More Than 50 Percent Of Clinical Trials Not Properly Registered
First, I need to apologize for not catching this JAMA study that came out in early September, since it is indeed serious stuff: a study by David Moher of the Ottawa Hospital Research Institute found that many researchers are in violation of a 2005 rule requiring that all clinical trials be registered in a public database, ClinicalTrials.gov. The rule was put in place by the International Committee of Medical Journal Editors and was a precondition for member journals publishing researchers’ studies.
In reviewing 323 published clinical trials published in 2008 involving cardiology, rheumatology, and gastroenterology, only 45.5 percent (147 trials) were registered before the end of the trial and had the trial’s outcome properly specified. That means that 54.5 percent of published clinical trials were not properly registered: 27.6 percent had no trial registration whatsoever; 13.9 percent were registered after trial completion; and, 12 percent were registered with no outcome measure or an unclear outcome measure.
Stunningly, 31 percent of the trials that were considered adequately registered had a discrepancy between the outcome measure at registration and what outcome measure was actually published. So much for integrity in science.
This is so unacceptable that I don’t have words for it. I’m sure there are many journal editors who are furious right now. And many researchers who look stupid.
Not that the media looks much better, since it almost entirely missed Moher’s study. The New York Times had a teensy mention of it deep inside the paper and the Ottawa Citizen had an article on the study. That’s it.
Lame all the way around.
Related posts:
Posted in Furious Seasons