Breastfeeding practice in the UK: midwives’ perspectives

Abstract

Despite breastfeeding prevalence increasing, many mothers in developed countries are dissatisfied with care provided by midwives. However, a paucity of research exists related to midwives’ experiences of supporting breastfeeding mothers. This study explored the experiences of English midwives’ during their breastfeeding support role. A qualitative study using grounded theory principles was used. Data were collected using in‐depth interviews and analysed using constant comparative techniques. The setting was two maternity hospitals in the North of England, UK. Thirty midwives who cared for normal, healthy babies participated. Volunteers were recruited using theoretical sampling techniques. The core category that emerged is called ‘surviving baby feeding’ and relates to midwives’ experiences when supporting mothers. The results reported in this paper refer to one category called ‘doing well with feeding’ which has three main themes: (1) communicating sensitively, (2) facilitating breastfeeding, and (3) reducing conflicting advice. Participating midwives reported practice that suggests that they valued breastfeeding, attempted to provide realistic information and advice, and tried to minimise confusion for mothers. However, some midwives used an authoritative manner when conversing with mothers. English midwives’ reported practice demonstrates that these midwives appreciated that breastfeeding mothers required specific support. However, breastfeeding education that encourages midwives to develop effective skills in ascertaining mother’s needs, but also encourages mothers to effectively participate in their care, should be provided. Further research is needed to clarify breastfeeding mother’s expectations and needs.

Introduction

Breastfeeding prevalence has steadily increased in many developed countries (Tappin et al. 2001; Yngve & Sjostrum 2001; Hamlyn et al. 2002; Leung et al. 2002; Ryan et al. 2002; Labbok et al. 2006). This is encouraging as research has shown that breastfeeding confers significant health benefits for both the mother and infant (Wambach et al. 2005). Reviews of evidence suggest that breastfeeding is important in tackling current public health challenges such as obesity (Owen et al. 2005), coronary heart disease (Dykes & Hall Moran 2006), and some cancers (Collaborative Group on Hormonal Factors in Breast Cancer 2002). Today, many health organizations around the world, including the World Health Organization (WHO), recommend that all babies are exclusively breastfed for the first 6 months of life, with continued breastfeeding for the following 6 months and beyond (WHO 2002; Department of Health 2003; American Academy of Pediatrics 2005). Yet, despite breastfeeding incidence at birth increasing in England (Bolling 2006), many mothers cease breastfeeding well before 6 months (Hamlyn et al. 2002). For example, in 2000, 71% of babies born in England and Wales were breastfed at birth, but fewer than 22% were still breastfed at 6 months (Hamlyn et al. 2002).

The breastfeeding process is complex. Physiologically, lactation is possible for most women (Akre 1989). Nonetheless, breastfeeding is affected by psychosocial and cultural influences in the proximate environment, as well as the healthcare practices encountered (Renfrew et al. 2000; McFadden & Toole 2006). Indeed, the provision of breastfeeding support by health professionals has often been criticized by mothers in developed countries (Stamp & Crowther 1994; Singh & Newburn 2000; Dykes 2006; Smale et al. 2006). Mothers have highlighted unhelpful practices, such as insensitive and unsupportive attitudes, insufficient information, conflicting advice, and hospital staff too busy to follow up mothers’ feeding progress from midwives, in Australia (Stamp & Crowther 1994; Hailes & Wellard 2000; Brown et al. 2005), Hong Kong (Tarrant et al. 2002), the UK (Rajan 1993; Hoddinott & Pill 2000; Murphy 2000; Dykes et al. 2003; Baxter 2006; Dykes 2006; McFadden & Toole 2006; Smale et al. 2006), and Sweden (Kvist et al. 2006). Although these studies have provided useful insights of mothers’ experiences of breastfeeding in the immediate postpartum period, the literature exploring the experiences of the health professionals who provide the breastfeeding support for these mothers at this time is not prolific. This study explored the experiences of English midwives’ during their breastfeeding support role. In this paper, we report on one topic that emerged from the results of the larger study.

Method

In this study, midwives are health professionals who have a legal role to provide postpartum care, including breastfeeding support, to mothers and their babies in the hospital and after transfer home in the UK [Nursing & Midwifery Council (NMC) 2004].

