Studies in medical journals are not always truthful

Nutrition Health Review, Summer, 2003

When you read about a clinical trial for a drug that has been approved by the U.S. Food and Drug Administration (FDA) in a prestigious medical journal, you might assume that all of the pros and cons of the medication have been weighed and analyzed to the point where nothing is left to chance and all safety concerns have been put to rest. You might expect that the drug has gone through rigorous testing to eliminate any doubt about its ability to treat the specified conditions. You would, unfortunately, be wrong.

Drug studies published in even the most prestigious of medical journals cannot always be relied upon for' accuracy, even if the FDA has approved the drug in question. Because the FDA relies on third-party test results to approve drugs, opportunities exist for misleading and outright false information to slip through unnoticed. These third-party test results often come from for-profit companies that have significant financial ties to the drug manufacturers. Bankrolled private physicians also regularly receive monetary gifts for locating patients to take part in the tests.

In 2000, A Southern Calironia drugmaker sought millions of dollars in damages after the University of California at San Francisco (UCSF) published a report in the Journal of the American Medical Association stating that the company's experimental drug for acquired immunodeficiency syndrome (AIDS) did not work as advertised. The drug company conceded that its initial assertion about the drug's abilities were false but claimed that a smaller study showed that the drug did have some beneficial results.

Dr. James Kahn, a UCSF AIDS clinician and principal investigator for the drug trial, claims that the drug company is trying to put its own perspective on bad information.

"The question we posed in the study is, 'Does it work?' and the answer is, 'No, it does not,' "he said. "There is no ambiguity here."

He also said that the data from the smaller study was analyzed but that no significant effect was found.

"Here is a company manipulating data to try to have a positive outcome," Dr. Kahn said.

He noted that this is a well-known tactic called "data dredging," which is prevalent in the industry.

In another case of money attempting to trump integrity, the manufacturer of a leading arthritis drug came under fire in 2002 for omitting critical data from its claims of superiority. The published report of the drug's safety, sponsored by the drug's manufacturer, concluded that it outperformed two older arthritis medications over a period of six months. The safety study claimed that people who received the medication for six months had lower rates of stomach and intestinal ulcers.

What the study did not say was that the trial lasted for a year. The manufacturer ignored the last half of the trial, in which it was discovered that the medication held no proven advantage over its competitors.

"Within a year of the publication of this trial, sales of [the drug] increased by almost a billion dollars," said Pete Juni, MD, an arthritis researcher with Switzerland's University of Berne.

"The economic incentives to manipulate trial data are huge. We are not talking about thousands of dollars here, or even millions. We are talking about billions."

Even though the flap was covered in The Washington Post and in several medical publications, Dr. Juni claims that the drug manufacturer got away with manipulating its data. He says that many doctors will still rely on the first, positive study instead of the recent, corrected ones.

An editorial in the New England Journal of Medicine blasted the alleged bias of clinical drug trials. It mentioned several occurrences of suppression of articles by drug companies worried about its products being seen in an unsavory light. The editorial claimed that often the only way to verify the accuracy of a drug trial was to contact the company that was being paid to promote it. It also stated that many physicians might receive up to $5,000 for each patient they recommend for the study, thus giving them an incentive to stretch the guidelines for who gets to be tested.

If the medical journals cannot believe in what they are printing, how are consumers supposed to tell the good drugs from the bad?

One solution is to make all information on drug testing available to the public in an effort to keep researchers honest. If a manufacturer's data were able to be analyzed, they would be less likely to become distorted.

Another solution is to have trials that do not involve the manufacturer in any way. Without the promise of huge kickbacks and "incentives," it might be possible to reverse the trend and go back to square one in medical practice: First, do no harm.