Immigrant and non-immigrant women’s experiences of maternity care: a systematic and comparative review of studies in five countries

1Increasing global migration has implications both for health care provision in receiving countries and for the health care experiences of immigrant populations. This is nowhere more apparent than in the experience of women giving birth post-migration. A systematic review of immigrant women’s perinatal outcomes published in 2010 identified very few studies over a ten-year period which described any aspect of immigrant women’s maternity care experiences in comparison with non-immigrant women. Some population-based studies of women’s experiences of maternity care conducted in a few countries do include limited data on immigrant and refugee women’s experiences of care for comparison with non-immigrant women, but immigrant women are commonly under-represented in these studies because of the formidable challenges of undertaking inclusive cross-cultural research that is population-based and large scale. These challenges include: sampling and recruitment issues, difficulties in translation and in assessment of validity with the use of standard research instruments, and increased research costs. Other studies have specifically investigated the experiences of individual groups of immigrant and refugee women, and to date these are mostly small and qualitative. Given the dearth of adequately-sized and appropriately conducted studies directly comparing representative immigrant and non-immigrant experiences of maternity care, a systematic review drawing on data in general population studies and in specific immigrant studies in the same countries, would seem to offer the best opportunity for drawing together and comparing what is known about immigrant and non-immigrant experiences, and what women want – and get – from their maternity care.

2Our purpose in selecting studies for this review was thus twofold. First, we aimed to identify and review all published population-based studies of women’s experiences of maternity care to determine what they say about what women want from care, including any data, if available, about immigrant women. Second, having identified the countries where such studies have been conducted, we aimed to investigate further what is known about the experiences of immigrant women in each of these countries, by identifying and reviewing studies focused specifically on immigrant women’s experiences of their maternity care. For the purposes of this review, we define immigrant women as those women not themselves born in the country in which they are giving birth.

3There were two review questions:

41. What do immigrant and non-immigrant women want from their maternity care?

52. How do immigrant and non-immigrant women’s experiences and ratings of care compare, both within and across included countries?

6Ovid was used to search the electronic databases Medline, CINAHL, Health Star, Embase and PsychInfo for the period 1989–2011. The search strategy was developed by MR with the assistance of the Health Sciences Librarian at La Trobe University in February 2010 and further searches were conducted to update the literature to December 2012. 1989 was chosen as the start year because the first population-based study of women’s experiences of maternity care was known to have been conducted in that year. Terms combined in the search included: emigration/emigrant, immigration/immigrant, migrant, ethnic group, ethnic minority, population groups, refugees, non-English speaking, women, view, opinion, attitude, experience, maternal health services, maternity care, perinatal care, prenatal/antenatal care, intrapartum care, postnatal care, delivery, obstetrics, midwifery. For an example of the search strategies used, see Additional file 1.

7Population-based studies of women’s experiences of care, defined as those with national or regional samples with representativeness assessed, were identified, retrieved and reviewed. Studies with a hospital-based or convenience sample or where representativeness could not be assessed were excluded. With these criteria, 12 studies from five countries were included. One national study was identified from Scotland, but subsequently excluded, as its overall population representativeness could not be assessed.

8Studies focusing specifically on immigrant women’s experiences of maternity care from these same five countries were then also identified, retrieved and reviewed. Studies of ethnic minorities who were not themselves immigrants or refugees were excluded, as were retrieved studies which on review, were found to focus only on cultural beliefs and practices around childbirth without investigating immigrant women’s actual experiences of the maternity care they received. For the immigrant studies, all retrieved studies were included (i.e. no quality criteria were applied), for two reasons. First, our purpose was to include as much data as possible about a diverse range of immigrant women’s experiences for comparison with data on non-immigrant women from the population-based studies. Second, the immigrant studies were relatively few across the included countries; and most were small and qualitative. Twenty-two studies of immigrant women’s experiences of care were identified, retrieved and reviewed across the five included countries.

