Does cultural context make a difference to women's experiences of maternity care? A qualitative study comparing the perspectives of breast-feeding women of Bangladeshi origin and health practitioners.

BACKGROUND: Maternity services struggle to provide culturally appropriate care that meets the needs of women from diverse populations. Problems include simplistic understandings of ethnicity and the role of culture in women's lives, and stereotypes held by health practitioners.
OBJECTIVE: To explore the extent to which cultural context makes a difference to experiences of breast-feeding support for women of Bangladeshi origin and to consider the implications for the provision of culturally appropriate care.
METHODS: The study comprised individual interviews with 23 women of Bangladeshi origin and four health service managers, and focus group discussions with 28 health practitioners between February and December 2008. Participants were recruited from four localities in northern England.
RESULTS: Women's rich descriptions of various facets of their identities were in contrast to practitioners' representations of women of Bangladeshi origin as homogenous. Practitioners did not recognize when the needs of women of Bangladeshi origin were similar to those of the majority white population, or where cultural context made a difference to their experiences of breast-feeding and breast-feeding support. Some practitioners used cultural stereotypes which, combined with organizational constraints, resulted in services not meeting many of the women's needs.
CONCLUSIONS: Implications for education, policy and practice include the need for training of health practitioners to work with diverse populations, implementing evidence-based practice and providing an organizational context which supports practitioners to respond to diversity without using cultural stereotypes.

Introduction

Simplistic understandings of ethnic identity and the role of culture prevalent in much health research and practice remain a barrier to understanding the health needs of diverse groups. Ethnic groups are frequently represented as homogenous and static with regard to their socio‐economic status, beliefs and customs. The role that racism can play in an individual’s health and ability to access services is often ignored, and the contribution of socio‐economic status, independent of cultural background, is not well understood. Unsophisticated understandings of ethnicity may lead to stereotyping by practitioners; consequently, those from minority ethnic groups are likely to be labelled as ‘high risk’ even in the absence of specific risk factors. Stereotyping and assumptions by practitioners are barriers to women making informed decisions about their care and having their individual needs met. ,

These multi‐faceted issues contribute to the poor quality maternity care experienced by many women from minority ethnic groups in the UK , , and in other developed countries , , and help to explain why health services often struggle to meet the needs of those from diverse backgrounds. , Improving the healthcare experiences of populations from disadvantaged minority ethnic groups requires policymakers and health practitioners to understand when cultural context makes a difference and when it does not.

Women of Bangladeshi origin comprise one of the most uniformly socio‐economically deprived ethnic groups in the UK. The majority of Bangladeshi families with dependent children are in the lowest income groups facing particular disadvantage in housing, employment and education. The Bangladeshi community is younger and has the largest average family size of all ethnic groups in the UK. These statistics, however, mask a diverse and changing population; for example, increasing numbers of women are born in the UK and participate in higher education.

Breast‐feeding has the potential to improve mother and infant health and reduce health inequalities. In the UK, women receive breast‐feeding support from midwives in hospital for the first 12–48 h, thereafter breast‐feeding support is offered by community midwives who make one to three home visits up to the tenth postnatal day. Following this, health visitors (public health nurses) visit mothers and babies at home at least once and offer ongoing support and growth monitoring in clinic settings. In some localities, additional dedicated breast‐feeding support is available, often including peer support and access to breast‐feeding support groups. Despite policy aspirations for increasing breast‐feeding. , rates remain low in many Western countries including the UK. The complex reasons for this include societal, health service and individual factors. In contrast to the majority white population, the relationship between breast‐feeding and socio‐economic status for non‐white groups in the UK is equivocal, suggesting that culture may play a role in breast‐feeding attitudes and practices.

