“It’s hard to think there wasn’t some mischief being done,” Jon Jureidini, a professor at the University of Adelaide in Australia, said in a conference call with reporters at the Washington Post.
Specifically, the original study reported only five instances of suicidal ideation out of total of 93 in children. By contrast, Jureidini said a review found the number was at least 12, which is high. The reason for the discrepancy was the miscoding of a serious suicide attempt as “emotional lability,” which was considered a temporary condition involving uncontrollable episodes of crying.
The re-evaluation of study 329 was just published in the British Medical Journal (Wednesday, September 16, 2015). Jureidini ‘s team drew their conclusions after reviewing 77,000 pages of documents that had only recently been made available and which were not included in the results originally presented. When the data were originally published in 2001, it enabled doctors to prescribe the SSRI to millions of adolescents.
However, shortly after its original publication, the study was heavily criticized as being seriously flawed. In this reanalysis, the authors note that the study appeared to play down harmful side effects including an increase in suicidal behavior. They also note that the study contained numerous transcription errors and other problems, which violated GSK’s own protocols for statistical analysis.
In 2003 and 2004, the Food and Drug Administration (FDA) was forced to issue warnings regarding an increase in suicidal behavior in some children and adolescents who took SSRIs, including Paxil and Zoloft.
Then, late in 2004, the FDA ordered makers of SSRIS to include a Black Box warning on all SSRIs, specifically mentioning the risk for suicidal behaviour among youth using them. This warning was expanded to include people 25 years of age and under in 2007 and in 2012 GSK was fined $3 billion for illegally marketing Paxil, among other drugs.