Suicide Patterns in Depression Unrelated to Antidepressant Use

SEATTLE, July 2 -- Increases in suicide rates in the first month of treatment for depression appear to be unrelated to the use of antidepressants, reported investigators here.

The patterns of suicide attempts among patients treated for depression were similar whether they received antidepressants, psychotherapy, or both, reported Gregory E. Simon, M.D., M.P.H., and James Savarino, Ph.D., of the Group Health Cooperative, a Washington state health plan.

Data on more than 130,000 new episodes of depression showed that regardless of treatment type, the number of suicide attempts was highest in the month before therapy, next highest in the first of month of therapy, and lowest thereafter, the investigators reported in the July issue of the American Journal of Psychiatry.

"Our study indicates that there's nothing specific to antidepressant medications that would either make large populations of people with depression start trying to kill themselves-or protect them from suicidal thoughts," said Dr. Simon.

"Instead," he said, "we think that, on average, starting any type of treatment-medication, psychotherapy, or both-helps most people of any age have fewer symptoms of depression, including thinking about suicide and attempting it."

The investigators reviewed outpatient insurance claims, and identified 131,788 new episodes of treatment for depression -- 70,368 beginning with a prescription for an antidepressant from a primary care physician, 7,297 with a prescription from a psychiatrist, and 54,123 with psychotherapy as the initial form of treatment.

The authors then looked at outpatient and inpatient claims to identify suicide attempts or possible attempts in a period ranging from 90 days before the start of treatment for depression to 180 days afterward.

They found that the overall incidence of suicide attempts was highest among those patients who received antidepressant prescriptions from psychiatrists, at 1,124 per 100,000 population, presumably because patients referred to psychiatrists are at highest risk for suicide, the authors said.

In contrast, patients started on psychotherapy had 778 suicide attempts per 100,000, and patients started on antidepressants by primary care providers made 301 attempts for every 100,000 treated.

The investigators also found that the pattern of attempts over time was the same in all three groups, with a spike in attempts the month before treatment began, a drop of about 50% over the first month of treatment, and a continued decline over six total months of therapy.

The relative odds for a suicide attempt among patients who received drugs from a primary care practitioner in the immediate pre-treatment period versus the first month of therapy were 1.90 (range 1.28-2.81). Among patients receiving drugs from a psychiatrist, the relative odds for an attempt in the month before treatment were 3.57 (range 1.70-7.48), and among patients on psychotherapy the odds were 3.35 (range 2.52-4.45).

The results were unchanged after the investigators censored the data for patients who received a combination of drugs and psychotherapy.

Adolescents and young adults had a higher overall incidence of suicide attempts compared with other age groups, the authors noted, but the time pattern remained the same across all three treatment groups in the teen-young adult category.

"These data offer some reassurance that overall risk of suicide attempts declines after depression treatment is started, with either medication or psychotherapy," the authors wrote.

"They also suggest that patterns of suicide risk early in depression treatment are more influenced by referral patterns and the expected improvement in depression after treatment is started than by specific effects of any particular treatment," they said.

The researchers noted that the study was limited by their inability to determine the severity of depression, and by a high number of dropouts (about one-third of patients treated with either drugs or psychotherapy within a few weeks of starting treatment), which may mask specific treatment effects.