Skip to main content
PMC home page
Canadian Journal of Public Health = Revue Canadienne de Santé Publique logoLink to Canadian Journal of Public Health = Revue Canadienne de Santé Publique
. 2019 Feb 14;110(4):440–452. doi: 10.17269/s41997-019-00182-8

Postpartum depression prevalence and risk factors among Indigenous, non-Indigenous and immigrant women in Canada

Nihaya Daoud 1,✉, Kristen O’Brien 2, Patricia O’Campo 2,3, Sarah Harney 4, Evelyn Harney 5, Kerry Bebee 5, Cheryllee Bourgeois 5, Janet Smylie 2,3
PMCID: PMC6964473  PMID: 30767191

Abstract

Objectives

The social position of different minority groups in contemporary societies suggests different risk factors for postpartum depression (PPD). In this study, we used two cut-offs of the Edinburgh Postpartum Depression Scale (EPDS) to examine prevalence and risk factors for PPD among mothers participating in the Canadian Maternity Experiences Survey (MES), and to compare Indigenous, Canadian-born non-Indigenous and immigrant mothers.

Methods

We used cross-sectional nationwide data from the 2006 MES (unweighted N = 6237, weighted N = 74,231) and conducted multivariate logistic regression models for EPDS ≥ 10 and EPDS ≥ 13 to explore risk factors for the total sample of mothers and in each study group.

Results

Prevalence (%, 95 % CI) of EPDS ≥ 10 and EPDS ≥ 13 was significantly higher among immigrant (12.2 %, 10.2–14.2 and 24.1 %, 21.5–26.7) and Indigenous (11.1 %, 7.5–14.7 and 21.2 %, 16.5–25.9) compared to Canadian-born non-Indigenous mothers (5.6 %, 4.9–6.3 and 12.9 %, 11.9–13.9). Multivariate analysis of the total sample showed similar risk factors for EPDS ≥ 10 and EPDS ≥ 13 (ethnicity, low education, ≥ low income cut-off, taking antidepressants, experiencing abuse, low social support). Stratification by study group revealed differing risk factors in each group for EPDS ≥ 10 and EPDS ≥ 13. Indigenous mothers had the most distinct risk factors, followed by immigrant mothers. Non-indigenous Canadian-born mothers had risk factors most similar to the total sample.

Conclusion

Differing prevalence and risk factors for PPD within and across study groups suggest that instead of a universal approach, tailored programs and services to prevent PPD in Indigenous, immigrant and non-Indigenous Canadian-born groups could better protect the mental health of Canadian mothers.

Keywords: Postpartum depression; Edinburgh Postnatal Depression Scale (EPDS); Risk factors; Canada; Indigenous, immigrant, and non-Indigenous Canadian-born mothers

Introduction

Postpartum depression (PPD), a common health disorder occurring after birth among women of childbearing age (O'Hara & McCabe, 2013), brings detrimental health effects for mothers (Beck, 2003), newborns (Gress-Smith et al., 2012), and partners (Falah-Hassani et al., 2015). PPD affects approximately 13 % of new mothers (O’Hara & Swain, 1996) and 7–19 % of mothers during the first 12 weeks postpartum (Cohen et al., 2010). However, PPD is not experienced evenly across population groups, with higher prevalence among mothers belonging to minority (Wei et al., 2008) and marginalized groups (Clare & John, 2012; Jesse et al., 2005)(Da-Silva et al., 1998; Hobfall et al., 1995; Sit & Wisner, 2009), including immigrant and Indigenous mothers (Falah-Hassani et al., 2015; Wei et al., 2008). Despite this, few studies in Canada have compared PPD risk factors among mothers of different minority groups. Previous research suggests that these groups occupy different social positions in society that might differentially affect the determinants of mental health and PPD. For Indigenous mothers in Canada, the social determinants of health that might impact PPD include experiences of historical trauma (Brave Heart, 2003), inequalities (Noël & Larocque, 2009; Smylie & Andersons, 2006; Smylie, 2008; Statistics Canada, 2011) and discrimination (CIHI, Canadian Institute for Health Information, 2009). Among immigrant mothers in Canada, risk factors for PPD include low income, language barriers, lack of social support (Dennis et al., 2017), being in Canada less than 2 years and living in neighbourhoods with high concentrations of immigrants (Dennis et al., 2017). However, there has not been a comparison in Canada of risk factors across immigrant, Indigenous, and Canadian-born non-Indigenous population groups. Most research into PPD, prevalence and risk factors in Canada have compared either Indigenous with non-Indigenous mothers or immigrant with non-immigrant mothers.

In general, research on PPD among Indigenous mothers has lagged behind that conducted among immigrant mothers (Bowen et al., 2014). For example, while a recent hospital-based study found higher PPD prevalence among refugees and recent immigrant mothers (Dennis et al., 2017), and another study identified PPD risk factors among immigrant mothers (Falah-Hassani et al., 2015), two studies that did examine PPD among Indigenous mothers were based only on small sample sizes, did not consider risk factors and reached contradictory results—one found significant differences in PPD among antenatal Indigenous and non-Indigenous mothers (Bowen & Muhajarine, 2006), while the other found none (Bowen et al., 2009).

The 2006 Canadian Maternity Experiences Survey (MES), which forms the basis of the current analysis, found a prevalence of 8.5 % minor risk for PPD in the general population of Canadian mothers (measured by the Edinburgh Postpartum Depression Scale (EPDS) with the cut-off of ≥ 10 to ≤ 12) and 8.7 % major risk of PPD (EPDS cut-off of ≥ 13) (Lanes et al., 2011). Risk factors for PPD identified in the study included young maternal age, immigrant status, antidepressant use, low household income, low availability of support after pregnancy, and high stress during pregnancy (Lanes et al., 2011). Rates of minor and major risk for PPD were also higher in geographical areas with a higher concentration of Indigenous and rural women (Lanes et al., 2011). A recent analysis of the MES data found that the prevalence of PPD is higher among each of three groups of Indigenous mothers (First Nations off-reserve, Inuit, and Métis) compared to non-Indigenous Canadian-born mothers (12.9 %, 10.6 %, and 9.1 %, respectively, compared to 5.6 %). These differences became non-significant after adjustment for different risk factors (Nelson et al., 2018). However, neither the MES data nor other data have been used to compare prevalence and risk factors across immigrant, Indigenous and Canadian-born non-Indigenous population groups. Conducting such a comparison is important for yielding evidence that would guide better programming to prevent PPD among mothers in different minority groups in Canada.

