A meta‐ethnographic synthesis of women's experience of breastfeeding

Abstract

Despite considerable evidence and effort, breastfeeding duration rates in resource‐rich countries such as Australia remain below World Health Organization recommendations. The literature on the experience of breastfeeding indicates that women construct and experience breastfeeding differently depending upon their own personal circumstances and the culture within which they live. Breastfeeding has also been described as a deeply personal experience, which can be associated with ‘moral’ decision‐making. The aim of this synthesis was to better understand the social phenomenon of breastfeeding by making the hidden obvious. Using a meta‐ethnographic approach, we analysed the findings from 17 qualitative studies exploring women's experience of breastfeeding. Commonly used metaphors, ideas and phrases across the national and international qualitative studies were identified. Two overarching themes emerged. Breastfeeding was described in terms of ‘expectation’ and ‘reality’, while the emotional aspects of breastfeeding were expressed in ‘connected’ or ‘disconnected’ terms. The prevalence of health professionals and public health discourses in the language women use to describe their experience, and the subsequent impact of this on maternal confidence and self‐assessment of breastfeeding are discussed. This synthesis provides insight into some of the subtle ways health professionals can build maternal confidence and improve the experience of early mothering.

Introduction

Breastfeeding is recognized as the healthiest form of infant feeding providing significant benefits for both mothers and infants [World Health Organization (WHO) & United Nations Children's Fund (UNICEF) 2003]. New mothers are encouraged to exclusively breastfeed their baby for the first 6 months of life and to continue to breastfeed until their infant is at least 2 years of age (National Health and Medical Research Council 2003; WHO & UNICEF 2003). Data from the Longitudinal Study of Australian Children recently revealed that while initiation rates in Australia are high, breastfeeding rates fall well below the WHO recommendations. Within a national cohort of 5000 children, only 14% of infants were exclusively breastfed at 6 months of age and approximately half were still receiving some breast milk at this age. A steady decline in breastfeeding was evident until at 12 months and 2 years, only 28% and 5% of infants, respectively, were receiving any breast milk (Australian Institute of Family Studies 2008, p. 15). Theses figures are consistent with previously documented national breastfeeding rates (Australian Bureau of Statistics 2003). International figures reveal that in other Western societies such as the United States of America (USA), Canada and the United Kingdom (UK), breastfeeding rates are similarly below the WHO recommendations (Clements et al. 1997; Callen & Pinelli 2004; Li et al. 2005; Bartington et al. 2006; Fein et al. 2008).

The variations in breastfeeding rates (Griffiths et al. 2005; Bartington et al. 2006; Kelly et al. 2006; Hawkins et al. 2008) attest to the fact that breastfeeding is more than the simple transfer of nutrients from mother to child; it is, we argue, a socially constructed practice (Blum 1993, p. 296; Hausman 2003). This assumption is supported by the extensive literature on breastfeeding that suggests it is a deeply personal experience embedded within a woman's specific social and cultural circumstances (Maclean 1990; Schmied & Lupton 2001; Bartlett 2002; Nelson 2006; Marshall et al. 2007; Spencer 2007; Crossley 2009). Decision‐making about infant feeding is thus a complex multifactorial process (Sheehan 2006) that is influenced by such things as family and social circumstances, the health of the baby and other dependants, return to work plans, the woman's own health needs, her previous experiences and the culture in which she lives (Papinczak & Turner 2000; Kirkland & Fein 2003; Gatrell 2007; Miller 2007; O'Brien et al. 2009).

It is also well recognized that the early postnatal period, in which the first breastfeeding experience occurs, represents a time of ‘redefining’ the self. Maternal self‐esteem and confidence can be especially vulnerable during this time (Mercer 1995; McVeigh & Smith 2000; Larsen et al. 2008). It is not surprising therefore, that how a woman experiences breastfeeding, at this time, has been found to impact upon her developing sense of herself as a mother (Maclean 1990; Hartrick 1997; Schmied & Lupton 2001; Sheehan 2006; Marshall et al. 2007). Nelson (2006) reports on this in her synthesis of 15 qualitative research papers on maternal breastfeeding experience. The synthesis yielded the universal translation of breastfeeding as an ‘engrossing personal journey’, and four underlying themes outlined the phases of the ‘journey’. Nelson concluded that improvements in clinician ‘sensitivity’ toward ‘the meaning and significance of breastfeeding to maternal self‐esteem’ are necessary (2006, p. 19). This is particularly important in the context of recent qualitative synthesis findings which demonstrate that health‐care services are currently ‘failing the new mother’ (McInnes & Chambers 2008, p. 423). Mc Innes & Chambers synthesized 47 qualitative research papers to ascertain maternal and health professional perceptions of breastfeeding support. The findings suggested a maternal preference for social or volunteer support networks and highlighted the need for ‘emotional’ support alongside ‘practical’ and ‘informational’ (McInnes & Chambers 2008, p. 422). In addition Britton et al. (2007) conducted a systematic review of 34 controlled trials to measure the effect of support interventions on the duration of breastfeeding. A combination of peer and professional support were identified as important factors in the maintenance of exclusive breastfeeding especially within the first 3 months (Britton et al. 2007).

Despite the extensive literature on breastfeeding and ongoing improvements in breastfeeding support, very few resource‐rich countries have breastfeeding rates that meet WHO recommendations. In order for health professionals to contribute to improve breastfeeding longevity, and, to ultimately meet national and international targets for maternal and infant health, further research into the barriers are necessary. This does not mean, however, that qualitative researchers have to continually ‘re‐invent the wheel’ by finding new and different ways to research breastfeeding experience (Sandelowski et al. 1997, p. 366) As demonstrated above, undertaking meta‐synthesis facilitates the gathering of new insights, through comparing and contrasting existing research findings, and has become a well‐established practice in health‐care research (Britten et al. 2002; Campbell et al. 2003; Pound et al. 2005; Nelson 2006; Downe 2008; Larsen et al. 2008). This approach ensures that individual research studies are not rendered ‘little islands of knowledge’ but rather, in combination, are contributing to the expansion of knowledge about the given area of interest (Sandelowski et al. 1997, p. 367).

