SSRI Prescribing Rates and Adolescent Suicide: Is the Black Box Hurting or Helping?

Suicide is the third leading cause of death in younger (10- to 14-year-old) adolescents in the United States and the leading cause of death in this age group in other countries, including China, Sweden, Ireland, Australia, and New Zealand.¹ The enormous human cost of suicide in youth makes research and prevention a na-tional priority. Biological, psychological, social, and cultural factors all affect the risk of suicide; however, about 90% of youth suicides in the United States are associated with psychiatric illness,^1-3 most commonly a mood disorder.^4-9

In the general population, there is compelling evidence of lack of continuing treatment for depression. While most of those who complete suicide have sought professional help within one month of death,¹⁰ the majority are untreated at the time of death.^11-14 Contrary to recent popular belief, very few children and adolescents who complete suicide are taking medication at the time.^15,16

Limited access to care and poor treatment compliance in young patients aggravates the problem. For example, in a study of 49 adolescents who completed suicide in Utah, antidepressants had been prescribed for 24%, but none tested positive for SSRIs at the time of their death.¹⁷ In a postmortem study of 54 youths who completed suicide, only 4 had antidepressants detected on serum toxicological testing within 3 days of death.¹⁵

Compared with completed suicide (12-month incidence, 8 of 100,000 youths aged 15 to 19 years),¹⁸ suicidal thinking and nonfatal attempts are much more common. Based on figures in the 2005 US Youth Risk Behavior Surveillance Report, 19% of teenagers aged 15 to 19 reported suicidal ideation, and 9% of teenagers reported making a suicide attempt during the past year.¹⁹ Depression annually occurs in 3% to 5% of youths and accounts for about 60% of suicides for all ages. Our best national estimates reveal that among youths receiving care for depression, 35% to 50% have made, or will make, a suicide attempt,^20-22 and 2% to 8% will commit suicide in a decade.^20,21,23 Thus, treatment of youths who are depressed is an important strategy for preventing suicide.

Over the past few years, warnings have been made about antidepressants potentially causing suicide. In October 2003, the FDA issued a public health advisory for all antidepressants used in treating children and teenagers that cautioned about a potentially higher risk for suicide—attempted and completed. In December 2003, the British Medicines and Healthcare Products Regulatory Agency (MHRA) issued a letter to all doctors advising against the use of almost all antidepressants in anyone younger than 18 years. A similar warning was issued by the European Medicines Agency.

In February 2004, the FDA issued a public health advisory that further cautioned physicians, patients, and families about a possible link between suicide and antidepressants in children and adolescents. In October 2004, the FDA issued a black-box warning for increased suicidality in pediatric patients taking antidepressants, a step just short of banning use in this population. The FDA decision was based on randomized controlled trial (RCT) data suggesting an increase from 2% to 4% in suicidal ideation and behavior when antidepressants, including SSRIs, were prescribed for youths who had not been actively suicidal.²⁴ In December 2006, the FDA convened another meeting of its scientific advisory committee, reviewed its RCT data for adults, and concluded that the black-box warning should be extended to young adults.²⁵

A major limitation of these studies is that no suicides were reported in the pediatric studies for either placebo or active medication, and only 8 completed suicides were reported in the adult studies, with no significant differences by condition (placebo, 2 of 36,049; test drug, 5 of 53,030; active control, 1 of 11,217). Also, the RCTs reviewed by the FDA excluded youths who had recently thought about or attempted suicide. As such, even the combination of the 372 studies and close to 100,000 patients provided little or no evidence about the effects of antidepressants on completed suicide.

Others have noted that the finding of increased suicidality in youths treated with antidepressants was limited to medical record data; there was no difference in suicidality obtained from standardized questionnaires.

In a recent reanalysis of the FDA data, augmented by additional studies that have become available since 2004, the difference between antidepressants and placebo in suicidal ideation (abstracted from medical records) was no longer statistically significant, whereas a pooled analysis of clinical response rates found a statistically significant beneficial effect for anxiety disorders and depression in children and adolescents that was not confined to any one antidepressant.²⁶ The result of this study is that antidepressants do show evidence of an antidepressant effect in the pediatric population and that the risk of nonfatal suicidal acts or suicidal ideation is less than that estimated by the FDA in its earlier analyses, yielding a clearly positive benefit-to-risk ratio.

The FDA's findings for children appear to be inconsistent with psychological autopsy studies of completed suicide that suggest very few pediatric suicides were associated with SSRIs.²⁷ They also are inconsistent with ecological studies on suicide completion¹⁰ and attempts that indicate an overall protective effect of SSRIs in pediatric and young adult populations.^28-30 Gibbons and colleagues³¹ found a significant inverse relationship between county-level SSRI prescription levels and adult suicide rates both within counties over time and between counties. This is consistent with a protective effect of SSRIs, both in the overall population and in a specific analysis of 5- to 14-year-olds in the United States.

Ludwig and Marcotte³² demonstrated an inverse association between SSRI prescription rates and SSRIs from 1980 to 1999 in 26 European countries; similar results have been found in the United States.³³ More recently, Bramness and colleagues³⁴ confirmed a significant inverse association between SSRI sales and suicide rates from 1980 to 2004 in a county-level analysis in Norway. Isacsson³⁵ hypothesized that an increase in antidepressant prescriptions in Sweden would be associated with a reduction in suicides. As a natural experiment, he found that a 3.5-fold increase in antidepressant prescriptions in Sweden over time was associated with a 19% decrease in suicides.