‘It should be the most natural thing in the world’: exploring first‐time mothers' breastfeeding difficulties in the UK using audio‐diaries and interviews

Abstract

Breastfeeding is a practice which is promoted and scrutinized in the UK and internationally. In this paper, we use interpretative phenomenological analysis to explore the experiences of eight British first‐time mothers who struggled with breastfeeding in the early post‐partum period. Participants kept audio‐diary accounts of their infant feeding experiences across a 7‐day period immediately following the birth of their infant and took part in related semi‐structured interviews a few days after completion of the diary. The overarching theme identified was of a tension between the participants' lived, embodied experience of struggling to breastfeed and the cultural construction of breastfeeding as ‘natural’ and trouble‐free. Participants reported particular difficulties interpreting the pain they experienced during feeds and their emerging maternal identities were threatened, often fluctuating considerably from feed to feed. We discuss some of the implications for breastfeeding promotion and argue for greater awareness and understanding of breastfeeding difficulties so that breastfeeding women are less likely to interpret these as a personal shortcoming in a manner which disempowers them. We also advocate the need to address proximal and distal influences around the breastfeeding dyad and, in particular, to consider the broader cultural context in the UK where breastfeeding is routinely promoted yet often constructed as a shameful act if performed in the public arena.

Introduction

Infant‐feeding decisions, and women's experiences in relation to these, take place within a context where breastfeeding is almost universally championed as the optimal form of infant nutrition. This is because breast milk yields considerable immunological, nutritional and developmental advantages to babies (Horta et al. 2007). Additionally, there is compelling evidence that breastfeeding conveys a number of benefits to maternal health (Blincoe 2005) and often considerable practical, environmental and economic advantages (Palmer 2009). As such, many countries have established targets for increased rates of breastfeeding initiation, maintenance and exclusivity (e.g. Department of Health 2003), and a series of health promotion campaigns have been developed, arguably creating a ‘moral imperative’ for mothers to at least attempt breastfeeding in the early post‐partum period (Dykes 2005; Crossley 2009).

In the UK, patterns of infant feeding are monitored regularly, most notably through the quinquennial Infant Feeding Survey (Bolling et al. 2007). Findings from the most recently published review show that while many mothers initiate breastfeeding, approximately 7% are exclusively breastfeeding at 6 months. Furthermore, only 34% of babies are receiving any breast milk at 4 months of age. These figures are well below the target set by public health agencies in the UK and internationally of exclusive breastfeeding for at least 6 months (Kramer & Kakuma 2002). However, the Infant Feeding Survey (Bolling et al. 2007) also found that 73% of mothers reported that they would have liked to have breastfed for longer.

This discrepancy between women's engagement with the practice of breastfeeding and their expressed intentions and desires to breastfeed has been frequently noted (e.g. Hoddinott & Pill 1999; Larsen et al. 2008). One reason for this is that difficulties with breastfeeding, especially among first‐time mothers, appear to be widespread in western societies (e.g. Binns & Scott 2002). Berridge et al. (2005) found that, among other problems, many women attending their drop‐in clinic reported problems latching the infant on effectively, painful and/or traumatized nipples and concerns over the adequacy of milk production or consumption. Given the links often made in contemporary discourse between breastfeeding and ‘good’ mothering (Wall 2001; Marshall et al. 2007), it would seem important to explore the experience and perspectives of such mothers further. This would not only improve our knowledge of the ways in which women manage such challenges and the barriers they face in doing so but would also enable exploration of the impact of breastfeeding difficulties on women's developing sense of themselves as mothers.

Although researchers have previously tended to prioritize evidencing the benefits of breastfeeding over explorations of women's feeding experiences (Spencer 2008), a number of studies have appeared that examine breastfeeding from the mother's perspective. In several of these, the authors have tried to provide a general account of breastfeeding rather than focusing exclusively on problems that may arise (e.g. Hauck & Irurita 2002; Dykes 2006a; Marshall et al. 2007). However, the difficulties experienced by a sizeable number of participants in these studies have consistently been evident. Nelson (2006) performed a meta‐synthesis of 15 qualitative research studies and concluded that breastfeeding is an ‘engrossing’ but challenging personal journey, requiring maternal self‐sacrifice, determination, resilience and adaptability. While, Nelson argues, the role of support from family members, health professionals and voluntary agencies can be important, breastfeeding primarily represents a ‘lonely responsibility’ for the mother. Some studies included within Nelson's review demonstrated that experiences of breastfeeding were highly heterogeneous. For example, Schmied & Barclay (1999) found that some women perceived breastfeeding as a rewarding and intimate sensual connection with their babies, while others reported a powerful sense of estrangement from both self and infant. For the latter group, breastfeeding was viewed as a distortion of ‘natural’ functioning. They spoke of feeling ‘invaded’ by their child and described breastfeeding as unpleasant and distressing. Other researchers have also reported that some participants found breastfeeding painful (Berridge et al. 2005), experienced concerns about insufficient milk (Marshall et al. 2007) or described anxiety and embarrassment relating to breastfeeding in public (Scott & Mostyn 2003).

