Evidence implicating 5-HT in suicide or violence is compelling. Reduced CSF 5-HIAA concentrations correlate highly with completed suicides in patients with depression (Edman et al. 1986; Ninan et al. 1984). In vitro binding assays have shown an increased density (B_max) of 5-HT₂ receptors in individuals with depression and suicidal tendencies (Pandey et al. 1990). Both observations are consistent with a relative state of 5-HT depletion among subjects with suicidal tendencies. The American College of Neuropsychopharmacology (1992) reviewed evidence showing that antidepressants result in substantial improvement or remission of suicidal ideation and impulses in the vast majority of patients. SSRIs were thought to potentially "carry a lower risk for suicide than older tricyclic antidepressants" (p. 181) when taken in overdose. Furthermore, the task force stated that no evidence indicated that SSRIs triggered emergent suicidal ideation above base rates associated with depression. In addition, Warshaw and Keller (1996) determined that fluoxetine use did not increase the rate of suicide in a group of 654 patients with anxiety disorders. In a large retrospective review of patients receiving one or more of 10 antidepressants (including fluoxetine), Jick et al. (1995) concluded that the risk for suicide was similar among all agents.