but not when they are at risk. Specific life events may precipitate or provide the occasion for suicidal behavior, but they do not tell us who is likely to make those attempts. The course or persistence of vulnerability factors over time and an individual’s reactions to life events and stressors are important influences. To accurately predict suicidal behavior, a better understanding of the interplay between vulnerability factors and stresses is needed.
Since 90 percent of suicide occurs in people with mental disorders, it is thought that treating the underlying disorder could reduce suicide risk. For some medications there is evidence that the effects on suicidality may be independent from the effects on the mental disorder. This section reviews the evidence that medications used to treat mental disorders can influence the risk of suicide.
Mood stabilizers are used to treat bipolar illness. These drugs fall into two major classes. The first is lithium, a naturally occurring salt, which is effective in reducing the manic and depressive symptoms. Another group of medications proven effective for bipolar disorder is the anticonvulsants (e.g., carbamazepine and valproic acid).
Evidence suggests that lithium treatment of bipolar disorder significantly reduces suicide rates (Baldessarini et al., 1999). In fact, lithium may have specific anti-suicide effects for people with this disorder since these effects may be separate from its antidepressant and antimanic effects. A prospective, randomized controlled clinical trial (Thies-Flechtner et al., 1996) in patients with bipolar illness found that lithium carbonate significantly reduced suicidal acts per patient, relative to patient years. A series of reviews and a meta-analysis of the effect of lithium on suicidality by Tondo and colleagues (Tondo et al., 1997; 1998; 2001) supported the finding that lithium reduced the rate of both suicides and suicide attempts in bipolar patients. The meta-analysis of 12 studies on lithium reported that the risk ratio in favor of a therapeutic lithium effect on suicide is 8.85 (confidence interval=4.14-19.1) (Tondo et al., 2001). This estimate, if correct, would make lithium the most potent therapeutic agent so far identified. However, the protective effects of lithium are not consistent across studies (see Bowden, 2000; Brodersen et al., 2000), and some method-