Study design and setting

This was a qualitative, exploratory study based on grounded theory principles (Glaser 1998). The setting was two maternity hospitals in the north of England. Neither setting had achieved baby‐friendly status (UK Baby Friendly Initiative 1998). Permission to start the study was obtained from both university and hospital research ethics committees, and midwifery managers.

Sample recruitment and selection

The inclusion criteria were midwives who cared for mothers who had given birth to normal, healthy babies. Midwives who cared for babies who were ill or had specialized feeding requirements were excluded. Participants were recruited by a poster exhibited in the clinical areas where midwives worked. Each volunteer was given a written information sheet explaining the study rationale and research design. Volunteers’ names were placed on a list, along with their contact details and a short summary of their clinical experience, which was later used for recruitment for interview using theoretical sampling procedures. This process involves the researcher making a ‘theoretically informed’ decision on which participant to recruit next, based on the results that emerged from the analysis of the previous participants’ interviews. The theoretical sampling process enables data to be generated that are focused on the study objective, yet take account of the variation and scope of ideas (Glaser 1998). For example, midwives who highlighted challenges experienced supporting mothers with breastfeeding were asked how they managed these situations. The ideas generated from the initial interviews were followed up with midwives working in all clinical areas, those who worked only during the day, or at night. As the study progressed, midwives who had undertaken post‐registration breastfeeding education were theoretically sampled from the list of volunteers for their ideas on managing breastfeeding challenges in order to increase the range of ideas.

The main researcher (C.M.F.) is a midwife educator who was relatively well known at both maternity units. Indeed, the research questions arose from C.M.F.’s experiences in providing breastfeeding education to qualified health professionals. In order to maintain a neutral stance, C.M.F. adopted a reflexive approach to identify her own personal values, and any possible power differentials, role conflicts and the interests of the reciprocators. Reflections were made throughout the study, which enabled C.M.F. to remain sensitive to her own thoughts as well as those of the participants.

Data collection

Data were collected by C.M.F. using unstructured audiotaped interviews with midwives who worked between Autumn 1999 and Autumn 2001. Participants signed a consent form at the beginning of the interview and were advised that they could withdraw at any time. Theoretical sampling principles also meant that some midwives who originally volunteered were not interviewed because they were not theoretically ‘suitable’. These midwives were thanked for volunteering and provided with a written explanation of why they were not approached. On completion of the study, participants received a copy of the results.

Each participant completed a short questionnaire outlining, for example, their clinical experience and midwifery education at the outset of the interview. Field notes were made after each interview of any relevant contextual factors. The data collection period lasted 2 years because C.M.F. worked part time, and data collection and analysis were concurrent to confer with grounded theory principles (Glaser 1998). Unstructured interviews were used because of their ability to enable the researcher to introduce the topic and for the participant to openly express what is most relevant to themselves (Holloway & Fulbrook 2001). Each participant was asked the same question ‘Tell me about your views on feeding?’, and then probing questions were used to explore the participant’s responses relevant to the study objective (Kvale 1996). The flexibility inherent in the unstructured interview, and use of theoretical sampling, enabled a thorough exploration of ideas that had emerged from previous interviews (Glaser 1998). Interviewing ceased after 30 interviews had been completed and saturation (no new ideas emerged from the data) was achieved (Glaser 1998). Average interview duration was 1 h and 15 min. Transcripts were returned to each participant for review (none of the participants queried these).

Approach to analysis

During the analysis process, each participant’s ‘data set’ of transcript and field notes was initially read, and then coded using line‐by‐line analysis (Glaser 1998). These codes, and subsequent codes from other interviews, were constantly compared with each other. Further codes, or conceptualisations, were made from the relationships that the codes had with each other (Glaser 1998). By doing this, the codes became arranged according to their similar content into ‘categories’ that were labelled according to their broad conceptual descriptions. Each category had several subcategories. For example, one category labelled ‘knowing about feeding’ had five subcategories: conflicting advice, beliefs about breastfeeding, deciding how to feed, equipment and feeding, and information about feeding for women. The computer software nudist (Qualitative Research & Solutions Pty. Ltd. 1997) was used to store and manage the data.