A few studies have examined cultural factors related to breast‐feeding for women of Bangladeshi origin. These identified the influential role of grandmothers, different expectations of women and practitioners of postnatal care, negative attitudes to colostrum and the need to maintain modesty. Studies of the attitudes of health practitioners to provision of breast‐feeding support for women from minority ethnic groups have highlighted the prevalence of stereotypes, poor communication and organizational problems such as lack of provision of interpreters. Much of this literature, however, has methodological and conceptual problems including failure to acknowledge both diversity within populations and commonalities between minority groups and the majority population. ,

Our study aimed to examine the extent to which cultural factors make a difference to experiences of breast‐feeding and breast‐feeding support for women of Bangladeshi origin by comparing their views and experiences with those reported in the general literature. , We also aimed to explore the understandings and experiences of health practitioners of providing breast‐feeding support for women of Bangladeshi origin and to consider the implications for the provision of individualized culturally appropriate healthcare.

Methods

Research design

This qualitative study used in‐depth interviews and focus group discussions. The study took place between February and December 2008 in four localities in West Yorkshire and North East England.

Participants

Twenty‐three women who self‐identified as ‘of Bangladeshi origin’ and had breastfed a child within the previous 5 years were purposively recruited through individuals from local Bangladeshi communities. Additional sampling categories included maternal age, birth place and length of time in the UK, language, socio‐economic status, parity and age of youngest child to try and include characteristics that might influence experiences of breast‐feeding and breast‐feeding support , Table 1 summarizes participant demographic and socio‐economic characteristics. Eleven women were in paid employment, mostly working part‐time in low‐paid public service roles. The majority of the women’s partners also worked in low paid jobs, frequently in the catering sector. Five women lived in rented accommodation with the remainder living in owner‐occupied housing or with family members.

Table 1

Summary of women’s characteristics, including age, migration history, education and household structure

	Age	Birthplace	Age at migration	Woman’s education*	Household structure	No. of children	Age of youngest child
1E Barjan	33	Bangladesh	13	GCSE	Nuclear	2	3 years
2E	34	Bangladesh	15	GCSE	Nuclear	2	3 years
3E Rokeya	27	Bangladesh	6	GCSE	Nuclear	1	6 years
4E	27	Bangladesh	11	NVQ	Nuclear + 1 brother‐in‐law	3	5 months
5E Khaleda	30	Bangladesh	2	Degree	Nuclear	1	9 months
6E Makbula	29	Bangladesh	3	GCSE	Nuclear	2	4 months
7E Razia	30	UK	N/A	GCSE	Nuclear	5	6 months
8S Shorifa	30	Bangladesh	19	10‐year‐schooling in Bangladesh	Nuclear	5	13 months
9S Dilara	21	Bangladesh	14	10‐year‐schooling in Bangladesh	Nuclear	1	20 months
10E Zoreena	32	Bangladesh	2	A Level	Nuclear	2	13 months
11S	?40	Bangladesh	?25	2‐year‐schooling in Bangladesh	Single parent extended family	6	2 years
12S Farhana	25	Bangladesh	10	Left school age 16 in UK no qualifications	Extended (same household as W11)	3	8 months
13S	34	Bangladesh	18	Schooling in Bangladesh	Nuclear	6	20 months
14S	23	Bangladesh	18	GCSE	Extended	2	2 years
15S	21	Bangladesh	21	Graduated from religious school in Bangladesh	Extended Same household as W13	1	3 weeks
16S Jharna	30	Bangladesh	11	GCSE	Extended	4	10 months
17S Nasreen	?	Bangladesh	22	Primary school in Bangladesh	Nuclear	3	3 years
18S Nuzhat	33	Bangladesh	18	None	Single parent	6	5 years
19E	34	Bangladesh	2	A Level	Nuclear	2	2 years
20E	31	UK	N/A	PGCE	Nuclear + 1 brother‐in‐law	1	20 months
21E Afia	28	UK	N/A	Diploma	Nuclear	1	23 months
22E	31	UK	N/A	GCSE	Nuclear	1	4 months
23E Rehana	32	Bangladesh	8 months	GCSE	Nuclear	2	2 years

E, interviewed in English; GCSE, general certificate of secondary education; S, interviewed in Sylheti; PGCE, post‐graduate certificate of education.

*Highest qualification or number of years schooling.

Participant information sheets were translated into Bengali and checked for accuracy and appropriateness by a bilingual community worker. Thirteen women were interviewed in English by AM and ten in Sylheti by a bilingual researcher. Seventeen women were interviewed at home and six in community centres.

