Abstract
Research question Do newer generation antidepressants increase the risk of suicidal thoughts and behaviours among adolescents and children?
Answer Probably, but existing data are poor and imprecise
Why did the authors do the study? Regulatory agencies in the UK and US have issued warnings about the increased risk of suicidal thoughts and behaviours among children and adolescents treated with newer generation antidepressants. These authors wanted to reanalyse and update the data that led to the warnings and give doctors a clearer idea of the absolute risks associated with these drugs in young people.
What did they do? They reanalysed data from randomised placebo controlled trials previously reviewed by the UK Committee on Safety of Medicines in 2004, and looked for new trials in the Medline, PsychINFO, and Cochrane databases and pharmaceutical industry websites. They found none. The known trials, many of which are unpublished, evaluated short term treatment with fluoxetine, sertraline, citalopram, paroxetine, venlafaxine, or mirtazapine.
These authors extended the committee's review by estimating the combined risks for any new generation drug (the committee analysed each drug separately), and by doing separate analyses for attempted suicide, suicidal thoughts, or self harm (the committee looked only at suicide related events combined). They also reanalysed data from the largest and most recent trial of fluoxetine in depressed adolescents, excluding those participants who had had cognitive behaviour therapy as well as fluoxetine. Cognitive behaviour therapy may protect young people from the adverse psychological effects of antidepressants.
What did they find? None of the trials reported any completed suicides. 4.8% (71/1487) of children and adolescents taking a newer generation drug had a suicide related event compared with 3% (38/1254) of those taking placebo (odds ratio 1.7, 95% CI 1.13 to 2.54), a significant increase in risk that adds up to one extra event for every 57 children treated for 8 to 12 weeks. The antidepressants were associated with small increases in the risk of suicidal thoughts (1.2% v 0.8%), self harm (3.3% v 2.6%), and suicide attempts (1.9% v 1.2%), but none of these increases was significant. The results for individual drugs were also inconclusive. Only venlafaxine was associated with a significant increase in the risk of suicide related events (7.7% v 0.6%; odds ratio 14.83, 1.93 to 114).
What does it mean? These results support the findings of previous reviews by UK and US regulatory agencies. They suggest that newer generation antidepressants could cause vulnerable children and adolescents to think about or attempt suicide or self harm. The absolute risks seem small—about two extra events for every 100 children treated—and must be balanced against the better established risks associated with major depression.
The trials in this and previous reviews tended to exclude children at high risk of suicide, so their findings may not translate well into the real world of clinical practice, where these children must be treated. The data on suicidal thoughts and behaviours are generally poor, badly defined, and incomplete, making interpretation even harder. We still can't say with any certainty which drugs, if any, are safe. Big, well designed, and inclusive trials are still needed.
References
- Dubicka B et al. Suicide behaviour in youths with depression treated with new generation antidepressants. British Journal of Psychiatry 2006;189:393-8 doi: 10.1192/bjp.bp.105.011833 [DOI] [PubMed]