Clinical notes from a trial of the antidepressant paroxetine, reported as finding it to be both safe and effective for adolescents, in fact showed the opposite, researchers have claimed. The largest re-examination of clinical data from a high profile trial has found a terrifyingly high rate of suicidal behavior that failed to be reported in the controversial study. The revelation could influence the prescription of similar drugs and will add to pressure to change the conduct of clinical drug trials, particularly for antidepressants.

In 2001 the pharmaceutical company SmithKline Beecham (now GlaxoSmithKline) published a paper claiming that a trial, known as Study 329, had found the antidepressant paroxetine was safe for adolescents with depression, and more effective than either a placebo, or the competing drug imipramine.

Also marketed as Aropax, Paxil and Seroxat, paroxetine is a Selective Serotonin Reuptake Inhibitor (SSRI), and was one of the most widely prescribed antidepressants in America. Although never approved for adolescents, Study 329 led to widespread “off-label” prescriptions, but was heavily criticized.

Now, however, Professor Jon Jureidini of the University of Adelaide has disputed the claims of Study 329 in the prestigious British Medical Journal.

Jureidini told IFLScience that on first reading the study he found it inconsistent. The widespread criticism it attracted eventually contributed to stronger warning on SSRI boxes and when the recent Restoring Invisible and Abandoned Trials – an initiative to have data from important trials re-examined – created an opening, Jureidini and his co-authors decided to take another look.

GlaxoSmithKline made the original clinical record forms available, but did so in a manner Jureidini says made re-examining the data much harder than anticipated. “This was the first time something like this has been done on this scale,” he told IFLScience.

“Although concerns had already been raised about Study 329, and the way it was reported, the data was not previously made available so researchers and clinicians weren’t able to identify all of the errors in the published report,” said Professor Jureidini in a statement. “It wasn’t until the data was made available for re-examination that it became apparent that paroxetine was linked to serious adverse reactions, with 11 of the patients taking paroxetine engaging in suicidal or self-harming behaviors compared to only one person in the group of patients who took the placebo.”

Moreover, Jureidini claims that correct analysis of the original data indicates that paroxetine provides no benefits to adolescents to justify this danger. Even the original paper admits, “Neither paroxetine nor imipramine differed significantly from placebo on parent-or self-rating measures,” but this was ignored because investigator assessments of the patients showed a statistically significant difference between those subjects given paroxetine and the control group.

“The investigator assessments always end up looking more favorable to the drug than those from the patients,” Jureidini told IFLScience. “There is a lot of debate about why this is, but the most worrying explanation is that blind is broken.” In other words, investigators are able to tell which patients are being given antidepressants rather than placebos, and their assessments are consciously or unconsciously influenced by this knowledge.

Jureidini suggested that the drugs' side effects can make it easy to work out who is taking them. Potentially this may cast doubt, not just on this individual study, but on all clinical trials of SSRIs, one of the most widely prescribed classes of drugs in the world.

Publicity from Study 329 contributed to paroxetine being prescribed to “hundreds of thousands” of adolescents, Jureidini said. “We’re only talking about one study, but if this accurately reflects the effects we would expect that many children engaged in suicidal behaviour as a result.”

Jureidini told IFLScience: “The safety of paroxetine for adults is unclear. The safety analysis may be biased. But we know that the younger you are and the less severe the depression the more likely it is that the harms of SSRIs will outweigh the benefits.”

In 2012 GlaxoSmithKilne was fined $3 billion (£1.9 billion) for fraudulently promoting drugs for diabetes and mental illness, including paroxetine. However, the New York Times reports that the authors of the original study stand by their research, despite the findings of Jureidini's team.