washingtonpost.com
Suicide-Risk Tests for Teens Debated

By Shankar Vedantam
Washington Post Staff Writer
Friday, June 16, 2006

A growing number of U.S. schools are screening teenagers for suicidal tendencies or signs of mental illness, triggering a debate between those who seek to reduce the toll of youthful suicides and others who say the tests are unreliable and intrude on family privacy.

The trend is being aggressively promoted by those who say screening can reduce the tragedy of the more than 1,700 suicides committed by children and adolescents each year in the United States. Many of the most passionate supporters have lost children to suicide -- among them Sen. Gordon Smith (R-Ore.), whose son Garrett died in 2003.

One screening program, TeenScreen, developed by Columbia University, has been administered to more than 150,000 children in 42 states and the District. The state of New York plans to start screening 400,000 children a year, and the federal government is directing tens of millions of dollars to expand screening nationwide.

Use of the psychological evaluations is growing even though there is little hard evidence that they prevent suicides. A panel of government experts concluded two years ago that the evidence to justify suicide screening was weak and that such programs, although well intentioned, had potential adverse consequences.

The growing use of screening has coincided with a rapid increase in the number of youngsters being prescribed powerful antipsychotic medications such as Risperdal and Zyprexa that have not been specifically approved for use by children. There was a fivefold increase in the use of these drugs in children between 1993 and 2002, according to one analysis published this month in the Archives of General Psychiatry, and a 73 percent increase in such prescriptions between 2001 and 2005, according to Medco, a firm that manages pharmacy benefits.

Proponents of screening say that it is no different than having health checkups or visiting a dentist, and that the potential benefits are incalculable. After Smith's son killed himself, the Republican bucked the objections of several conservative groups to push into a law an $82 million effort to expand programs such as TeenScreen.

"Without any doubt, had TeenScreen been available to us as Garrett's parents, I am convinced we would have been empowered to save his life," Smith said in an interview. "Logic tells me the more you know, the more you are able to help."

Garrett Smith died one day shy of his 22nd birthday. He had seen a psychiatrist shortly before he committed suicide and was given a prescription for an antidepressant. Sen. Smith said the family did not know whether Garrett took the medication. Later, Smith said, several experts concluded that Garrett probably had bipolar disorder, also known as manic-depression. Antidepressants are not recommended for this condition, and Smith said his son had probably concealed his symptoms during his single visit with the psychiatrist. Still, he said, if the family had known that Garrett had bipolar disorder, they could have acted years earlier.

The controversy over screening has become emotional. Opponents say such programs have turned into fronts for the pharmaceutical industry to boost sales. Advocates, meanwhile, say those against screening are often driven by anti-psychiatry ideologies such as Scientology.

"It is industrial psychology at its worst," said Michael D. Ostrolenk, a family therapist with the Eagle Forum, a conservative group founded by commentator Phyllis Schlafly. "We think it is inappropriate to turn state schools into laboratories for psychiatry." He added that the group is also concerned that screening violates family privacy.

But screening has wide support among both Republicans and Democrats. In 2004, President Bush signed into law the Garrett Lee Smith Memorial Act to boost funding for suicide screening, and the President's New Freedom Commission on Mental Health has been broadly supportive.

The debate over screening also turns on the scientific paradoxes of suicide. It is rare enough that it is difficult to study by conventional scientific trials, but common enough to claim the lives of more than 30,000 Americans each year -- far more than those who die by homicide. There were 1,737 suicides by children and adolescents in 2003, the last year for which national statistics are available.

Among those younger than 20, the suicide rate is 2.14 per 100,000, a fraction of the 14.6 per 100,000 rate for those older than 50. But national surveys suggest that about 1 in 12 high school students tries to harm himself or herself each year with an eye to committing suicide.

Because suicide victims often turn out to have had mental disorders such as depression and bipolar disorder, David Shaffer of Columbia University, who developed the TeenScreen questionnaire, and other specialists say identifying and treating youngsters with such disorders may reduce the number of suicides.

"If the only product of screening was to predict who is going to commit suicide, you could argue about its utility," he said. "But the risk factors for suicide are other treatable psychiatric disorders."

Laurie Flynn, national executive director for TeenScreen, the largest of several such programs nationwide, said annual physical exams are less likely than mental health checkups to reveal problems. Moreover, she said, suicide screening can reveal problems that parents may never detect. Flynn's daughter attempted suicide when she was 17. When the school phoned Flynn with the news, she said, her initial reaction was "You have the wrong number."

Shaffer and Flynn said the goal is not to put children on medication but to alert parents to a problem, which they can then discuss with a pediatrician, a psychiatrist or a clergy member. Flynn said TeenScreen is supported by private donors and receives no money from the drug industry. (Much of the initial funding came from the late William J. Ruane, a former board member of The Washington Post Co.) Shaffer said the screening test he developed is now in the public domain and he does not profit from its use.

In New York state, where 70 to 80 children commit suicide each year, Sharon Carpinello, commissioner of the Office of Mental Health, said officials plan to spend more than $60 million to expand youth suicide prevention initiatives such as TeenScreen.

Although the argument that treating mental disorders would reduce suicides is intuitively appealing, the U.S. Preventive Services Task Force, a federal panel of independent experts, concluded in 2004 that there was insufficient evidence either for or against general physicians screening the public for suicide risk. Ned Calonge, chairman of the task force, established to assess the evidence for various disease-prevention strategies, said the panel would reach the same conclusion today.

"Whether or not we like to admit it, there are no interventions that have no harms," said Calonge, who is also chief medical officer for the Colorado Department of Public Health and Environment.

There is weak evidence that screening can distinguish people who will commit suicide from those who will not, he said. And screening inevitably leads to treating some people who do not need it.

Such interventions have consequences beyond side effects from drugs or other treatments, he said. Unnecessary care drives up the cost of insurance, causing some people to lose coverage altogether. For every 1 percent increase in premiums in Colorado, Calonge calculated, 2,500 people lose their health insurance.

The same panel had concluded that there is sufficient evidence to recommend screening adults for depression. This is in part because a variety of medications have proved effective in treating adults. Only one drug, Prozac, has been proved effective in clinical trials for treating depression in children.

Steven E. Hyman, a former director of the National Institute of Mental Health and now provost at Harvard University, said he favors developing screening questionnaires and treatments for children to reduce the number of suicides, but he is skeptical that such tools currently exist.

"By and large, brief diagnostic tests -- especially doing broad screening in children -- are not well validated, and one has to be concerned about missing real illness or, conversely, interpreting transient life troubles as a mental illness requiring intervention," Hyman said.

"It doesn't mean ignorance is good," he added. "But if your instrument is poor, or you don't know how to intervene to prevent a condition like suicide, there is actually a risk of harm. Besides cost and intrusiveness, there is a risk of harm in terms of stigmatization, but also interventions that backfire."

View all comments that have been posted about this article.

© 2006 The Washington Post Company