Skip to main content

Transforming the understanding
and treatment of mental illnesses.

Celebrating 75 Years! Learn More >>

 Archived Content

The National Institute of Mental Health archives materials that are over 4 years old and no longer being updated. The content on this page is provided for historical reference purposes only and may not reflect current knowledge or information.

Searching for Risk Factors of Suicidal Events During Antidepressant Treatment

Science Update

teen couple

A new set of analyses of the NIMH-funded Treatment for Adolescents with Depression Study (TADS) were conducted to better understand what may predict the development of suicidal events during treatment. The analyses, which were published in the May 2009 issue of the Journal of Clinical Psychiatry, showed that youths with suicidal thoughts and more severe depression prior to treatment were at higher risk for suicidal events while undergoing treatment.


Depression is a major risk factor for suicidal behavior. Antidepressant medications, though effective in decreasing depression, may increase the risk for suicidal ideation and attempts in adolescents. In TADS, more adolescents developed suicide ideation or behavior on medication than on placebo or psychotherapy. It is unknown how these medications may prompt some teens to consider or attempt suicide. It has been proposed that some adolescents will get negatively activated by the antidepressant, with emergence of anxiety, agitation and irritability, which, in turn, may trigger suicidality. To add further complication, suicidality may emerge even when depression begins to lift and patients begin to improve. For these reasons, doctors treating teens with depression are strongly encouraged to closely monitor their patients for at least the first month of treatment.

Benedetto Vitiello, M.D., of NIMH, and colleagues analyzed data from the 36-week TADS to determine if they could identify possible predictors of suicidal events among the 439 participants. The primary results of TADS, which compared medication, psychotherapy, and their combination in treating teens with depression, have been previously reported.

Results of the Study

Although no suicides occurred during the study, about 10 percent of participants experienced at least one suicidal event, defined as serious suicidal thinking or a suicide attempt. In addition, events occurred anywhere between the first week and the 31st week of the trial, indicating that the risk for suicidal events did not decrease after the first month of treatment. The researchers also found no differences in suicidal event timing among participants receiving antidepressant medications compared to participants not receiving medication.

Rather, participants who showed serious suicidal thinking and severe depressive symptoms prior to study treatment were more likely to have a suicidal event during treatment. In addition, interpersonal stressors, such as conflicts with family members, preceded suicidal events in 73 percent of cases.


Because the timing of suicidal events varied, the researchers suggest that careful clinical monitoring should continue past the first month of treatment. In addition, because the teens typically did not show increased irritability, insomnia or agitation before an event—common signs prior to a suicide attempt—the researchers concluded that this pattern did not suggest that suicidality was triggered by medication—induced negative activation. Rather, the analysis showed that the teens that experienced a suicidal event tended to do so in the context of persistent depression and difficult interpersonal problems.

What's Next

More research is needed to better understand individual characteristics and identify specific predictors of suicidal actions, especially among adolescents with no history of suicidal thinking or behavior.


Vitiello B, Silva S, Rohde P, Kratochvil C, Kennard B, Reinecke M, Mayes T, Posner K, May D, March J. Suicidal events in the Treatment for Adolescents with Depression Study (TADS). Journal of Clinical Psychiatry. 2009 May;70(5):741-747.