Our paper addresses this gap in the literature by analyzing the MES data to:

  1. Compare prevalence of and risk factors for EPDS ≥ 10 and EPDS ≥ 13 in three groups: Indigenous, Canadian-born non-Indigenous, and immigrant mothers in Canada; and

  2. Compare risk factors for each group with those of the total sample. Using two EPDS cut-offs can reveal different risk factors among these three groups of mothers based on the severity of PPD, which might be important for maternity clinicians.

Methods

Data and participants

We obtained data from the 2006/7 Maternity Experiences Survey (MES) (detailed elsewhere) (Dzakpasu et al., 2008; Chalmers et al., 2008). The MES, conducted by Statistics Canada, is the first nationwide survey on pregnancy, birth and early postpartum in Canada. Eligible mothers (age 15 and older, with a live singleton birth) were randomly selected from the 2006 Census using a stratified sampling procedure. Women were proportionally sampled according to their geographic distribution in the general Canadian population. Information about the study was mailed to mothers in the sample, after which trained female interviewers contacted them by phone. Interviews lasted about 45 min and were conducted in English, French and 13 other languages. The final sample included 6421 pairs of babies and their mothers who were 5 to 14 months after birth. After removing missing information on the main outcome and key predictor variables and applying weights, we arrived at 74,231 participants: N = 3152 Indigenous, N = 53,803 Canadian born non-Indigenous and N = 17,276 immigrant mothers. Indigenous mothers included First Nations, Inuit, and Métis mothers from all provinces and territories. For operational reasons, however, this group excluded First Nations mothers living on reserve (Dzakpasu et al., 2008). Indigenous mothers were sampled according to the proportion of the population they made up in geographic areas across Canadian provinces. However, they were not oversampled. Therefore, we were cautious in interpreting results related to these mothers and generalizing from these to all Indigenous mothers in Canada (Dzakpasu et al., 2008). The overall response rate for the MES was 77.9 % (Chalmers et al., 2008). The response rate by our study subgroups was not presented in the MES study report. However, the response rate varied by province, ranging from highest (86 %) in Nova Scotia to lowest (64 %) in Nunavut, where many Indigenous mothers live. Overall, the MES is considered a representative sample of mothers in Canada (Dzakpasu et al., 2008).

Ethics approval

The MES was approved by Statistics Canada (CIHR, 2007). Our secondary analysis was approved by St. Michael’s Hospital’s Research Ethics Board. In keeping with standards for secondary data analysis of Indigenous peoples, our analysis was conducted in collaboration with the Native Women’s Association of Canada and the National Aboriginal Council of Midwives (CIHR, 2007; Royal Commission on Aboriginal Peoples, 1993).

Main outcome measures

We measured PPD using the Edinburgh Postnatal Depression Scale (EPDS), which includes 10 questions evaluating a woman’s feelings over the past week (Cox et al., 1987). Answers range from 0 to 3, and the total questionnaire score ranges from 0 to 30. A cut-off point of 10 or less represents a minor risk for PPD; a cut-off of 13 and above represents medium to high risk. The EPDS is sensitive to PPD (Cox et al., 1987; Murray & Carothers, 1990) and has been shown to be a reliable tool among Indigenous women in Australia (Campbell et al., 2008) and Canada (Clarke, 2008). In this study, we used two cut-offs for the EPDS: EPDS ≥ 10 for total scores ≥ 10 and EPDS ≥ 13 for total scores ≥ 13. Internal consistency of the EPDS was high for the three study groups (see Cronbach’s alpha in Table 2a). Among Indigenous mothers, Cronbach’s alpha for EPDS ≥ 13 was also high (Table 2b).

Table 2.

Distribution of EPDS ≥ 13 and EPDS ≥ 10 and Cronbach’s alpha for subgroups of women in the study

a. Distribution of EPDS ≥ 13 and EPDS ≥ 10 among Indigenous, non-Indigenous Canadian-born and immigrant women (total weighted N = 74,231)**
Indigenous Canadian-born Non-Indigenous Canadian-born Immigrant p value
Weighted N = 3152 Weighted N = 53,803 Weighted N = 17,276
Cronbach’s alpha (EPDS-13) 0.832 0.836 0.793
EPDS ≥ 13 N % N % N % < 0.001
 < 13 2802 88.9 50,797 94.4 15,173 87.8
 ≥ 13 350 11.1 3006 5.6 2103 12.2
EPDS ≥ 10 N % N % N % < 0.001
 < 10 2484 78.8 46,863 87.1 13,110 75.9
 ≥ 10 668 21.2 6940 12.9 4166 24.1
b. Distribution of EPDS ≥ 13 and EPDS ≥ 10 among Indigenous women subgroups (weighted N = 3059)
First Nations Métis Inuit p value
Total weighted N = 1423 Total weighted N = 1428 Total weighted N = 208
Cronbach’s alpha* (EPDS-13) 0.835 0.830 0.807
EPDS ≥ 13 N % N % N % 0.2424
 < 13 1237 86.9 1309 91.6 187 90.1
 ≥ 13 186 13.1 120 8.4 20 9.9
EPDS ≥ 10 N % N % N % 0.1997
 < 10 1082 76.0 1182 82.8 167 80.2
 ≥ 10 342 24.0 246 17.2 41 19.8
Open in a new tab

*Cronbach’s alpha was calculated based on weighted data

**Total unweighted N = 6237

Independent variables

Women’s study groups by ethnicity and immigrant status were determined by first asking participants ‘Were you born in Canada?’ We categorized mothers responding negatively as immigrants (living in Canada for less than 5 years). Next, women born in Canada were asked ‘Are you an Aboriginal person, that is, First Nations, Métis or Inuit?’ (Statistics Canada, 2006a).

Mother’s age at childbirth: we categorized age into ‘15–24 years’ and ‘25–50 years’.

Marital status: we dichotomized this status into ‘lone’ (not married, single, divorced, separated or widowed) or ‘cohabitating’ (married or common-law).

Education included three levels of achievement: ‘less than high school diploma’, ‘high school diploma or vocational training’ and ‘greater than high school diploma or vocational certificate.’

Low income cut-off after tax (LICO-AT): women could be ‘above LICO’, ‘at or below LICO’, or this information could be ‘missing’. The LICO-AT reflects whether respondents lived in a household spending 20 percentage points more of their after-tax income than the average family on food, shelter and clothing, thus leaving less income for other expenses (Statistics Canada, 2006b).

Experience of abuse: either a woman reported experiencing at least one of ten acts of physical or sexual violence during the past 2 years at the hands of a spouse, partner or anyone else, or a woman reported no exposure to abuse during the past 2 years (Statistics Canada, 2006a; Statistics Canada, 1993).

Parity: number of times a respondent gave birth to a live baby, including the birth of the last baby. Answers were dichotomized into ‘one child’ and ‘two or more children’.

