Psychiatrists say a new study should help convince the public that newer antidepressants do not appear to be associated with a risk of suicide.
In the study of about 60,000 enrollees at Group Health Cooperative (GHC), a nonprofit health care system based in Seattle, the risk of suicide fell by 60% in the first month after treatment began and continued to drop in the next 5 months (Am. J. Psychiatry 2006;163:41–7). The risk of suicide was highest in the month before treatment.
Looking at patients both before and after treatment gives a clearer look at the whole picture, said Dr. Greg Simon, the study's lead investigator, who noted that no other study has examined suicide risk before a patient began treatment. “If you didn't look at the months before treatment, the risk after treatment would look very high,” Dr. Simon said in an interview.
He said the study seems to debunk the idea that starting a medication somehow leads a depressed person to act out suicidal behaviors more aggressively than before treatment.
Dr. Darrel Regier, director of the American Psychiatric Association's research division, said one of the study's strengths was its examination of patients before medication started. And because it was observational, it included people with more severe mental disorders, who are normally excluded from randomized trials, he said.
The study “confirms that people who are partially treated are going to have a higher risk of suicide, and the longer they are in treatment, the risk of suicide decreases, which is the opposite of what you would see if the risk of the medication was greater than the illness,” Dr. Regier said in an interview. The GHC result “shows the opposite: The risk of the illness was far greater than the risk of treatment,” he said.
Dr. Simon emphasized, however, that the observational study could not definitively answer whether the decline in suicide was directly attributable to a medication. “It's quite probable that the risk might go down when people get placebo, because they are starting treatment,” he said.
But to truly gauge a medication's particular effects would require a clinical trial with 300,000 patients, “and that will never occur,” he said. The study did, however, address the risk of dying by suicide—something the Food and Drug Administration could not do with the data it used as the basis for its black box warning on use of antidepressants in adolescents because so few were studied in those trials, he said.
Dr. Simon and his coinvestigators at Harvard Medical School and Brigham and Women's Hospital, Boston, acknowledged other limitations of the study, including the possible underestimation of suicide risk before treatment and overestimation in the first month after starting because of misclassifications in prescription and hospitalization records.
Both Dr. Regier and Dr. David Fassler agreed, however, that the study was important because it took all comers and examined patients of all ages treated in real world settings.
“From my perspective, the findings are consistent with previous reports” that newer antidepressants are not associated with higher rates of suicide, said Dr. Fassler, a child and adolescent psychiatrist in Burlington, Vt.
“What we really need are more long-term studies with real world samples,” added Dr. Fassler, also of the University of Vermont, Burlington.
Dr. Simon and his colleagues examined pharmacy records and computerized records from GHC physicians, as well as hospital discharge data and mortality records. They found 65,103 members—70% female—who were treated for depression during the 10.5-year study period. The mean age at time of prescription was 44 years, but ranged from 5 to 105.
They examined several issues: risk of death by suicide and serious suicide attempt during acute-phase treatment with antidepressants, increased risk of death by suicide or serious suicide attempt during the month after starting, and the possibility that newer drugs were associated with a higher risk of death than were older antidepressants.
Overall, the researchers found that the rate of attempts in the first 6 months of treatment was 78 per 100,000 in adults and 314 per 100,000 in adolescents.
The higher rate in adolescents does not come as a surprise, explained Dr. Fassler, who noted that, according to the Centers for Disease Control and Prevention, 1 in 6 adolescents thinks about suicide each year, and by the end of high school, 1 in 10 has made an attempt.
Over the last decade, the rate of adolescent suicide has actually declined by one-third. However, he said, “I'm quite concerned that we could start to see a reversal in that very encouraging trend” if there is reduced access to “appropriate and effective treatment.”