Study Reveals a Bias In Which Antidepressant Trials Get Published
Positive antidepressant trial results were more likely to be published than negative ones, making it look like the drugs performed better than they really did, according to a new study in the New England Journal of Medicine.
In published trials, about 60 percent of participants who took the drugs experienced significant relief from depression compared to 40 percent of those taking a placebo who experienced significant relief. When the unpublished, less positive, trials were included in the analysis, the drugs still out-performed placebo, but the difference between the two shrank to a more modest figure.
The analysis also found that while 94 percent of positive studies ended up in print, only 14 percent of the less favorable ones had been published.
This new analysis, which reviewed 74 trials involving 12 drugs, is the most thorough to date.
The study authors were unable to determine if the bias resulted from the studies not being submitted for publication, from the journals not electing to publish them or from a mixture of both.
Comments
I read about someone who was trying to get a paper published on a negative result on fish oil for depression. I can’t remember the exact paramaters of the test, but what sticks in my mind is that she had difficulty getting the paper published. She kept being rejected and had to persevere before the paper was finally published. She felt that information on what doesn’t work is as important as information about what does work, and I agree!