The purpose of this meta‐ethnographic synthesis was to conduct an interpretive enquiry into the descriptive phrases and metaphors used by women when explaining the experience of breastfeeding. By focusing on the language and discourses women used to describe their experience, the synthesis aimed to make explicit the events and interactions which shape breastfeeding experience, and to highlight ways for health professionals to better engage with breastfeeding women. Nelson (2006) presented her meta‐synthesis findings using quotes from the women in the original studies. We became intrigued as to whether women use similar descriptive language across the national and international studies which focus on the experience of breastfeeding. While Nelson (2006, p. 15) translated the synthesis findings into ‘the embodied reality’, ‘becoming a breastfeeding mother’, ‘a need for support’ and ‘the journey must end’, we wondered if women's own descriptive language might reveal more of the diversity of the experience for mothers. In particular, we were interested in building upon the implications for clinical practice, identified by Nelson (2006, p. 19) by examining in greater depth, the subjective experience of breastfeeding. It was hoped the focus on the language used to describe experience might offer greater insight into individual and group beliefs and practices, (Fairclough 1992, p. 63; Weedon 1997, p. 21) and highlight the dominant discourses impacting on breastfeeding mothers. We have used a Foucauldian understanding of discourses as ways of articulating and developing knowledge about a given topic which can be influenced by power and control (Foucault 1972, p. 22, 216). Women are exposed to various discourses about breastfeeding and infant feeding in general, and issues of power and control are often central to this endeavour. Some discourses seek to control behaviour and can be deemed important, or not, based on an assessment of the source (Foucault 1972; Fairclough 1992; Parker 1992; Lupton & Barclay 1997). In this way, professional discourses may be deemed more powerful than social discourses however, each influence the woman's decision‐making and actions in subtle ways. Subjective understandings and internalization of these various discourses and their powerful effects, will be revealed in the language used by women to describe their experiences. The study of language as discourse has gained increasing popularity in health research in recent times as language is understood to reflect, and simultaneously construct, our experience of the world (Weedon 1997). In this synthesis, we have used an analysis of language to gain insight into what motivates action, what restricts action and what influences thinking (Weedon 1997, p. 40).

Key messages

• • Dominant western sociocultural and health professional discourses can be contributing to the sense of ‘disillusionment’, ‘guilt’ and ‘failure’ which some breastfeeding women describe.
• • Biomedical descriptions of female bodily experiences can suppress women's own embodied descriptions of breastfeeding and have the potential to impact upon the developing mother/infant relationship.
• • Opportunities for women to explore and articulate their own experience of breastfeeding can influence the way health professionals represent breastfeeding during pregnancy and in the postnatal period. Further research into the facilitative and inhibitive components of health professional language and practices is necessary in order to build supportive discourses around breastfeeding.

Method

The synthesis presented in this paper replicated the approach to meta‐ethnography advocated by Noblit & Hare (1988). These authors argue that meta‐ethnographic enquiry is driven by the desire to develop interpretive explanations and understanding from multiple cases of a given study phenomenon by utilizing research which is ‘grounded’ in the experiences of participants (Noblit & Hare 1988, p. 12). Critically examining the literature by systematically undertaking cross‐case comparison can facilitate new insights and has the potential to make the ‘hidden’ obvious (Noblit & Hare 1988, pp. 13, 17–18).

Noblit & Hare (1988, pp. 26–29) outlined seven phases of meta‐ethnographic synthesis. These seven phases provided the framework for the synthesis presented in this paper. Phase one involved identifying the area of interest and formulating the synthesis question (Downe 2008, p. 5); ‘How do women describe breastfeeding and their experience of it?’

Phase two involved determining which research studies should be included in the synthesis. The literature search was conducted in May 2008 using the CINAHL, MEDLINE and PsycINFO databases with the key words: ‘breast feed’, ‘experience’ and ‘qualitative’. In addition, some key frequently cited qualitative research studies, which included women's narratives on breastfeeding, were also obtained using the SCOPUS database. In total, the search yielded 236 possible research papers. The inclusion criteria were set at: peer‐reviewed journal articles reporting research using qualitative methodology, published in English, reflecting participant's experience of breastfeeding and presented with extensive quotes from women throughout the text. The quality criteria applied to the research papers was consistent with the summary score advocated by Downe et al. (2007). The included studies scored a ‘C’ or above which reflected an acceptable level of credibility, transferability, dependability and confirmability (Downe et al. 2007, p. 132) Saturation was apparent after 17 qualitative papers had been synthesized. Sixteen of the included studies utilized interviews or focus groups as the data collection method. One study reported the open‐ended responses to a questionnaire and was included as it provided rich descriptive data for synthesis. A summary of the 17 research papers can be found in Table 1.

Table 1.