Five focus group discussions with 28 health practitioners and interviews with four service managers were conducted to locate women’s experiences within a service context. Of the 28 participants, half were midwives, five were health visitors and the remaining had a range of roles connected with either breast‐feeding support or services for women from diverse ethnic groups. Twenty‐three participants were of white British origin and only one, an interpreter, was of Bangladeshi origin.

Ethics

The study was approved by the North West NHS Research Ethics Committee. Bilingual recruitment facilitators explained the study to women and the researchers obtained written informed consent. Confidentiality and anonymity were assured. All names appearing hereafter are fictional.

Interviews and focus group discussions

Key topics for interviews with women included family background, breast‐feeding history and experiences, infant feeding practices in the UK and Bangladesh and suggestions for how breast‐feeding support could be improved. The interviews with women lasted for 40–80 min. Key topics for the practitioners and managers were experiences of providing breast‐feeding support for diverse communities and women of Bangladeshi origin and the role of families. The focus group discussions lasted approximately 1 h. The managers’ interviews lasted from 30 to 50 min.

All research encounters were audio‐recorded and transcribed. Translation by the bilingual researcher focussed on conveying meaning rather than literal translation, keeping as close as possible to participants’ words to produce authentic accounts. We have indicated where quotes are translated.

Analysis of research material

Analysis of the research material focussed on giving voice to participants while considering the context in which the accounts were produced. Detailed analysis began with the women’s interviews. Each transcript was coded using Atlas Ti software™. Initial coding was open and inductive, then codes were reorganized to form a logical framework. The complete data set was structured around women’s journeys from before their first pregnancies until they ceased breast‐feeding. Material from research encounters with health practitioners was analysed in the context of women’s accounts allowing comparison.

Findings

We present women’s experiences first, starting with ethnic identities, followed by experiences of breast‐feeding support in the early postnatal days. We then highlight key themes in women’s experiences of breast‐feeding in the home. The same structure is used to present practitioners’ experiences to show where they contrasted with women’s accounts and highlight where practitioners used cultural stereotypes and assumptions.

Women’s experiences

Ethnic identities

While all the women self‐identified as ‘Bangladeshi’ and ‘Muslim’, there were diverse interpretations of what that meant. Understanding this offers important context in which to interpret our findings. Razia, born in the UK, identified herself as ‘Bangladeshi’ because she ‘loved Bangladesh and felt it was home’. Shorifa, who migrated to the UK at 19 years old, felt British because she lived and had her children in the UK. The contextual nature of ethnic identities was illustrated as some women’s sense of identity shifted during interviews. Zoreena, who migrated to the UK at 2 years old and self‐identified as ‘Bangladeshi’, later referred to herself as ‘a modern British mother’ in contrast to her mother who ‘still held onto what she grew up with’. There was evidence of divergent interpretation of the influence of Islam on breast‐feeding practices; for example, some women described a religious obligation to breastfeed while others represented breast‐feeding as a personal choice.

Breast‐feeding support in the early postnatal period

Hospital support

All 23 women had given birth in hospital and all but one had commenced breast‐feeding in hospital. Most women described feeling weak or ill following birth and expected support from practitioners with all aspects of their own and their babies’ care. Jharna, who was satisfied with her care, provided insight into the type of support women might expect:

The support, well I had this big illness, having a baby is like having an illness isn’t it? [….]. They helped me with feeding and the nappy change, and there is sometimes when a first child is born they carry out the bathing and when I was weak they supported by holding me and walking with me (age 30, migrated aged 11, four children, all breastfed).

However, Nasreen suggested that hospital staff were not always sympathetic to her needs.

When this one was born if I asked them to help me they would get angry with me making an angry face. I just had the baby and I asked the midwife I mean the nurse to hold me to get up and she did not hold me (unsure of age, migrated aged 22, three children, all breastfed for a few days). Translated

According to women, hospital staff expected them to return to normal activity immediately following birth. This contrasts with postnatal support in Bangladesh where, like many parts of the world, new mothers receive practical support to help them rest, nurturing and special status. , This is significant because without such rest and nurturing, women may doubt their ability to breastfeed. Regarding support for breast‐feeding, women’s greatest concern was how to attach the baby at the breast and many received ineffective support. Rehana said:

They said ‘just breastfeed’ as though I know how to do it and I thought OK so I tried and I couldn’t even get the baby’s mouth on the teat [nipple] it was that bad. The next morning, the midwife came, showed me how to do it and she latched him onto the breast and he was sucking for a few seconds, soon as the lady went he came off it and I couldn’t put him back on (age 32, migrated aged 8 months, two children, breastfed for few days).

