boston.com Your Life your connection to The Boston Globe

It's premature to blame FDA for suicide rise

The front page headline in the Washington Post was alarming: "Youth suicides increased as antidepressant use fell." A new study argued that a record increase in youth suicides in 2004 may have been the unintended consequence of federal warnings that antidepressants such as Prozac can trigger suicidal thoughts in children. Media outlets across the country reported the painful irony that the Food and Drug Administration's attempt to prevent suicides seemed to have increased them by discouraging doctors from properly treating depressed young people.

"We may have inadvertently created a problem," lamented Thomas Insel, director of the National Institute of Mental Health, to the Post earlier this month.

But a closer look at the numbers suggests that the suicide fears are at least premature, if not baseless, say people who specialize in health statistics.

Suicide is so rare among young people that even the record increase reported by the Centers for Disease Control and Prevention - based on 2004 statistics - represents just 248 extra deaths among 61.5 million youths in the United States ages 5 to 19. And almost half the extra deaths involved 18- and 19-year-olds who were not included in the antidepressant warning. As a result, the suicide rise is so small that statisticians will not be able to say whether it's a real trend or just bad luck - at least until 2005 totals are available later this year.

Even if 2004 does mark the beginning of a bad trend in youth suicides, it's unlikely that a drop in antidepressant use had much to do with it. Drug industry data show that antidepressant prescription rates for people under 20 stayed nearly the same in 2004 as in 2003, and did not fall significantly until 2005, after the FDA required antidepressant makers to add a "black box" warning to pill bottles about the suicide risk in October 2004.

"It's too early to interpret" the rise in suicides "as the start of a dangerous trend that could have been caused by the warnings," said Andrew C. Leon, a psychiatric researcher at Weill Medical College of Cornell University in New York who served on the FDA panel that recommended the black box warning. "We need a couple more years of data."

Robert D. Gibbons, a University of Illinois at Chicago psychiatric researcher who wrote the paper blaming FDA policy for the suicide jump, admits that the rise in suicides and drop in antidepressant prescriptions happened in different years, but said it doesn't undermine his conclusions. He said youth suicide in the Netherlands increased from 34 per year to 51 as antidepressant prescription rates went down nationwide from 2003 to 2005, giving him confidence that the same thing is happening here even though the data in the United States are less conclusive.

"Seeing a record increase in teen suicides is certainly an important thing to get people thinking about," said Gibbons.

The fight over the safety of antidepressants for young people dates to October 2003 when the FDA issued the first of several warnings that, in a small percentage of patients, the drugs seem to cause an increase in suicidal thoughts and actions. No child had killed him- or herself, but several studies showed that children taking antidepressants were twice as likely to consider suicide or begin preparations than children not on the drugs.

Since most studies had shown that antidepressants don't significantly help depressed children, the FDA concluded the drugs may not be worth the suicide risk for some and that all children should be monitored when they first start taking antidepressants.

Prior to 2004, suicide rates among people ages 5 to 19 had declined in most years since 1988, reaching a low of 1,737 suicide deaths nationwide in 2003, in a trend that some analysts attributed to the steady decline in youth alcohol and drug abuse. Then, in 2004, the number of suicides for that age group rose by 14 percent to 1,985, the biggest yearly increase since the federal government began keeping the statistics in 1979.

The suicide increase was much more dramatic among girls than boys. The number of girls ages 10 to 19 who killed themselves increased from 321 to 449, a 40 percent jump, while the comparable number for boys increased 8 percent.

Centers for Disease Control officials called the increase for girls "dramatic and huge," but said they needed more study to understand why it happened.

Gibbons argued that the increase for both genders reflected an FDA warning that scared parents and doctors away from antidepressants. Though the overall prescription rate for young people declined by less than 3 percent in 2004, Gibbons speculated that the young people who seemed most suicidal were the ones most likely to have their prescriptions terminated when the bad publicity about antidepressants began. His paper in the American Journal of Psychiatry - partially funded by the maker of the antidepressant Zoloft - suggested that the FDA warning "may have left some of the most vulnerable youths untreated."

But Leon suggests that Gibbons overstated the importance of antidepressants as a factor in youth suicide. Leon's review of all the young people who committed suicide in New York City from 1993 to 2002 found only five victims out of 108 who had antidepressants in their blood.

In other words, in the years before the FDA's black box warning, relatively few young people who killed themselves were taking antidepressants. If they weren't taking the drugs in the first place, less availability of antidepressants would have no effect on them.

Leon and others acknowledge that Gibbons could eventually be proved right if the 2005 suicide figures show a continued increase in a year that antidepressant prescriptions for young people dropped by more than 10 percent, but he could also be proven wrong.

Gibbons said his research team has already had a positive impact by revealing the weakness of the science behind the FDA antidepressant warning.

"A lot of this started because people thought [antidepressant use] led to suicide. If nothing else, we have shown that that assumption is totally wrong," Gibbons said in an interview. "Is the cutback leading to an epidemic of completed suicides? That is the question we have now."

Scott Allen can be reached at allen@globe.com.

More from Boston.com

'Related'

1,737

people under

age 20 in

the United

States killed

themselves in

2003,

the lowest

number since

at least 1998.

1,985

people under

age 20 killed

themselves in

2004,

the biggest

increase since

the federal

government

began keeping

track in 1979.

NOTE: 2005 figure won't be available until later this year.

SOURCE: Centers for Disease Control and Prevention

SEARCH THE ARCHIVES