AbstractThis article argues that a new mental health policy and programmes are needed to deal with the major mental disorders (schizophrenia, major depression and bipolar disorder). Evidence has now accumulated to show that many of the persons who are afflicted with these disorders continue to suffer throughout their adult lives, despite treatment. In addition to their own suffering, their mental disorders lead to unmeasurable suffering for their families which often include young children. Not only do these individuals present all of the symptoms and social impairments usually associated with the major disorders, they are also at increased risk for premature death, substance abuse/dependence, criminality, violence, homelessness, and infectious disease. This situation cannot be left to continue. New policy and programmes designed to prevent the major mental disorders are needed. Two consistent findings suggest that prevention may be possible: 1) many of the children at risk for the major mental disorders can be identified by their family history of mental disorder; and 2) non - genetic factors, biological and/or psychosocial, can limit the expression of the hereditary factors associated with each of these disorders. Given what we know about the fate of children within these high risk families, it may be unethical to not intervene.This article reviews recent findings on the major mental disorders (schizophrenia, major depression, and bipolar disorder). The results demonstrate that these disorders, in most cases, are chronic, devastating, and debilitating, and that they are associated with increased risk of premature death, alcohol and/or drug abuse, criminality, violence, homelessness, and infectious disease. Since the implementation of deinstitutionalization policies along with widespread use of antipsychotic and antidepressant medications, the social problems associated with the major disorders have increased substantially. However, as other findings clearly demonstrate, there is no reason for the current situation to persist. The prevention of the major disorders is a realistic goal because: 1) a large part of the population at risk for these disorders can be identified with relative accuracy; and 2) twin studies have consistently demonstrated that some individuals who carry the genetic predisposition for one or other of these disorders, never develop the disorder; some non - genetic factor protects them. Identification of these protective factors and of the factors which exacerbate the genetic predisposition could lead to the development of prevention programmes. Given current knowledge in this field, Canada needs a new mental health policy which emphasizes prevention of the major mental disorders, and an allocation of resources which would allow for the implementation and evaluation of experimental prevention programmes.A BRIEF REVIEW OF CURRENT FINDINGS ON THE MAJOR MENTAL DISORDERSPrevalenceOnset of the major mental disorders occurs, in most cases, in late adolescence or early adulthood. Schizophrenia afflicts about 1.0% of men and women (Robins & Regier, 1991), and bipolar disorder about 1.6% of men and women (Kessler et al., 1994). The Epidemiologic Catchment Area project which evaluated a random, stratified sample of more than 15,000 U.S. citizens in the early 1980s, documented the prevalence of major depression to be 2.6% among men and 7.0% among women (Robins & Regier, 1991). A similar epidemiological investigation conducted in the early 1990s, documented rates of 12.7% among men and 21.3% among women (Kessler et al., 1994). Several investigations, conducted in a number of different Western industrialized countries, have documented increasing prevalence rates of major depression in cohorts born since the 1940s (Klerman & Weissman, 1992).Validity and reliability of diagnosesThere is now good consensus, and more than adequate reliability and validity, for the diagnoses of schizophrenia and bipolar disorder that are obtained using structured, standardized, diagnostic instruments administered by experienced clinicians trained in their use (see for example, Spitzer, Williams, Gibbon, & First, 1992). …