Search strategy

Four databases were searched; CINAHL (EBSCO), Maternity and Infant Care (Ovid), MEDLINE (EBSCO), and PsycINFO (ProQuest). The searches in MEDLINE and CINAHL were undertaken in August 2017, PsycINFO in Sept 2017, and Maternity and Infant Care in October 2017. The search strategy comprised four facets with terms relating to: (1) the perinatal period, (2) ethnicity or possible countries of origin (3) mental health, and (4) research undertaken in Europe. All terms were searched for in the title and abstract fields and, in addition, the fourth facet was searched for in the country of publication field or the nearest equivalent. Controlled vocabulary terms were used where available. The Boolean operators AND and OR were used, alongside truncation, phrase searching and proximity operators. The search syntax was adapted for each database. The full search strategy, as applied in MEDLINE (EBSCO interface) is provided in Supporting Information S2 File.

Inclusion criteria

Studies were eligible for inclusion if they were primary qualitative, quantitative or mixed methods studies. Studies must have explored perinatal mental health disorders, help-seeking or use of mental health services in women from ethnic minority populations living in Europe. Any interventions targeting this group were eligible for inclusion provided the paper reported self-reported experiences, preferences or health-seeking behaviours associated with perinatal mental health from ethnic minority women's perspectives. Studies were not excluded on the basis of language or publication date.

Study selection

All papers were screened by title and abstract to determine their eligibility for inclusion in the review. The full-texts of all remaining papers were screened to determine their inclusion/exclusion from the review. Ten percent of papers were independently double checked by a second reviewer and discrepancies in judgement regarding eligibility discussed and resolved within the research team. Reference lists of included papers were checked for further relevant papers.

Quality assessment

The quality of each paper was assessed, however, no papers were excluded on the basis of low quality scores as there is no empirical evidence to justify this approach [26]. A total of 10% of the articles were appraised by a second reviewer with referral to a third reviewer if disagreement was not resolved. All included qualitative articles were quality appraised using the qualitative National Institute for Health and Care Excellence (NICE) appraisal tool [27]. An overall rating of the quality of the article was classified as high (++) if the majority of the appraisal criteria were met and the study was judged to be trustworthy and reliable and there was significant evidence of author reflexivity. An acceptable quality score (+) required that most criteria were met and identified that there may be some flaws in the study resulting in a lack of rigor. A low-quality score (-) was assigned if either most criteria were not met, or it was judged that there were significant flaws in the study design.

Quality of the included quantitative papers was similarly appraised using the critical appraisal checklist for a questionnaire-based survey study adopted by the National Institute for Health and Clinical Excellence (NICE) [28] for evidence synthesis in guideline development. A high quality score (++) was assigned if the majority of the criteria were met, and it was judged that there was little or no risk of bias. An acceptable quality score (+) required that most criteria were met, and that there may be some flaws in the study with an associated risk of bias. A low quality score (-) was assigned if either most criteria were not met, or it was judged that there were significant flaws relating to key aspects of study design.

Data extraction

Data extraction was managed in Microsoft Excel using a data extraction proforma developed specifically for this review. Extracted data included author/s, year of publication, study aim, setting, sample size, participant demographics, data collection method, method of analysis and outcomes. Relevant outcome data were extracted from sections of the qualitative studies entitled findings, results, discussion, interpretations or conclusions and were imported directly into NVivo-11 software (QSR International).

Synthesis

A convergent approach to data synthesis was adopted [29]. This approach was undergirded by the belief that both qualitative and quantitative studies can address the same research questions and produce data that can be readily transformed into each other and that differences between the studies do not warrant separate analyses [30]. In this case quantitative findings were transformed into qualitative form as themes, so that they could be combined with the qualitative findings [29,30] using a thematic synthesis approach [31]. Thematic synthesis included line by line coding, organising codes into descriptive themes and generating analytic themes. The findings from 20% of the papers were independently synthesised by two reviewers in order to ensure a consistent approach.

Confidence in the findings

The confidence in the findings was assessed using the Confidence in the Evidence from Reviews of Qualitative Research (CERQual) approach [32]. This systematic approach enables the reviewers to produce a score ranging from low to high confidence in the review findings by considering; the methodological limitations of the individual studies, relevance to the review question, coherence and adequacy of the data [32]. The CERQual scores were assessed by two reviewers and then discussed amongst the review team to ensure agreement.

