Table 2.

Characteristics of included studies

Country	Publication	NHMRC level of evidence	Method	Evidence of cultural concerns regarding EPDS
Australia	Freeman et al. 2017 New South Wales	Level III-3 Reported OR > 1, p ≤ 0.05, increased risk of postnatal service affecting EPDS acceptance Unreported sample size calculation	Retrospective cohort study (prognosis) 424 Aboriginal children and their mothers (n = 215)	Women: yes. 57% (152 of 267 offered) rejected the EPDS. Increased frequency of CFHN interaction positively associated with mother accepting EPDS screening. HCP: unreported.
	Gausia et al. 2013 Nation-wide	Level III-3 Reported OR > 1, p < 0.05, Aboriginal, birthing and care plans, financial, < 4 antenatal visits increased risk of missing emotional wellbeing screening	Retrospective cohort study (prognosis) 36 primary health centre (all Australian states) 674/797 pregnant Aboriginal women	Women: unreported. HCP: yes. Language perceived as a concern for Aboriginal mothers. Midwives reported women required assistance when completing the EPDS (anecdotal).
	Hayes et al. 2010 Queensland	Level III-3 EPDS > 9 (not at risk) vs EPDS > 12 (at risk)	Cross-sectional case-control study 92/110 Aboriginal mothers	Women: yes. Application of the translated TAIHS EPDS. HCP: yes. Reference groups revised and altered specific words or simplification of sentence structure.
	WAPMHU and WNHS 2011 Western Australia	Level III-3 Used baseline as comparison point after the service was implemented Over multiple time-points (32 months)	Longitudinal mixed-methods, comparative interrupted time series 34 Aboriginal women 14 HCP	Women: unreported. HCP: no. Midwives perceived that mothers genuinely responded to the EPDS.
	Campbell et al. 2008 Queensland	Level IV Did not use reference standard Reported z-scores, 9 women did both measurements at both time-points	Case-series validation study (diagnostic accuracy) 210 Aboriginal women	Women: yes. TAIHS and MIT EPDS modified to be meaningful for women. HCP: yes. Words and sentence structures altered by reference group.
	Highet and Goddard 2014 Nation-wide	Level IV Unreported total n of HCP surveyed Cross-sectional, undetermined casual relationship	Mixed-methods, interviews 82 services surveyed, 26% HCP were Aboriginal 28 HCP interviewed, 35% were Aboriginal	Women: unreported. HCP: yes. Some HCP thought EPDS was not meaningful to use, insulting, and “white fella ask questions”.
	Marley et al. 2017 Western Australia	Level IV Used GP diagnosis Cross-sectional, undetermined casual relationship	Case-series validation study (diagnostic accuracy) 97 Aboriginal women 15 HCP	Women: yes. 98% accepted KMMS. unreported % accepted EPDS. HCP: yes. 8/9 HCP reported KMMS was more useful and superior to the EPDS.
	Carlin et al. 2019a Western Australia	Level V Consultations with women with a recent experience of receiving perinatal care in the Pilbara region	Qualitative yarning methodology 15 Aboriginal perinatal women	Women: yes. Supported KMMS more than EPDS. HCP: Unreported.
	Hayes et al. 2005 Queensland	Level V Consultations with reference group members (5 TAIHS, 12 MTI, 5 PI)	Expert opinions 141 antenatal women 127 postnatal women	Women: yes. Piloted translated EPDS versions. HCP: yes. Reference groups modified language, cultural images and colours that were more meaningful to women.
	Kotz et al. 2016 Western Australia	Level V Consultations with midwives and CHNs, Aboriginal women from community as advisory group	Expert opinions through yarning 100 Aboriginal people (8 language groups) 72 HCP	Women: yes. Selected preference in wording and provided additional suggestions. HCP: yes. Advisory group modified wording and formatting.
	Queensland Health 2013 Queensland	Level V Small working party in North QLD Consultations with state-wide partnerships	Expert opinions 43 HCP (11 Aboriginal health workers, 26 registered nurse/midwife)	Women: unreported. HCP: yes. EPDS culturally inappropriate due to language (30%), not asking the right questions (47%), screener lacking cultural expertise (42%).
Canada	Clarke 2008b Saskatchewan	Level IV Used Clinical Interviews DSM-IV diagnosis Reported OR > 1 but p > 0.05	Case-series validation study (diagnostic accuracy) 103 First Nations and Metis women	Women: no. Validated standard EPDS with PDSS and BDI-II. HCP: unreported.
	Clarke 2008a Saskatchewan	Level V Expert opinion, unpublished doctoral thesis Used Structured Clinical Interviews DSM-IV diagnosis	Qualitative interview methods 9 Aboriginal postpartum women	Women: no. Congruence between high EPDS score, diagnosis, and feelings of disconnect with baby. HCP: unreported.
USA	Heck 2018 Nation-wide	Level IV Inclusion criteria was unclear Studies in this review were mainly Level C1 Used DSM, SCID, MINI, ICD-10, CIDI, some unreported reference standard	Systematic review of EPDS (n = 54) and PHQ-9 (n = 7) validation studies with American Indian and Alaska Native (AI/AN) women	Women: no. Validated EPDS with AI/AN women. HCP: unreported.

OR odds ratio that quantities the strengths of the relationship between two variables, CFHN Child and Family Health Nurse, TAIHS Townsville Aboriginal and Islanders Health Services (an Aboriginal community controlled health service in Queensland), WAPMHU Western Australian Perinatal Mental Health Unit, WNHS Women and Newborn Health Service, MTI Mount Isa (city in Queensland), KMMS Kimberley Mums Mood Scale, PI Palm Island, CHN Child Health Nurse, PDSS Postpartum Depression Screening Scale, BDI-II Beck Depression Inventory-II, DSM Diagnostic and Statistical Manual of Mental Disorders, SCID Structured Clinical Interview for DSM, MINI Mini International Neuropsychiatric Interview, ICD-10 The International Classification of Diseases, 10th Revision, CIDI Composite International Diagnostics Interview, PHQ-9 Patient Health Questionnaire-9

Level C of qualitative studies, descriptive, correlational, integrative reviews, and systematic review/RCT with inconsistent results on AACN Levels of Evidence (Armola et al. 2009). Heck’s (2018) review consisted of Level 3 studies on American Association of Critical-Care Nurse’s Evidence-Leveling System (Armola et al. 2009), equivalent to a low rating of NHRMC III-3 and IV level of evidence