Rigor was maintained by C.M.F. constantly recording ideas from the data on memos (Glaser 1998) in order to keep track of thoughts, and provide a rationale for coding and theoretical sampling over the 2‐year period. An experienced qualitative researcher, not involved with this study, checked coding of the data at an early stage. Adherence to the constant comparison process ensured that the data ‘fitted’ the emerging codes (Glaser 1998). Finally, all participants were invited to a presentation of the results. Those who attended assured the authors that the results ‘fitted’ their ideas, and represented their experiences of their breastfeeding support role. Each midwife was assigned a pseudonym to maintain anonymity; this was used during the research process and reporting of the results. All data were safely secured, and audio tapes destroyed once the study was complete to ensure confidentiality.

Results

Characteristics of participants

All participants were female (n = 30). Length of clinical midwifery experience varied from 8 months to 31 years. Twenty‐one midwives worked only in the hospital, including antenatal clinics, labour wards and post‐natal wards, covering either day or night shifts. Another six were community midwives visiting mothers and babies at home. Two worked in ‘midwifery teams’ (Allen et al. 1997), providing holistic midwifery care in both the hospital and community. One midwife was a manager with responsibility for hospital and community. These midwives had gained their registration by varying routes. Twenty‐six were registered nurses who had completed midwifery education, and four had registered after completing midwifery education for non‐nurses in the 1990s. Twenty‐three midwives had completed further study, after their initial education, including midwifery diploma/degree awards. Seven midwives had not undertaken any further study since midwifery qualification; however, all but one of these participants had completed their education in the preceding 2 years. All but six midwives had attended study days related to breastfeeding. Two had studied breastfeeding post registration; one was studying for the International Board Certified Lactation Consultant qualification, and another had completed an English National Board‐approved course ¹ in breastfeeding.

Summary of categories

A core category called ‘surviving baby feeding’ emerged, which relates to the processes that these midwives dealt with when supporting mothers. ‘Surviving baby feeding’ consists of four main categories: altering proximities of feeding, emotionalizing feeding, struggling with feeding, and directing feeding. Each of these categories has several subcategories, but here we report on one labelled ‘doing well with feeding’ that is part of the ‘directing feeding’ category. Other results from this study can be accessed elsewhere (Furber 2004; Furber & Thomson 2006a, 2006b; Furber & Thomson 2007).

‘Doing well with feeding’ subcategory

Data suggested how these midwives dealt with feeding in their midwifery practice. Three main themes related to practice were apparent in this subcategory:

• • communicating sensitively,
• • facilitating breastfeeding, and
• • reducing conflicting advice.

Communicating sensitively

Several midwives explained how they motivated mothers to persevere with breastfeeding, especially when mothers were struggling with the frequency of feeds, and felt like giving up. These midwives illustrated that they recognized that mothers required reassurance when initiating breastfeeding, so spent time explaining what was happening, and what to anticipate during the time in their care:

The easier option at 6 o’clock in the morning is to give a bottle (which the Mum asks for). Whereas I would say if you want to give your baby a bottle that is up to you, but I have to tell you that it could have this effect, this effect, and this effect on breastfeeding. If you look at the physiology of breastfeeding and discuss it with her . . . it is normal for a baby to feed little and often at this point because of the size of its stomach. Then I find that they say, ‘Oh well, I’ll carry on and I’ll be all right’, and then by day 3 the milk comes in and they feel a lot better. (Andrea, hospital midwife)

Another participant confirmed that she visited mothers up to 28 days after the birth (NMC 2004), providing breastfeeding support after the health visitor had become involved:

I say things will get better usually within 2 weeks, when most women are quite happy with breastfeeding. They’re over the worst of the problems, start feeling as if they’re really getting ahead with things. I find that . . . you’re keeping them on [after women are normally transferred to sole health visitor care] because they’re needing that extra support . . . with breastfeeding. Just to keep them going with it in many ways because two weeks they’re just beginning to settle down but they might just need that little bit . . . Somebody there to say ‘Yeah, that’s going well, keep going’. (Ann, community midwife)

Another midwife explained that she encouraged mothers to articulate their needs and reflect on how the breastfeeding process was progressing:

I always say to them ‘Think logically, take a step back if the breastfeeding isn’t going according to plan. What would you like me to do? How can I help you? What do you want?’. (Sue, hospital midwife)