Taking antidepressants before pregnancy: determined by whether the mother reported being prescribed antidepressants or diagnosed with depression before pregnancy or reported not having been prescribed antidepressants or diagnosed with depression in that period.

Social support was measured based on the following question: “People sometimes look to others for companionship, assistance or other types of support. During your pregnancy, how often was support available to you when you needed it?” Answers were dichotomized into ‘no support’ (none of the time, a little of the time, or some of the time) or ‘some support’ (most of the time or all of the time).

Stressful life events in the past 12 months: measured based on how respondents characterized the stress of most days during the 12 months before the most recent childbirth. Responses were categorized into ‘not stressful’, ‘somewhat stressful’, or ‘very stressful’.

Statistical analysis

After applying weights, we calculated two cut-off points for EPDS ≥ 10 and ≥ 13 and examined the internal consistency of each outcome for each study group. For prevalence, we estimated the percentage (and 95 % confidence intervals, CIs) of EPDS ≥ 10 and EPDS ≥ 13 scores above the cut-offs for the weighted subsamples and compared estimates between the study groups (Canadian-born Indigenous and non-Indigenous and immigrant), and among the three groups of Indigenous mothers (First Nations, Inuit and Métis) using a chi-square test. To determine risk factors for EPDS ≥ 10 and EPDS ≥ 13 for the total sample, we fit various multivariate logistic regression models while adjusting for independent variables that showed significant differences in PPD (p < 0.05) between study groups in the univariate associations. The first model was unadjusted and examined the association between ethnicity and PPD. Subsequent models sequentially adjusted for different independent variables are as follows: model 2 adjusted for age, model 3 added socio-economic status (SES—education and LICO-AT), model 4 added marital status, model 5 added antidepressant use before pregnancy, model 6 added experiences of abuse, and the final model considered social support in addition to all previous variables. Finally, to explore the specific risk factors for depression as measured by EPDS ≥ 10 and EPDS ≥ 13 in each study group (Indigenous and non-Indigenous Canadian-born and immigrant), we conducted multivariate logistic regression analysis while considering all independent variables (except ethnicity and immigrant status). We used SAS version 9.4 (SAS Institute Inc., Cary, NC, USA) for all analyses. We followed the MES Users’ Guide and did not report estimates based on cell counts < 5 (Statistics Canada, 2006c). We report adjusted odds ratios and 95 % CI that we calculated while applying bootstraps to adjust for the complex study design.

Results

We found significant differences between the three study groups with respect to the independent variables (Table 1). Indigenous mothers were younger, less likely to report cohabitation, more likely to be living at or below LICO-AT, more likely to have taken antidepressants before pregnancy and to have experienced abuse in the past 2 years, and had lower levels of education compared to immigrant and non-Indigenous Canadian-born mothers. Immigrant mothers reported lower social support than the two Canadian-born groups of mothers. Parity and stressful life events were not significantly different across study groups, although, compared to others, more Indigenous mothers responded that most days in the past 12 months were ‘very stressful’.

Table 1.

Distribution of study variables (weighted N = 74,231)

Total weighted N = 74,231 Indigenous Canadian-born weighted N = 3152 Non-Indigenous Canadian-born weighted N = 53,803 Immigrant weighted N = 17,276 p value
N % (95% CI) N % (95% CI) N % (95% CI) N % (95% CI)
Age < 0.001
 15–24 years 11,728 15.8 (14.8, 16.8) 1227 38.9 (33.2, 44.7) 8759 16.3 (15.2, 17.4) 1742 10.1 (8.3, 11.9)
 25–50 years 62,503 84.2 (83.2, 85.2) 1925 61.1 (55.3, 66.8) 45,045 83.7 (84.8, 72.1) 15,534 89.9 (88.1, 91.7)
Marital status < 0.001
 Lone 6154 8.3 (7.6, 9.0) 799 25.3 (20.2, 30.5) 4523 8.4 (7.6, 9.2) 832 4.8 (3.5, 6.1)
 Cohabitating 68,077 91.7 (91.0, 92.4) 2353 74.7 (69.5, 79.8) 49,280 91.6 (90.8, 92.4) 16,444 95.2 (93.9, 96.5)
Parity 0.09
 1 child 33,794 45.6 (44.2, 46.9) 1410 44.7 (38.8, 50.7) 24,965 46.4 (44.9, 48.0) 7419 43.0 (40.0, 46.0)
 2 children or more 40,370 54.4 (53.1, 55.8) 1742 55.3 (49.3, 61.2) 28,785 53.6 (52.0, 55.1) 9842 57.0 (54.0, 60.0)
Education < 0.001
 Less than HS 5552 7.5 (6.8, 8.2) 749 23.7 (18.8, 28.7) 3588 6.7 (5.9, 7.4) 1216 7.0 (5.5, 8.6)
 HS diploma/vocational 19,658 26.5 (25.3, 27.7) 1380 43.8 (37.8, 49.7) 14,937 27.8 (26.4, 29.1) 3341 19.3 (17.0, 21.7)
 Greater than HS diploma/vocational 49,020 66.0 (64.8, 67.3) 1023 32.5 (26.8, 38.1) 35,279 65.6 (64.1, 67.0) 12,718 73.6 (71.0, 76.3)
Household income < 0.001
 $0–$29,999 11,763 15.8 (14.9, 16.8) 1136 36.0 (30.4, 41.7) 6964 12.9 (11.9, 14.0) 3663 21.2 (18.7, 23.7)
 $30,000–$59,999 21,550 29.0 (27.8, 30.3) 822 26.1 (20.8, 31.3) 15,407 28.6 (27.2, 30.0) 5321 30.8 (28.0, 33.6)
  ≥ $60,000 37,029 49.9 (48.5, 51.2) 869 27.6 (22.1, 33.0) 29,387 54.6 (53.1, 56.2) 6774 39.2 (36.3, 42.2)
 Missing 3889 5.2 (4.6, 5.8) 326 10.3 (6.8, 13.9) 2045 3.8 (3.2, 4.4) 1518 8.8 (7.1, 10.5)
LICO-AT < 0.001
 At/below LICO 13,512 18.2 (17.2, 19.2) 1193 37.9 (32.1, 43.6) 7436 13.8 (12.8, 14.9) 4882 28.3 (25.5, 31.0)
 Above LICO 54,667 73.6 (72.4, 74.8) 1534 48.7 (42.7, 54.6) 42,866 79.7 (78.4, 80.9) 10,266 59.4 (56.5, 62.4)
 Missing 6053 8.2 (7.4, 8.9) 425 13.5 (9.4, 17.5) 3501 6.5 (5.7, 7.3) 2128 12.3 (10.3, 14.3)
Antidepressant use before pregnancy < 0.001
 No 62,708 84.5 (83.5, 85.4) 2399 76.1 (70.9, 81.3) 44,304 82.3 (81.2, 83.5) 16,005 92.6 (91.1, 94.2)
 Yes 11,523 15.5 (14.6, 16.5) 753 23.9 (18.7, 29.1) 9499 17.7 (16.5, 18.8) 1271 7.4 (5.8, 8.9)
Experiencing any abuse < 0.001
 No 66,026 89.0 (88.2, 89.9) 2172 69.3 (63.8, 74.7) 47,557 88.5 (87.5, 89.4) 16,297 94.5 (93.1, 95.8)
 Yes 8124 11.0 (10.1, 11.8) 964 30.7 (25.3, 36.2) 6210 11.5 (10.5, 12.5) 950 5.5 (4.2, 6.9)
Stressful life events 0.08
 Not stressful 32,016 43.3 (41.9, 44.6) 1232 39.1 (33.2, 44.9) 22,881 42.7 (41.2, 44.2) 7903 45.9 (42.9, 48.9)
 Somewhat stressful 33,102 44.7 (43.4, 46.1) 1418 45.0 (39.1, 50.9) 24,347 45.4 (43.8, 46.9) 7338 42.6 (39.6, 45.5)
 Very stressful 8900 12.0 (11.1, 12.9) 502 15.9 (11.6, 20.3) 6418 12.0 (10.9, 13.0) 1980 11.5 (9.6, 13.4)
Social support during pregnancy < 0.001
 No support or some 31,356 42.3 (41.0, 43.7) 1264 40.3 (34.4, 46.2) 20,749 38.6 (37.1, 40.2) 9342 54.3 (51.3, 57.3)
 All times 42,699 57.7 (56.3, 59.0) 1873 59.7 (53.8, 65.6) 32,963 61.4 (59.8, 62.9) 7863 45.7 (42.7, 48.7)
Open in a new tab