Some researchers have focused specifically on women who have found breastfeeding less straightforward (see Dykes & Williams 1999; Mozingo et al. 2000; Hauck et al. 2002). A general finding from these studies has been that women are often surprised by the extent of their difficulties in breastfeeding and find the situation of struggling or ‘failing’ to breastfeed upsetting, anxiety provoking and damaging to their sense of self‐worth.

These and other studies are contributing to a better understanding of the ways in which breastfeeding can breakdown and the kinds of support that might be useful to breastfeeding women. However, our understanding of the impact of breastfeeding difficulties on mothers and on their developing relationship with their child remains limited. This is in part because the focus of much of the breastfeeding literature has, understandably, been on trying to identify barriers to breastfeeding in the hope that this might inform interventions that help women to manage the challenges they might face, thereby increasing breastfeeding rates. That difficulties breastfeeding might also have a significant impact on women and their identities as mothers is suggested by the way in which successful breastfeeding has become inexorably linked to ‘good’ mothering because of the dominance of the ‘breast is best’ discourse within ideologies of motherhood (Murphy 1999; Marshall et al. 2007). A closer examination of discourse around breastfeeding also highlights the complex and potentially contradictory nature of much of this, suggesting that the experience of breastfeeding is likely to be emotionally complex, whether or not lactation is straightforward. For example, breastfeeding is simultaneously constructed both as a quintessentially natural process and as one requiring close ‘expert’ supervision from lactation specialists (Locke 2009), and as both a mechanical process (Dykes 2005) and an intimate mutuality of mother and child (McCarter‐Spaulding 2008). This may mean that women struggling to breastfeed experience their bodies, selves and difficulties in contradictory ways, and suggests that meaning‐making in respect to problematic experiences of breastfeeding warrants further exploration.

One final point to make about the literature on breastfeeding experiences to date is that most research has been based on retrospective accounts obtained via interviews. New mothers often find the first few days of breastfeeding the most challenging, when they are learning how to attach the infant to the breast and may be struggling to achieve this with engorged breasts and sore nipples (Renfrew et al. 2000; Lawrence & Lawrence 2011). Although not all authors report the exact timing of data collection, it appears that, with a few exceptions, (e.g. Schmied & Barclay 1999; Dykes 2005) participants have not usually been interviewed about their early difficulties in breastfeeding until at least several weeks, after they occurred.

This study therefore aims to extend our understanding of the potential impact on women of finding breastfeeding challenging by focusing on a group who experienced considerable difficulties in the early post‐partum period. Our study aims to build on previous research in a number of ways, most notably through the inclusion of daily audio‐diaries alongside standard interviews. This enables experiences to be captured as they occur, and wholly in women's own words, providing a unique and dynamic commentary on the process of engaging with breastfeeding challenges and the day‐to‐day impact of these. We also aim to consider the experience of breastfeeding difficulties in the context of cultural discourses around breastfeeding and motherhood within the UK. The data analysed in this paper are drawn from the first phase of a short‐term longitudinal study in which first‐time mothers recounted their breastfeeding experiences during the first month post‐partum.

Key messages

• • Breastfeeding is frequently challenging for first‐time mothers who may have difficulties in interpreting and managing unpleasant and painful sensations relating to feeding.
• • Approaches to breastfeeding promotion and support that construct breastfeeding as ‘natural’, straightforward and a non‐negotiable sign of ‘good mothering’ can serve to isolate and disempower women struggling with breastfeeding. Such messages may have a deleterious impact on these women's psychological well‐being and emerging maternal identities.
• • Audio‐diaries can facilitate the collection of data in real time and context in a relatively convenient manner and provide a useful tool for researchers exploring the lived experience of breastfeeding.

Method

Participants

Twenty‐two women completed both a diary and interview, eight of whom form the sample on which the present analysis is based. As interpretative phenomenological analysis (IPA) aims for an in‐depth exploration of experience and should be performed intensively on a small and relatively homogenous sample of participants (Eatough & Smith 2008); we purposely limited the analysis to the accounts of the first eight women in the study who reported experiencing significant difficulties with feeding in the first week post‐partum. To protect their anonymity, we asked all participants to select pseudonyms that have been used throughout.

The participants were first‐time mothers with singleton infants born at a gestational age of between 38 and 42 weeks without significant child or maternal illness. All eight were white, aged between 25 and 36 years of age, and either married or cohabiting with long‐term male partners who were the father of the infant. All spoke English as a first language. Two of the participants (Erica and Queenie) had experienced home births, and two had delivered their babies via Caesarean section (Arabella and Robin). The remaining deliveries were vaginal births in hospital. In three cases (Caitlin, Gina and Uma), ventouse vacuum extraction was used in the delivery. To be eligible to participate in the study, all participants had to have declared an intention to breastfeed their infant for at least 1 month.

