acts in most studies was a consequence of the exclusion of suicidal patients for safety reasons (see Chapter 10). Some investigators (Khan et al., 2001; Khan et al., 2000) also note that the increased contact with mental health professionals for both treatment and placebo groups confounds the observed relationships, and could possibly represent a kind of treatment in itself.
As mentioned earlier, however, increasing prescription rates for antidepressants, in particular SSRIs, has correlated with declines in suicide rates observed in a number of countries including Sweden, Finland, Hungary, and the United States (Isacsson, 2000; Ohberg et al., 1998; Rich and Isacsson, 1997; Rihmer et al., 2001). With access to national health data, it was found that with a doubling of the number of SSRI prescriptions, the suicide rate was reduced by 25 percent in Sweden (Isacsson et al., 1992). A similar result was reported in Italy, but the effect was confined almost entirely to females (Barbui et al., 1999). Though these correlations do not determine causality, they suggest the potential for antidepressants, particularly SSRIs, to reduce suicide rates. This is further supported by the findings of psychological autopsies and toxicological analyses that frequently have found that suicide victims with a mood disorder were taking inadequate therapeutic amounts of antidepressants (Blazer et al., 2000; Isacsson et al., 1994; Isacsson et al., 1992; 1997; Marzuk et al., 1995; Ohberg et al., 1996; Rich and Isacsson, 1997).
The tricyclic antidepressants are effective for the treatment of depressive symptoms. A tricyclic such as amitriptyline may be chosen in cases of suicidality due to its sedative effects, but the high risk of fatal outcome in overdose of tricyclics is a particular concern with regard to suicidal patients. Soloff et al. (1986) found that amitriptyline non-responders made more suicidal communications than placebo non-responders in a group of 29 borderline personality disorder patients.
In terms of actual suicidal behavior, a prospective long-term, placebocontrolled treatment study of 1141 patients found more suicide attempts, including suicides in the group treated with the norepinephrine reuptake inhibitor maprotiline compared with placebo (Rouillon et al., 1989). While maprotiline was an effective antidepressant, it was associated with increased suicide attempts.