Substance Dependence Among Bipolar, Unipolar Depression and Psychotic Homeless: A Canadian National Study

Abstract

Introduction: Homeless individuals are often mischaracterized as members of a homogeneous population that suffers from a wide mental health and addiction issues, with little consideration of potentially important differences within or between samples. The aim of the present study was to investigate the comorbidy of alcohol and/or substance dependence (ASD) and major psychiatric diagnoses (bipolar disorder, unipolar depression, and psychotic disorder) in a large Canadian sample of homeless individuals, and to examine potential sources of variability including location and ethnicity. Materials and Methods: A sample of 1,585 homeless individuals were assessed for alcohol and/or substance dependence and bipolar disorder, unipolar depression and psychotic disorder with the Mini-International Neuropsychiatric Interview (version 6.0). Regional and ethnic differences in major psychiatric diagnoses between homeless with and without ASD were examined using univariate (i.e., chi-square) and multivariate (i.e., logistic regression) statistics. Results: Members of the sample with ASD were found to be younger, Aboriginal, less well-educated, and born in the Americas. They were more significantly more prevalent in Western Canada and less prevalent in Central and Eastern Canada. The odds of having ASD were higher among people affected by bipolar disorder and (to a less extent) unipolar depression. Limitations: Data collected were self-reported and no urinalyses were performed. We considered diagnosis of ASD according to the previous 12 months only. Conclusions: Homeless people with major mental illness are at high risk for concurrent ASD, however the prevalence of ASD varies significantly between cities, and based on ethnicity and specific psychiatric diagnosis (with greater prevalence in individuals affected by bipolar disorder and, to a less extent, unipolar depression). Clinicians, administrators and policy makers should develop and deliver services based on careful assessment of the local population.