Suicide as a cause ofdeath is one ofthe eventual outcomes ofmental illness, and the estimated risk ofsuicide for specific disorders is substantial, ranging from 4% to 10% for schizophrenia and up to 15% for severe affective disorders (1-6). The prevalence ofmental disorder in cases ofsuicide is 80% to 90% or higher (7-11). As a generalization, almost every substance abuse and severe mental disorder carries an increased risk of suicide (12,13). It will, therefore, inevitably be a cause of death in the caseload of every facility involved in the care of patients with serious mental illness.Not unexpectedly, then, the reported suicide rates for psychiatric caseloads are higher than those for general populations. An overview of the literature presents rates for both individual hospitals and groups ofhospitals over different time periods from the 19th century to 1985 (14). The more contemporary studies from 1950 on indicate a wide range in suicide rates. The cumulative rates for studies lasting from 3 to 21 years range from 79 to 684 per 100 000 patients treated or admitted; the rates for 5-year periods ranged from 35 to 629. The annual rates (in studies lasting 8 to 20 years) ranged from 49.6 to 208 per 100 000 patients. For schizophrenia alone, rates of 167, 411, and 456 per 100 000 patients have been reported (15)(16)(17). In several studies, these high rates are extrapolations from small numbers of actual suicide cases. Recent reports of inpatient suicide at Canadian hospitals do not cite the rates for registered patients at risk (18)(19)(20).Although inevitable and not uncommon within hospital caseloads, the occurrence of a suicide is generally very distressing to staff-often more so than a death from natural causes. To address both staffconcerns and more general quality assurance issues, many facilities conduct mandatory reviews ofall suicides in their caseloads. Such audits, including those conducted at the Clarke Institute, may address more than do traditional psychological autopsies, answering why the person died, what was the cause of death, and whether it was a suicide (21).The primary purpose of this report is to describe the experience with suicide in over 100 000 patients at a single psychiatric facility over a period ofmore than 30 years. The secondary purpose is to present an overview ofrecommendations made pursuant to audits of the documentation, management, and post-suicide response for each case. Although data on part of this caseload have been published previously (22,23), this is the first report of the entire cumulative experience with