Sept. 22 (Bloomberg) -- The first research to track every prescription drug approved by U.S. regulators over several years concluded that many studies finding the medicines ineffective were not published in medical journals.

To win approval from the Food and Drug Administration, companies must submit at least two studies demonstrating a drug is safe and effective, even if other clinical trials find it is not.

The researchers tracked all studies submitted to the FDA by drug companies seeking approval for new treatments from 1998 to 2000. By 2005, five years after the last FDA approvals of the drugs, most of the studies finding the drugs ineffective hadn't been published in journals. Other trials finding them effective were much more likely to be published.

``We found that there was indeed a pattern that favorable studies were more likely to be published than unfavorable trials,'' said Ida Sim, associate professor of internal medicine at UCSF and the lead author of the analysis published today in the Public Library of Science journal, PLoS Medicine. ``This is something that is essentially structural in the way clinical trial information is disseminated to the public.''

Failure to publish negative and positive results could skew doctors' opinions about a drug, the researchers said. While previous research has examined publication patterns of studies of antidepressants and pediatric drugs, this is the first to look at all drugs approved over a defined time span.

Some critics have said drug companies selectively publish results that show a drug works and fail to publish those suggesting it doesn't.

Relevant Studies

Ken Johnson, vice president of the Pharmaceutical Research and manufacturers Association, said in an e-mail that the FDA incorporates all relevant studies about safety and effectiveness of a drug in its decision to approve. That information is summarized in the drug's prescribing information used by doctors, he said.

Concern that publication bias -- the tendency for positive studies to be more frequently published than negative ones -- may mislead consumers led Congress last year to mandate the creation of a registry of clinical trials. Under that legislation, drug companies must submit the results of all trials they conduct for posting on a government Web site that lists trials and results.

Johnson said his organization supported the new legislation as a way to gain ``more comprehensive information about those trials.''

While the new law may help ensure that studies finding the drugs ineffective aren't buried, a brief summary of study results posted on a Web site isn't the same as full publication in a medical journal, Sim said in a telephone interview today.

Influence of Journals

``Medical journals are one of the most influential ways that clinicians and the public get evidence about which drugs work or don't,'' she said. That's partly because of the attention published studies attract from the media, she said.

The new law may give drug companies less motivation to submit studies to journals because they can argue the Web-site summaries are providing full disclosure.

`` If there's less of an incentive to publish a negative study, the ratio of positive to unfavorable results might actually increase,'' she said.

Sim and her colleagues searched the electronic databases of journals looking for published results of 909 trials submitted to the FDA in support of 90 drugs that were later approved for sale. They found that 57 percent of the trials, 515 in all, were never written about in studies published.

The study didn't examine whether trials were submitted and rejected by journal editors or simply weren't submitted at all. Sim said previous research has shown that the primary reason for publication bias is that companies or investigators don't submit them.

To contact the reporter on this story: Rob Waters in San Francisco at rwaters5@bloomberg.net.

Last Updated: September 22, 2008 21:41 EDT