The antidepressant drug Paxil was deemed safe for teenagers after its manufacturer, SmithKline Beecham (now part of GlaxoSmithKline), published a study in 2001 supporting its use for adolescents.
Since then, the antidepressant has been the subject of much controversy over its safety, particularly in light of some patients who took it with disastrous like suicide or suicide attempt. The FDA required Paxil and similar drug labels to bear strong messages (known as “black box” warnings) about the potential risk of suicide for children, adolescents and young adults. (The drugs don’t carry the same degree of risk for many adults older than 25.)
As reported last week in the New York Times, it’s still not clear who is at risk and who benefits most from these drugs. But what is clear, according to a new analysis by researchers published in the journal BMJ, the original study’s report was opposite of what the data actually show.
The BMJ researchers who reanalyzed the original data, The Times reported, included a psychiatrist who has been a paid expert witness in lawsuits against Glaxo. But the company agreed to let the analysts review Glaxo’s files on the initial study, which began in the late 1990s, when antidepressant makers started testing the drugs in young people.
Psychotropic drugs (those that affect the mind, emotions and behavior) are difficult to study in a clinical trial because as many as half of the subjects can show improvement even if they’re not in the group taking the drug, but instead taking a placebo (a fake pill). Such improvement is known as “the placebo effect.” As The Times’ story said, “Choices about how to measure improvement — and how to label side effects — can make all the difference in how good a drug looks.”
That’s the issue in the Paxil study. The original research tracked depression scores among three groups of adolescents, one taking Paxil, one on placebo pills and one taking an older, generic depression drug. The Paxil group did no better than the other groups according to the primary measure of effectiveness, but did rate higher on secondary measures, which are akin to circumstantial evidence — possibly useful but not as compelling as the primary measure.
Still, the manufacturer submitted the trial results to federal regulators, who said that the drug was looking good for approved use by adolescents.
Critics weren’t happy after the trial results were published in the Journal of the American Academy of Child and Adolescent Psychiatry, claiming that the evidence wasn’t strong and that the serious side effects had been minimized.
The study authors said at the time that the testing of antidepressants in young people was new, that the paper was clear about its use of secondary measures and that the criticism was baseless.
“Prescriptions of antidepressants to young people surged in the wake of the study,” according to The Times, “increasing by 36% between 2002 and 2003, …. The growth slowed after regulators ordered the black-box warnings on labels.”
The BMJ report from last week echoes the original criticisms — Paxil is not clearly effective, and mislabels the severity of the side effects. One of the study’s authors, Dr. David Healy, a professor of psychiatry at Bangor University in Wales, told The Times that 5 of 6 adverse events labeled “emotional lability” in the original study involved suicidal thinking or behavior but weren’t described that way.
Real stories of patient harm included a teenager who overdosed on Paxil and other medications after a dispute with her mother. Others suffered “severe suicidal ideation.”
Brian Nosek, a professor of psychology at the University of Virginia who wasn’t involved in either the initial study or the BMJ reanalysis of data, told The Times, “This paper is alarming, but its existence is a good thing. It signals that the community is waking up, checking its work and doing what science is supposed to do — self-correct.”
Better late than never.