It was clear that many women expected practitioners to provide ‘hands‐on’ support for breast‐feeding. Nuzhat commented on the support she received in hospital:

It is helpful but if they had tried to help practically by holding the baby onto it [nipples] then it would have been good but all they do is give the bottle (age 33, migrated aged 18, six children, all mostly formula fed). Translated

Another common theme in the early postnatal days was concern about producing sufficient milk. Nasreen asked the interviewer:

I want to ask this thing that at that time there is no milk, during the first few days, 2, 3 days after the birth there is no milk so why do they keep on telling me to get them to suck and feed (unsure of age, migrated aged 22, three children, all breastfed for a few days). Translated

One consequence of women not receiving the support they expected following birth, struggling to establish breast‐feeding and doubting the adequacy of breastmilk in the early days, was that they introduced formula feeds to meet their babies’ needs. It appeared that hospital staff readily complied with or suggested this strategy without explaining the implications for establishing breast‐feeding:

Barjan: First 2 days because I didn’t have anything [breastmilk] and they [staff] did give him bottle because he was crying, he was hungry. So they said give him, like don’t let him get used to it because you are going to breastfeed him (age 33, migrated aged 13, two children, breastfed son for 1 year).

Community support

Women were generally satisfied with the breast‐feeding support they received in the early postnatal period from health professionals at home. Women particularly appreciated help with positioning the baby at the breast:

Khaleda: When I came back home I got midwives coming every other day […] She said ‘do you, do you want to do it in front of us so you know you’re doing it properly’ coz I said ‘I don’t know if he’s feeding properly or not’. […] She said can I touch you to show you how to do it and she moved the baby around so that it’s more comfortable (age 30 years, migrated aged 2 years, first child, breastfed for 8 months).

However, seven women commented that there were too few homes visits or that practitioners appeared rushed. Afia said:

It’s a matter of ticking boxes now and just getting in and out, as many visits as you can in a day and trying to fit everybody in but that’s not the way it should be. Every individual is different and every individual should be treated differently rather than a tick box (age 28, born in UK, first child, breastfed for 3 weeks).

In addition to postnatal home visits, seven women were informed about breast‐feeding support groups. None had attended because they did not have time, did not need support or did not want to breastfeed outside the home as in the following examples:

Q: Did you attend this breastfeeding support group?

Dilara: No I didn’t because I didn’t have much time. I was tied with him so didn’t get the time (age 21, migrated aged 14, first child, breastfed for 8 months). Translated

Q: And what about the support group, were you interested in anything like that?

Makbula: No I didn’t. No because I wouldn’t breastfeed in front of public. That’s the main problem. So if I was to go to one of these, you never know, she might be hungry and I might need to breastfeed her (age 29, migrated aged three, second child, breast‐feeding at 3 months).

Home context of breast‐feeding

In extended households, grandmothers appeared to play a key role in infant feeding and care. Zoreena described her experience with her first baby:

When (1st child) was born it was everybody took over […] and I think Mum did all, Mum felt she had to look after the baby coz I was newly‐married and my first child and I know 20 doesn’t seem young but it felt it (age 32, migrated aged two, breastfed first child for 1 week).

Several grandmothers were said to have influenced women to formula feed.

Rokeya depicted her mother’s attitude to breast‐feeding:

I don’t think my mum was quite happy with the idea of breastfeeding because I did it on the night and daytime I was feeding bottle and then one afternoon I forgot to make some milk and I thought OK I’ll give her this [breastfeed] and then my Mum was rushing to make me bottle and I went to my Mum it’s OK you don’t need to make any milk. She was like no, […] but then I started breastfeeding and Mum wasn’t too happy with that at all (age 27, migrated aged six, breastfed first child for 6 weeks).

