Antidepressants and Suicide Risk in Kids: Are They All The Same?

William T. Basco, Jr., MD, MS


April 25, 2014

Antidepressants and Suicide Attempts in Children

Cooper WO, Callahan ST, Shintani A, et al
Pediatrics. 2014;133:204-210

Study Summary

This study used a large claims database from Tennessee, using a retrospective cohort approach to evaluate differences in suicide risk among individual antidepressant medications, with specific emphasis on selective serotonin reuptake inhibitors (SSRIs) and serotonin-norepinephrine reuptake inhibitors (SNRIs). Previous studies have hinted at potential differences in suicide risk among the preparations, but the clinical trials often did not contain enough participants or did not compare drugs with one another.

Patient-linked Medicaid data from Tennessee state vital records were used to identify children on antidepressants with a later medically treated suicide attempt or completed suicide. The children were enrolled from 1995 through 2006, and all were aged 6-18 years.

New users of several different antidepressants were identified, including fluoxetine, sertraline, paroxetine, citalopram, escitalopram, and venlafaxine. All drugs were compared with the outcomes of fluoxetine, the drug with US Food and Drug Administration (FDA) approval for treatment of depression in children and adolescents. The investigators defined a "new user" as someone who was prescribed an antidepressant medication without having been prescribed any other antidepressant for a 12-month period before the new prescription. The evaluation excluded children with potential comorbidities and other major mental illnesses, such as psychoses or bipolar disorder.

The investigators looked for outcomes of interest by using established diagnostic codes in the data first, and then verifying each outcome with further review of the child's medical records. They identified more than 36,000 children who qualified as "new users" of one of the study medications.

The final cohort included 419 children (59% girls; mean age, 14 years) who had a confirmed suicide attempt (4 completed suicides). To minimize the bias associated with nonrandom assignment to the drugs, a propensity score that included demographic and medical care variables was used.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: