Kara Jaye-Anne Otter, 12, had been on the antidepressant Paxil for seven months when she committed suicide. “I was told the worst side effects would be flulike symptoms,” recalls her mother, Shannon Baker. “But after three weeks she had begun to cop an attitude. Her grades started falling. Then she didn’t care what she looked like, and she was fighting with everybody.” Baker says her daughter developed rashes and dark circles under her eyes and had trouble sleeping. Then, on June 7, 2001, Kara pinned a note to her chest reading, “By the time you find me I’ll be dead. I love you with all my heart. Don’t worry, Jesus is with me.” She hooked a bungee cord onto a plate hanger on the wall, wrapped the cord around her neck and pulled against it until she passed out. Within minutes she was dead.
Baker is convinced Paxil is what killed her daughter, and that’s what she’ll tell a U.S. Food and Drug Administration panel meeting this week in Bethesda, Md. For years a small but vocal group of patients and doctors have insisted that certain antidepressants, including Paxil, Zoloft, Prozac and other medications known as selective serotonin reuptake inhibitors (SSRIs), carry an unacceptable risk of antisocial behavior and suicide in kids who take them. Many clinicians and most pharmaceutical companies disagree. Major depression is a dangerous illness that in itself can lead to suicide, and they insist that the benefits of these drugs outweigh the risks. But drug firms also refuse to release all their research some of which tends to undercut their claims. “It’s a real shell game here,” says Vera Sharav, president of the Alliance for Human Research Protection, an advocacy group.
The issue is coming to a head. By last December, the British Medicines and Healthcare Products Regulatory Agency, the equivalent of the FDA, had declared Celexa, Effexor, Lexapro, Luvox, Paxil and Zoloft (but not Prozac) too risky for kids under 18. Two weeks ago, a task force of the American College of Neuropsychopharmacology fired back, saying that despite the risk of potentially serious side effects, it’s riskier to withhold the drugs.
Now comes the FDA probe, which began with an investigation of Paxil launched in the fall of 2002. Last June the FDA cautioned doctors against prescribing the drug to young people because of a possible increase in suicidal thinking. In October the FDA issued a broader public-health advisory alerting physicians that in clinical trials, Paxil, along with about half a dozen other antidepressants, wasn’t any more effective than a placebo at treating depression in kids. Indeed, no antidepressant other than Prozac has been formally approved for young people. But that’s not the same as a ban. Once a drug is legal, doctors can prescribe it at will.
The British, on the other hand, actually banned the use of Paxil (under the drug’s British trade name, Seroxat) for kids last June, and went on to ban the others in the following six months. Why the difference, when British authorities were looking at the same data as the FDA? One reason is that most studies on effectiveness aren’t really definitive. They don’t prove one way or the other whether the drugs work significantly better than placebos and the Brits went with the more conservative interpretation. That, argues Dr. Graham Emslie, co-chair of the American panel and the author of several studies on SSRIs, is shortsighted. “A failure to show effectiveness is not the same as proving ineffectiveness.”
Moreover, he argues, the evidence linking SSRIs to suicide is very weak. Autopsies on people who have taken their lives show that most victims hadn’t taken an antidepressant, or had taken an overdose, in the hours before their death. The British cited “suicide-related events,” not actual suicide, as the reason for the ban and there’s obviously a big difference. “About 2 million teens in the U.S. have clinical depression,” says Dr. Harold Koplewicz, a psychiatrist and director of the NYU Child Study Center. “But 3 million kids have suicidal thoughts. They seem to be part and parcel of the adolescent experience.” And, he points out, only a fraction of either group actually attempts suicide.
Finally, say critics, the British medical authorities looked at the evidence in an unsophisticated way. Many of the unpublished studies they reviewed came from the drug companies themselves, and at least some of those had been carried out for a very specific purpose. Under FDA rules, any company that tests its medications on young people at the FDA’s request wins an extra six months’ worth of patent protection, whether or not the results are positive. Since these studies were done for financial gain and weren’t reviewed by independent scientists, they probably shouldn’t be given full weight. Making them public, as activists demand, might muddy the waters rather than help families make informed decisions.
Nonsense, say scientists who favor sharing such data. The reason some studies remain unpublished is that the drug companies don’t like to admit negative results. According to Dr. Martin Teicher, a researcher at McLean Hospital outside Boston who has published case reports linking suicide and Prozac use, an internal Eli Lilly study of adults taking the drug showed “a substantially greater incidence of suicide attempts in people in the study receiving Prozac than in people receiving placebo or other antidepressants.”
But that study wasn’t published, he says, while research showing no increase in suicide attempts was. Says Dr. Richard Harrington, an expert in child and adolescent psychiatry at England’s University of Manchester: “It’s very important that things get peer reviewed. There is no question about that. But if you have to make decisions about giving drugs to children, you might sometimes have to go on unpublished reports.”
Starting this week, the FDA will try to make sense of it all: the studies, published and unpublished; expert testimony from both sides; and the personal stories of ordinary people like Mark Taylor, 19, wounded in the Columbine school shootings by Eric Harris, 18, who was taking the antidepressant Luvox at the time.
The agency could come to a decision by next summer, but it might not be as well informed as some would like. The truth, suspect many psychiatrists, is complicated: SSRIs help some people and hurt others. Says Teicher: “To figure out what impact [SSRIs] have on a side effect like suicidal thoughts that might only affect a couple of percent or even less than 1% of people in a study, you need a much, much larger study.” And, says Teicher, nobody is doing those studies yet. Besides, argues Koplewicz, SSRIs are most often prescribed for kids by pediatricians or family practitioners, not by experts in child psychiatry. “We need to train these physicians properly,” he says, “not ban drugs that are clearly effective.”