Wherever possible, we informed potential participants about the research and asked them to register an initial interest several weeks before the birth of their infant. This was achieved through advertising the study in general practitioner surgeries and at antenatal classes and clinics. Participants were then recruited on to the study in one of two ways. First, those women who had previously expressed a willingness to participate were contacted by the research assistant shortly after the birth of the infant and invited to formally sign up to the study. Second, women who had not previously been made aware of the study were approached on the ward shortly after the birth of their infant and provided with an information sheet. In these latter instances, participants were given at least 24 hours to consider potential participation and were then contacted again by the research assistant and, where appropriate, formally enrolled.

Methods of data collection

Participants were asked to keep an audio‐diary for a 7‐day period beginning as soon as possible following the birth of their infant, and they subsequently participated in a follow‐up interview typically 3 or 4 days after the last diary entry. Although diary techniques have a long history within research (Alaszewski 2006), audio‐diaries (using portable recording equipment) are not frequently employed in health research. However, they offer a practical ‘hands‐free’ method for participants to provide accounts of experience in real time and context (Bolger et al. 2003). In our study, the use of audio‐diaries meant that once participants had received training in how to use the equipment, data entries could be made whenever convenient and in the home environment. Participants were provided with (but not required to use) some general prompts to help them to construct accounts. Examples included ‘How is the feed going?’ and ‘How do you feel about this feed?’ In all cases the research assistant or the first author listened to the participant's diary data before the interview was carried out. This facilitated the inclusion of additional questions within the interview where appropriate. Through the use of a semi‐structured interview schedule, we explored participants' motives for breastfeeding, both positive and negative experiences of breastfeeding and the impact of these, relationships with others in relation to breastfeeding and future feeding intentions. Thus, the diaries and interviews represent a synergistic form of methodological triangulation. The diaries allowed for accounts of experiences to be captured as they unfolded, while the subsequent interviews allowed for further and more holistic exploration of important issues and later reflection on the meanings and implications of events.

Procedure

Participants were visited by our research assistant as soon after the birth of the baby as practicable and ethically appropriate (typically immediately following discharge from hospital) and were asked to make twice daily diary entries for a 7‐day period using simple voice‐recording equipment. In all cases, this comprised a portable, battery‐operated mini‐disk recorder with external microphone. The women were asked to make entries every day either during or shortly after two (or more) feeding sessions. At the end of this period, participants were visited by the research assistant who collected the recordings. Once these had been reviewed on an individual basis, the research assistant returned to the participant's home to carry out an interview.

Approval for the research was given by both a university ethics committee and the appropriate National Health Service Regional Ethics Committee, and the study was undertaken during 2006 and 2007.

Analytic approach

Data were transcribed in full and analysed using IPA (Smith et al. 2009). It has been argued that hermeneutic phenomenological approaches are particularly well suited to women's descriptions of breastfeeding experiences, especially where interpretations of individual accounts are located within wider socio‐cultural discourses (Spencer 2008). IPA represents a flexible method for analysing phenomenological data drawn from both diary and interview methods (Smith et al. 2009) and has become popular within European health research (Brocki & Wearden 2006). The paradigm involves the application of a ‘double hermeneutic’ process through which the researchers ‘give voice’ to the participants and then seek to make sense of their accounts (Larkin et al. 2006). The analysis therefore represents a co‐construction of meaning‐making between participant and researcher (Smith et al. 2009). In keeping with the position articulated by Larkin et al. (2006), we argue that what distinguishes IPA is a detailed focus on individual experience and meaning‐making, but that the researcher's interpretation of this experience may draw on varied theoretical traditions. One of our theoretical assumptions was that individual experience is mediated through wider cultural discourses (Willig 2000). Therefore, in making sense of the women's experiences as socially and culturally constituted, we paid particular attention to the availability of discursive resources for enabling and limiting their sense‐making.

We read each of the data sets several times before coding began. Each participant was treated idiographically, and ideas were coded and grouped to identify and label a full set of superordinate themes for each individual. We then compared these across participants through the construction of master themes, and appropriate consideration was given to where participants' accounts converged and how they differed (Smith et al. 2009). We discussed the initial set of master themes within the research team, and a second wave of interpretative work was applied at this point to produce the final analysis that considered the women's experiences in the context of prior theory and research, particularly with regard to the wider cultural construction of breastfeeding.

Reflexivity

The research team involved in this project comprised four psychologists from different traditions (two health psychologists, one clinical psychologist and one developmental psychologist) and a research assistant with experience of working in family welfare contexts. Three members of the research team are parents with personal experiences of breastfeeding, some of which were problematic. Thus, our team consisted of a combination of ‘insiders’ and ‘outsiders’ to the behaviour under scrutiny (Langdridge 2007). Our views on issues around breastfeeding advocacy and promotion vary in a number of ways, but we believe that such diversity enriched the ways in which data were scrutinized and interpreted, in particular allowing for an integration of hermeneutics based on empathy and scepticism (Ricoeur 1976).

