team (1998) found that the annual suicide rate for 30,000 patients with schizophrenia and schizoaffective disorders was 63.1 per 100,000 patients (between 1993 and 1995), approximately five times higher than in the general population. In contrast, only one suicide occurred in 6 years among patients treated with clozapine who were of similar diagnosis, age, and sex (a yearly rate of about 12.7 per 100,000 patients). Similarly, the suicide rate was found to be 15.7 per 100,000 patients per year in all United States patients treated with clozapine based on the clozapine national registry system maintained by Novartis Pharmaceutical Corporation, the United States manufacturer of clozapine. Similar analyses with other novel antipsychotic medications have been initiated, and preliminary results suggest that they may also have some beneficial effect in reducing suicide rates.

A number of investigators worldwide have recently reviewed outcomes across large populations showing that a decrease in suicides correlates with the increase of antidepressant use in various European countries (Isacsson et al., 1996; Markowitz, 2001; Ohberg et al., 1998; Rich, 1999; Rihmer et al., 1998) and that suicidal behavior correlates with the inadequate prescription of antidepressants (Henriksson et al., 2001; Oquendo et al., 1999). Such population-based changes in the suicide rate may be due to numerous causes in addition to the increase in antidepressant prescriptions, but these findings suggest a benefit from receiving antidepressants which may be related to appropriate treatment of the underlying depression.

Psychological autopsy studies suggest that the rate of adequate treatment with antidepressants of depressed suicide victims is about 6–14 percent, and toxicological analyses indicate the presence of antidepressants and other prescription psychotropics in about 8–17 percent of suicides, with the frequency in men being about half that of women, and in Blacks and Hispanics being half that of Caucasians (Blazer et al., 2000; Isacsson et al., 1999; Marzuk et al., 1995; Rich and Isacsson, 1997). In general, surveys of university teaching hospitals indicate that most depressed outpatients, even in such academic centers, are either not treated or are under-treated with antidepressant medications (Keller et al., 1986; Oquendo et al., 1999). Oquendo and colleagues (1999) showed that this was just as frequent a problem for those depressed patients with a history of suicidal behavior as for those without.

A variation on the epidemiological studies is the examination of the benefits of an educational intervention. Gotland, an island province of Sweden with a population of 58,000, is a single epidemiological catch-