Qualitative papers included in synthesis

Number	1	2	3	4	5		6	7	8	9	10	11	12	13	14	15	16	17
Author	Bottorff (1990)	Dykes (2005)	Dykes & Williams (1999)	Hauck et al. (2002)	Hegney et al. (2008)		Kelleher (2006)	Manhire et al. (2007)	Mozingo et al. (2000)	Schmied & Barclay (1999)	Sheehan et al. (2003)	Earle (2000)	Raisler (2000)	Shakespeare et al. (2004)	Hauck & Irurita (2003)	Hoddinott & Pill (1999)	Baker et al. (2005)	Leff et al. (1994)
Country	Canada	England	England	Australia	Australia		USA & Canada	New Zealand	USA	Australia	Australia	England	USA	England	Australia	Wales	England	USA
Participants	Three women, SES not provided	61 women SES ranged from high to low	Ten women, primip, exclusive breastfeeding and motivated to breastfeed, SES ranged from high to low	Ten women having difficulty with breastfeeding, attending breastfeeding centre, high SES	40 women, SES ranged from high to low		33 women, SES ranged from high to low	153 women, SES ranged from high to low	Nine women, SES based on education attainment was ranged from high to low.	25 women, SES ranged from high to low	29 women, SES ranged from high to low	19 women, SES ranged from high to low	42 women, low SES	39 women, SES ranged from high to low	33 women, SES ranged from high to low	21 women, Low SES	24 women, SES ranged from high to low	26 women, SES ranged from low to high
Methodology	Phenomenology based on face‐to‐face interviews	Ethnography based on observation and interviews	Phenomenology based on in‐depth interviews	Phenomenology based on interviews	Retrospective, case–controlled, 20 participants in each group – continuing cohort and non‐continuing cohort, both groups experienced ‘extraordinary difficulties’. Face‐to‐face semi‐structured interviews. Comparative content and thematic analysis.		Retrospective semi‐structured in‐depth interviews, 90% breastfeeding at interview.	Content analysis of open‐ended responses on a semi‐structured questionnaire‐analysed using Van Manen thematic analysis.	Phenomenology based on interviews.	Discourse analysis based on interviews	Grounded theory based on in‐depth semi‐structured interviews	Prospective descriptive design using unstructured interviews,	Ethnographic content analysis based on seven focus groups	Thematic description based on in‐depth interviews with women who had either breastfeeding difficulties, postnatal depression, or both.	Grounded theory based on in‐depth interviews with women, questionnaires from nine partners and a focus group and interview with child health nurses.	Grounded theory based on interviews with mothers and two focus groups with support people. Discourse analysis applied to transcripts.	Retrospective descriptive study based on in‐depth, semi‐structured interviews with women on four topics: (1) birth of first child, (2) birth of subsequent children, (3) experience of motherhood, and (4) woman's menstrual cycle. Feminist analysis.	Phenomenology, based on semi‐structured interviews.
Themes	1. Experience of persistence 2. Deciding to breastfeed 3. When it is not easy 4. Giving 5. Being committed  • The ties of commitment  • The talk of others  • The talk of self 6. Choosing a time to stop	1. Providing 2. Supplying 3. Demanding 4. Controlling	1. Quest to quantify and visualize: ‘I'll try’, compare breast milk and artificial – quality, quantity, weighing 2. Dietary concerns –‘My milk is what I eat’ 3. Breastfeeding as a challenging journey ‘Falling by the Wayside’ 4. Giving out' and the need for support, nurturing and replenishment.	1. Path of determination:  • Searching for answers  • Presence of physical and emotional exhaustion 2. Staying on the path  • Encouragement  • Individualized assessment and advice  • Seeing signs of improvement  • Consequences of staying on the path 3. Coming off the path  • Encumbrances  • Standardized advice  • No signs of improvement  • Consequences of coming off the path  • Overcoming breastfeeding difficulties	Continuing 1. Expectations  • Realistic or idealized  • Disillusionment 2. Support issues  • Partner support  • Trusted health professional  • Peer support 3. Feelings about breastfeeding  • Breast is best  • Bonding  • Automatic decision  • Faith in nature 4. Psychological distress and breaking point 5. Coping strategies  • Goal setting  • Positive self talk  • Determination  • Optimism/ perseverance  • Cue from baby 6. Pride	Non‐continuing 1. Expectations  • Idealized  • Disillusionment 2. Early breastfeeding problems 3. Reluctance to seek help 4. Support issues  • Partner support  • Feelings of isolation  • Family culture  • Social pressure to breastfeed 5. Feelings about breastfeeding Breast is best Bonding Automatic decision Need/desire to feed Supply concerns 6. Psychological distress and breaking point 7. Guilt  • Dissonance and justification	1. No‐one tells you 2. Sore as hell 3. Scared of the pain 4. Really intimidating 5. Packing it in or coping	1. Persistence and determination 2. Confidence 3. Satisfaction 4. Pain and limitation of mothering activities 5. Conflicting advice and professionalism 6. Others expectations	1. Idealized expectations 2. Clash with reality 3. Personal feelings of discomfort 4. Inadequate or inappropriate assistance 5. Incremental disillusionment and cessation of breastfeeding 6. Relief vs. guilt/shame/sense of failure 7. Lingering self‐doubt vs. resolution	1. Breastfeeding as a connected, harmonious and intimate embodiment 2. Breastfeeding: the disrupted, distorted and disconnected experience	1. Assuming I'll breastfeed 2. I'm definitely breastfeeding 3. I'm going to play it by ear 4. I'm going to bottle feed	1. Baby feeding: making decisions 2. ‘Breast is best’: knowledge of breastfeeding 3. Bottle feeding and the role of the father	1. Health system factors  • Prenatal care  • The hospital experience  • Post‐partum encounters  • Influence of the WIC  • Effect of breastfeeding peer counsellor 2. Breastfeeding beyond the Health‐care system  • Getting on with my life  • The physical bond of breastfeeding  • Modesty and physical exposure  • Returning to work or school	1. Commitment to breastfeeding with high expectation of success 2. Unexpected breastfeeding difficulties 3. Seeking professional support for difficulties 4. Finding a way to cope 5. Guilt	1. Expectations of mother 2. Expectations of others 3. Incompatibility consequences of incompatible expectations  • Confusion  • Self‐doubt  • Guilt	1. Exposure to breastfeeding and influence on feeding intention 2. Women's conversation as a reflection of their level of exposure	1. Perception of control 2. Staff attitudes and behaviours 3. Resources 4. Feeding	1. Infant health 2. Infant satisfaction 3. Maternal enjoyment 4. Attainment of desired maternal role. 5. Lifestyle compatibility 6. Core concept – working in harmony.

SES, socioeconomic status; WIC, Women, Infants, and Children.

Phases three to six outlined the analysis process (Noblit & Hare 1988, p. 28). A period of reading and re‐reading the papers commenced the process. The themes from each individual paper were identified, and quotes from women were grouped within each relevant theme. Initial gathering of ideas and concepts occurred across the texts and the relationships between the studies were identified. A process of unravelling the studies to determine similarities and differences was then commenced. This fifth phase involved identifying ‘key metaphors, phrases, ideas, and/or concepts’ which were similar across the studies (Noblit & Hare 1988, p. 28). Noblit and Hare described these as ‘reciprocal translations’. These translations represented more than simply a summary of all the similarities between the studies. Instead, as Noblit and Hare advocate, by finding the relationships between the translations we were able to uncover the links and in the process gain a deeper understanding of the phenomenon being reviewed. Noblit & Hare termed this process, of building inference, the ‘Line of Argument Synthesis’ (Noblit & Hare 1988, pp. 62–4; Thorne et al. 2004, p. 1349). Our ‘line of argument’ therefore, represented an interpretive reading of the identified translations. Phase seven, writing up and presentation of the findings, allowed for further clarification and consolidation of the synthesis.

In their original work, Noblit & Hare (1988) advocated the synthesis of researcher interpretations or what Schutz (see Schutz & Natanson 1990) termed ‘second order’ interpretations. This was said to yield, what Britten et al. (2002, p. 213) have named, ‘third order interpretations’ which represent the integrated interpretive findings of the synthesizer. In some cases, however, researchers using this method have found that the derived themes were too abstract to allow for comparison between studies. In these circumstances, using ‘first order constructs’ or the direct participant quotes/data for comparison, albeit within the context of the original researcher interpretations, was necessary (Pound et al. 2005; Atkins et al. 2008, pp. 11, 23; Garside et al. 2008). Researchers argue that this approach is appropriate when the expression of second order constructs is largely descriptive rather than interpretive which can limit the translations during synthesis (Walsh & Downe 2005; Garside et al. 2008). In addition, there is an expectation that interpretive synthesis will be ‘grounded in the data’ reported in original studies (Dixon‐Woods et al. 2005, p. 46).

To answer the question posed for this meta‐ethnographic synthesis, it was necessary to use the direct quotes/data from women in the included studies. Interpretations by the original author(s) were utilized to ensure that the quotes were examined in context. It was assumed that the quotes the authors used in presenting their findings represented examples of the opinions of other participants and were reflective of the broader interpretive theme. We do acknowledge, however, that original authors' decision‐making regarding which quote was included represented a level of interpretation of the original data. As such, quotes cannot be viewed as reflective of the ‘totality of participant experience’ (Atkins et al. 2008, p. 9; Downe 2008, p. 7).

Findings

The 17 studies included in this meta‐ethnographic synthesis represent the experience of breastfeeding (including decision‐making) from over 500 women in six Western countries. Although the question posed for this synthesis necessitated the inclusion of areas of similarities and differences across the studies, very few differences or ‘refutational translations’ were uncovered. Instead the synthesis generated metaphors, phrases and concepts which were predominantly consistent across all the studies.

This meta‐ethnographic synthesis revealed two main themes. The first, labelled ‘Expectations and reality’, grouped together the phrases and metaphors that described women's expectations of breastfeeding as a natural process which was best for the baby and was aligned with being a good mother. Subsequently, women perceived that it was important to get breastfeeding right. The description of reality however, reflected a sense that breastfeeding was not necessarily easy but was ‘demanding’ and required perseverance. Ceasing breastfeeding was often associated with guilt and failure.

