A review of 372 trials involving nearly 100,000 people who took antidepressants showed that the drugs increase the risk for suicide in people younger than 25, have no effect in those 25 to 64 and reduce risk in those 65 and older. A report on the findings is published online on 12 August in BMJ.
Information on the suicide risk linked to antidepressant use was posted on the FDA Web site some time ago, but "we thought it needed to be in a peer-reviewed publication, which would make it more useful to professionals," said Dr Marc Stone, a senior medical reviewer in the agency's Center for Drug Evaluation and Research and lead author of the BMJ report.
The risk of suicide among younger users of antidepressant drugs became prominent in 2003, when the FDA cited studies showing such an increased risk of suicide attempts or suicide-related behaviour among children and adolescents taking the widely-used selective serotonin reuptake inhibitors (SSRIs), such as Zoloft and Prozac.
A year later, the agency had the companies that market those drugs revise labeling to include a black-box warning so that health-care providers would be alert to any indication of suicide-related behaviour in young users.
In 2006, analysis of the studies described in the new journal report led to a labeling change that expanded the warning to young adults.
The studies show "a bigger picture, where we see a declining rate in older patients that is essentially zero in middle-age adults, and suggest a beneficial effect which lowers the risk of suicidal thoughts in older people," Stone said.
The analysis revealed eight completed suicides, 134 suicide attempts and 378 instances of suicidal thoughts that did not lead to action. The odds ratio for suicidal behaviour declined at a rate of 4.6% for every year of age, according to the study.
Trials by drug companies
The report was criticised by Dr John Geddes, a professor of epidemiological psychiatry at the University of Oxford in England and a co-author of an accompanying editorial in BMJ, who said it relied too much on data from trials sponsored by drug companies.
"In these trials, they compare a drug with a placebo," Geddes said. A placebo is an inactive substance. "These are hard trials to do so they exclude people who are very ill," he said. Because of this, he said, "they can't possibly observe how the drugs affect such people."
"One of the things that we need to be clear about is that we can't rely on placebo-control trials of new medications done by the industry," Geddes said. "They can't answer all the clinical questions. We need more active comparison trials which don't have placebo in them and larger trials aimed at antidepressants in clinical practice."
Geddes also criticised the report for lumping together all SSRIs. "Sometimes people are wary of saying that one drug is safer than another drug in the same class," he said.
But, he said, the studies indicate that suicide risk varies widely among SSRIs. "Sertraline is best in terms of tolerability," he said.
Studies cited in the report indicate that sertraline, the generic name for Zoloft, carries only half the risk for suicidal thoughts and behaviour of some other SSRIs, Geddes said. (HealthDay News, August 2009)