Characteristics of the included studies

The characteristics of the 15 included studies [33–47] can be seen in Table 1. Twelve of the papers contained qualitative data obtained using interviews [35,36,38,40,41,44,46,47], focus groups [35,37,39,43,46] and observation [33]. Three of the studies collected data using surveys [34,42,45]; all included some narrative summary of their findings, and two presented descriptive statistics [42,45] which were converted to qualitative themes for purpose of this review (See Supporting Information S1 Table). Although the search strategy included all European countries, all of the studies which met the inclusion criteria were undertaken in the UK. The date of publication of the studies ranged from 1984 to 2016, with all but one study [35], published since 2003. The ethnicity or nationality of study participants were categorised as; Bangladeshi [33,39,42,43,46,47], Pakistani [42,47], Indian [42,46,47], South Asian [41], Asian [45], Other Asian [46], Mixed Asian/British [44], Pathan [35], Black [45], Black African [44], Nigerian [38], Ghanaian [38], Black Caribbean [36,37], Portuguese [46], White American [44], White Australian [44], White other [44], other ethnicity [45,46], unspecified black or minority ethnicity [34], and unspecified ethnicity in the case of a group of asylum seekers presumed to be non-native [40]. Although some studies included data from English [33] or White British participants [44,45], this was excluded during data extraction where possible. The studies were of mixed quality, two were judged to be of high quality [45,47], nine of acceptable quality [36–39,41–44,46] and four of low quality [33–35,40]. A summary of the quality scores for each paper are shown in Table 1 with full details in Supporting Information files S2 Table and S3 Table.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Characteristics of included studies.

https://doi.org/10.1371/journal.pone.0210587.t001

Awareness and beliefs about mental ill health

Several studies demonstrated that some women from minority ethnic groups were not aware of perinatal mental health disorders [34–36,38,39,42,43,46]. Despite being familiar with or experiencing perinatal mental health symptoms, they didn't consider them to be an illness [36,38,39], didn't have the language to describe the collective symptoms as a disorder [34,39,47], or adopted alternative explanations for the symptoms which included practical problems [43], lack of rest [43], isolation [38] lack of support [38,43], the influence of evil spirits [39] or something that would just go away [46].

"…don’t know what call it, in Pakistan where used to live, they don’t have depression there. There is no word in Urdu for depression, I don’t know what is happening to me, never seen it before, only when moved here." South Asia mother ([47], p.486])

"I’ve had two babies now, I don’t know what postnatal depression is supposed to be, how you’re supposed to feel, look or whatever, I don’t know. I have no idea. What exactly is postnatal depression? What are you supposed to be doing, saying, or whatever?" Black Caribbean mother ([36], p.18-19)

Other women who were able to recognise their symptoms as a mental health disorder identified contributing factors that included financial problems [36], difficulties in personal relationships [36], hormonal factors [36], pregnancy-related physical health problems [36], racist incidents [35], living in a different country [38,47], stress [38], lack of community [38] and issues related to being an asylum-seeker such as the loss of homes, communities, jobs and family members [40].

"I think it is about the stress……. and the [lack of] community." Black West African mother ([38], p.758)

Influence of culture

Black Caribbean and South Asian women explained that depression was culturally unacceptable because of its impact on women fulfilling their role in society [36,43]. Women explained that Black Caribbean culture has been influenced by the impact of their history of slavery, and women are expected to be strong and hold the family unit together and are not allowed to have depression as this would be a sign of weakness or not coping [36]. The women reported that they had inherited a cultural legacy and ability to be strong and to cope [36]. Similarly amongst Bangladeshi women, depression was seen as implying weakness [43].

"I do think that Black people get depression, but I don’t think we’re allowed to have depression. I think it’s quite a matriarchal society and therefore you’ve got to cope. You’ve got to sort your family out, and so therefore you are not allowed to be depressed." Black Caribbean woman ([36], p.19)

Many of the women described that it was culturally unacceptable to talk about problems, feelings or emotional issues to people outside the family or home [36,38,39,43,46], and that if these things were revealed it would result in stigma [38].

"I didn't just……. open up totally…… to them. I wouldn't want to… I know it's just the stigma…. It's just you know oh… look at the girl….. I think it's just, it's just that I don't want the stigma to just keep following me around." Black West African mother ([38], p.760)

Pathan women explained that as they believed that their health and illness, happiness and unhappiness was all in God's hands, these things should not concern them [35].

Symptoms and coping strategies

Women who had experience of perinatal mental health disorders described symptoms that included anxiety [40,47], low mood and sadness [40,47], prolonged crying [40], negative thoughts [47], pain in the chest or stomach [46], lethargy [46], behaviour changes such as difficulty leaving the house [47], and appetite and sleep problems [47].

The women responded to their symptoms by minimising their feelings and self-silencing [38,44]. They reported that they adopted coping mechanisms which included; drawing on their inner strength [36,38], solving it themselves [36,43], putting up with it [43], seeking spiritual treatment from a religious leader [39], and other sources of spiritual support such as personal faith [38], private prayer and accepting prayer from others [37,38,43,47].