The second overarching theme was labelled ‘Discourses of connection and of disconnected activity’. Women who articulated their enjoyment of breastfeeding referred to the special relationship it afforded them with their infant. Maternal confidence and appropriate support were key factors expressed by women who positively articulated the pleasures of breastfeeding. In contrast, women who experienced difficulties and described negative experiences with breastfeeding expressed a lack of confidence in their body and their baby. The level of appropriate support was also an area of critical reflection.

The themes from this analysis will be presented with the words women used to describe their experience. All participant quotes will be acknowledged within quotation marks. For readers' ease, the studies will be referred to according to the number allocated to them in Table 1.

Expectations and reality

Breastfeeding is the ‘best’

The single most influential motivator women gave for choosing to breastfeed was their desire to give their baby the ‘best’ (1, 5, 8–11, 13–15). Women consistently acknowledged the benefits of breastfeeding for the baby (1, 2, 10, 14) including the nutritional or nourishing aspects (3, 7, 10) and the provision of immunity (2, 10, 11, 13). The dominant public health discourse, ‘breast is best’, had not been lost on those women who chose to bottle feed either. Research reporting interviews with women who either intended to bottle feed, or who were currently bottle feeding, identified their belief in the health benefits of breastfeeding for babies and mothers (10, 11). There were, however, some who questioned whether breastfeeding was always best for baby. In the most part, this was based on a belief that discomfort with breastfeeding can negatively impact upon the infant and that scientific advances have made infant formula an acceptable alternative (10, 11).

Breastfeeding is ‘natural’

The majority of women identified breastfeeding as ‘natural’ (2, 3, 8–11, 17) and important for ‘bonding’ (2, 3, 5, 8–11, 15, 17). A link between the expectation that breastfeeding would be easy because it is natural was apparent (1, 7, 8, 13). Many women expressed complete surprise at the realization that breastfeeding may not be problem‐free (1, 2, 4–6, 8, 13). While the analysis suggests that for some, ‘natural’ equated to ‘automatic’ (8 p. 123), and therefore easy, there were women who identified breastfeeding as a learned skill (1, 4, 5, 8). Reading books, leaflets and attending classes were mentioned as strategies to enhance breastfeeding readiness (8, 12). These actions were, however, also acknowledged by one participant in Mozingo et al. (2000, p. 123) study as not protecting against having ‘trouble breastfeeding’ (8).

‘Good’ mothers breastfeed

The notion or belief that breastfeeding was inextricably equated with being a ‘mother’, and more importantly to being a ‘good’ mother (1, 4, 8, 10–16), stood out in many of the women's narrative. This was not only a powerful motivator in the decision to breastfeed but also to persevere with breastfeeding when experiencing difficulties (1, 8, 10, 12–14, 16). For example one woman stated, ‘What good mother wouldn't want what's best for her baby?’ (8 p. 122). Similarly, but from the reverse perspective, the data and interpretations presented by Sheehan et al. (2003) attested that women were also conscious that their decision not to breastfeed may prompt health professionals to view them as, ‘the worst mother in the world’ for not breastfeeding (10 p. 263). Participants articulated that professional discourses imposed considerable ‘pressure’ on them to breastfeed (10). As a result, women used words and phrases such as ‘bad’, ‘awful’, ‘horrible’ or ‘irresponsible mother’ if not breastfeeding (4, 11, 14, 16). In contrast, some women who ceased breastfeeding cited reasons related to being ‘more of a mother’ (8 p. 124) as motivating their desire to cease breastfeeding because they felt (or others suggested) their experience of pain and discomfort was inhibiting their ability to mother effectively (6, 8, 11).

It is important to get breastfeeding ‘right’

The expectation and/or desire to get breastfeeding ‘right’ was a dominant theme (4, 6, 7, 8, 11, 13). Women emphasized the importance of having a heath‐care professional check their attachment to see if they were ‘doing it right’ (3, 4, 6, 7, 13, 14) and expressed dissatisfaction if staff didn't do this often (3). The desire to do it ‘right’ was related not only to women's self esteem as a mother but also to the more tangible physical aspect of preventing nipple pain. Breast and nipple pain were repeatedly mentioned as negative aspects of breastfeeding (1, 2, 4–7, 9, 13, 15). One participant in the Kelleher (2006) study questioned the notion that the ‘right latch’ would protect against nipple pain stating, ‘right latch or not it is going to be sore’ (6 p. 2731). Nipple pain for a number of women was unexpected, and the descriptions of pain experienced ranged from ‘uncomfortable’ to ‘excruciating’ (4, 5, 6, 7, 8).

Breastfeeding is not as ‘easy’ as it looks

While there were women who expected and/or predicted that breastfeeding may not ‘be easy’ (4, 10), there were many accounts demonstrating women's overwhelming surprise at the ‘scale’ of difficulties experienced (1, 2, 4–8, 13). The sense of having been disappointed by the reality of breastfeeding was a persistent theme in women's stories. Women described a sense of ‘silence’ around what the experience would be like; ‘no‐one really tells what the body will feel like’ (6 p. 2730). Many women reported reaching a breaking point from which some recovered and continued to breastfeed whereas, others saw the challenges as insurmountable and ceased (4, 5, 6, 8). In several studies, breastfeeding was described as an ‘awful’ or ‘horrible’ experience (6, 8, 9). In many instances, women described becoming ‘fearful’ and ‘dreading’ breastfeeding as a result of problems such as nipple pain (5, 6, 8, 13, 16, 17).

Breastfeeding is ‘demanding’

In a number of papers, breastfeeding was depicted as ‘demanding’ or ‘wearing’ (2, 3, 9, 13). A participant in the study by Shakespeare et al. (2004) stated ‘I didn't find breastfeeding very easy . . . from time to time you give a bottle I mean it is, it's so liberating, because you are free of this thing . . . your friend or your partner can take the baby and give it the milk . . . It's just so demanding of you, you know’ (13 p. 258). The concept of a baby ‘demanding’ a feed was described as an ‘inconvenience’ by some. For example, one woman in the Baker et al. (2005) study said, ‘he wanted feeding no matter where you were and to me I didn't feel like breastfeeding’ (16 p. 330). In Bottorff's (1990) study, one woman described the baby as an abuser; for this mother, the experience was in direct contrast to the ‘beautiful pictures of a mother breastfeeding’ which she had imagined or seen (1 p. 205). Similarly, a participant in the more recent work of Kelleher (2006), described the breast pain she had experienced as causing her breasts to ‘tak(e) a beating’ (6 p. 2731). For a small number of women in the Schmied & Barclay (1999) research, descriptions of breastfeeding included language such as a ‘battleground’, a ‘fight’ and ‘violent’ (9).