Total unweighted N = 6237

Prevalence of EPDS ≥ 13 for the total sample was 7.4 % (95 % CI = 6.6, 8.1), and prevalence of EPDS ≥ 10 was 15.9 % (95 % CI = 14.9, 16.8). Across study groups, prevalence of both EPDS ≥ 13 and EPDS ≥ 10 was significantly higher among immigrant (12.2 % and 24.1 %, respectively) and Indigenous mothers (11.1 % and 21.2 %, respectively) compared to non-Indigenous Canadian-born mothers (5.6 % and 12.9 %, respectively) (Table 2a).

We found no significant differences when comparing prevalence of EPDS ≥ 10 and EPDS ≥ 13 across First Nations, Inuit and Métis groups, although a trend existed towards higher PPD among First Nations mothers compared to Inuit and Métis mothers (Table 2b).

Results of the multivariate analysis for the total sample (Table 3) show that the unadjusted odds ratios (ORs) for EPDS ≥ 13 were a little more than twice as high among Indigenous and immigrant mothers compared to non-Indigenous Canadian-born mothers (model 1). In addition to immigrant and Indigenous identity, risk factors for EPDS ≥ 13 included low education, living at or below LICO-AT, taking antidepressants before pregnancy, experiencing abuse, and some or no social support. We found similar trends and risk factors in the multivariate results for EPDS ≥ 10 (Table 4). However, the associations (ORs) were stronger for EPDS ≥ 13 than for EPDS ≥ 10.

Table 3.

Multivariable models for EPDS ≥ 13 among participants in the Canadian MES (weighted N = 74,231)

Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7
OR (95% CI) OR (95% CI) OR (95% CI) OR (95% CI) OR (95% CI) OR (95% CI) OR (95% CI)
Ethnicity
 Indigenous 2.11 (1.44, 3.09) 1.87 (1.28, 2.74) 1.47 (1.00, 2.17) 1.45 (0.99, 2.12) 1.40 (0.95, 2.05) 1.24 (0.84, 1.85) 1.27 (0.85, 1.90)
 Immigrant 2.34 (1.87, 2.93) 2.43 (1.94, 3.05) 2.19 (1.74, 2.77) 2.24 (1.77, 2.83) 2.65 (2.09, 3.37) 2.85 (2.24, 3.64) 2.53 (1.97, 3.24)
 Non-Indigenous Canadian-born 1 1 1 1 1 1 1
Age
 15–24 years 1.63 (1.29, 2.07) 1.15 (0.88, 1.52) 1.12 (0.84, 1.49) 1.11 (0.83, 1.48) 0.99 (0.73, 1.34) 1.07 (0.79, 1.45)
 25–50 years 1 1 1 1 1
Education
 Less than HS 1.70 (1.18, 2.45) 1.68 (1.16, 2.41) 1.63 (1.13, 2.36) 1.55 (1.07, 2.26) 1.59 (1.09, 2.30)
 HS diploma/vocational 1.38 (1.08, 1.76) 1.38 (1.08, 1.76) 1.34 (1.05, 1.72) 1.32 (1.02, 1.69) 1.30 (1.01, 1.68)
 Greater than HS diploma/vocational 1 1 1 1 1
LICO-AT
 At/below LICO 1.89 (1.47, 2.44) 1.82 (1.41, 2.35) 1.81 (1.40, 2.34) 1.73 (1.33, 2.25) 1.67 (1.28, 2.18)
 Above LICO 1 1 1 1 1
 Missing 1.29 (0.90, 1.86) 1.25 (0.86, 1.80) 1.24 (0.85, 1.81) 1.28 (0.88, 1.86) 1.27 (0.86, 1.86)
Marital status
 Lone 1.25 (0.91, 1.72) 1.19 (0.86, 1.64) 0.99 (0.70, 1.39) 1.00 (0.72, 1.41)
 Cohabitating 1 1 1 1
Antidepressant use before pregnancy
 Yes 2.86 (2.26, 3.63) 2.69 (2.12, 3.42) 2.58 (2.02, 3.29)
 No 1 1 1
Experiencing any abuse
 Yes 2.42 (1.85, 3.15) 2.17 (1.65, 2.84)
 No 1 1
Social support
 No support or some 2.42 (1.95, 3.02)
 All times 1
Open in a new tab

Unweighted N = 6237

Table 4.

