Journal Articles on Drug Trials Often Offer Limited Information for Doctors, Study Says

According to a study in the current issue of PLoS Medicine, medical journal articles on drug trials often do not provide doctors with complete information necessary to evaluate the potential benefits and risks of new drugs, the San Francisco Chronicle reports.

To obtain approval for a new drug, companies must present FDA with results from various clinical trials. After gaining FDA approval, the companies typically fund written accounts of such trials to appear in medical journals. The firms can employ third-parties to prepare journal articles on newly approved drugs. Such articles can carry the names of doctors who participated in the trials and can be written by drug company doctors who helped develop the drug, the Chronicle reports. Many doctors use the articles to determine whether, and how, they will use the drugs.

University of California-San Francisco researchers, led by Lisa Bero of UCSF's Philip R. Lee Institute for Health Policy Studies, examined 164 drug trials involving as many as 1,500 patients over a two-year span. Researchers looked at write-ups of these trials as they appeared in medical journals, as well as trials whose results remained unpublished. They found that not all trial results submitted to FDA as part of drug approval bids were published in medical journals, and those that were written up often contained discrepancies between the results sent to FDA and the published data.

According to the study, drug trials showing "favorable" results were more than five times as likely to be published than those showing "unfavorable results." Results from one-quarter of the trials remained unpublished five years after FDA had approved the drugs, the study shows. In addition, the team found that new drug information available to doctors in medical journals is often "incomplete and potentially biased."

Kristin Rising, a physician who took part in the study but has since moved to Boston University Medical Center, said, "We found really important information from the official trial reports that were either not published at all or that stressed mostly the positive results of trials in the published versions," adding, "Doctors who prescribe new drugs -- or even older ones -- for their patients should have complete and unbiased information on those medicines before prescribing them."

Bero said, "I do think our findings are important for patients because their physicians may not have full and accurate information about the drugs they prescribe." Doctors frequently complain that they have access only to limited information from drugmakers, she added. A commentary by Mayo Clinic physician and drug approval policy expert An-Wen Chan, published alongside the report, states that the UCSF findings show "bias, spin and misreporting" by the industry.

Ken Johnson, senior vice president of Pharmaceutical Research and Manufacturers of America, said doctors can make "appropriate prescribing decisions" on new drugs using FDA-approved drug labels, which contain information based on every study the drug has been subject to. A policy statement from the industry said that drugmakers "have a long-standing commitment to ensure that physicians and patients have access to all relevant information about the medicines we discover, consistent with regulatory requirements, so that our products can be used safely and effectively" (Perlman, San Francisco Chronicle, 12/14).

The study is available online.

The commentary on the study findings also is available online.

FDA Action, Clinicaltrials.gov
FDA now requires that researchers post basic results publicly on the federal online registry maintained by the National Library of Medicine. Previously, researchers were required only to report the start of a new clinical trial. Researchers who fail to post their results online at clinicaltrials.gov "in a timely way" can be fined $10,000 daily and lose their federal research funding. In addition, urging from medical journal editors and settlements in five major lawsuits have led researchers to increase the number of new studies registered each week from 25 to about 350, the Wall Street Journal reports (Hotz, Wall Street Journal, 12/12).