Several midwives were aware of the personal and private nature of assisting mothers with breastfeeding, and that some mothers may feel affronted if their breasts were touched inappropriately. See this example:

I think some women can be put off by excessive or insensitive handling of the breasts. I think that this can be sometimes quite insensitively done and women don’t like that. You can tell from their body language. (Connie, community midwife)

Facilitating feeding

These midwives spoke at length of the measures that they employed when helping mothers initiate breastfeeding. Several described explanations that they used when observing mothers attaching their baby onto their breasts:

I’ll explain to them as they’re going along, if they’re just opening [babies’ mouth] and I see them pushing [the breast in baby’s mouth], ‘I say it’s better to wait until they get a great big mouth before you put them on because then they’re taking more off the areola . . . much better’. (Joanne, community midwife)

You have to tell the woman ‘Wait until the baby opens its mouth, try and get the nipple inside on top of the tongue and hold the baby close’. We all have that habit of pulling the breast away; you just have to tell them that doesn’t really matter. I find [with] large‐breasted women . . . I say to them ‘Sometimes, you might have to support underneath [the breast]’. (Mary, hospital midwife)

Another midwife described how she positioned her own body so that the mother observes the baby’s position from a similar position when she will be attaching her baby onto her breast herself:

I would normally pick the baby up and show them how to hold the baby. I would show them how to fix the baby onto my own breast, not exposed! I would stand with my back to them so they can see from their viewpoint what they are looking for, how to get baby to open its mouth wide, how to manipulate the baby on to the breast and how for them to put the baby on the breast rather than allow the baby to take the breast and then to change position. (Carol, hospital midwife)

Carol then went on to describe how she encourages the baby to open his/her mouth for attachment:

I rub the baby’s nose around the nipple area, and just keep teasing the baby and keep moving the face away from the breast tissue. As it goes away it loses the smell of breast milk and therefore will go to root . . . make sure it is actually lower on the breast than above the nipple. . . . It doesn’t happen immediately, it does take time. And the first few times you go to do it baby will take a while to get into the habit of doing this. Particularly if it has some feeding practice when it has just been taking the end of the nipple ’cos that is what he will want to do straight away rather than actually open its mouth. The ‘trick’ is to get the bottom lip on first and to place the baby’s nose at the top of the nipple and it’s getting through to Mum that you are moving the baby on to the breast rather than allowing baby to ‘take to’ the breast. I think the position of the baby at the breast for the mother and that particular baby, is a bit of an art. There’s always a slightly different position for that Mum and that baby which is optimum for her and that attachment. It’s being able to give the mother skills to obtain that. The mother has got to be able to do that for herself, and doing a ‘hands‐off’ approach is very difficult. But it’s not always got to be a hands‐off approach, sometimes you have to position Mum, and show Mum, and then sometimes go back and show her how to use her hands to position baby.

Other midwives described how they used differing maternal and neonatal positions to ensure that the mother was comfortable, and that breastfeeding was optimal:

. . . they are lying down [after their caesarean section]. I try to get women in a position that they’re comfortable, or I put the baby on a pillow. (Dorothy, hospital midwife)

The rugby ball [hold], women find that comfortable, it can also drain off different areas of the breast as well if it’s a bit engorged on one [breast]. (Victoria, community midwife)

When mothers had sore nipples, these midwives described how they managed these situations:

I check and re‐check the position and say ‘If you’re getting sore nipples, every feed get somebody to check your positioning. A slight alteration in position can make all the difference. Make sure baby’s latched on properly’. (Lucy, hospital midwife)

Just massage in a little bit of breast milk in [the areola and nipple area]. (Nancy, hospital midwife)

Several midwives described how they used skin‐to‐skin contact:

At least four women changed their mind. They were going to bottle feed . . . after skin‐to‐skin contact was initiated after delivery they changed their mind and went on to breastfeed. (Kathie, hospital midwife)

I use skin‐to‐skin contact especially if they have a section. I tuck them up in bed with them, put them safely with Mum, just next to the breast. (Sue, hospital midwife)

Another participant described how she had started to facilitate skin‐to‐skin contact with mothers and babies in the post‐natal ward:

I have been practising it on the ward. It’s something that’s a bit different rather than actively trying to latch the baby on. I’ve let the baby lie there. I have encouraged her to expose her breast to have the baby there so it can smell and feel, and hopefully eventually ‘cotton on’ itself. . . . I’ve only done it three times since I saw the film [a reference to observing a baby ‘crawling’ to its mother’s breast in a video film] last week. (Nancy, hospital midwife)

When mothers at home were experiencing breastfeeding problems, these midwives encouraged them to call the midwife at feeding times so that he/she could visit and supervise the feed:

I get them to ring when the baby’s woken up . . . and then I’ll go back and supervise the feeding. (Kirsty, community midwife)

Reducing conflicting advice

The results indicate that hospital and community‐based midwives were trying very hard to minimize conflicting advice. These excerpts illustrate some of the strategies devised to share ideas with each other, and to find out what others had advised mothers:

As a team, we meet everyday and . . . the most common one is weight. That’s something that we would discuss with each other. What do you think? This baby has lost so much weight and it’s not gained any what shall I do? (Liz, community midwife)

I always make a point of saying to the Mums that just because I’ve said something different doesn’t mean it’s necessarily conflicting. It’s just that you’ve moved on in the plan of care so where [the other midwife] might have suggested one thing yesterday we tried that, and that’s not worked. I’m telling her something different for this particular time. (Andrea, hospital midwife)

Discussion

This study has identified midwifery‐reported practice outlining how midwives support mothers with breastfeeding. To some extent, these results contradict some mothers’ experiences of midwifery support (Stamp & Crowther 1994; Hailes & Wellard 2000; Hoddinott & Pill 2000; Singh & Newburn 2000; Tarrant et al. 2002; Dykes 2006; McFadden & Toole 2006; Smale et al. 2006). Successive studies in developed countries such as Australia, Canada, the UK, and the United States of America have shown that breastfeeding mothers want more attention from the midwife – practical advice and support such as being shown the best positions for attachment (Stamp & Crowther 1994; Raisler 2000; Graffy & Taylor 2005; Nelson 2006), and emotional encouragement (Stamp & Crowther 1994; Raisler 2000; Hauck & Irurita 2003; Graffy & Taylor 2005; Nelson & Sethi 2005). Studies with mothers exploring health professionals’ responses to the need for breastfeeding support suggest that this is often a low priority in their workload in hospitals in both Australia (Brown et al. 2005) and the UK (Singh & Newburn 2000; Graffy & Taylor 2005; Dykes 2006). In the present study, midwives illustrated that they valued breastfeeding, attempted to provide realistic information and advice, and purposefully made efforts to encourage mothers.

The explanations of facilitating breastfeeding suggest that these midwives have specific knowledge of the attachment process and the technical nature of breastfeeding (Dykes 2006), especially in relation to the need for the baby’s mouth to be open wide and use of the baby’s innate reflexes in opening the mouth, maternal positions that may encourage the flow of breast milk (Henschel & Inch 1996; Royal College of Midwives 2002), and practices for managing sore breasts (Renfrew et al. 2000). The use of skin‐to‐skin contact described here is particularly heartening as a systematic review indicates that this practice may improve breastfeeding outcomes (Anderson et al. 2003). Nancy’s experience facilitating skin‐to‐skin contact demonstrates the benefits of ‘seeing’ effective practice in an educational session, and the impact that this may have on changing practice.

Many authors recommend that ‘hands‐off’ practice (mothers are taught to attach the baby using verbal techniques supplemented with written information) is better when supporting breastfeeding mothers, than ‘hands‐on’ (baby is attached to the breast by the health professional) (Henschel & Inch 1996; RCM 2002; Inch 2003). In this study, ‘hands‐on’ practice, or even a ‘combined approach’ (Carol), was apparent. It was pleasing that some midwives were sensitive to the feelings that mothers may have about being touched in intimate parts of their body. Mothers have often complained of health professionals ‘manhandling’ their breasts (Green et al. 1998; Hoddinott & Pill 2000; Dykes 2006), which can lead to embarrassment (D’Anzi 1998) or even reminders of past abusive encounters (Price 1988).