Breast‐feeding was also influenced by women’s household responsibilities. Shorifa described her experiences when living with in‐laws:

My second daughter I didn’t feed her at all because I was very busy in the home. I had a mother‐in‐law, brothers‐in‐law and everyone together in the same house so I had to cook and tidy up and do the housework, so all day I was continuously busy (age 30, migrated aged 19, breastfed five children for varying lengths of time). Translated

In contrast, women living in nuclear households appeared to have more control over their time. Here, Shorifa talks about breast‐feeding in a nuclear household:

Like now if I sit and feed my son all day no one will say ‘go and do this or do that’ (Breast‐feeding fifth child at 13 months) Translated

Women living with their mothers had more help with household chores but were as unlikely to breastfeed. This was possibly because breast‐feeding had to take place in private and the baby was seen as belonging to the family, as Rehana described:

I was living with my Mum, and two of my sisters so there was quite a few of us and I can never breastfeed in front of Mum, I definitely couldn’t so I have to go upstairs in my room when it just felt like as soon as you go upstairs somebody’s calling you ‘oh somebody’s here to see the baby’ and you have to come back down (age 32, migrated aged 8 months, two children, breastfed for few days).

Women expressed diverse attitudes to breast‐feeding in front of others. Most, like Farhana, said they would only breastfeed in the home:

We can’t have people seeing us because we are Muslim ladies. You have to keep your breast hidden in parda [covered]. It’s not right to feed outside but you can do it at home inside (age 25, migrated aged 10, three children all breastfed). Translated

Three women said they would not breastfeed in front of anyone including their husbands. Only three women had breastfed outside the home.

Thus, household structure, the role of grandmothers and embarrassment at feeding in front of others combined to make breast‐feeding problematic. Facing these difficulties, many women turned to formula feeding, either exclusively or in combination with breast‐feeding. This contributes to explanations for high rates of mixed feeding among women of Bangladeshi origin.

Practitioners’ experiences

Many practitioners viewed the Bangladeshi community as fixed and homogenous. This was typified by statements like ‘they’re a culture that think babies should sleep all night’ or that all women of Bangladeshi origin lived in extended families and were socio‐economically deprived. In only one focus group were diversity of socio‐economic circumstances and family structure acknowledged. While the complex nature of ethno‐religious identities was reflected in practitioners’ confusion around ethnicity, religion and language, there was a lack of insight that this underlined complex, contingent identities. There were examples of practitioners conflating Bangladeshi women’s identities with those of women of Indian and Pakistani origin as ‘Asian’, and confusion around religious categories. Many practitioners viewed language as the main identifying characteristic and barrier to meeting women’s needs. Nearly all practitioners and managers described the diversity of the population they served in terms of language.

Supporting women to breastfeed in the early postnatal period

Practitioners appeared to have little understanding of the expectations of postnatal support of women of Bangladeshi origin. Consequently, they interpreted women’s reluctance to mobilize following birth or to breastfeed without ‘hands‐on’ help as confirming stereotypes of women as passive:

Midwife 1: They’re very passive, aren’t they?

Midwife 2: Yes

Midwife 1: They’ll let us attach the baby for them to the breast but although they’re very shy, but they’ll actually, it’s the whole thing about us, they expect us to do things for them so they’ll let us attach the baby to the breast for them, just like they’ll let us care for the baby but they won’t do it themselves (extract from focus group five)

Practitioners confirmed women’s accounts that breastfed babies of mothers of Bangladeshi origin were frequently formula fed in hospital but practitioners’ ascribed this to a cultural or religious belief that colostrum is unclean:

Health visitor 1: A lot still feel that the colostrum is [dirty] that they need to get rid of that […]. Particularly the Bangladeshi community groups still have that.

Health Visitor 2: It’s the Muslim women, isn’t it? Muslim women will discard the colostrum because it is felt to be unclean so they will bottle feed for the first 3 days and then breastfeed quite successfully, some (extract from focus group 1).

In contrast, women reported giving formula feeds because they were struggling to establish breast‐feeding and doubted that colostrum was adequate to meet their babies’ needs.

