Healthy Skepticism Library item: 12454

Warning: This library includes all items relevant to health product marketing that we are aware of regardless of quality. Often we do not agree with all or part of the contents.

Publication type: news

Waters R.
Antidepressant Studies More Likely to Get Published If Positive
Bloomberg.com 2008 Jan 17
http://www.bloomberg.com/apps/news?pid=20601103&sid=aYbGr9ma0J9E&refer=us

Full text:

Studies of antidepressants that concluded the medicines were effective were more likely to be published in medical journals than studies showing failed results, according to researchers.

The study compared published research on 12 medicines, including Wyeth’s Effexor and Forest Laboratories Inc.‘s Lexapro, to reviews of the same drugs by the U.S. Food and Drug Administration. The analysis found that 94 percent of published research showed positive results, while FDA reviewers found just half of studies, both published and unpublished, were positive.

Many doctors rely on medical journals to keep abreast of research on the medications they prescribe. Researchers and drug safety advocates contend that so-called “publication bias” — the tendency for articles that get published to reflect chiefly positive findings about drugs and medical devices — gives doctors and patients a distorted view of their value.

“We know publication bias exists,” said Jeffrey Drazen, editor-in-chief of the New England Journal of Medicine, which published the study, in a Jan. 14 telephone interview. The article “shows that what’s reported is really a much more rosy situation than actually exists.”

Publication bias can occur either because companies don’t submit negative studies for publication or because journal editors aren’t interested in publishing negative results, said Erick Turner, a psychiatrist at Oregon Health and Science University in Portland, who led the study.

Negative Findings

Drazen said that the New England Journal has “made a conscious effort since 2002” to publish more negative studies and that an analysis by the journal’s staff found that one-third of the trials published in 2003 had negative findings.

Turner, a former FDA medical reviewer, began his career as a researcher at the National Institute of Mental Health and has maintained a small private practice over the years. Turner said that during his time at the NIMH, he assumed, as did most of his colleagues, that antidepressant studies were overwhelmingly positive.

“Then I got to the FDA and suddenly learned about all of these studies that were not positive,” he said, in a Jan. 14 telephone interview. “I thought, ‘Here’s this secret that I’ve been let in on.’ I’d never seen a negative study.”

Turner’s research found that two studies of New York-based Pfizer Inc.‘s Zoloft were published and both were presented as positive. However, one of those studies was viewed as questionable by FDA reviewers and three others that weren’t published were assessed as failures by the agency’s staff.

Pfizer, Glaxo Respond

“Pfizer is committed to the communication of results of all registered clinical studies, regardless of outcome,” spokesman Jack Cox said today in an e-mail. The company has made a commitment to disclose results within one year of their completion for all its products, he said.

In the case of London-based GlaxoSmithKline’s Paxil, 10 studies were published and six were not, Turner found. Eight of the published studies presented positive results and two presented results that included negative findings. The FDA considered one of the studies published as positive to be questionable and all of the unpublished studies to be negative.

Glaxo “agrees that public disclosure of clinical trial results for marketed medicines is essential,” spokeswoman Mary Ann Rhyne said in an e-mail today.

The company has posted on the Internet results from all clinical trials completed since December, 2000, as well as older results if they “inform medical judgment,” Rhyne said. Glaxo lists all trials it undertakes on clinicaltrials.gov, a Web Site run by the U.S. National Institutes of Health, she said.

Turner’s research reviewed 74 trials. The study found 11 cases in which journals portrayed studies as positive that the FDA considered negative.

Disclosure Steps

While Turner’s research looked only at studies involving adults taking antidepressants, the problem of publication bias is not limited to them, he said.

Steps are being taken to make some unpublished information more available to doctors and patients. Editors of major journals won’t consider publishing results of a trial that wasn’t listed at the onset in a registry, Drazen said. That means there will at least be a record that a trial was undertaken if the company sponsoring it decides not to submit it for publication.

Legislation approved last year by Congress and signed by President George W. Bush requires disclosure of the results of clinical trials. Supporters of the law said it will make it harder for drugmakers to withhold results that show dangers.

“Things are changing, but I don’t think we’re there yet,” Turner said. “There are still various ways you can elect not to publish things if you like.”