Authors’ reply to Sasich and Linden
BMJ 2015; 351 doi: https://doi.org/10.1136/bmj.h5412 (Published 14 October 2015) Cite this as: BMJ 2015;351:h5412- Joanna Le Noury, research psychologist1,
- John M Nardo, retired clinical assistant professor2,
- David Healy, professor1,
- Jon Jureidini, clinical professor3,
- Melissa Raven, postdoctoral fellow3,
- Catalin Tufanaru, research associate4,
- Elia Abi-Jaoude, staff psychiatrist5
- 1School of Medical Sciences, Bangor University, Bangor, Wales, UK
- 2Emory University, Atlanta, GA, USA
- 3Critical and Ethical Mental Health Research Group, Robinson Research Institute, University of Adelaide, Adelaide, SA, Australia
- 4Joanna Briggs Institute, Faculty of Health Sciences, University of Adelaide
- 5Department of Psychiatry, Hospital for Sick Children, University of Toronto, Toronto, ON, Canada
- Jon.Jureidini{at}adelaide.edu.au
Restoring Study 329 was important for several reasons. Firstly, in October 2002 the Food and Drug Administration issued an approvable letter to GlaxoSmithKline’s application to license paroxetine for paediatric major depressive disorder (MDD) and obsessive-compulsive disorder despite noting “we agree that the results from . . . [Study] 329 . . . failed to demonstrate the efficacy of Paxil in pediatric populations with MDD” [but] “we agree that it would not be useful to describe these negative trials in the labelling.”1 2 The knowledge that there was a problem with this trial did not come from the FDA.3
Secondly, Keller and colleagues’ study remains embedded in guidelines recommending antidepressant use in children and adolescents. FDA reviews make no difference to these or any other guidelines.4
Thirdly, the FDA still does not have an accurate tally of the suicidal events in this study.
Finally, our paper is about both data access and authorship. Linden’s response shows that there can be a range of interpretations of the data and that our interpretation of the harms in Study 329 deviates from a computer default position.5 Deviations will almost always occur when data are analysed by people. Others can judge whether this deviation is appropriate if they can access the data.
Our contention is that the data put in the public domain by Keller and colleagues was radically flawed and its interpretation lay beyond the bounds of what was acceptable by the standards of 2001 or any other time.
Notes
Cite this as: BMJ 2015;351:h5412
Footnotes
Competing interests: See original paper at: www.bmj.com/content/351/bmj.h4320.
Log in
Log in using your username and password
Log in through your institution
Subscribe from £173 *
Subscribe and get access to all BMJ articles, and much more.
* For online subscription
Access this article for 1 day for:
£38 / $45 / €42 (excludes VAT)
You can download a PDF version for your personal record.