This study has emphasized the different techniques used by midwives to facilitate attachment of the baby on the breast. The literature, nonetheless, is ambiguous. Mothers state that they want to be ‘shown’ how to attach their baby to their breast, but it is not clear whether they mean ‘hands‐on’ or ‘hands‐off’ assistance (Hoddinott & Pill 2000; Graffy & Taylor 2005). Despite several research studies from Australia (Duffy et al. 1997; Fletcher & Harris 2000; Henderson et al. 2001; Forster et al. 2004) and the UK (Ingram et al. 2002; Woods et al. 2002; Wallace et al. 2006) evaluating ‘hands‐off’ vs. ‘hands‐on’ practice, the results are equivocal because of research design limitations (Wallace et al. 2006). Wallace et al. (2006) suggest that distinct ‘hands‐off’ care at the first feed may not be as important as initially considered, and that appropriate supervision of the breastfeed (including ‘hands‐on’ care) is more important. Further evidence is therefore required to support the method of facilitating breastfeeding that is preferred by mothers and is most effective.

Conflicting advice disempowers mothers, and weakens their self‐confidence and ability to breastfeed (Simmons 2002). Midwives in the present study tried hard to minimize this, and were aware of the basis of conflicting advice, such as being cared for by midwives who do not communicate with each other, and variations in their knowledge. They also recognized that conflicting advice does not necessarily mean ‘incorrect advice’ (Andrea), and that inconsistent and inappropriate advice may be detrimental (Cox & Turnbull 2000).

Unfortunately, the language used by some midwives indicates an authoritarian stance. This approach could be overbearing for some mothers (Hewison 1995), and undermine their ability and confidence in continuing breastfeeding. ‘Telling’ mothers what to expect implies an imposing tone (Andrea), whereas asking the mother ‘for her reasons for bottle use’ demonstrates respect for the mother’s opinions and views (Hunter 2006). Suggesting that breastfeeding may take 2 weeks to improve (Ann) may seem a long time to a new mother, but encouragement without time frames may be more reassuring. Carol’s ‘manipulation’ of the baby’s mouth onto the breast, rather than waiting for the baby to open his/her mouth and latch onto the breast naturally, indicates an assertive, even aggressive, stance. Some care appears ‘commanding’, and devoid of discussion with the woman regarding her preferences and needs, suggesting that the midwife prefers to control the feeding interaction and a power imbalance in the midwife–mother relationship exists: Lucy –‘make sure that the baby is latched on properly’, Kirsty’s desire to ‘supervise’ feeds, and Sue –‘I use . . .’.

Limitations

It cannot be presumed that these quotes are representative of all midwives’ breastfeeding practice, especially as the voluntary recruitment process attracted midwives with a special interest in breastfeeding. Breastfeeding practice is most likely to be variable among midwives, as in the wider study, some participants disclosed practices that were not research based (Furber 2004; Furber & Thomson 2006a), including some of the participants quoted in this paper. It also cannot be assured that what these midwives ‘said they did’ is how they execute their practice. Therefore, observation of the midwives in practice and data collected from mothers receiving the care may have provided more insight into midwives’ care and communication with mothers.

Conclusion

This study provides insight into the knowledge and attitudes of midwives towards breastfeeding support. However, improvements are still required with communication skills. Specific education designed around the WHO/UNICEF Breastfeeding Guidelines is influential in improving midwives’ knowledge and confidence in supporting breastfeeding mothers (Dinwoodie et al. 2000; Wissett et al. 2000; Hall Moran et al. 2004), and should be promoted. Breastfeeding education that equips midwives with the skills to ascertain needs and anxieties, and facilitates participatory approaches of mothers, in their care (Laverack 2005), may go some way to address the current mismatch of mothers’ expectations of their care with what many health professionals provide (Smale et al. 2006). However, support for midwives is crucial, especially in relation to improving skills and knowledge, and maintaining these in practice. One way forward in the UK may be for statutory midwifery supervision (NMC 2004) to be more proactive in monitoring, and supporting breastfeeding support. There is still need for rigorous research exploring midwives’ breastfeeding practices, and their rationale for care, but for this to be most helpful, data collection that includes observation of their practice and mothers’ opinions of the care received is needed. Further research and clarification of mothers’ needs and expectations will also enable midwives to prepare their care more appropriately.