Even More Suicide Attempts in Clinical Trials with Paroxetine…

The full title of the article published in BMC Psychiatry is “Even More Suicide Attempts in Clinical Trials with Paroxetine Randomised Against Placebo.” The authors are Aursnes, Tvete, Gaasemyr, & Natvig. They performed a Bayesian analysis (a form of statistical analysis) on whether paroxetine (Paxil/Seroxat) is related to an increase in suicides in clinical trials. Here’s what they had to say, starting with some background:

“Last year we wrote a paper ‘Suicide attempts in clinical trials with paroxetine randomised against placebo’ [1] that hit the front pages of newspapers worldwide [2,3]. Our publication demonstrated an increased intensity of suicide attempts per year when using paroxetine compared to placebo, and caused GlaxoSmithKline (GSK) to come up with a comment [4]. Since then GSK has provided additional data to the American Food and Drug Administration (FDA), as the agency required new documentation on paroxetine. This also resulted in a Briefing Document from GSK [5] in which they admit that there is an increased risk for suicide attempts associated with paroxetine.”

Then the authors mentioned that, in the current study, “We analyzed the data GSK presented in their latest report by the same Bayesian approach used by us in our article. We included only the double blind, parallel design studies with patients randomized to either paroxetine or placebo, as recognized by GSK in the Briefing Document. These 19 studies contained 3455 and 1978 patients to the treatment and placebo groups, respectively. They resulted in 11 and 1 suicide attempts, respectively, as compared to 7 and 1 in our study. The studies lasted 6 – 12 weeks, and we obtained 601 and 333 patient years in the treatment and placebo groups, respectively.”

What did they find?

“We found that the posterior probability that medication with paroxetine is associated with an increased intensity per year of a suicide attempt is 0.99 with the pessimistic prior, 0.98 with the slightly optimistic prior and 0.99 with the slightly pessimistic prior. Hence, we can be at least 98% sure that paroxetine increases suicide attempts. This is stronger evidence than the p value equal to 0.058 given in the Briefing Document [5].”

I am not a Bayesian statistician, so I can’t necessarily critique their method very well. I can tell you that these conclusions match fairly well with evidence marshaled by Healy and others that paroxetine (as well as other SSRIs) are indeed related to an increased suicide risk. Feel free to read other posts (here, here, here, and here) and their related sources. At that point, I think you will see that, based on clinical trial data, there is indeed a greater risk for suicide on SSRI than placebo.