Six community-based studies (Table 3) met the selection criteria of this meta-analysis. These studies were published between 1988 and 2020. All these studies involved cross-sectional assessments on adult participants. These studies were based on five community surveys—Epidemiologic Catchment Area study (ECA) in the USA (65, 71), the Singapore Mental Health Studies 2010 and 2016 (69, 70), a population-based study from Iran (68), and the British National Psychiatry Morbidity Survey 2000 (66). A direct comparison between epidemiologic/ community- and clinic-based samples was not made, due to the stark difference in numbers, as also, more fundamentally, study conceptualisation and design. Reports from the ECA survey highlight the instability in primary and comorbid diagnoses in community surveys (65, 71), perhaps influenced by recall bias and reporting variations due to stigma (69). The ECA studies (65, 71) reported comorbidities including mood disorders, anxiety disorders, substance use disorders, schizophrenia, and schizophreniform disorders. Those with a stable diagnosis across waves 1 and 2 had lower age of onset and higher rates of depressive and anxiety disorders, substance use and schizophrenia (71). The Singapore national mental health