Results

The analysis presented here is based around two master themes –difficulties with breastfeeding as a threat to maternal identity and interpreting and responding to pain, which were developed through close engagement with the data. Several other themes were developed within the analysis, but these two themes appeared to represent particularly powerful and salient aspects of these women's breastfeeding experiences and appeared consistently within and across the women's accounts. Together, the two themes suggested an overarching experience of tension between the women's lived embodied experience of their struggle to breastfeed and a prior expectation that breastfeeding would be relatively straightforward by virtue of being ‘natural’. We present and discuss the overarching theme of a tension between experience and prior expectation first, followed by each of the master themes.

Breastfeeding as ‘natural’ vs. the lived embodied struggle to feed

Although other women within the larger sample of participants reported finding breastfeeding relatively unproblematic, all eight women discussed here experienced the early days of breastfeeding as difficult. For several of them, it was a daily, painful struggle to synchronize two bodies in the act of breastfeeding – with sore nipples, difficulties finding a feeding position that worked and a baby either disinterested in feeding or distressed and seemingly unable to feed:

‘cos you just feel like you're gonna crack, you know, because it's so difficult. Every feed is such a battleground to try and get it to work (Gina, interview)

try and get him in the right position, try and get his arms out the way . . . , you're trying to hold and support his head which wobbles, and getting him to open his mouth wide, and it's just so much to do. I know it sounds pathetic, it must be, it should be the most natural thing in the world, but . . . so difficult isn't it, baby boy? (Queenie, diary, day 1)

Like Queenie, several women commented on the way in which this daily struggle was at odds with their expectation that breastfeeding would be ‘natural’:

‘cause we'd been to these classes, read about, heard about how natural it is, you know . . . seeing all these happy mothers breastfeeding on videos and things. (Gina, interview)

You just assume, as a woman, you can do this, and you can't. And that's a very sobering experience. So, to have to be taught that, and to get through the whole pain barrier of doing it . . . people don't tell you these – I suppose they don't want to put you off . . . but it's really, you know, it's hard going. (Uma, interview)

That breastfeeding is a struggle for a significant number of women, particularly in the first few days, is evident in other studies (e.g. Kelleher 2006; Palmer et al. 2010). Moreover, other researchers (e.g. Hauck et al. 2002; Marshall et al. 2007) have also noted the discrepancy that many women experience between the difficult reality of breastfeeding and cultural images of breastfeeding as ‘natural’ and hence straightforward, which are often emphasized in educational and promotional literature for new mothers. In her discourse analysis of Canadian literature on breastfeeding, Wall (2001) noted the dominance of themes related to nature and the natural. As she points out, cultural constructions of nature as ‘pure, unsullied, ancient, wise and deserving of respect’ (p. 596) serve to make breastfeeding non‐negotiable, whatever the mother and baby's particular experience. Wall also argues that by drawing on the ideology of natural motherhood and the notion of biologically driven maternal instincts, talk of breastfeeding as ‘natural’ minimizes the possibility that women might experience difficulties breastfeeding, and hence makes it hard to accept and talk about these problems. The following themes illuminate in more detail the particular ways in which it was difficult for our participants to manage the tension between constructions of breastfeeding as a natural and unproblematic process and their day‐to‐day struggle to breastfeed.

Difficulties with breastfeeding as a threat to maternal identity

Several of the women talked explicitly about a sense of pride in being able to nourish their baby with their own body during the more ‘successful’ feeding sessions. Having produced and being able to feed their babies themselves was for them, in part, what defined them as mothers. It was not only a natural but an essential part of motherhood.

I'm doing it, it's from me that I've made this baby and I'm feeding it and she's gonna grow and she's gonna get big and she's gonna get strong because of what I'm giving her and that's a big thing. (Uma, diary, day 3)

It is amazing that you can actually feed somebody else from your own body . . . I feel much more self contained and able to look after him than if I was dependent on a bottle. I'd feel like I could be anybody really I wouldn't, might not be his mother, I could be any old Tom, Dick or Harry. (Erica, interview)

However, all eight women found breastfeeding difficult much of the time and some supplemented breast milk with formula feeding, perceiving this as their only option for providing the baby with adequate nutrition. Given their construction of breastfeeding as a natural and essential part of motherhood, this situation had problematic implications for their identity as a mother. Seven of the eight women referred explicitly to their struggle to breastfeed or consequent use of formula as failure or inadequacy. They saw themselves as unable to do something that, according to dominant cultural discourses, mothers should be able to do with their bodies:

I just cried and cried and cried ‘cos it was just such a big disappointment I felt like I had failed really, almost as a woman really, you feel like this is a natural thing, why can I not do this? . . . my baby would die if he was in a country where they didn't have bottles. (Queenie, interview)

And you feel really blamed, if you don't [breastfeed], you feel that society's judging you for not being able to do it, and you're not normal. (Gina, interview)

