Data Refute Link Between Suicide, Antidepressants

Researchers in Washington state have found that the risk of suicide and serious suicide attempts is highest in the 30 days before patients begin taking an antidepressant medication, rather than the month following initiation of drug therapy. With continued antidepressant use, the researchers said, risk of suicide death remains fairly constant through the first six months of treatment, while risk of a suicide attempt serious enough to result in hospitalization steadily decreases.

The new report involved an in-depth review of computerized health plan records at Seattle-based Group Health Cooperative (GHC). Between January 1, 1992, and June 30, 2003, researchers led by Gregory Simon, M.D., M.P.H., identified 65,103 GHC patients who received 82,285 prescriptions for new antidepressant treatment (many patients were prescribed trials of more than one antidepressant). Within that population, during the six months following initiation of new antidepressant treatment, 31 suicide deaths occurred, and patients were hospitalized for a serious suicide attempt 76 times.

The report, funded by grants from the National Institute of Mental Health, appears in the January American Journal of Psychiatry.

Suicide and serious suicide attempts are “tragic but fortunately rare events,” said Simon, an assistant professor of psychiatry at the University of Washington and a staff physician and researcher at GHC's Center for Health Studies.

“We don't find any evidence to support the widely held belief that suicide risk increases when people start taking antidepressant medications,” Simon told Psychiatric News. “What we did find was that suicide risk actually decreases when people start taking antidepressants.”

Simon added that the results “do not mean that there may not be individuals who somehow have adverse reactions to these medications, but certainly on average, risk goes down [for patients on medication].”

Simon and his colleagues used GHC computerized pharmacy records and outpatient visit and hospital discharge data for all GHC health plan members during the study period. Mortality records were created for each study year by linking GHC membership rolls with state and national death-certificate data. Mortality data were included for every GHC member during the 10.5-year period, regardless of whether the member remained covered by GHC at the time of death.

All episodes of antidepressant treatment (the “index prescription”) were identified, which represented a new outpatient prescription for an antidepressant medication to any patient who had no record of any antidepressant prescription filled in the previous six months. These patients were diagnosed with unipolar major depressive disorder, dysthymia, or depressive disorder not otherwise specified during the 30 days prior to or following the prescription of the antidepressant.