Multivariable models for EPDS ≥ 10 among participants in the Canadian MES (weighted N = 74,231)

Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7
OR (95% CI) OR (95% CI) OR (95% CI) OR (95% CI) OR (95% CI) OR (95% CI) OR (95% CI)
Ethnicity
 Indigenous 1.82 (1.36, 2.42) 1.69 (1.26, 2.25) 1.36 (1.02, 1.83) 1.34 (1.00, 1.80) 1.30 (0.97, 1.74) 1.17 (0.86, 1.59) 1.18 (0.86, 1.63)
 Immigrant 2.15 (1.82, 2.53) 2.20 (1.86, 2.59) 1.99 (1.68, 2.36) 2.03 (1.72, 2.41) 2.34 (1.97, 2.79) 2.50 (2.10, 2.98) 2.23 (1.87, 2.67)
 Non-Indigenous Canadian-born 1 1 1 1 1 1 1
Age
 15–24 years 1.38 (1.16, 1.66) 1.00 (0.82, 1.23) 0.97 (0.79, 1.20) 0.96 (0.78, 1.19) 0.86 (0.69, 1.08) 0.92 (0.73, 1.14)
 25–50 years 1 1 1 1 1
Education
 Less than HS 1.53 (1.16, 2.03) 1.50 (1.13, 1.99) 1.47 (1.11, 1.96) 1.41 (1.06, 1.88) 1.42 (1.06, 1.89)
 HS diploma/vocational 1.34 (1.13, 1.59) 1.34 (1.13, 1.59) 1.31 (1.10, 1.56) 1.29 (1.08, 1.53) 1.28 (1.08, 1.53)
 Greater than HS diploma/vocational 1 1 1 1 1
LICO-AT
 At/below LICO 1.76 (1.46, 2.12) 1.68 (1.39, 2.04) 1.67 (1.38, 2.03) 1.59 (1.30, 1.94) 1.55 (1.27, 1.90)
 Above LICO 1 1 1 1 1
 Missing 1.47 (1.14, 1.90) 1.42 (1.09, 1.84) 1.43 (1.10, 1.86) 1.48 (1.13, 1.92) 1.47 (1.12, 1.93)
Marital status
 Lone 1.28 (1.00, 1.63) 1.22 (0.95, 1.56) 1.02 (0.78, 1.33) 1.04 (0.80, 1.35)
 Cohabitating 1 1 1 1
Antidepressant use before pregnancy
 Yes 2.66 (2.23, 3.17) 2.53 (2.12, 3.03) 2.46 (2.05, 2.95)
 No 1 1 1
Experiencing any abuse
 Yes 2.48 (2.01, 3.05) 2.27 (1.84, 2.81)
 No 1 1
Social support during pregnancy
 No support or some 2.18 (1.87, 2.54)
 All times 1
Open in a new tab

Unweighted N = 6237

Looking at the risk factors for EPDS ≥ 10 and EPDS ≥ 13, we found differences within and between each of the three study groups (Table 5). Indigenous mothers had distinct risk factors compared to the total population and the other two study groups. Risk factors for Indigenous mothers also differed for EPDS ≥ 13 and EPDS ≥ 10: for EPDS≥13, risk factors included antidepressant use prior to pregnancy and experiencing abuse, while for EPDS ≥ 10, risk factors were low education and stressful life events in addition to antidepressant use.

Table 5.

Multivariable models for EPDS ≥ 13 and EPDS ≥ 10 among Indigenous, immigrant and non-Indigenous Canadian-born mothers in the Canadian MES (total weighted N = 74,231)

EPDS ≥ 13 EPDS ≥ 10
Indigenous weighted N = 3152 Non-Indigenous Canadian-born weighted N = 53,803 Immigrant weighted N = 17,276 Indigenous weighted N = 3152 Non-Indigenous Canadian-born weighted N = 53,803 Immigrant weighted N = 17,276
OR (95% CI) OR (95% CI) OR (95% CI) OR (95% CI) OR (95% CI) OR (95% CI)
Age
 15–24 years 1.64 (0.69, 3.92) 1.24 (0.83, 1.86) 1.18 (0.60, 2.35) 1.08 (0.47, 2.47) 1.11 (0.84, 1.48) 0.69 (0.39, 1.23)
 25–50 years 1 1 1 1 1 1
Education
 Less than HS 3.18 (0.78, 12.90) 1.85 (1.09, 3.11) 1.07 (0.49, 2.34) 5.45 (1.94, 15.36) 1.70 (1.17, 2.48) 0.95 (0.50, 1.79)
 HS diploma/vocational 0.76 (0.25, 2.28) 1.71 (1.24, 2.34) 1.01 (0.62, 1.65) 1.91 (0.81, 4.49) 1.40 (1.12, 1.75) 1.21 (0.84, 1.74)
 Greater than HS diploma 1 1 1 1 1 1
LICO-AT
 At/below LICO 0.72 (0.25, 2.12) 1.41 (0.95, 2.08) 2.35 (1.50, 3.69) 0.53 (0.24, 1.17) 1.41 (1.06, 1.87) 2.15 (1.50, 3.07)
 Above LICO 1 1 1 1 1 1
 Missing 0.38 (0.08, 1.87) 0.77 (0.40, 1.48) 2.30 (1.28, 4.13) 0.47 (0.16, 1.33) 1.17 (0.76, 1.79) 2.44 (1.53, 3.89)
Marital status
 Lone 2.46 (0.99, 6.14) 0.72 (0.45, 1.17) 0.89 (0.42, 1.92) 1.82 (0.84, 3.95) 0.80 (0.56, 1.14) 1.04 (0.54, 2.00)
 Cohabitating 1 1 1 1 1 1
Parity
 1 child 0.92 (0.34, 2.49) 0.85 (0.64, 1.13) 1.02 (0.67, 1.55) 1.14 (0.48, 2.71) 0.94 (0.77, 1.14) 1.13 (0.82, 1.55)
 2 children or more 1 1 1 1 1 1
Antidepressant use before pregnancy
 Yes 2.74 (1.12, 6.70) 2.33 (1.73, 3.15) 1.51 (0.78, 2.90) 3.25 (1.56, 6.77) 2.17 (1.76, 2.68) 1.52 (0.90, 2.55)
 No 1 1 1 1 1 1
Experiencing any abuse
 Yes 2.39 (1.02, 5.57) 1.79 (1.26, 2.55) 1.54 (0.71, 3.34) 1.43 (0.72, 2.85) 2.04 (1.57, 2.64) 1.81 (1.02, 3.23)
 No 1 1 1 1 1 1
Stressful events
 Not stressful 1 1 1 1 1 1
 Somewhat stressful 1.96 (0.64, 5.99) 2.62 (1.77, 3.88) 1.65 (1.04, 2.61) 2.73 (1.23, 6.04) 2.58 (2.04, 3.27) 2.00 (1.43, 2.80)
 Very stressful 2.88 (0.74, 11.27) 7.26 (4.66, 11.31) 4.38 (2.47, 7.74) 4.99 (1.83, 13.63) 4.93 (3.66, 6.64) 4.34 (2.66, 7.06)
Social support during pregnancy
 No support or some 1.61 (0.60, 4.28) 2.31 (1.74, 3.08) 2.09 (1.37, 3.10) 1.58 (0.79, 3.16) 1.94 (1.60, 2.36) 2.20 (1.63, 2.98)
 All times 1 1 1 1 1 1
Open in a new tab

Unweighted N = 6237

Immigrant mothers also had different risk factors for EPDS ≥ 13 and EPDS ≥ 10 compared to the total sample and other groups. Risk factors for EPDS ≥ 13 among immigrant women included living at or below LICO-AT, stressful life events, and low social support. For EPDS ≥ 10, we found the same risk factors, as well as that of experiencing abuse.

