Child Suicide: No Antidepressant Link?

A new study questions whether giving antidepressant drugs to children really increases their risk of suicide.

The FDA in 2004 put its strongest warning label -- the "black box" -- on antidepressants. It warns that children and teens who take the medications may be at higher risk of suicide.

A new study questions whether this warning is needed for the class of antidepressants known as selective serotonin uptake inhibitors (SSRIs). SSRI drugs include Prozac, Celexa, Zoloft, and Paxil.

The study's methods are controversial. It doesn't look at whether children who did or didn't kill themselves actually took antidepressants. It simply looks at the number of SSRI prescriptions written for every outpatient -- regardless of age -- in every county in the U.S. Then it compares this prescription rate to the number of child suicides in each county.

The result: There are fewer suicides among children aged 5-14 in U.S. counties where doctors prescribe the most antidepressants. An earlier study showed the same is true for adults.

It's not proof. But it's reassuring all the same, says study researcher Robert D. Gibbons, PhD, professor of psychiatry and director of the Center for Health Statistics at the University of Illinois at Chicago.

"In the general population, there is not an increase in suicide completions in children as you go to higher and higher levels of exposure to SSRIs," Gibbons tells WebMD. "In fact, you have the opposite relation. It looks as if SSRIs are treating depression and decreasing the rate of suicide in children."

Gibbons and colleagues report their findings in the November issue of the American Journal of Psychiatry.

'Some Reassurance'

Gibbons is quick to add that the study merely points to a link between decreased child suicide rates in U.S. counties with increased rates of antidepressant use. He stresses that the study does not offer any information on whether an antidepressant can cause a particular child to commit suicide.

Psychiatrist Gregory E. Simon, MD, MPH, a researcher at Seattle's Group Health Cooperative, also finds the study reassuring -- especially as other recent studies fail to link antidepressants to suicide.

"These studies don't prove to us that prescribing more antidepressants reduces suicide risk -- but as rates of antidepressant use have gone up, we don't see that rates of suicide have gone up at the same time," Simon tells WebMD. "So there is some reassurance, at least, that we are not creating a big suicide problem by prescribing these drugs."
Opposing Voices

James A. Kaye, MD, DrPH, does not find the study reassuring. Kaye is a researcher at Boston University's Collaborative Drug Surveillance Program, where he studies drug safety and efficacy.

"This is really a poorly designed study for drawing any conclusions about antidepressant use and suicide in children and adolescents," Kaye tells WebMD. "It tells you nothing about the use of SSRIs in an individual who did or did not commit suicide."

Yes, Kaye says, the study does demonstrate a link between antidepressant prescriptions written for everyone in a county and the number of child suicides in that county.

"This is interesting -- but it can mean a lot of different things," he says. "It is certainly not strong evidence -- and I hesitate to say it even provides much evidence at all -- that antidepressant use decreases the risk of suicide."

Antidepressant Prescriptions and Child Suicide

Gibbons knows all about the FDA's warning that antidepressants may cause suicide in children. He sat on the advisory panel that recommended it. He notes that the warning was based on data from clinical trials suggesting that children and adolescents taking antidepressants are twice as likely to think about suicide or engage in "suicidal behavior" than those taking inactive placebo pills -- a relatively small increase in risk from 2% to %.

However, not one child in these studies actually committed suicide. Moreover, Gibbons says, the evidence that the drugs actually caused increased thoughts of suicide is weak. And the FDA panel, he says, looked harder at the drugs' risks than at their benefits.

"Many of the panel members who voted for the black-box warning felt we really need better monitoring -- and that is a very good idea," Gibbons says. "But many doctors now look at the black box and say, 'This is too dangerous, I am not going to prescribe these any more.' If that happens, you have a lot of untreated depression. And since the warning, the rate of SSRI prescriptions for children has gone down 30%. That is huge."

Gibbons is very worried about the consequences of this, especially in light of his research team's findings.

"Our study provides a signal that if you cut off the supply of SSRIs -- which we are now doing in response to the black-box warning -- that there is a very good chance that the rate of completed suicides in children will go up," he says.

Detecting Vulnerable Patients

Is Gibbons right? Simon says emerging data support the idea that antidepressants don't increase the risk of suicide -- except for particularly vulnerable patients.

"I look at these large data sets, and I don't see evidence antidepressants cause people to commit suicide," Simon says. "But like every psychiatrist, I have patients who tell me, 'I never thought about this kind of thing before, but when I took this medicine I wanted to jump out of the window.' These patients are very convincing. So maybe these medicines have an overall good effect, and in some few people they have a bad effect."

The real question, Simon suggests, is how to detect these vulnerable patients. Right now, there's no way. So he prescribes caution.

"Here is a very brief summary of what I tell patients," Simon says. "I say, 'There is a concern that antidepressant medicines may cause people to have suicidal ideas they didn't have before. This is based on very complicated information, and it's not clear what this means. But it seems there are people who have these bad reactions, and if you have them, you have to call me right away."

Simon's editorial appears alongside the Gibbons study.

SOURCES: Gibbons, R.D. American Journal of Psychiatry, November 2006; vol 163: pp 1989-1904. Simon, G.E. American Journal of Psychiatry, November 2006; vol 163: pp 1861-1863. Robert D. Gibbons, PhD, professor of psychiatry and health statistics and director, Center for Health Statistics, University of Illinois at Chicago. Gregory E. Simon, MD, MPH, senior investigator, Group Health Cooperative, Seattle. James A. Kaye, MD, DrPH, Boston Collaborative Drug Surveillance Program, Boston University School of Public Health.