9Papers were read and the findings summarised, noting (where available) overall ratings of care and key conclusions about what women wanted from care (RS, MR and TS). The country, year of study, sample size and study type (e.g., population-based postal survey, qualitative interview study) were also noted. For the population-based studies, the main findings were recorded separately for non-immigrant and immigrant women, except when the data did not distinguish these groups of interest (the three US studies and two of the UK studies). Study findings were tabulated for ease of discussion and interpretation (MR and RS) and a descriptive thematic analysis of the extracted data was undertaken. Two authors independently developed codes for describing the data (MR and RS) and a third author (TS) reviewed these. The resulting interpretation of the data was then reviewed and revised by all authors.

10Figure 1 provides a flow diagram of the review process and the selection of studies.

11Three population-based studies from the state of Victoria (1989, 1994, 2000) and seven studies of immigrant women (including Vietnamese, Chinese, Cambodian, Laotian, Thai, Korean, Filipino, Turkish, Muslim women from a range of countries) were reviewed.

12One national survey (2006) and four studies of immigrant women (including Somali, South Asian, Punjabi, Muslim women from various countries) were reviewed.

13One national study (1999–2000) and two studies of immigrant women (including immigrant Somali, Eritrean and Sudanese women) were reviewed.

14Four national surveys (1995, 2006 and 2007) and six studies of immigrant women (including immigrant South Asian, Somali, Indian, Pakistani and Bangladeshi women) were reviewed.

15Three national surveys (2002, 2006 and 2013) and four studies of immigrant women (including Somali, Hmong, Puerto Rican and ‘Hispanic’ immigrant women) were reviewed. Although Puerto Rico is an unincorporated US Territory, not a separate country, Puerto Rican women coming to the US have been considered ‘immigrants’ for the purposes of this review.

16These 12 population-based studies from five countries were conducted in the period 1989–2013 and involved 55,495 women (range 790–26,325). In four of the studies (involving 31,887 women), it was not possible to determine women’s country of birth in order to calculate the number of women who were immigrants. For the remaining eight studies (involving 23,608 women) there were 2,682 women (8.3%) who were immigrants and 15,593 women who were non-immigrants. For the 22 specific studies of immigrant women, sample sizes ranged from 6 to 432, with a total of 2,498 immigrant women involved, with studies published between 1990 and 2012.

17The key findings from the population-based studies about what non-immigrant women appreciate and want from their maternity care proved remarkably similar across the included countries, as can be seen in the study summaries provided in Table 1. Most of these population-based studies assessed women’s overall ratings of care for each of the three phases of care: during pregnancy, during labour and birth and during the postpartum hospital stay. The exceptions to this were: the Canadian survey, in which women were asked to rate their satisfaction with six aspects of their interaction with health care providers during the entire pregnancy, labour and birth, and immediate postpartum period, and the US surveys, where women were not asked to give overall ratings of their care except in response to a question in the 2005 and 2013 surveys asking women their view about the maternity care system overall, with 35% and 36% rating it as excellent, 47% and 47% as good, and 16% and 17% as poor, respectively.

18Women commonly reported problems in pregnancy care with long waiting times, staff not taking time to attend to individual concerns and provide enough information, staff seeming rushed, and lack of continuity of care. Seeing fewer caregivers during antenatal visits was associated with more positive experiences of care, or was seen as important by women in most studies. The need for adequate and consistent information, being treated as an individual, and having effective interaction with caregivers were also commonly reported to be important in shaping positive experiences about pregnancy care.

19Dissatisfaction with intrapartum care in the population based studies was consistently associated with lack of sufficient information during labour, the perception that caregivers were not kind and understanding, caregivers being unhelpful, and not having an active say in making decisions,.