Breastfeeding requires ‘perseverance’

Persistence (4, 7) and perseverance (1, 2, 7, 13, 15) were recurrent themes throughout the literature. Commonly, women reported that in the first few weeks ‘it's so easy to give up’ (1 p. 202). Continuing to breastfeed was perceived by women to require strong levels of commitment and determination (1, 4, 5, 8, 10, 14). A participant in Hegney et al. (2008) anticipated difficulties with establishing breastfeeding and described it as a ‘real battle’ requiring commitment (5 p. 1185). A number of women were motivated to persevere due to a determination to achieve their preset breastfeeding targets (4, 5, 8, 13, 14). To do this, women used strategies such as, ‘express(ing)’ (5 p. 1188) to get through an uncomfortable period, and long term goal setting such as breastfeeding until baby is ‘12 months’ old (14 p. 71). Perseverance with breastfeeding despite difficulties was also influenced by a faith in the temporal improvements which accompany any newly learned skill (1, 3, 4, 5, 7).

I felt ‘guilty’ for stopping breastfeeding

Women described feeling ‘pressured’ to breastfeed (10, 11, 13, 16); or ‘forced’ to attach their baby to their breast (11, 16) and predicted feelings of guilt if breastfeeding didn't ‘work’ (10 p. 262). Guilt associated with depriving the baby of ‘the best’ and resultant feelings of selfishness were persistent themes around breastfeeding cessation (1, 6, 8, 14). Feelings of having given up too easily or not tried ‘hard’ enough were apparent (4, 6, 8, 14). Concern that ceasing breastfeeding may have harmed the baby in some way, further exacerbated feelings of guilt (8). Shame and/or embarrassment associated with not breastfeeding was also apparent as women sought to hide the fact that they were bottle feeding, or, intended to bottle feed (8, 10, 11). Guilt associated with feelings of having let the baby, or others, down was evident (5, 13, 14, 16). Some women experienced a sense of ongoing guilt. For example, ‘I felt very, very guilty about not being able to breastfeed . . . every time R*** has a patch of eczema I attribute it to the fact that I didn't breastfeed her’ (13 p. 258).

I felt a ‘failure’

Self‐blame (2, 4, 7, 8, 13) and the use of the word ‘failure’ (5, 7, 8, 13, 14, 16) to describe unmet breastfeeding goals were consistent findings. The sadness and confusion women felt if breastfeeding had not proceeded as they anticipated was evidenced by quotes such as ‘Mentally I felt really let down; all I wanted to was to breastfeed and failed’ (7 p. 376) and ‘I just burst into tears. I felt like a real failure’ (5 p. 1186). A participant from Shakespeare et al. (2004) describes the intensity of the emotion around breastfeeding difficulties, ‘So I remember . . . sitting on the sofa . . . not having success and crying because I couldn't do it and thinking “Oh God, I'm failing and I can't do this” ’ (13 p. 256). The impact of these feelings on the developing relationship with the baby have also been highlighted (6, 13). In addition, some women described a sense of ‘rejection’ if their infant did not breastfeed as they had anticipated (8, 17).

Discourses of connection

‘I Love it’

For many women, the positive breastfeeding relationship experienced with their infant was expressed in terms of incredible ‘closeness’ (7, 9, 12, 17) and ‘connectedness’ (9). Words and phrases women used to describe what they loved about breastfeeding included ‘intimate’ (2), ‘bonding’ (7, 11) and ‘my favourite time’ (9). Many women disclosed their delight at the ‘special time’ (7, 9) breastfeeding afforded them with their infant, and the way it ensured they had ‘time out’ (7) to sit and just enjoy the baby. The feeling of ‘being needed’ (9) gave many women satisfaction as did the ‘privilege’ of sharing their body with their baby (9, 10). When, despite difficulties, breastfeeding had been established, maintained and goals had been met, women expressed an overwhelming sense of achievement, joy and pride (4, 5, 13). Not surprisingly, when women experienced breastfeeding this way they often found it difficult to part with when it came to an end (1, 9).

‘I feel confident’

Antenatally, women could predict breastfeeding as dependent upon having confidence in one's body and oneself; ‘. . . if you are confident then you'll be able to breastfeed’ (10 p. 262). Postnatally, women reported feelings of confidence (2, 3, 17) or a desire to be confident (2) with breastfeeding. Women who expressed confidence with breastfeeding, identified various cues from the baby such as ‘settling between feeds’ (2 p. 2287), a baby who was ‘. . . happy and contented’ (3 p. 237) and had plenty of ‘wet nappies’ (5 p. 1188) as indicative that breastfeeding was progressing well.

Women also identified a faith in their body to provide all that was necessary for their baby's needs, for example, ‘my body tells me what I want (to eat) really’ (3 p. 237) or ‘to me it doesn't make sense that your body wouldn't be able to make enough [milk]’ (5 p. 1187). These cues further enhanced the levels of assurance that breastfeeding was going well for themselves and their baby.

Strategies utilized by women to improve their confidence with breastfeeding included positive self‐talk (1, 5), prioritizing breastfeeding (3), goal setting (5), ‘taking . . . cue[s] from the baby’ (2, 5 p. 1188), being flexible (3, 5), faith in improvements with time (4, 5, 7, 10), and support and assistance from family and staff (1, 3, 4, 5, 7).

‘I just felt so supported’

According to Bottorff (1990), women relied on the ‘talk of others’ to persist with breastfeeding as the words from others brought ‘hope and comfort’ (1 pp. 206–7). Women acknowledged the positive impact of the support they received (3–8, 13, 14). Feelings of connectedness with other mothers who were experiencing the same emotions and challenges were apparent in quotes from women (5, 4). The quality of the relationship which developed between a peer‐support worker and a breastfeeding woman was reported as resembling a friendship rather than a worker/client relationship which facilitated improved support (12). Breastfeeding clinics were reported by women as offering opportunities for supportive interactions with health professionals (4, 13). The communal nature of a breastfeeding clinic environment also brought advantages such as meeting other women in the same situation and normalizing the need for extra assistance (13). Women in the Kelleher (2006) study reported a ‘helpline’ service kept them connected to ‘information and guidance’ (6 p. 2734), while others found having a face to face consultation with a health professional extremely beneficial (4). The desire to actually develop a connection with the health professional was apparent in references to the importance of continuity of carer, individualized advice, and a warm and supportive interaction (4, 7, 13).

The importance of having a supportive partner was a recurring theme in women's stories (4, 5, 6, 7, 11). Women sometimes referred to ‘we’ when reflecting on decisions made about infant feeding (7, 8). For some, breastfeeding was viewed as part of the broader relationship between mother, father and baby (4, 5). One participant in Manhire et al. (2007) summed up a consensus opinion about breastfeeding apparent in the literature saying ‘It takes perseverance, dedication and a supportive partner to get through the first weeks’ (7 p. 376). Having a mother who breastfed and was supportive also facilitated sustainable breastfeeding for women (3, 4).