"It was my belief and faith in God, cause I kept praying, my church prayed for me at all times and all that…and I believed that I would be well again, it was in my head…so it was… my faith in God." Black West African mother ([38], p.761)

Women also coped by having time for themselves [47], keeping busy [38] and distracting themselves with activities such as work or shopping or cleaning [38].

"For me when I am down, I just want to do something, go out for window shopping or sometimes tidy the house. But the best thing for me is to go out." Black West African mother ([38], p.760)

Isolation and seeking support

Many women described that they felt emotionally isolated [38,44,47] and that this was exacerbated when they felt misunderstood by their partner, family members or friends [44,47], were physically separated from family members [38,47] or lacked friendships [40,44].

"My husband just don’t understand how I feel, everybody just keep saying Dimaak kharaab hai [mind is not working properly]." South Asian mother ([47], p.486])

"The thing is I have nobody, I feel totally alone, I have no support and this makes me feel worse." South Asian mother ([47], p.487)

Women described that support from friends [33,38], family [34,38,47], neighbours [46], the spiritual community [36] and work colleagues [46] helped them to cope. Women particularly valued and sought out support from others with the same experiences of poor perinatal mental health in person [38,44,46] or through an online, virtual connection [44].

Accessing mental health services

Some women were not aware of the support that was available for women with perinatal mental health problems [34,36,38,46,47] and access to support services was influenced by a number of different factors.

Experiences of perinatal mental health services

What women want

Women identified a number of ways that perinatal mental health service provision for women from ethnic minority backgrounds could be improved. They highlighted that services need to address the full range of mental ill health during the perinatal period [37] and that better multi-agency working was needed [34,37], particularly between the health service and voluntary organisations [34]. They suggested that health professionals would benefit from cultural competency training [41], and that culturally specific support should be provided [47]. Women wanted the opportunity to have early assessment and intervention, and to be provided with information about perinatal mental health problems and services that were available [47]. Women wanted interventions that were regular [44], designed around women's individual preferences [37] and which could be flexible in terms of length, as they felt that some problems required support for longer periods of time than was currently offered [41]. Women also identified that they wanted the opportunity to meet other mothers in similar situations either in group settings [44,47] or in community-based peer support schemes [37,44].

Confidence in the review findings

A summary of the CERQual scores can be seen in Table 2 with full details in Supporting Information S4 Table. A CERQual score of high confidence in the review findings was applied to five review findings, a further five findings were judged to score moderately, and a score of low confidence was attributed to two findings where the predominant contributing factor that reduced the confidence scores was concern about adequacy as these were based on a small number of studies.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. CERQual scores.

https://doi.org/10.1371/journal.pone.0210587.t002

Summary of the review findings

This review has demonstrated the complex issues surrounding ethnic minority women's experiences of perinatal mental ill health, and experiences of seeking and accessing support. Women from ethnic minority backgrounds represent a diverse population with varied beliefs and levels of awareness of perinatal mental health problems and different cultural expectations about their management. For many women the cultural discourse around mental health and the cultural expectations of the role and identity of women in society leaves women feeling isolated, fearful, stigmatised and suffering in silence. This is compounded by culturally inappropriate and fragmented health care services and culturally incompetent health care providers who demonstrate conscious or unconscious bias and poor attitudes towards women seeking help. The lack of published evidence about these issues from other European countries beyond the UK is concerning, particularly in the context of the serious implications of poor perinatal mental health for women, children and their families [8,12].

The unique experience of poor perinatal mental health for ethnic minority women

The unique aspects of experiencing perinatal mental ill health for women from minority ethnic groups that are identified in this review are further highlighted below by drawing on wider evidence, including that concerning the experiences of majority, white women who are experiencing poor mental health in the context of becoming a mother in the UK.

Review strengths and limitations

To our knowledge this is the first systematic review of literature focussing on the perinatal mental health experiences of women from ethnic minorities in Europe. This review is strengthened by the adoption of a systematic approach in adherence to the a priori protocol, which included a comprehensive search strategy that did not limit study eligibility by language, date or study type. The search identified a relatively high number of relevant studies; however, as all the eligible studies were set in the UK, which limits our findings and may not represent experiences of ethnic minority women from other parts of Europe. The identified gap in the high quality literature from other European countries in such an important area of research and practice, is however one of the main findings of this review. Independent, dual assessment of data extraction and quality assessment of included studies enhanced trustworthiness of the findings. The involvement of a multidisciplinary study team strengthened the development of the search strategy, the final analytic themes, and the assessment of confidence in the findings.