Because the women felt that they should, as women and mothers, be able to breastfeed, they located the problem within themselves, seeing it as their failure. Even where they identified a number of contextual constraints on the success of breastfeeding, they still had a tendency to feel as if the situation was somehow their fault:

It makes you feel terribly inadequate and have low self esteem when you haven't been told that you might not have milk immediately, that it's painful and that if your baby fidgets like mine fidgets and puts his hands up all the time, you might find it takes 20, 25 minutes to get him properly latched on. (Robin, diary day, 5)

This sense of inadequacy expressed by Robin and several others was related to their perception of problems with bodily functioning, that is, with their milk supply. As such, there was, in some of the accounts, an absence of what has been described elsewhere as the usual Western dualism between mind or self and body (e.g. Gillies et al. 2004). This dualism has previously been noted by other researchers in the talk of breastfeeding mothers (e.g. Dykes 2005), where use of a techno‐medical model of childbirth by mothers has constructed a separation between the self and faulty ‘body‐machine’. However, in contrast to this, several of the women in the present study appeared to blame themselves for what their bodies did not produce. The implication was that the ‘good mother’ was an embodied entity who produced plenty of milk and was able to sustain a child from her maternal body. From their perspective of struggling to breastfeed, they were their inadequately lactating bodies and as such were inadequate as mothers:

It [‘topping up’ with a bottle of formula milk] also made me feel very, um, just like a really crap mother, to be honest . . . I just felt that I couldn't um, produce what she was needing . . . It just made me feel very inadequate. (Caitlin, interview)

The extent to which the women blamed themselves for the breakdown in breastfeeding can also be understood in the context of the drive to increase breastfeeding participation in recent years. Much of the literature on breastfeeding for new mothers reinforces the idea that the vast majority of women can breastfeed, while providing limited information about what might go wrong with breastfeeding, even temporarily (Hauck & Irurita 2003). Kukla (2006) also notes the way in which campaigns to improve breastfeeding rates often start from the position that it is women who need to change, emphasizing individual knowledge and ‘choices’, while paying scant attention to possible contextual barriers to, and constraints on, breastfeeding. As such mothers can be positioned as deviant and blameworthy if they have difficulty with breastfeeding, or choose not to do it.

For the women in our study, seeing themselves as failing to achieve something that was meant to be natural and routine, meant that they felt obliged to persist in trying to overcome obstacles to successful breastfeeding. Many of the diaries included lengthy accounts of the varied ways in which the women were struggling day to day to make breastfeeding work, even if this meant forgoing their own needs. For example:

You want to persevere so you try to carry on and you want to do the right thing and you don't . . . I didn't want to appear . . . oh a failure is a strong word. I didn't really want to appear like a failure and give in to bottles but I had no, I had no choice. . . . As I say I shall give her two ounces of bottle and then stick her on breast and . . . grin and bear the pain for a while and hopefully things shall improve tomorrow. (Uma, diary, day 1)

The acute distress experienced by some of the women in the study when breastfeeding sessions were breaking down was captured vividly in some of the diary entries, and participants were often tearful while narrating.

It's really, really hard. (.) You just don't know what you're supposed to do. [(Queenie crying)] Come on little baby, come on little baby, there's a good boy. Upsetting your mummy, you worry her. Please just have a little bit, Zak. (. . .) Come on (. . .) . . . Come on (.) Teasing him on his nose, around his mouth, with my nipple, but he's just not having any of it. (. . .) Oh dear. I'm sorry to be crying, it's just hard. (Queenie diary, day 3)

As Queenie indicated, the women's distress was also related to anxiety about being able to provide their baby with sufficient nutrition while being concerned that the formula feeding alternative was not just second best but actually harmful. Caitlin experienced such difficulties with breastfeeding that she switched to feeding her infant exclusively with formula milk during the diary‐keeping period. This impacted very negatively on her at this time.

. . . it was such a hard decision to go onto bottles all the time, cos the way they [voluntary support group] . . . talked about it, it was something that was really, really bad . . . so then it makes you feel as well that you are almost poisoning them . . . so for about 2 or 3 days I was in tears. (Caitlin, interview)

The diaries also showed how precarious maternal identity was on a day‐to‐day basis and the way in which a sense of maternal pride or failure, and hence psychological well‐being, was closely linked to the success or otherwise of breastfeeding from one feeding session to the next. Two entries from Gina's diary illustrate the relationship between emotional well‐being and the success or otherwise of breastfeeding episodes:

I felt really good about the feeding, I'm just finally realising that I'm normal and the baby's normal. (Gina, diary, day 1)

His [Gina's husband's] family found me in floods of tears and a bit concerned I think . . . My baby was really hungry ‘cos I hadn't fed him properly and we resorted to giving him some formula feed . . . and then once they'd gone, um, we were just in crisis point my husband and I. We were both crying and sort of, and couldn't, I don't know, what happened? (Gina, diary, day 4)

Hoddinott & Pill (1999) concluded from their interview study of new mothers that women's confidence as mothers is often intimately related to their babies' feeding behaviour. The diaries in the present study showed the extent to which well‐being could fluctuate on a daily basis (or even moment by moment) in relation to the perceived ‘success’ or ‘failure’ of feeds.