Risk factors for PPD among non-Indigenous Canadian-born mothers were most similar to those of the total sample. For EPDS ≥ 13, risk factors included low education, antidepressant use before pregnancy, experiencing abuse, stressful life events, and low social support. We found the same risk factors for EPDS ≥ 10, with the addition of living at or below LICO-AT. Notably, age and parity were not significant risk factors in any of the study populations. A summary of the multivariable associations can be found in Appendix 1.

Discussion

Our study revealed that Indigenous and immigrant mothers in Canada are at higher risk for PPD compared to non-Indigenous Canadian-born mothers (almost twofold EPDS ≥ 10 and EPDS ≥ 13), with immigrant women experiencing slightly higher risk for PPD than Indigenous mothers. In the total sample of mothers, we found similar risk factors for EPDS ≥ 13 and EPDS ≥ 10, including (in addition to ethnicity) low education, ≥ LICO-AT, antidepressant use before pregnancy, experiencing abuse, and low social support. This might suggest that both EPDS ≥ 13 and EPDS ≥ 10 capture similar PPD concepts, given that the risk factors were also the same for the total sample of mothers.

Although the EPDS has been validated for use as a screening tool among young mothers (Friesen et al., 2017) and Indigenous mothers in Canada (Clarke, 2008), it is important to consider issues of sensitivity and specificity of the two EPDS cut-offs among different groups of mothers in Canada, as the multivariate associations for EPDS ≥ 10 and EPDS ≥ 13 revealed different risk factors within and between study groups. While risk factors for non-Indigenous Canadian-born mothers were most similar with those of the total sample—a reasonable result, considering that these mothers compose a high percentage of the sample—risk factors for Indigenous mothers were most distinct, and their risk factors were different for the two EPDS cut-offs. In the Indigenous group, while antidepressant use before pregnancy was a common risk factor for both EPDS cut-offs, any abuse was a significant risk factor for EPDS ≥ 13 only, while low education and stressful life events were important only for EPDS ≥ 10. Further, some PPD risk factors we identified in the total population (low social support and at or below LICO-AT) were not found among Indigenous mothers. Last, among immigrant mothers, risk factors for EPDS ≥ 13 included ≥ LICO-AT, stressful life events, and low social support. For EPDS ≥ 10, the risk factors for this subgroup included any abuse in addition to the previous risk factors identified for EPDS ≥ 13.

While our results found prevalence rates similar to those arrived at in other Canadian studies, they are novel because research on PPD in Canada has compared either Indigenous or immigrant mothers with Canadian-born mothers but not these two minority groups with the majority group. Meanwhile, the higher PPD prevalence found among immigrant mothers aligns with results that compared immigrant with non-immigrant mothers in Canada (Dennis et al., 2017; Lanes et al., 2011). Since we applied different EPDS cut-offs for PPD, we found slightly different risk factors from those identified in previous research based on MES data (Lanes et al., 2011). Our results regarding risk factors support previous evidence showing that low SES, psychological distress (Davey et al., 2011), low social support during pregnancy and after birth (Lanes et al., 2011), and intimate partner violence (Dennis & Vigod, 2013) are risk factors for PPD among Canadian mothers. Also, our results with regard to the risk factors for PPD among immigrant women are consistent with results of a prospective study in Ontario that found that low social support, low income, living in Canada for less than 2 years, and living in communities with a high concentration of immigrants are risk factors for PPD among immigrant women (Ganann et al., 2016). As for Indigenous mothers, few studies in Canada have examined prevalence and risk factors for PPD for this group. Drawing on the same MES data that we used in this study, Lanes et al. (2011) (Lanes et al., 2011) found that the highest rates of PPD were among mothers living in the territories, where there is a high concentration of Indigenous Canadians, compared to other provinces (Lanes et al., 2011). However, these researchers did not examine risk factors within this group of mothers. Our results regarding the unique risks among Indigenous mothers for each EPDS cut-off suggest that using only one such cut-off might miss groups of Indigenous mothers, a consideration when screening for PPD. This result is supported by Nelson et al.’s 2018 study (Nelson et al., 2018), which used the MES data to compare the prevalence and risk factors for PPD within three groups of Indigenous mothers with non-Indigenous mothers in Canada (Nelson et al., 2018). The distinct risk factors for PPD that we identified for Indigenous mothers might relate to higher prevalence of mental health problems in the general Indigenous population in Canada and elsewhere (Bowen et al., 2014; King et al., 2009; Kirmayer et al., 2000). The fact that low social support was not a risk factor for PPD among Indigenous mothers might suggest a different perception of this concept in this population. This needs to be studied in future research.

Limitations

We used cross-sectional data from the MES. These data were collected retrospectively, 6 to 9 months after birth, and up to 14 months for Indigenous mothers, and were based on mothers’ reports of PPD symptoms and risk factors. Given that the risk for PPD can differ for different periods after birth, this data collection method might have affected reports by mothers on both PPD symptoms and risk factors. As well, PPD risk factors among immigrant mothers could differ by country of origin and length of stay in Canada (Ganann et al., 2016), but we did not analyze these variables, as they were not relevant to and could not be compared with the other two groups. Next, the sample of Indigenous mothers was relatively small compared to the other two groups and did not include women living on reserve, who have generally lower SES. Indigenous mothers living on reserve were excluded from the MES for practical reasons (Chalmers et al., 2008). Future research on PPD should include a larger and more representative sample of Indigenous mothers. This would enable the research to yield more generalizable results for this population group. Previous research on Indigenous mothers suggests several contextual factors influencing PPD that were not included in the MES data, including historical trauma, colonialism, discrimination, and the legacy of multi-sectoral, impoverishing, discriminatory policies (King et al., 2009). Research on PPD among Indigenous mothers must consider these variables. Last, because the MES did not collect data about use of health care services for PPD, we could not consider service usage in our analysis. Previous research has shown a gap in access to maternal health care services for Indigenous mothers (Beking, 2013). Further, although the MES includes a representative sample of mothers in Canada, it included only singleton mothers with live babies. Other mothers (with twins or multiples, not live baby, etc.) were excluded, which might have led to a selection bias. The MES collected data via phone calls, and mothers who do not have a telephone at home (e.g., with low socio-economic status) might be less represented in the survey.

