A highly-cited trial from 2007 used by international health agencies to support the use of medication of adolescents with depression, has been critically reviewed by a team from the University of Adelaide.
Professor Jon Jureidini, a child psychiatrist and head of the University's Critical and Ethical Mental Health research group, and team reanalysed Duke's Clinical Research Institute's Treatment for Adolescents with Depression Study (TADS), with the findings published in Journal of Risk and Safety in Medicine.
"TADS was a multicentre randomised controlled trial examining the impact of the selective serotonin reuptake inhibitor (SSRI), fluoxetine on 439 adolescents aged 12 to 17," says the University of Adelaide's Professor Jureidini, Adelaide Medical School.
"Participants were randomised into four treatment arms for 12 weeks - fluoxetine only, cognitive behavioural therapy (CBT) only, a combination of fluoxetine and CBT, or a placebo pill.
The original trial found the combination treatment had a superior result and its initial publication has now been cited more than 1900 times as an "evidence-based treatment" for adolescent major depressive disorder (MDD).
"The overall objective of our reanalysis, under the Restoring Invisible and Abandoned Trials (RIAT) methodology, was to accurately report protocol-specified effectiveness outcomes and harms in TADS to evaluate whether the trial justifies the widespread use of antidepressants prescribed for adolescents with depression," says lead author and University of Adelaide Research Fellow Dr Natalie Aboustate, Adelaide Medical School.
"We confirmed what was already known: combination treatment was superior and the study failed to convincingly demonstrate any clinically meaningful advantage for fluoxetine over the placebo.
"More importantly, we found that serious adverse events, including suicidal behaviour were more common for fluoxetine than were reported by the TADS team.
"We found documentation of 32 serious adverse events in the TADs dataset, including at least three suicide attempts that weren't clearly reported by the original authors."
"Overall, our findings demonstrate harm was clinically significantly more prevalent in participants taking fluoxetine compared to participants in the CBT and placebo arms of the study", says Professor Jureidini.
