Psychiatric Morbidity in Indigenous Peoples From Australia and the Americas: Unexpected Findings From a Systematic Review and Meta-Analysis

Abstract

<p><b>Background:</b> Indigenous populations are often considered at high risk of psychiatric disorder, but many studies do not include direct comparisons with the general population.</p> <p><b>Objectives:</b> To undertake a meta-analysis of studies comparing the prevalence rates of psychiatric disorders in Indigenous populations in the Americas and Australia with those of other groups.</p> <p><b>Method:</b> A systematic search of MEDLINE, PsycInfo, EMBASE and article bibliographies. We included empirical quantitative comparisons of the 12-month or lifetime prevalence of any psychiatric disorder in Indigenous and non-Indigenous populations.</p> <p><b>Findings:</b> We found 17 studies (<i>n</i> = 72,419) from Australia, Latin America, Canada and the United States. Indigenous people were at greater risk of concurrent posttraumatic stress disorder (odds ratio [OR] = 1.62; 95% confidence interval [CI] = 1.20, 2.1), alcohol use disorders (OR = 1.93; 95% CI = 1.50, 2.48) and substance use (OR = 2.02; 95% CI = 1.40, 2.93). However, there were no differences between Indigenous and non-Indigenous groups in the prevalence of a range of depressive and anxiety disorders. In some studies, Indigenous rates were lower. The results did not vary greatly by continent or setting (e.g. urban/rural). There were no differences in the lifetime prevalence rates of any disorders.</p> <p><b>Conclusions:</b> The reasons for these results are unclear. One explanation might be that assessment tools may not accurately measure psychiatric symptoms in Indigenous populations. Future research therefore must ensure that diagnostic instruments are validated for use with Indigenous people. Another possibility might be that risk factors for psychiatric illness are a complex interaction of educational, economic and socio-cultural factors that may vary from disorder to disorder. Interventions need to take into account that disadvantage is rarely due to one factor.</p>