lost 15 pounds of fat mass in 12 weeks, while those taking a placebo lost 10. In an internal memorandum accompanying the study, a Muscletech official warned, "None of these results can be deemed significant," adding that "Hydroxycut can't be claimed as superior" to the placebo. To get around that, the official proposed that copy writers simply say, "Lose 15 lbs. of fat in 12 weeks with Hydroxycut and exercise!"
Another study, by Carlon Colker of Peak Wellness, a Greenwich, Conn., practice that runs trials on dietary supplements, found that Hydroxycut had no apparent effect whatsoever. That study simply went unpublished.
"We are quite surprised at the lack of positive results from this study," Dr. Colker wrote to Muscletech after his trial of a new product that added green tea extract, a caffeine source, to Hydroxycut's ephedra and caffeine recipe.
Dr. Colker declined to comment, other than to say through his lawyer that he stood behind his work.
'The Data Are Buried'
Suppression of study results is a common practice in the dietary supplement industry, said Mr. Almada, the consultant. "You wouldn't believe how many studies are done on products like that and the data are buried," he said.
Jeff Bauer of Springfield, Mo., the lawyer who sued Muscletech in the Oklahoma case, agreed. "They don't have to turn the data over to anyone," he said. "It's the Wild West of D.S.H.E.A." a reference to the 1994 law on dietary supplements, the Dietary Supplement Health and Education Act. "If they had people dying, I don't think they'd have to tell anyone."
Muscletech was also unhappy with another study, by the University of Guelph in Ontario, that found Hydroxycut did not "burn fat," a claim made by many companies that sell ephedra. The company was unable to keep Guelph from publishing the results, but demanded that its product name be removed from publications.
The most problematic study came next, court records show. Four subjects dropped out and were replaced, complaining of symptoms like heart palpitations and high blood pressure. Four more had to lower their doses, two under doctors' orders.
When this study generated no strong scientific claims Hydroxycut users lost more weight, but the numbers were not statistically significant the company demanded new data omitting the dropouts. That kind of analysis is widely considered scientifically inferior. Although the principal researcher, Douglas Kalman, at first resisted, he was eventually persuaded.
"That's unacceptable," Dr. Marcia Angell, a former editor of The New England Journal of Medicine, said in a recent interview.
"You don't change your protocol as you go along," added Dr. Angell, who
reviewed the court records at the request of The
In the end, the published abstract ignored the four dropouts and said the product was safe precisely because it had not raised blood pressure or heart rates.
Those problems with the study were also initially hidden from the court in documents that, days before trial, were found to have been altered. The company's obfuscation so angered Judge Robin Cauthron of district court that she refused to honor a confidentiality agreement when the suit was settled in January.
Muscletech declined to comment on most specific issues, but said through its lawyer, Thomas Ringe III of Philadelphia, "Muscletech is proud of its record of providing to its customers a range of high-quality dietary supplements that have enabled thousands and perhaps millions to achieve a better level of fitness."
Jon Whitcomb, Mr. Kalman's lawyer, defended the process. "It is common during the course of a study to analyze data and results in different ways," he said in an e-mail message, "and in this instance Muscletech requested more than one evaluation."
Mr. Kalman and Dr. Colker were also involved in the Cytodyne study that the California judge said he believed had been designed to justify the money spent by the company. In that study, reported in the journal Current Therapeutic Research in April 2000, Mr. Kalman said Xenadrine users lost significantly more weight than people who took placebos. But the statistician's report on which it was based, produced at the trial, showed that the weight loss was not significant, or only marginally so.
Mr. Kalman said at trial that the claims of significance were typographical errors. He testified that he remembered sending a correction to the journal, but Mr. Whitcomb said he later acknowledged he had not.
More important, several of the more seriously overweight subjects dropped out during the course of the study. The statistician left them out of his analysis of average weight loss, according to testimony. Mr. Kalman, in a table in the study, used those overweight subjects to calculate the average weight at the beginning of the study, but not at the end serving, the judge said, to exaggerate the appearance of weight loss among Xenadrine users.
Mr. Whitcomb said the judge misunderstood the calculations, and cited another judge's decision in a 1999 lawsuit, which found the analysis legitimate. But experts said Mr. Kalman's methods were scientifically indefensible.
"That's not justifiable on any statistical basis," said Dr. Stan Heshka, a methodology and statistics specialist at the New York Obesity Research Center and co-author of a 2000 study on the science of weight-loss alternatives. He added, "I would say it would be a deceptive analysis."
Mr. Whitcomb said: "There's no such thing as a perfect study. Despite any of those perceived deficiencies, the bottom line is the conclusions still hold up. The product was safe and effective within the confines of the study."
The exaggerated in Judge Styn's estimation, false assessment of weight loss made it into dozens of Xenadrine advertisements, and proved critical to his false-advertising ruling.
"The statements of 9 percent decrease in weight are false and misleading and were known or should have been known by defendant to be so," Judge Styn said. "It appears that Mr. Kalman went through and picked and chose the data which would give the most favorable results."