The controversy over the pediatric use of antidepressants is just one
example of concerns over the validity of the data from randomized controlled
trials (RCTs) that medicine uses as its evidence base. The concern occurs
A major source of state-of-the-art treatment data is the new research
findings unveiled at medical meetings. These meetings present thousands of
abstracts each year, many with data from the nearly 50,000 RCTs conducted in
the U.S. annually.
But after the meeting, what happens to all that preliminary data? A
mid-1990s study of research presented at an emergency medicine meeting found
that of 223 unpublished abstracts, only 20 percent were published in
peer-reviewed journals within the following three years.
A review last year found that of 510 abstracts of large, phase 3 RCTs
presented at meetings of a major oncology group between 1989 and 1998, 81
percent were published within five years of the meeting at which they were
presented; the remainder were not published within five years. Industry-funded
RCTs with a positive outcome were significantly more likely to be
Another review this year again found that funding and outcome are
significantly related to publication. Of 332 RCTs (158 on drugs, 87 on
surgical treatments, 87 on "other therapies") 37 percent
acknowledged industry funding. Overall, those that were funded by industry
were 1.9 times more likely to be positive toward the treatment studied.
Surgical trials were eight times more likely to be positive, and drug trials
were five times more likely to have pro-industry results.
Though psychiatry was at the forefront of the call for a clinical trials
registry to help track RCT data, response has been broadly supportive from the
AMA, National Institutes of Health, Food and Drug Administration, and
Pharmaceutical Research and Manufacturers of America (the industry's main
lobbying group). All seemingly agree that it's time to change the current