20The nature of women’s interactions with caregivers appears to be a critical factor for women’s experiences at all stages of care. The earliest Australian survey conducted in 1989 revealed a four to sixfold increase in dissatisfaction if women had not received sufficient information from caregivers. Likewise, women who described their caregivers as not being very kind and understanding were four to five times more likely to be dissatisfied with their care; and caregivers regarded as being unhelpful was associated with significant dissatisfaction with intrapartum care. The 2008 national survey in England reported that women were more satisfied with intrapartum care when they received individualised care, enough information and explanations, and were cared for by kind and understanding staff. Involvement in decisions about care and having an ‘active say’ also seem to be consistently important factors associated with more positive experiences of care in labour and birth.

21Women were less positive about their postpartum care compared with the care they received in pregnancy, or during labour and birth in all three Australian surveys, in the four UK surveys and also in the Swedish study.

22The factors that seem to be important in women’s experiences of their postpartum care are focused on the attitudes and behaviour of staff: caregivers being sensitive and understanding, providing support and advice, and the helpfulness of that advice and support. Factors associated with women’s negative experiences of postnatal care included: when their concerns and anxieties were not taken seriously, staff being rushed and too busy to spend time with them, staff not being sensitive and understanding, and not providing enough advice and support about baby care. Another important factor was receiving enough support and advice about women’s own health and recovery. In the national Swedish study, content analysis of responses to open-ended questions regarding women’s negative experiences of postpartum hospital care two months and one year after the birth showed that the aspects of care women were most dissatisfied with were: shortages of staff and staff being rushed, staff behaviour, lack of attention to women’s concerns, inadequate support and advice, and lack of sufficient information and explanation regarding baby care and women’s own physical and emotional health after birth.

23Drawing on the common themes emerging across the population-based studies from these five countries, we propose the ‘QUICK’ summary, where ‘QUICK’ is a mnemonic that captures the essence of what women want from their maternity care:

24Q = Quality care that promotes wellbeing for mothers and babies with a focus on individual needs.

25U = Unrushed caregivers with enough time to give information, explanations and support.

26I = Involvement in decision-making about care and procedures.

27C = Continuity of care with caregivers who get to know and understand women’s individual needs and who communicate effectively.

28K = Kindness and respect.

29When one or more of these aspects of care was lacking, women were likely to be less happy with their care.

30Where data were available for immigrant women in the population-based studies, the key findings have also been included in Table 1. The immigrant women born in countries where English was not the principal language spoken who responded to the three Australian surveys – although unlikely to be representative of all immigrant women, given English language requirements for participation – were less happy with their care than non-immigrant women and more likely to have difficulties with getting the information and support they required. In the Canadian and Swedish studies, similar levels of satisfaction with care were found for immigrant and non-immigrant women, although language issues are acknowledged to have excluded many immigrant women from participation in the Swedish study, and almost one in five immigrant participants in the Canadian study reported not receiving care in a language they could understand. Only two of the UK studies provided data on immigrant women, with comparisons made for black and minority women without reference to country of birth in the others. Immigrant women of black and minority ethnicity were less likely to feel spoken to with respect and understanding, and in a way they could understand; to feel they had options in care or adequate information; and were less likely to describe care providers positively. Findings for immigrant mothers were not reported in the US surveys – the third survey did give the numbers of immigrant women participating, but did not report their experiences separately.

31The findings about what immigrant women value in their maternity care from studies conducted to investigate specific groups of immigrant women’s experiences are summarised in Table 2, and are organised by each receiving country.

32Table 2 shows that the findings from these studies are not only quite consistent across immigrant groups originating from very different cultures and countries, but also that the ‘QUICK’ summary elements found in the population studies, appear also to be central in the accounts of immigrant women from these immigrant-specific studies, again regardless of women’s country or culture of origin, or of the country to which they had migrated.

33However, additional challenges associated with negative impacts on women’s experiences of care emerge from the studies of immigrant women. First, language difficulties clearly hamper good communication and understanding between immigrant women and their caregivers when women are not fluent in the language of the receiving country. Communication difficulties were identified as a key problem in almost all the immigrant studies. Lack of information in community languages and insufficient access to interpreters when needed were also commonly reported and a few studies noted that even when interpreters were available, women did not always feel that they were competent. Lack of familiarity with how care is provided or not receiving adequate information about what options for care exist, were also common problems for immigrant women. Several studies also reported immigrant women feeling they were not welcomed, or were made to feel anxious, when they came to hospital in labour.

