Jay D. Amsterdam: The Paroxetine 352 Bipolar Study Ethical Conduct


The paroxetine 352 bipolar trial: A study in medical ghostwriting


        Using federal court documents from paroxetine litigation, this study demonstrates how medical ghostwriters from Scientific Therapeutics Information, Inc. (STI)  together with academic key opinion leaders from major medical research universities contributed to the production and publication of  a widely circulated article in the American Journal of Psychiatry that used manipulated outcome data to favor Glaxo-SmithKline’s  (GSK) proprietary antidepressant, paroxetine, for the treatment of bipolar major depression.

        The article was ghost managed and written by GSK employees and ghost writers from STI, without acknowledging their contribution in the published article. The so-called authors named in the byline of the article were also designated by GSK and STI employees, and all had undisclosed financial ties to GSK. They had little or no direct involvement in the conduct of the paroxetine 352 bipolar trial, and most had not reviewed any of the trial results or the manuscript drafts. The manuscript was originally rejected in peer review by the American Journal of Psychiatry; but was ultimately accepted for publication via the journal editor who also had financial ties to GSK.

        Thus, this study shows how GSK was able to take an under-powered and non-informative trial with negative results and present it as a positive marketing vehicle for off-label promotion of paroxetine for bipolar depression. In addition to the commercial spin of paroxetine efficacy, important protocol-designated safety data went unreported that may have shown paroxetine to produce potentially harmful adverse events.




November 3, 2022