Discourses of disconnected activity

‘My body was out of control’

This synthesis revealed that for some women, breastfeeding was experienced as a negative activity. Examples of women's disconnected references to their own body, and/or to their baby, highlighted the lack of control which a number of women felt accompanied breastfeeding. Antenatally, a few women were ‘repulsed’ (10 p. 263) by the idea of breastfeeding. Women in the Schmied & Barclay (1999) study reported being amazed at their body's ability to produce milk and yet, for some this was followed by a disgust at their lack of control over their body (8, 9, 13). References to ‘messy’ (8 p. 124), leaky, ‘sticky’ (9 p. 331) breast milk were given as evidence that the body was ‘out of control’ (8 p. 124, 9, 13). The desire to return to normal was a recurring theme in women's stories (2, 9, 10, 13) with several women expressing a need to get their ‘body back’ (9 p. 331, 10 p. 263, 17 p. 102). Comparing oneself to a ‘machine’ (9 p. 330) or a ‘cow’ (9 p. 330, 10 p. 262, 17 p. 102) was a feature of some of the research findings from the USA and Australia.

A lack of faith in the capacity of the female body to adequately nourish an infant was a further perceptible theme in this synthesis. The concerns expressed by women about their own ability to produce ‘enough’ milk (1, 2, 3, 4, 6, 8) and milk of sufficient quality (3, 4, 7) were abundant. This lack of trust in the body extended to concerns about enough milk ‘leaving my body’ (2 p. 2287), worry about ability to ‘keep up with him’ (1 p. 204) and babies crying because they're not ‘getting enough’ (2 p. 2287, 8 p. 123). Women questioned whether their diet was good enough to sustain their baby (3) and the notion that a breastfeeding mother has to continually ‘watch what you eat all the time’ (1 p. 206 ), seemed to be overwhelming for some (3). Weighing babies to ascertain how healthy they were further reinforced the fear women had of potentially not providing sufficiently for their infant (3, 4, 7). Similarly when women compared breast milk to artificial formula, it created a sense of uncertainty that breast milk was of equal quantity and quality (2, 3). A participant in the Dykes & Williams (1999) study highlighted this point: ‘My milk looked sort of bluey grey and thin and watery. When you see the baby milk in bottles it's white and frothy and it looks really thick and healthy. It made me think well my own is not much in comparison’ (3 p. 236).

My baby ‘didn't know how to feed’

This analysis revealed that a number of women were keen to assert the reason breastfeeding did not establish was because the baby was not able to do it (8, 12, 16). One mother kept a bottle of breast milk in the fridge as ‘confirmation that . . . it wasn't MY problem’ (8 p. 126). Comments regarding babies' inability to feed or ‘latch on’ were apparent (8, 12, 16). Examples of the tendency to explain difficulties with establishing breastfeeding by referring to the baby include, ‘She didn't know how to feed’ (8 p. 125), and ‘. . . she was always hungry and wasn't putting on weight very well’ (7 p. 377).

The establishment of an enjoyable fulfilling breastfeeding relationship was, for a number of women, dependent upon the baby behaving in a socially acceptable, appropriate or ‘civilised’ manner (2, 5, 7–9, 13). Women reported feelings of being restricted (12) and expressed frustration when babies woke for a feed early (1), fed all the time (4, 12), ‘feed whenever they want to’ (2 p. 2288), were ‘always hungry’ (7 p. 377, 16 p. 103), ‘gorge[d]’ (2 p. 2287 ) themselves when they did feed, cried ‘for nothing really’ (2 p. 2288) or wanted to feed ‘. . . no matter where you were . . .’ (16 p. 330). These behaviours seemed to be considered ‘uncivilised’ rather than normal infant behaviour. A few women linked feelings of disconnectedness to their baby with the suboptimal acquisition of a breastfeeding relationship (6, 9).

Potential support people were ‘very unhelpful’

Women reported that at times, their partner or family member recommended the time to wean (7, 14, 5, 12) or the time to supplement with formula (3) which inevitably influenced their decision‐making, and/or, caused conflict. Breastfeeding in public was described by many women as restrictive and ‘embarrassing’ (10, 11, 12, 15), and avoidance of breastfeeding in public was often for fear of ‘offending’ others (4) or ‘. . . because of other people's embarrassment . . . .’ (15 p. 34). A number of women predicted embarrassment if breastfeeding in front of particular people such as their father‐in‐law (10) or their partners' male friends (15). In general, a lack of community support for breastfeeding in public influenced women's comfort with this and in some instances women chose to bottle feed rather than compromise their modesty (12).

A recurring theme in this synthesis was that health professionals in Western hospitals had no time to spend supporting and educating women about infant feeding (4, 7, 8). Further, some health professionals were reportedly intimidating (6), rude (7), gave technical, inconsistent and conflicting advice (7, 8), had an ‘unhelpful attitude’ (16 p. 331), and were pushing breastfeeding even at the ‘. . . expense of the mothers emotional health’ (7 p. 379, 13, 16). Women reported health professionals touched and grabbed at their breasts without permission (6, 8, 12). With a number of women expressing a preference to be assisted to breastfeed independently (6) and, to have practical support from health professionals to achieve this (8).

Health professionals reportedly advised some women to cease breastfeeding and commence bottle feeding for a variety of reasons (3, 4) or gave babies bottles of formula or ‘sugar water’ without permission from parents (8, 12). At times a lack of respect for individual choice and decision‐making was evident (16) as a participant in the Raisler (2000) study indicated ‘I said no bottles and they would like force bottles on me . . . they gave her a bottle without even waking me up to ask me . . . I had made it specifically clear! . . . I want this baby completely breastfed’ (12 p. 256).

Discussion

This synthesis, which draws on the findings of 17 qualitative studies exploring the experience(s) of breastfeeding for women, has revealed some of the dominant discourses which are potentially impacting on breastfeeding women's experience. Historically, breastfeeding knowledge had been passed down by women from one generation to the next within communities and within families (Dettwyler 1995; Fildes 1995; Ryan & Grace 2001). We argue, like Bartlett, that there has been a ‘cultural shift in authority’ away from women's own shared embodied knowledge towards a ‘biomedical narrative’ (Bartlett 2002, p. 376). In Western societies, breastfeeding knowledge is predominantly delivered by experts who often position mothers as ‘novitiates in need of tuition on how to breastfeed’ (Bartlett 2002, p. 376). Yet recent synthesis findings have demonstrated that professional knowledge and support are not always viewed positively by women and peer‐support is sometimes favoured because of the use of lay language and practical suggestions (Britton et al. 2007; McInnes & Chambers 2008, p. 421). This has prompted further consideration of the extent to which biomedical and public health messages may contribute to the sense of disillusionment and resultant negative experience which some women describe. In this discussion, we will highlight some of the phrases and metaphors women used to describe their experiences. As advocated by Noblit & Hare, we have utilized these ‘reciprocal translations’ to build a ‘line of argument’ synthesis (Noblit & Hare 1988, pp. 74–5) which reveals the subtle (or hidden) influence of public health, biomedical and health professional discourses on maternal experiences of breastfeeding.