Interpreting and responding to pain

Another threat to these eight participants' well‐being was breastfeeding‐related pain, which all of them experienced, and which was often severe. They used words and phrases such as ‘extremely, excruciatingly painful’ (Arabella), ‘toe‐curling’ (Robin, Caitlin), ‘very painful’ (Erica, Gina), ‘just excruciating’ (Uma), ‘horrific’ (Imogen) and ‘really tender sore breasts’ (Queenie). References were also made to bleeding, cracked and blistered nipples – a phenomenon that has been recorded elsewhere as affecting over a third of mothers during the first week of breastfeeding (e.g. Foxman et al. 2002). Although some of our participants had expected a degree of initial discomfort associated with breastfeeding, all were surprised by the intensity and duration of the pain, as participants in other studies have also been (e.g. Schmied & Barclay 1999; Kelleher 2006). This left several of them confused about the amount of pain they ought to experience and whether or not their discomfort signalled a problem or was to be expected (and thereby tolerated), especially as they were offered varying interpretations of their pain by others (typically midwives or more ‘experienced’ mothers). Thus, representations of initial breastfeeding as being inevitably painful were juxtaposed with notions that breastfeeding should be ‘naturally’ painless if performed correctly.

Well, people generally said that it wouldn't be sore and if it was hurting then you were doing something wrong, but friends of mine who have had babies said just keep with it because the first two weeks it's going to be really difficult and then it will all change so if you persevere and just get through that you would be fine. (Erica, interview)

I guess, because you are always told if they're latched on properly it doesn't hurt, you expect that. Yet when you are in the post‐natal ward you are there trying to latch them on and it's really getting, and they say ‘course it's going to hurt’, you know, so you kind of think yes, it might kind of hurt a little bit to start with but then it will be pain‐free . . . and he'll be filling up. (Arabella, interview).

Although the experience of pain sometimes led to useful advice about changing position and latching the baby on, the possibility that the pain signalled a problem also undermined confidence in breastfeeding. Uma described her experiences in both her diary and interview:

Nobody tells you it hurts or what kind of pain it's gonna be and that it should kind of hurt a bit but that if you're doing it wrong that it should be excruciating. (Uma, diary, day 3)

I had no confidence in it, in the hospital, ‘cos I thought for the first feed or two I thought, oh, you know, this is fine, then it started to get painful, then it starts to get really painful and you think, ‘Oh I'm not doing it right’. Your confidence just hits the floor and you think, ‘Oh God’. (Uma, interview)

It appeared to be primarily the women's prior assumption that breastfeeding was a normative and natural part of motherhood, and therefore should be relatively pain free that made it difficult for them to make sense of the pain. As Kelleher (2006) notes, the downplaying of problems related to breastfeeding in the promotional literature can mean that women do not have the cultural resources to enable them to make sense of difficult breastfeeding experiences. This uncertainty, along with a desire to avoid the sense of failure as a mother that they equated with abandoning breastfeeding meant that most of the women felt the need to endure the pain and continue in spite of it. Although most of the women supplemented breastfeeding with formula, only one woman (Caitlin) changed to exclusive formula feeding and one (Arabella) to exclusive feeding of expressed breast milk during this phase of the study.

Not surprisingly, some of the women indicated that there were times when they dreaded the next feed. Two participants spoke more explicitly about the effect of the pain and other difficulties with breastfeeding on their relationship with the infant. Arabella talked about ‘automatically recoiling’ from her son when he wanted feeding because she associated him with pain.

It absolutely kills still. I'm actually starting to feel nauseous now when I feed him and I'm dreading every feed time now which is every 2 ½ hours which is not a nice feeling because I should be enjoying these moments. I am starting to hate it more and more. (Arabella, diary, day 4)

Gina experienced very sore nipples and had difficulty finding a breastfeeding position that did not hurt.

It's started to affect my relationship with the baby, it's um, all the difficulties . . . it's lovely once he latches on and I am feeding and it's nice and gentle and he's all snugly and everything, but just before that, quite often he's just like, you know, wriggling and red and angry and I'm just so stressed, it's so difficult and, when I was feeling so low in the last couple of days it was difficult to sort of want to pick him up and stuff. (Gina, diary, day 4)

The literature addressing the physicality of breastfeeding has tended to emphasize positive aspects such as the role of skin‐to‐skin contact in bonding and the benefits of the release of hormones such as oxytocin when breastfeeding (e.g. Bartlett 2005). However, as Schmied & Barclay (1999) also noted from their participants, a number of women find breastfeeding physically unpleasant. Such reactions are difficult for women to articulate. A mother's disclosure of negative or ambivalent emotions in relation to her child remains highly taboo in most societies, and it is likely that many women might be reluctant to acknowledge such feelings in a research context (Shelton & Johnson 2006). It is perhaps of relevance that the only one of our participants who mentioned experiencing negative feelings towards the baby at length (Gina) did so in the diary component rather than the interview.