Support in community settings

Peer support and breast‐feeding support groups were prominent in practitioners’ discussions of how they would like to develop services for women from minority ethnic communities. The practitioners and managers recognized the need to provide culturally appropriate community support tailored to the needs of diverse populations. One manager stated:

One of the things that we’ve learnt from Children’s Centres working in community development approaches is the baby café approach may not be the first way to do breastfeeding support.

There were several examples of unsuccessful service initiatives for women of Bangladeshi origin because women’s circumstances had not been taken into account:

We’ve tried to run something called [inaudible] workshops for Bengali women 4/5 years ago and we got an interpreter on board and we got someone to go in a taxi to collect them. The first week we had about four or five, it was quite good. Just went round hunting people and got people out but after that we never got anybody. Just one or two, you know (practitioner focus group four).

Practitioners’ explanations of why such initiatives were unsuccessful rested on assumptions about women’s control over their lives, for example one community midwife said:

I think there’s the permission thing, as well. Often you have to go by custom, you know, mother‐in‐law for them to get permission to go to things (practitioner focus group four).

It was only when prompted that practitioners acknowledged that low‐income white women also did not attend breast‐feeding support groups. Practitioners and managers expressed frustration in their attempts to provide community services for women from diverse backgrounds.

Practitioners were aware of the struggles some women faced combining breast‐feeding with household responsibilities and maintaining modesty. They were also aware of the involvement of grandmothers in women’s daily lives and caring for babies. However, this appeared to confirm stereotypes of women’s passivity. Women were described as ‘timid’ or ‘rolling over backwards’ when pressured by grandmothers to give formula feeds. For many practitioners, the role of grandmothers was problematic. One midwife spoke of not wanting to ‘tackle the grandmother’. Another expressed feelings of frustration:

You can be there for a long time giving them all the information about exclusive breastfeeding and establishing breastfeeding and because of, you know there’s a lot more exterior family who are going to be putting pressure on her to, you know, the baby’s hungry, give a bottle. I’ve got to say I do sometimes think, ‘Well, I’ve only got another half an hour,’ I’ve got another so many people to get round and my motivation sometimes dwindles a bit. You kind of think, ‘I know, this afternoon, you’re going to give the baby a bottle anyway’ (practitioner focus group four).

Discussing opportunities to engage with families, most practitioners said they lacked time or would not feel comfortable discussing breast‐feeding with families. Thus, many practitioners felt they had little influence on women’s infant feeding decisions.

Organizational context

From practitioners’ accounts there were organizational barriers to providing culturally appropriate breast‐feeding support. Some of these such as time and lack of priority by management applied to supporting all women. Specific to providing culturally appropriate care were lack of training and language barriers.

Practitioners had received recent equality and diversity training in only one locality. It was evident that, by focussing on cultural practices, previous training could potentially reinforce cultural stereotypes rather than challenge them. One manager recognized the challenges of providing training:

What we want to think through is that we are not pigeon‐holing the training but rather think more broadly around what are the common themes coming through that so that people aren’t attending homelessness‐awareness on 1 day and then gypsy and traveller‐awareness the next day, but to just think through what are the needs of the minority and also vulnerable communities.

Unlike women’s accounts, language barriers were a key feature of practitioners’ discussions with descriptions of organizational barriers and cost constraints to provision of interpreting services:

Midwife: We can book interpreters but the Trust doesn’t like us to use them too often.

Facilitator: Why is that?

Midwife: I mean, I was thinking of cost (extract from focus group one).

Discussion

This study provides an insight into the challenges of providing culturally appropriate maternity care that meets the needs and improves the experiences of women from diverse backgrounds. It highlights the extent to which cultural context makes a difference to women’s experiences of breast‐feeding and breast‐feeding support compared to women from the majority population. The starting point for this is recognition of the complex and contingent nature of ethnic identities. Women’s rich descriptions of various facets of their identities can be contrasted with practitioners’ representations of women of Bangladeshi origin as homogenous. Practitioners neither recognized diversity within the Bangladeshi population nor similarities between women of Bangladeshi origin and the majority population.