Conclusion

This is the first study we know of to compare the prevalence and risk factors of PPD within and between three groups of mothers in Canada: Indigenous, immigrant and non-immigrant Canadian-born. Our findings show that mothers belonging to these different population groups experience different prevalence and risk factors for PPD. This highlights the importance of tailoring prevention and treatment programs to meet the unique needs of groups of mothers in order to protect them from developing PPD and to prevent PPD consequences for both mothers and infants. Our work also shows the need for more adequately powered research into PPD prevalence and risk with large sample sizes of these population groups.

Acknowledgements

We thank the Research Data Centre (RDC) team of the University of Toronto (UoT).

Appendix 1- Summary of multivariate associations of risk factors for EPDS≥13 and EPDS≥10

EPDS≥13 EPDS≥10
Indigenous Non-Indigenous Canadian-born Immigrant Total Indigenous Non-Indigenous Canadian-born Immigrant Total
Ethnicity and immigrant status + +
Age - - - - - - - -
Education - + - + + + - +
≤LICO - - + + - + + +
Marital status - - - - - - -
Parity - - - NR - - - NR
Anti-depressants + + - + + + - +
Experiencing any abuse + + - + - + + +
Stressful life events - + + NR + + + NR
Social support - + + + - + + +
Open in a new tab

NR – not included in the model

Compliance with ethical standards

Conflict of interest

The authors declare that they have no conflict of interest.