34Despite evidence that immigrant women want to be involved in decisions about their care, some studies found that immigrant women were at times reluctant to make their wishes known. Experiences of discrimination, and/or cultural stereotyping were also commonly reported in the immigrant studies from all five countries. Studies of Somali immigrants in Canada, Sweden and the UK also found that women felt staff were insensitive to their experiences of pain in labour and responded inappropriately to traditional female genital cutting, demonstrating a lack of knowledge about this issue.

35Some studies noted particular cultural issues that immigrant women felt were not well understood during their maternity care and about which they desired more understanding from their caregivers. One US study of Hmong women described women's fears of being touched by doctors and nurses because of beliefs about the causes of miscarriage. Some studies reported women's preference for female caregivers, with Muslim women in particular expressing this preference. It is worth noting however that this question is rarely asked in studies of non-immigrant, or non-minority women, so whether immigrant women are more likely to prefer female caregivers than non-immigrant women is not readily known. Several Australian studies found that women sometimes found it difficult to follow traditional cultural practices in hospital (for example food preferences, not showering after birth), and women reported that they were rarely asked by caregivers about their postnatal practice preferences.

36Interestingly though, lack of attention to cultural issues or restrictions on traditional cultural practices by caregivers were not the principal focus of women's descriptions of negative aspects of the maternity care they received post migration. Communication problems and discriminatory or negative caregiver attitudes appear to be the more critical areas of concern reported by women in the studies reviewed here, just as immigrant women's positive experiences of care centred around appreciation of being treated with kindness and respect and having their individual concerns addressed competently and sympathetically.

37Two published systematic reviews of studies of immigrant women’s experiences of childbirth and maternity care broadly support the findings about immigrant women’s experiences from our five included countries. The first is a recent systematic review which included 16 qualitative studies from six European countries (Greece, Ireland, Norway, Sweden, Switzerland and the UK). It aimed to investigate immigrant women’s needs and experiences of pregnancy and childbirth and found as we did, that good communication and information, an understanding of how care operates in their new homeland, caregivers who are respectful, non-discriminatory and kind, and achieving a safe pregnancy and birth are key aspects of what immigrant women wanted from their maternity care. The second review included 40 qualitative studies from Australia, Canada, Denmark, Ireland, Israel, Japan, Norway, South Africa, Sweden, and the USA. Aiming to explore aspects of intercultural caring from immigrant women’s perspectives of their maternity care, the review concludes that addressing communication problems, providing continuity of care, addressing racism and discrimination and providing flexibility in care to accommodate individual and cultural diversity are likely to enhance immigrant women’s experiences of maternity care. What the current review additionally offers is a comparison with non-immigrant women, previously missing in the literature.

38This review has drawn together the available population-based studies of women’s experiences of maternity care in order to assess what is known about immigrant compared with non-immigrant women’s experiences. As immigrant women have often been under-represented in population-based research, we supplemented our review of these studies with the findings from studies focused on specific groups of immigrant women in each of the countries where population-based studies were identified. This is both a strength, and a limitation. It could be said that we are not comparing like with like, and that is true. Most of the specific immigrant studies are small and qualitative in design and the representativeness of the immigrant participants cannot be ascertained. On the other hand, synthesising the evidence from a range of study types for immigrant women, in an area where assembling representative samples is particularly difficult, has proved informative, particularly given the consistency that has emerged in the findings from both the population-based and the qualitative studies. Examining studies drawn from the same receiving countries is also a strength of this review. Had factors associated with different maternity care systems been important in shaping women’s experiences of care, then this should have become apparent in comparisons of women’s experiences in the different countries. It is significant that at least in relation to care in Australia, Canada, Sweden, the UK and the United States of America, women identify the same problems with care and articulate very similar wishes in relation to what they want from care when giving birth. We are not aware of other reviews that have as yet attempted to directly compare immigrant and non-immigrant women’s experiences of care within and across countries, as we have done here.