Discussion

The analysis presented in this paper is based on the accounts of a purposively sampled group of first‐time mothers who were finding breastfeeding particularly difficult in this first post‐partum week. It should be noted that while several other participants within the lager sample reported similar problems, we also had accounts from women who reported finding breastfeeding enjoyable and rewarding. Nevertheless, the findings support those of other researchers who have explored breastfeeding difficulties (e.g. Hoddinott & Pill 1999; Hauck et al. 2002; Marshall et al. 2007) and highlight the schism for some new mothers between their lived experience of struggling to breastfeed and dominant cultural discourses which construct breastfeeding as both natural and synonymous with ‘good mothering’. In their meta‐synthesis of seven studies, Larsen et al. (2008) use the metaphor of ‘shattered expectations’. They argue that dominant discourses of ‘natural breastfeeding’ and mechanized views of the maternal body serve to isolate and disempower those women who struggle with breastfeeding. By including a novel method of data collection alongside interviews, our analysis illuminates further just how problematic it can be for women to experience difficulties in breastfeeding within a culture which, at least in part, sustains a romanticized approach to the practice, especially if they perceive their difficulties as a sign of personal deviance or inadequacy.

‘Failing’ at something which is deemed a ‘natural’ part of motherhood can be seen as calling into question one's identity as a mother and as a woman, particularly within the context of ideologies of natural motherhood which locate mothering within biological processes and construct the notion of maternal instincts (Wall 2001). As Taylor & Wallace (in press) have pointed out, a sense of a deficient self because of ‘failing’ to breastfeed can be more problematic than simply feeling guilty about not providing breast milk. They argue that such experiences might be better understood as shame rather than guilt – an emotional experience which has been viewed by emotion theorists (e.g. Tangney & Dearing 2002) as potentially more problematic as it can involve secrecy, inferiority, inhibition and a debilitating focus on a damaged and incapable self. It is noteworthy that recent research on the transition to motherhood has highlighted that difficulties with breastfeeding may be a contributing factor to post‐natal depression, particularly where it is at odds with women's prior positive expectations (Shakespeare et al. 2004; Homewood et al. 2009).

Although the women in our study were able, to some extent, to talk about their sense of ‘failure’, a feeling of inadequacy or shame about one's ‘failure’ to breastfeed may make it less likely that some women would seek help to overcome breastfeeding difficulties and less likely that they would talk about these openly or explore solutions to the challenges they face.

It is highly likely that some of the women in our study could in fact have been assisted further to adapt their feeding techniques so that breastfeeding was less painful and their babies able to feed more easily (see Renfrew et al. 2000). There is an important role to be played by midwives and other breastfeeding support workers who can provide detailed and sustained one‐to‐one tuition and support for mothers beginning to breastfeed. As we have reported elsewhere (Williamson et al. 2007), this is what many of our participants were seeking, although they reported varying degrees of success in obtaining such support. However, while it is important to acknowledge that many women can overcome breastfeeding difficulties, we also need to ensure that ‘failure’ to do so is not inappropriately located with individual women, particularly where maternity services are not resourced to provide optimal support.

A particular irony of our data is that despite our participants' characterization of their difficulties as a deviation from ‘normal’ expected motherhood, practitioners, researchers and policy makers are aware that it is not uncommon for women to experience difficulty in breastfeeding, especially in the early days (e.g. Renfrew et al. 2000; Lawrence & Lawrence 2011). In fact major public health initiatives around breastfeeding such as the World Health Organization (WHO)/United Nations Children's Fund (UNICEF) Baby Friendly Hospital Initiative and its ‘Ten steps to successful breastfeeding’ are premised on the notion that breastfeeding is not necessarily straightforward and requires a particular supportive environment (Vallenas & Savage 1998). Unfortunately our participants were not prepared for the difficulties that they encountered, possibly because breastfeeding promotion campaigns have often reminded mothers that breastfeeding is ‘natural’ and not something they should expect to be difficult (Carter 1995; Kukla 2006), even, as Taylor & Wallace (in press) note, to the point of asserting that all women can breastfeed. Such breastfeeding promotion messages can be understood in the context of concerns about the increased commercialization and medicalization of infant feeding, and hence the erosion of women's confidence in their ability to feed their infants (Renfrew et al. 2000).