Footnotes

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References

  1. Beck CT. Recognizing and screening for postpartum depression in mothers of NICU infants. Advances in Neonatal Care. 2003;3:37–46. doi: 10.1053/adnc.2003.50013. [DOI] [PubMed] [Google Scholar]
  2. Beking D. Access to maternal health care for native Canadians on reserves in Northern Canada. Interdisciplinary Journal of Health Sciences. 2013;3:21–25. [Google Scholar]
  3. Bowen A, Muhajarine N. Prevalence of depressive symptoms in an antenatal outreach program in Canada. JOGNN. 2006;35:492–498. doi: 10.1111/j.1552-6909.2006.00064.x. [DOI] [PubMed] [Google Scholar]
  4. Bowen A, Stewart N, Baetz M, Muhajarine N. Antenatal depression in socially high-risk women in Canada. Journal of Epidemiology and Community Health. 2009;63:414–416. doi: 10.1136/jech.2008.078832. [DOI] [PubMed] [Google Scholar]
  5. Bowen A, Duncan V, Peacock S, et al. Mood and anxiety problems in perinatal Indigenous women in Australia, New Zealand, Canada, and the United States: a critical review of the literature. Transcultural Psychiatry. 2014;51:93–111. doi: 10.1177/1363461513501712. [DOI] [PubMed] [Google Scholar]
  6. Brave Heart MY. The historical trauma response among natives and its relationship with substance abuse: a Lakota illustration. Journal of Psychoactive Drugs. 2003;35:7–13. doi: 10.1080/02791072.2003.10399988. [DOI] [PubMed] [Google Scholar]
  7. Campbell A, Hayes B, Buckby B. Aboriginal and Torres Strait islander women's experience when interacting with the Edinburgh postnatal depression scale: a brief note. The Australian journal of rural health. 2008;16:124–131. doi: 10.1111/j.1440-1584.2007.00930.x. [DOI] [PubMed] [Google Scholar]
  8. Chalmers B, Dzakpasu S, Heaman M, Kaczorowski J, For the maternity experiences study Group of the Canadian Perinatal Surveillance System, Public Health Agency of Canada The Canadian maternity experiences survey: an overview of findings. Journal of Obstetrics and Gynaecology Canada. 2008;30:217–228. doi: 10.1016/S1701-2163(16)32758-X. [DOI] [PubMed] [Google Scholar]
  9. CIHI, Canadian Institute for Health Information. (2009). Mentally Healthy Communities: Aboriginal Perspectives. Ottawa, Ontario, Canada.
  10. CIHR. (2007) CIHR Guidelines for Health Research Involving Aboriginal Peoples. http://www.cihr-irsc.gc.ca/e/documents/ethics_aboriginal_guidelines_e.pdf. Accessed 2 July 2017.
  11. Clare CA, John Y. Postpartum depression in special populations: a review. Obstetrical and Gynecological Survey. 2012;67:313–323. doi: 10.1097/OGX.0b013e318259cb52. [DOI] [PubMed] [Google Scholar]
  12. Clarke PJ. Validation of two postpartum depression screening scales with a sample of First Nations and Metis women. The Canadian journal of nursing research = Revue canadienne de recherche en sciences infirmieres. 2008;40:113–125. [PubMed] [Google Scholar]
  13. Cohen L, Wang B, Nonacs R, et al. Treatment of mood disorders during pregnancy and postpartum. Psychiatric Clinics of North America. 2010;33:273–293. doi: 10.1016/j.psc.2010.02.001. [DOI] [PubMed] [Google Scholar]
  14. Cox J, Holden J, Sagovsky R. Detection of postnatal depression. Development of the 10-item Edinburgh Postnatal Depression Scale. The British Journal of Psychiatry. 1987;150:782–786. doi: 10.1192/bjp.150.6.782. [DOI] [PubMed] [Google Scholar]
  15. Da-Silva V, Moraes-Santos A, Carvalho M, et al. Prenatal and postnatal depression among low income Brazilian women. Brazilian Journal of Medical and Biological Research. 1998;31:799–804. doi: 10.1590/S0100-879X1998000600012. [DOI] [PubMed] [Google Scholar]
  16. Davey HL, Tough SC, Adair CE, Benzies KM. Risk factors for sub-clinical and major postpartum depression among a community cohort of Canadian women. Matern Child Health J. 2011;15:866–875. doi: 10.1007/s10995-008-0314-8. [DOI] [PubMed] [Google Scholar]
  17. Dennis CL, Vigod S. The relationship between postpartum depression, domestic violence, childhood violence, and substance use: epidemiologic study of a large community sample. Violence Against Women. 2013;19:503–517. doi: 10.1177/1077801213487057. [DOI] [PubMed] [Google Scholar]
  18. Dennis CL, Merry L, Gagnon AJ. Postpartum depression risk factors among recent refugee, asylum-seeking, non-refugee immigrant, and Canadian-born women: results from a prospective cohort study. Social Psychiatry and Psychiatric Epidemiology. 2017;52:411–422. doi: 10.1007/s00127-017-1353-5. [DOI] [PubMed] [Google Scholar]
  19. Dzakpasu S, Kaczorowski J, Chalmers B, Heaman M, Duggan J, Neusy E, for the maternity experiences study group of the Canadian prerinatal surveillance system, Public Health Agency of Canada The Canadian maternity experiences survey: design and methods. Journal of Obstetrics and Gynaecology Canada. 2008;30:207–216. doi: 10.1016/S1701-2163(16)32757-8. [DOI] [PubMed] [Google Scholar]
  20. Falah-Hassani K, Shiri R, Vigod S, Dennis C-L. Prevalence of postpartum depression among immigrant women: a systematic review and meta-analysis. Journal of Psychiatric Research. 2015;70:67–82. doi: 10.1016/j.jpsychires.2015.08.010. [DOI] [PubMed] [Google Scholar]
  21. Friesen K, Peterson WE, Squires J, Fortier C. Validation of the Edinburgh Postnatal Depression Scale for use with young childbearing women. J Nurs Meas. 2017;25:1–16. doi: 10.1891/1061-3749.25.1.E1. [DOI] [PubMed] [Google Scholar]
  22. Ganann R, Sword W, Thabane L, Newbold B, Black M. Predictors of postpartum depression among immigrant women in the year after childbirth. Journal of Women's Health (2002) 2016;25:155–165. doi: 10.1089/jwh.2015.5292. [DOI] [PubMed] [Google Scholar]
  23. Gress-Smith J, Luecken L, Lemery-Chalfant K, Howe R. Postpartum depression prevalence and impact on infant health, weight, and sleep in low-income and ethnic minority women and infants. Maternal and Child Health Journal. 2012;16:887–893. doi: 10.1007/s10995-011-0812-y. [DOI] [PubMed] [Google Scholar]
  24. Hobfall E, Ritter C, Lavin J. Depression prevalence and incidence among inner city pregnant and post partum women. Journal of Consulting and Clinical Psychology. 1995;63:445–453. doi: 10.1037/0022-006X.63.3.445. [DOI] [PubMed] [Google Scholar]
  25. Jesse D, Walcott-McQuig J, Mariella A, et al. Risks and protective factors associated with symptoms of depression in low-income African American and Caucasian women during pregnancy. Journal of Midwifery and Women's Health. 2005;50:405–410. doi: 10.1016/j.jmwh.2005.05.001. [DOI] [PubMed] [Google Scholar]
  26. King M, Smith A, Gracey M. Indigenous health part 2: the underlying causes of the health gap. Lancet. 2009;374:76–85. doi: 10.1016/S0140-6736(09)60827-8. [DOI] [PubMed] [Google Scholar]
  27. Kirmayer LJ, Brass GM, Tait CL. The mental health of Aboriginal peoples: transformations of identity and community. Canadian Journal of Psychiatry:Revue canadienne de psychiatrie. 2000;45:607–616. doi: 10.1177/070674370004500702. [DOI] [PubMed] [Google Scholar]
  28. Lanes A, Kuk JL, Tamim H. Prevalence and characteristics of postpartum depression symptomatology among Canadian women: a cross-sectional study. BMC Public Health. 2011;11:302. doi: 10.1186/1471-2458-11-302. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Murray L, Carothers AD. The validation of the Edinburgh post-natal depression scale on a community sample. The British Journal of Psychiatry. 1990;157:288–290. doi: 10.1192/bjp.157.2.288. [DOI] [PubMed] [Google Scholar]
  30. Nelson C, Lawford K, Otterman V, Darling E. Mental health indicators among pregnant Aboriginal women in Canada: results from the maternity experiences survey. Health Promotion and Chronic Disease Prevention in Canada. 2018;38:269–276. doi: 10.24095/hpcdp.38.7/8.01. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Noël, A., & Larocque, F. (2009). Aboriginal peoples and poverty in canada: can provincial governments make a difference? Paper prepared for the Annual Meeting of the International Sociological Association’s Research Committee 19 (RC19), Montréal, QC, August 20, 2009.
  32. O’Hara M, Swain A. Rates and risk of postpartum depression: a meta-analysis. International Review of Psychiatry. 1996;8:37–54. doi: 10.3109/09540269609037816. [DOI] [Google Scholar]
  33. O'Hara MW, McCabe JE. Postpartum depression: current status and future directions. Annual Review of Clinical Psychology. 2013;9:379–407. doi: 10.1146/annurev-clinpsy-050212-185612. [DOI] [PubMed] [Google Scholar]
  34. Royal Commission on Aboriginal Peoples . Appendix B: Ethical guidelines for research. Ottawa, ON: RCAP; 1993. [Google Scholar]
  35. Sit D, Wisner K. Identification of postpartum depression. Clinics in Obstetrics and Gynaecology. 2009;52:456–468. doi: 10.1097/GRF.0b013e3181b5a57c. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Smylie J. The health of the aboriginal people. In: Raphael D, editor. Scoial Determinants in Health-Canadian Perspectives. Toronto: Canadian Scholars Press; 2008. [Google Scholar]
  37. Smylie J, Andersons M. Understanding the health of the indigenous people in Canada: Key methodological and conceptual challanges. Canadian Journal of Public Health. 2006;175:602–605. doi: 10.1503/cmaj.060940. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Statistics Canada. (1993) Public Health Agency of Canada. Violence Against Women Survey Ottawa, ON: Statistics Canada.
  39. Statistics Canada. (2006a) Public Health Agency of Canada. Maternity Experiences Survey, 2006 Questionnaire. Ottawa, ON Statistics Canada.
  40. Statistics Canada . Census dictionary catalogue no.92-566-X 2010. Minister of Industry: Ottawa; 2006. [Google Scholar]
  41. Statistics Canada. (2006c) Public Health Agency of Canada. Maternity Experiences Survey, share file. Ottawa, ON Statistics Canada
  42. Statistics Canada. (2011) Aboriginal affairs and northern development canada: aboriginal people and communitties. http://www.aadnc-aandc.gc.ca/eng/1100100013785/1304467449155. Accessed 2 July 2017.
  43. Wei G, Greaver L, Marson S, Herndon C, Rogers J, Robeson Healthcare Corporation Postpartum depression: racial differences and ethnic disparities in a tri-racial and bi-ethnic population. Maternal and Child Health Journal. 2008;12:699–707. doi: 10.1007/s10995-007-0287-z. [DOI] [PubMed] [Google Scholar]

Articles from Canadian Journal of Public Health = Revue Canadienne de Santé Publique are provided here courtesy of Springer

RESOURCES