International Coalition for Drug Awareness
www.drugawareness.org & www.ssristories.drugawareness.org
Glaxo Is Testing Paxil on 7-Year-Olds Despite WellKnown Suicide Risks
By Jim Edwards | May 21, 2010
It was established years ago that Paxil carries a risk of suicide in children and teens, but GlaxoSmithKline (GSK) has for the last 18 months been conducting a study of the antidepressant in kids as young as seven — in Japan. It’s not clear why the company would want to draw more attention to its already controversial pill, but it appears as if GSK might be hoping to see a reduced suicide risk in a small population of users — a result the company could use to cast doubt on the Paxil-equals-teen-suicide meme that dominates discussion of the drug.
GSK didn’t immediately respond to a request for comment. A staffer on GSK’s trials hotline confirmed the study was ongoing, however. The drug carries a “black box” warning on its patient information sheet, warning doctors and consumers that the antidepressant is twice as likely to generate lethal thoughts than a placebo.
The trial criteria listed on ClinicalTrials.gov, however, provide an interesting lesson in how managers can carefully design drug trials designed to flatter their products — something good companies don’t do.
The primary aim of the study is not to find out why Paxil makes some children kill themselves. Rather, it’s yet another efficacy study, which the drug doesn’t need because it was approved years ago — we already know the drug works.
Paxil is being tested against a placebo, so the results won’t be very surprising — even terrible drugs work better than sugar pills.
To what degree Paxil triggers suicide is only a secondary aim of the study. If the results suggest a lower suicide risk, expect GSK to play them up. If they’re bad, expect the company to dismiss them in favor of the primary endpoint results.
About 130 children have been enrolled, according to ClinicalTrials.gov, which puts about 65 patients in each arm. That means the results won’t be too statistically robust — there only need to be two or three outlier results to skew the numbers by several percentage points.
The trial will wrap up in September.
676 total views, 2 views today