Consequently, practitioners did not recognize when breast‐feeding support needs of women of Bangladeshi origin were similar to those of the majority population. For example, it is frequently reported that many women have difficulties with positioning and attachment of the baby at the breast and doubt the adequacy of breastmilk in the early days. , From our findings, implementing evidence‐based practice, which includes proactive support with positioning and attachment and exploring women’s understandings of the adequacy of colostrum to meet a baby’s needs, for all women regardless of ethnic background is the most important step towards improving breast‐feeding support for women of Bangladeshi origin.

However, our study also suggests that cultural factors make a difference to the experiences of breast‐feeding and breast‐feeding support of women of Bangladeshi origin. For example, practitioners’ assumptions about women’s attitudes towards colostrum were a barrier to women making informed decisions about giving their babies formula feeds in hospital. A further example was women’s expectations of support in the early postnatal period. This contrasted with practitioners’ interpretations of women’s behaviour as confirming stereotypes of passivity. These cultural misunderstandings underpinned a failure to respond to women’s individual needs. Antenatal discussion of what to expect in a UK hospital and greater sensitivity to women’s expectations postnatally could address this.

Our research confirmed findings that current provision of community breast‐feeding support may be inappropriate and inaccessible to many women from minority ethnic backgrounds. However, this may also be true for many low‐income white women. , There was evidence that service initiatives failed to consider women’s circumstances in general. Involving women and families in service development could avoid this. Practitioners could work alongside community projects to develop more accessible and appropriate services.

Our study suggested that cultural factors also influenced women’s experiences of breast‐feeding in the home context. For some women, breast‐feeding was constrained by the combination of household responsibilities, having to breastfeed in private and the influence of grandmothers. This is congruent with previous studies of the position of women of South Asian origin living in extended households. , However, women in our study had diverse living arrangements and attitudes towards breast‐feeding in front of others. Many practitioners did not recognize this diversity and thereby stereotypes of women as passive were reinforced. Our findings suggest that women’s experiences would be improved if practitioners found ways to engage with wider family members.

To provide culturally appropriate breast‐feeding support, practitioners need to explore women’s individual needs and expectations regardless of ethnic background. To move beyond cultural stereotypes while remaining aware that culture plays a role in women’s infant feeding experiences requires practitioners to develop sensitivity to the cultural factors suggested by our study. There was a striking absence of training to increase practitioners’ understanding of these complex issues. Training should move away from offering prescriptive cultural descriptions, purporting to explain and manage ‘ethnicity’. Instead, it should offer general principles which contextualize diversity and difference, without recourse to simplistic explanations and naïve solutions, potentially perpetuating disadvantage and discrimination.

Training alone will not improve women’s experiences of health services if other organizational constraints are not addressed. Barriers to the provision of language services were highlighted confirming other research. However, our research suggested practitioners overemphasized their impact on women’s experiences.

We believe this study is unique in its reflexive exploration of women’s experiences of breast‐feeding within the context of their daily lives and encounters with health services. The comparison of women’s understandings with practitioners’ perspectives has highlighted how cultural stereotypes can be perpetuated. However, there are limitations to the study. These include the small number of participants, the socio‐cultural distance between the researcher and women participants, and the problems associated with generating data in two languages. Nevertheless, the study has raised important issues regarding provision of culturally appropriate care and demonstrated where provision of breast‐feeding support for women of Bangladeshi origin could be improved.

Conclusion

This study focused on complex topic areas which are challenging for health services. It demonstrates how practitioner assumptions and cultural stereotypes resulted in services not meeting women’s needs. Implications for education, policy and practice include the need to train health practitioners to work with diverse populations, implement evidence‐based practice and provide an organizational context which supports practitioners, enabling them to respond appropriately and sensitively to the diverse needs of their client or patient population.

Funding

This work is produced by Alison McFadden under the terms of Researcher Development Award research training fellowship issued by the NIHR. The views expressed in this publication are those of the authors and not necessarily those of the NHS, The National Institute for Health Research or the Department of Health.

Authors’ contributions

All authors contributed to the design of the study. AM collected and analysed the data, and drafted the manuscript. All authors contributed to the critical revision of the manuscript and approved the final version.

Declarations

There are no conflicts of interest.