BMJ  2006;332 (20 May), doi:10.1136/bmj.332.7551.0-e

Filler

bmjupdates⁺

Serotonin selective reuptake inhibitors are associated with increased risk of suicide in older people

Research question Is there a link between suicide and selective serotonin reuptake inhibitors (SSRIs) in older people?

Answer Yes, but only during the first month of treatment.

Why did the authors do the study? SSRIs have been linked to suicidal thoughts in adolescents. These authors wanted to explore any link in older adults because they are often prescribed antidepressants and may be particularly vulnerable to suicide.

What did they do? The authors did a large case-control study starting with all adults aged at least 66 who had committed suicide in Ontario between 1992 and 2000. They closely matched each of the 1138 cases to four population based controls who had not committed suicide, then looked at the antidepressant drug histories of both groups. They used a technique called propensity scoring to ensure that cases and controls had similar demographic characteristics and illness profiles.

The cases came from records kept by Ontario's chief coroner. Data on antidepressant prescriptions came from the Ontario Drug Benefit Program, which covers all residents over 65. The authors say it is 99% complete. Clinical information and demographic data for the propensity scores came from hospital records, doctors' diagnosis claims, and outpatient prescription claims.

The authors used conditional logistic regression to estimate the odds of suicide associated with starting treatment with any class of antidepressant except monoamine oxidase inhibitors, which are rarely used. Their analysis included the SSRIs fluoxetine, fluvoxamine, paroxetine, sertraline, and citalopram.

What did they find? Starting treatment with any SSRI was associated with a risk of suicide five times higher than the risk associated with any other class of antidepressant (adjusted odds ratio for suicide 4.8, 95% CI 1.9 to12.2). The increased risk was apparent only during the first month of treatment and was dominated by an increased risk of violent suicide among men.

The absolute risk of suicide in the first month of treatment with any antidepressant was low, however—an estimated 1 in 3353 for SSRIs and 1 in 16 037 for patients taking other antidepressants. 68% of the 1329 suicide cases in this study had received no antidepressants in the six months before death. Venlafaxine was not included in the main analysis.

What does it mean? SSRIs are associated with an increased risk of suicide during the first month of treatment compared with other antidepressants. This could be because doctors prescribe SSRIs to the patients most likely to commit suicide, although the authors think this unlikely given their careful propensity matching. It could also be because SSRIs don't reduce the risk of suicide as much as other antidepressants or because SSRIs increase the risk of suicide in vulnerable people. It's impossible to say from this analysis, but the authors urge doctors not to stop prescribing SSRIs to depressed older people. The risks of undertreatment are likely to be worse than the early risks associated with SSRIs, which are still low. More than two thirds of the cases in this study had no treatment in the 6 months before they committed suicide.

---

Juurlink DM, et al. The risk of suicide with selective serotonin reuptake inhibitors in the elderly. Am J Psychiatry 2006;163: 813-21[Abstract/Free Full Text].

This summarises a paper that has been selected by bmjupdates. To register for bmjupdates (free email alerts about high quality new papers in your favourite subjects) go to http://bmjupdates.com/