It must also be noted that the difficulties described by the women in our study (e.g. pain and discomfort around feeding, difficulties latching the baby to the breast) likely reflect a broader cultural ambivalence around breastfeeding that is often associated with industrialized countries. Like many developed nations, the UK has a history of relatively low rates of breastfeeding initiation and maintenance, and in many communities (especially those with the most acute levels of economic and cultural deprivation) the bottle‐feeding of formula milk remains the accepted norm, and therefore women have limited exposure to effective breastfeeding practices (Hoddinott & Pill 1999; Scott & Mostyn 2003). Thus, although breastfeeding is promoted, romanticized and idealized, competing discourses construct breastfeeding outside the home environment as taboo and as a potentially shameful act (Stewart‐Knox et al. 2003). Until very recently there has been no legal protection for mothers in the UK who wish to breastfeed their infants in public spaces. Breastfeeding has therefore been typically constructed as something which ought to be hidden. This rendering of breastfeeding as a private act further serves to isolate women, most especially those who are experiencing difficulties. Moreover, the discomfort and cultural ambivalence with which the visible performance of breastfeeding is viewed is also likely to serve to weaken potential sources of support such as partners, peers and health professionals.

Therefore, professionals and policymakers who aim to promote breastfeeding face something of a dilemma. On the one hand, asserting that breastfeeding is ‘natural’ offers a means of resisting the notion that breastfeeding is a taboo and marginal activity which should be kept out of sight. Such messages might be expected to empower women to breastfeed. However, as our data and those from other studies demonstrate, presenting a view of breastfeeding as ‘natural’ and ‘unproblematic’ can in fact be profoundly disempowering for women who do not find breastfeeding straightforward. In fact, a number of our participants and several women in other studies (e.g. Hoddinott & Pill 1999; Hauck & Irurita 2003) said that they would have coped better if they had been forewarned of possible difficulties. What is required therefore is a carefully balanced approach to breastfeeding advocacy that explains the benefits of breastfeeding for mothers and infants and asserts breastfeeding as a reproductive right, but which also discusses potential barriers to breastfeeding and openly informs women about solutions to possible problems. This is not an easy task. Such an approach would emphasize dissemination of general information about the ‘mechanics’ of breastfeeding as much as persuasion that ‘breast is best’. In this way, women who do struggle with breastfeeding may not only have easier access to useful advice but also feel less need to resort to the notion that they are in some way abnormal or deficient in order to account for their difficulties.

Some of the breastfeeding promotional and educational literature goes some way towards achieving this. However, the emphasis in much of the literature is still on educating women about the benefits of a somewhat idealized image of breastfeeding and the potential threat to infants where mothers are unable and/or unwilling to breastfeed (Kukla 2006). This focus on women's choices individualizes breastfeeding and obscures the contexts within which women engage with or disengage from breastfeeding. Drawing on the work of Tiedje et al. (2002) and Dykes (2006b) among others, we believe that there is a need to move further away from individualistic and dyadic (mother–infant) understandings and representations of breastfeeding towards a more integrative biopsychosocial approach that acknowledges that embodied experiences of breastfeeding (both positive and negative) occur within a dynamic and particular socio‐cultural context.

Thus, constructive responses to breastfeeding difficulties might require that practitioners and significant others in the mother's life acknowledge and address their own limitations with regard to supporting breastfeeding, so that suboptimal outcomes are located properly within the social and health care systems of which new mothers are a part, rather than within the individual mother's presumed deficiencies (Labbok 2008).

The present study utilized a phenomenological approach in which data were captured contemporaneously through the use of audio‐diaries. Thus, women were able to record their thoughts and emotions while feeding their infants in real time, in the home environment and at a point in their transition to motherhood when other methods of data collection might have been unacceptably burdensome or intrusive. This method facilitated the acquisition of an element of immediacy in the participants' accounts which interview methods alone are unable to access. The benefit of combining the diaries with individually adapted interviews ensured that we were also able to see how the women had subsequently reflected upon these experiences.

A key principle of the epistemological position of IPA is the belief that participants are able to represent salient experiences and the sense that they are making of these through their talk (Eatough & Smith 2008). However, it is important to note that accounts are inherently contextually embedded and likely to vary with different audiences. Breastfeeding is a highly intimate embodied act, elements of which participants might have been unable or unwilling to articulate (Schmied & Barclay 1999).

Nonetheless, we believe that our findings provide compelling evidence of the extent of isolation, self‐doubt and distress that women who have difficulties with early breastfeeding experience and reinforce the need to prioritize addressing these difficulties and those factors which exacerbate them at individual, community and societal levels. Altering established attitudes, discourses and restrictive practices around breastfeeding in order to empower all mothers to breastfeed across social and occupational milieux will require profound socio‐cultural changes at many levels, and over many years. In the meantime, more can certainly be done to help individual women who are experiencing breastfeeding difficulties by enhancing both professional and peer support. As Taylor & Wallace (in press), among others, argue, this might include encouraging women to share actual experiences of breastfeeding, such as those of the participants above, ‘be they good, bad, or even ugly’. In this way, individual women may be better able to begin to contextualize, normalize and engage with their own struggles to breastfeed.

Source of funding

The research was funded by The British Academy, London.

Conflicts of interest

The authors declare that they have no conflicts of interest.

Contributions

All four named authors contributed to all aspects of the research and write‐up of the journal article.