Nature discourse

It is evident that women have heard the ‘public health’ message that ‘breast is best’ and all women can and should do it. The desire to give one's baby the ‘best’ clearly motivated the majority of women to choose to breastfeed. However, for many, giving one's baby ‘the best’ unpredictably required the acquisition of new skills, overcoming challenges and the need for support from others. The sense of disillusionment which some women experienced when establishing breastfeeding was apparent. The expectation that breastfeeding would be ‘easy’ has been linked to a discourse of ‘nature’ where breastfeeding is represented as instinctive. Hall and Hauck report that in Australia breastfeeding is still depicted as a ‘ “natural” romanticized, problem‐free experience’ (Hall & Hauck 2007, p. 794). Wall (2001) argues that breastfeeding educational material aimed at pregnant women emphasizes the natural aspects of breastfeeding and implies that breastfeeding is easy. Antenatal preparation tends to focus upon portraying all the positive, natural aspects of breastfeeding in an effort to ensure that as many women as possible commit to breastfeeding (Schneider 2001; Wall 2001). Lavender et al. (2005) speculated that an antenatal education session aimed at informing women of the benefits of breastfeeding actually contributed to the ‘clash with reality’ which women subsequently experienced. Lavender and colleagues concluded that the provision of a single antenatal breastfeeding education session overemphasized the positives of breastfeeding without also providing information on overcoming difficulties and challenging the ‘peer and societal pressure’ for short‐term breastfeeding (Lavender et al. 2005, p. 1052). Research has shown women with realistic expectations during the early post‐partum period breastfeed for longer when compared to women with unrealistic expectations (Whelan & Lupton 1998; Hauck & Irurita 2003; Hegney et al. 2008).

Scientific discourse

Advances in technology have enabled the viewing of the internal mechanism of breastfeeding (Henderson & Scobbie 2006, p. 515; Jacobs et al. 2007) to the extent where the minute detailing of every aspect of ‘correct’ attachment is defined. The discovery that ‘positioning and attachment play[s] a crucial role in the successful ejection and transfer of milk’ (Henderson & Scobbie 2006, p. 515) has meant that much professional discourse and practice is focused on the acquisition of optimum attachment. Some breastfeeding educators and lactation consultants advocate the teaching of the science and ‘physics’ of correct attachment to new mothers (Powers 2008). The use of this type of ‘scientific’ discourse implies that there is one ‘right’ way to breastfeed. Women have accepted these discourses and express a strong desire to get breastfeeding ‘right’ (Hall & Hauck 2007). Midwives inadvertently contribute to the preoccupation with the ‘right’ latch when they have a preference for physically attaching the baby, (Mozingo et al. 2000; Kelleher 2006) and/or repeatedly checking and rechecking, detaching and reattaching the baby in their effort to achieve that picture perfect ‘latch’. In addition, the focus on the science of breastfeeding, results in an obsession with weighing babies to seek confirmation that optimum growth is being achieved (Dykes & Williams 1999; Sachs et al. 2005). Relatively little is written about women's embodied knowledge and women's explanatory language around the experience of breastfeeding; for example, the language used when breastfeeding feels ‘right’ or when it feels ‘wrong’ (Beasley 1991; Ryan & Grace 2001; Spencer 2007).

Success/failure and the good mother discourses

The ‘successful breastfeeding’ discourses in the public health and professional literature (Leff et al. 1994, pp. 99–104; WHO 1998; Dykes & Williams 1999, pp. 232–246; Hauck & Irurita 2003, pp. 62–78) are reflected in the tendency to describe breastfeeding experiences in terms of success or failure (Hawkins et al. 1987; Bottorff 1990, pp. 201–209; Driscoll 1992; Hoddinott & Pill 1999, p. 34; Mozingo et al. 2000, pp. 120–126; Hauck & Irurita 2003, pp. 62–78; Shakespeare et al. 2004, pp. 251–26; Baker et al. 2005, pp. 315–342; Hall & Hauck 2007, p. 791; Manhire et al. 2007, pp. 372–381; Hegney et al. 2008, pp. 1182–1192; Larsen et al. 2008). For example the WHO Baby‐friendly Hospital Initiative statement on breastfeeding includes, ‘The ten steps to successful breastfeeding’ (WHO 1998). Research on breastfeeding ‘success’ has been measured by duration, by women's own self‐assessment, and by infant factors (Leff et al. 1994; Lothian 1995; Hauck & Reinbold 1996; Whelan & Lupton 1998). It can be argued that the use of a language of ‘success’ inadvertently alienates women who have not been deemed successful in this area. The use of this language is additionally problematic when authors unintentionally slip into the use of further alienating terms such as ‘successful breastfeeders’ (Leff et al. 1994, p. 102; McInnes & Chambers 2008) or ‘successful mothers’ and even ‘successful women’, when describing those who have established breastfeeding (Dykes & Williams 1999, pp. 239 & 242; Nelson 2006). If women are construed as ‘unsuccessful’, both as mothers and as women, for not establishing a satisfactory breastfeeding relationship, the discourse of the ‘good mother’ as a ‘breastfeeding mother’ is further reinforced.

If the ‘good’ and ‘successful’ mother breastfeeds in order to give her baby the ‘best’ (Blum 1999; Murphy 1999; Schmied & Lupton 2001), how does the mother who has not established breastfeeding assess herself? The ‘moral baggage’ (Murphy 1999; Shaw 2004) associated with infant feeding, has the potential to result in the feelings of guilt and low self–esteem, and potential for depression reported by women whose breastfeeding experience has not progressed as anticipated (Cooke et al. 2007). While culturally, the ‘good mother’ breastfeeds is contrasted by the fact that ‘good women’ do not expose their breasts in public (Maclean 1990; Henderson et al. 2000; Mahon‐Daly & Andrews 2002). The confusion that these sociocultural mixed messages cause not only restricts the performance of breastfeeding (Hoddinott & Pill 1999; Shaw 2004) for some but for others is a sufficient deterrent for them to avoid breastfeeding all together (Earle 2000; Raisler 2000; Sheehan et al. 2003). The long‐term impact of these discourses necessitates the health professional obligation to challenge the dominant ‘good mother’ parameters.

It has been speculated that feelings of having ‘failed’ impact upon maternal self‐esteem (Papinczak & Turner 2000). Women who exclusively breastfeed have higher ‘self concept’ when compared with those who exclusively bottle feed (Britton & Britton 2008). Women who have been deemed to have ‘failed’ (or consider themselves failures) search for reasons to rationalize why this may have happened. From this synthesis, it is apparent that women attribute blame to either their own bodily malfunction, their infants' inabilities and/or the inadequate support they received. The tendency for a number of mothers to ‘blame the baby’ when breastfeeding did not progress as planned was an interesting finding from this analysis.