39Finally, this review is limited by the studies that have been conducted to date. Globally, relatively few countries have undertaken population-based studies of women’s experiences of their maternity care. Of these, only the Canadian study has used a multi-language strategy in an attempt to address the under-representativeness of immigrant women in population studies, and the Australian research involved a companion study of three immigrant groups in tandem with one of the three population surveys undertaken there. It is also worth noting that the recent waves of migration between countries in the European Union and of refugee and asylum-seeking arrivals are not yet well represented in studies of women’s experiences of maternity care.

40This review has found that immigrant and non-immigrant women appear to have very similar ideas about what they want from their maternity care, notwithstanding the diversity of countries and cultures of origin of the women represented in the reviewed studies. In regard to women’s overall ratings of their maternity care however, immigrant women commonly gave poorer ratings of the care they received compared with non-immigrant women, and a range of additional challenges they faced tended to have negative impacts on their experiences of care. These chiefly included: communication difficulties due to language problems, lack of familiarity with how care was provided and experiences of discrimination.

41Authors of the studies of immigrant women often recommended the need for more culturally sensitive care, with cultural competency training for maternity services staff seen as a means to this end. While in some studies immigrant women did comment on staff not understanding their cultural beliefs and practices, a careful examination of what women most commonly wanted – as shown in Table 2 – demonstrates that women themselves were focused more on the need for respectful care that was attentive to their individual needs, on assistance with communication difficulties and on receiving better information about how care is provided in their new country. Women in more than one study commented that staff cannot possibly ‘know’ every culture. Moreover, cultural beliefs and practices are not static phenomena, with considerable diversity among women from within any one culture with regard to adherence to particular traditions or beliefs, so that encouraging staff to ask all women about their childbirth preferences and beliefs is likely both to be more achievable, and also to result in more responsive care for all women, immigrant and non-immigrant alike.

42Notably in this review, women from a range of immigrant backgrounds in studies from all five receiving countries, reported problems with discrimination or prejudice in their experiences of care. If services are to take seriously what immigrant women say they want, then perhaps what is most needed to improve care is an enhanced focus on promoting equity and non-discriminatory attitudes in care provision, along with strategies aimed at improving communication (including training in working effectively with interpreters), and better recognition of the need to familiarise immigrant women with how maternity care is provided, so that they can more actively participate in decisions about their care and feel less anxious and disempowered about giving birth in their new country.

43What this review has revealed is that improvements in immigrant women’s often poorer ratings of care will only come if more attention is paid to addressing the additional challenges they face due to language difficulties, lack of familiarity with care systems and at times, exposure to discriminatory attitudes and poorer quality care. Proper recognition of these extra challenges is required in the provision of care. In addition, maternity staff need to be supported – with time, resources and training – to enable them to provide appropriate and non-discriminatory care to immigrant women, in accord with published declarations and standards of quality care for immigrant populations. More inclusive approaches to enable the involvement of immigrant women in future population-based studies of women’s experiences of maternity care would also ensure that care improvements for immigrant women can be appropriately evaluated over time.

44The authors declare that they have no competing interests.

45RS, CR, DK, AG and MH conceived the project; MR conducted the searches; MR, TS and RS reviewed the studies for inclusion; MR developed the first draft of the tables of studies and these were checked and modified by RS and TS, and subsequently by all authors. RS, MR, TS and CR were involved in drafting the manuscript and all authors (RS, CR, MR, TS, DK, CMcC, MH and AG) contributed to revising it critically for intellectual content and all approved the final manuscript.

46The pre-publication history for this paper can be accessed here:

47http://www.biomedcentral.com/1471-2393/14/152/prepub