For well over a decade now there have been suggestions that the selective serotonin reuptake inhibitor and other newer antidepressants might induce suicidal ideas or behaviour in some people. In 1990 details of six patients who had developed intense suicidal thoughts after starting fluoxetine (Prozac) were published in an article in the American Journal of Psychiatry, followed by similar case reports. A series of papers produced by Eli Lilly employees then claimed to find no connection, some even suggesting that Selective Serotonin Reuptake Inhibitors were associated with a lower rate of suicidal ideation than other antidepressants. However the efforts of campaigners kept the issue alive and in the early years of the 21st century drug-regulatory bodies in the United States and United Kingdom issued warnings of a possible link between antidepressants and suicidal behaviour first for children and then for adults.
However the connection remains controversial. Finding quantitative evidence of links between drugs and suicide is difficult because of the rarity of suicide. It is well established that there is a very high risk of someone committing suicide in the month after they have been prescribed an antide-pressant of any sort, but it is usually assumed that this is unrelated to the drug treatment. Two meta-analyses of data from randomised trials in adults indicated small increases in suicide attempts or self-harm in people on Selective Serotonin Reuptake Inhibitors compared with placebo. Several studies of children and adolescents found an increase in suicidal behaviour with at least some selective serotonin reuptake inhibitor antidepressants. Some analyses have found slightly increased rates of suicide or suicidal behaviour among people who were prescribed the Selective Serotonin Reuptake Inhibitors fluoxetine and paroxetine compared with other antidepressant drugs, but the differences did not reach levels of statistical significance. However other studies found no difference between different classes of antidepressants. In addition, some meta-analyses did not find increased rates of suicide or suicide attempts associated with use of Selective Serotonin Reuptake Inhibitors compared with placebo.
David Healy has questioned the data from placebo-controlled trials on the basis that people in the placebo group have usually been withdrawn from an antidepressant before starting the placebo. Since there is a suggestion that withdrawal from some antidepressants might also increase the risk of suicide, he argues that the suicide rate in the placebo group might be higher than would be expected in a group of people who had had no drug treatment, at least early on in the study. Comparisons between Selective Serotonin Reuptake Inhibitors and people who had not had previous drug treatment might indicate higher differences in suicidal behaviour. When Healy and Whitaker (2003) analysed Khan et al.’s (2003) data distinguishing suicidal acts that occurred in the early placebo washout phase of the trial and adding some further data, they produced figures that suggested a statistically and clinically significant increase in suicide and suicidal acts in patients on Selective Serotonin Reuptake Inhibitors compared with placebo. The odds of suicide while taking an selective serotonin reuptake inhibitor were increased more than four times and the odds of non-fatal suicidal acts by more than twice.
Several authors, including David Healy, have suggested that the induction of suicidal ideas is related to the ability of Selective Serotonin Reuptake Inhibitors to induce activation or an akathisia-like state. The idea is that the agitation makes people do desperate things. Acts of violence and hostility have also been linked to use of Selective Serotonin Reuptake Inhibitors. Again, quantitative evidence is difficult to find because, like suicide, extreme violence is rare. However evidence from case reports of violent incidents, including legal reports and data from drug-monitoring agencies suggest that a link between Selective Serotonin Reuptake Inhibitors and violence is at least a possibility. The association, if it exists, may again be attributable to activation or agitation; or it may be due to emotional blunting effects, whether these be specific to Selective Serotonin Reuptake Inhibitors or generic to all psychoactive drugs.
The arguments are likely to continue and may be difficult to resolve quantitatively, although evidence suggesting a link between antidepressants and suicidal behaviour, at least in children, seems to be accumulating. The case reports, especially those where restlessness, agitation or akathisia were clearly involved and suggest a comprehensible mechanism for suicidal or dangerous acts, are also compelling. It may always be difficult to be certain about the connection, or to estimate the prevalence of selective serotonin reuptake inhibitor-induced suicidal or violent acts. However because the events are so serious by nature, even a small and uncertain effect should make people cautious about the use of these drugs.