In order for any drug to be approved by America’s Food and Drug Administration (FDA) only two randomly controlled trials showing drug superiority are required. Otto and Nierenberg (2002) found that despite the quantity of trials showing no significant improvements in symptoms, researchers can continue to alter remedy conditions and hypothesis and selectively choose the measurements that give the outcome their sponsors desire (Rising, Bachetti, Bero, 2008). Turner et al. (2008) reviewed 74 trials of 12 anti-depressant studies and in their meta-analysis found that anti-depressant studies with favorable outcomes were 16 times more likely to be published than those with unfavorable ones.
The study by Piggott, Leventhal, Alter and Boren (2009) conducted meta-analyses on efficacy trials that were submitted to the FDA including the STAR*D (Sequenced ?emedyAlternatives to Relieve Depression) which is the largest and most expensive trial ever conducted. They found that the effectiveness of anti-depressants was even lower than the modest one reported by the study authors. The trial failed to report that the drop out rate was slightly higher than the group of participants, with drop rates increasing progressively across each study phase, due to negative side effects or suicidal ideation. It was also discovered that the measure of relapse required participants to be even more depressed than when they first commenced the trials, compromising the true rates of relapse.
Despite beliefs that anti-depressants are more enduring than placebo groups, Kirsch et al. (2002) found that while patients’ initial positive response to both decrease over time, the correlation was higher for anti-depressants (r=-0.84) than placebos (r=-0.62), indicating that the effects of anti-depressants diminish more rapidly than those of the placebo.
All of these studies aim to encourage the FDA to allow public access to all trials of a new drug, whether it has the sponsor’s desired outcome or not. They also specify that measures should not be changed in order to fit the hypothesis, and that a hypothesis is determined before the trials are carried out, to avoid publication bias (Piggott, Leventhal, Alter & Borren, 2010).
Kirsch, I., Moore, T., Scorbia, A., Nichols, S: The emperor’s new drugs: an analysis of antidepressant medication data submitted to the Food and Drug Administration. Prev Treat 2002; 5: article 23.
Otto, M., & Nierenberg, A: Assay sensitivity failed clinical trials for psychiatric disorders. Psychother Psychosom 2003:72: 115-127.
Piggott, E., Leventhal, A., Gregory, A., & Boren, J. (2010) Efficacy and Effectiveness of Antidepressants: Current status of Research. Journal of Psychotherapy and Psychosomatics, 79, 267-279.
Rising K., Bachetti, P., Bero, L: Reporting bias in drug trials submitted to the Food and Drug Administration: A review of publication and presentation. PLoS Med 2008: 358: 252-260.
Turner E, Matthews A, Linardatos E, Rosenthal R: Selective publication of anti-depressants trials and its influence on apparent efficacy. N Eng J Med 2008: 358:252-260.