Biomedical discourses and ‘demand’ feeding

Biomedical discourses are evident in the language some women used to describe their body. Mechanistic and disconnected descriptions are consistent with the techno medical ‘body‐as‐machine’ metaphor highlighted by Davis‐Floyd (2001). The biomedical body is contained, controlled and stable and is prefaced on a male norm; in this context female bodily processes such as breastfeeding can tend to be pathologized (McDowell & Pringle 1992; Shildrick 1997, pp. 14–15; Davis‐Floyd 2001):S6). The use of disembodied language by women reveals the Cartesian dualist separation of mind and body which biomedicine advocates (Beasley 1991; Spencer 2007). The cultural acceptance of the ‘techno medical’ model of health (Davis‐Floyd 2001; Dykes 2006) is further evident when health professionals, family and community members espouse a lack of faith in the woman's body to produce sufficient milk (Dykes & Williams 1999; Mozingo et al. 2000; Hauck & Irurita 2003). Similarly the tendency to focus research on when things go ‘wrong’ (Mozingo et al. 2000; Hauck et al. 2002; Shakespeare et al. 2004; Kelleher 2006; Hegney et al. 2008), rather than on a more solution‐focused approach, is additional evidence of the influence of techno medicine on health professional practice (Davis‐Floyd 2001, p. S7).

The description of breastfeeding as ‘demanding’ (Dykes & Williams 1999, p. 238; Schmied & Barclay 1999; Shakespeare et al. 2004) may be a reflection of the force of a ‘demand’ feeding discourse. Examples of the experience of ‘demanding’ breastfeeding include the sense of ‘being completely worn by it’ (Shakespeare et al. 2004, p. 255) or describing the baby as being ‘always at me’ (Schmied & Barclay 1999).Dykes (2006) argued the use of the word ‘demand’ originated from the industrialization of Western societies where the focus was on supply/demand and production line output. Dykes identified that the demand‐feeding discourse has resulted in women describing breastfeeding as ‘breaching temporal’ and ‘bodily boundaries’ (Dykes 2006). The etymology of the word includes definitions such as ‘a forceful request’ (Rooney 2004) or ‘to ask for with authority’ (Delbridge 1991) and prompts the questions – firstly, do some women interpret ‘demand feeding’ as involving the use of force? and, secondly, do health professional discourses set women up to view themselves as passively responding to the ‘demands’ of the infant with little or no regard for their own needs or wants? (Shaw 2004; Dykes 2006). MacLean concludes in her book on women's experience of breastfeeding that ‘[W]omen, particularly first‐time mothers, must learn to accept that life with a newborn is demanding’ (MacLean 1990, p. 205). The impact of this type of language may be that the breastfeeding relationship becomes viewed by women as a ‘battle’ between themselves and the ‘demanding’ baby (Bottorff 1990; Schmied & Barclay 1999; Hegney et al. (2008). References to feeling ‘pressured’ (Sheehan et al. 2003) or ‘forced’ (Baker et al. 2005) to breastfeed may be further indications of the implications of this type of discourse.

Historically, research into maternal experience of breastfeeding has charted many similar findings as those revealed in this synthesis (MacLean 1990; Dignam 1995; Ryan & Grace 2001). MacLean (1990) documented women's experiences from 756 interviews and found many similar themes as those revealed in this endeavour, such as what motivates women to breastfeed, and the expectations and realities of breastfeeding. However, this synthesis, which drew on the findings of qualitative studies exploring the experience(s) of breastfeeding since 1990, and across six Western countries, has revealed that despite the extensive qualitative research on women's experience of breastfeeding many women are still reporting negative feelings about their body, their baby, themselves and about the quality of support they have received.

The public health and health professional messages such as breast is ‘best’, and ‘[The ten steps to] successful breastfeeding’, are reflected in the words women use to describe their experience. We would argue the highly technical, institutionalized health‐care environment potentially influences the discourses health professionals draw upon when supporting women who are establishing breastfeeding. The influence of biomedical discourses can also adversely influence how women view their own body and even how they view their baby. The findings from this synthesis are supported by two research papers published after May 2008. A Scandinavian meta‐synthesis of seven qualitative studies on breastfeeding experience identified that a woman's confidence with breastfeeding is influenced both by expectations, professional support and by the discourses of ‘nature’ and ‘body as machine’ (Larsen et al. 2008, p. 657) The authors concluded that breastfeeding should be viewed as a ‘competency to be attained’ rather than something which every woman can do ‘naturally’ (Larsen et al. 2008, p. 660) The second paper utilized women's own descriptions of their breastfeeding experience to reveal the impact of ‘expectations’, ‘infant behaviour’ and ‘support’ on maternal confidence. Women in this study conveyed, through their language, the importance of infant satisfaction and connection, to breastfeeding longevity (Grassley & Nelms 2008).

Ryan & Grace (2001, p. 494) have highlighted that the health professional use of ‘medicoscientific language’ has effectively robbed women of the opportunity to establish their own discourses on the subjective experience of breastfeeding. Health professionals contribute to this ‘suppression of alternative discourses’ (Ryan & Grace 2001, p. 494) by the paucity of scholarly enquiry revealing the embodied reality of breastfeeding (Schmied et al. 2001). Dignam (1995) alluded to a resurgence of interest in valuing embodied descriptions of breastfeeding however, as revealed in this synthesis, the language used by women is so heavily peppered with techno medical descriptors that the intimacy and relational aspects of breastfeeding continue to be largely suppressed (Ryan & Grace 2001; Schmied et al. 2001).This synthesis reinforces the need for health professionals to move away from biomedical breastfeeding discourses towards more holistic language where the mind and body are viewed as ‘inseparably intertwined’ and the embodied reality of breastfeeding is more clearly articulated (Davis‐Floyd 2001; Dykes 2002; Spencer 2007). These findings have prompted the research team to further investigate the influence of the discursive constructions of breastfeeding, and newborn infant behaviour, on women's experience of breastfeeding and their reflexive recount of this.

Limitations

Sandelowski (2006) warns of the limitations of qualitative meta‐syntheses. She encourages researchers to acknowledge that the interpretations presented are at least three times removed from the original data. Similarly, she reminds readers that the synthesis is only one ‘reading’ of the data where several alternative interpretations are likely to be possible.

Additional limitations from this meta‐ethnographic synthesis include the retrospective nature of some of the studies; the lack of uniformity of methodology used and, in some instances, an inadequate description of methodological approach (6, 11, 13): one paper focused on women who were motivated to breastfeed only (3), five papers were focused on those experiencing significant difficulties only (4, 5, 6, 8, 13), two papers included data from women from low socioeconomic groups only (12 & 15) and one paper was not solely about breastfeeding experience (or decision‐making), however, it included valuable data on this (16). The papers predominantly articulated white Caucasian women's experience of breastfeeding. Having said this, however, the meta‐ethnographic synthesis does highlight both the diversity of the experience of breastfeeding for women as well as the similarities in experience across Western societies.

Conclusion

Much has been written about the disillusionment some women feel when faced with breastfeeding challenges. Clearly, being confident and well supported enables women to experience breastfeeding as a joyous, connected relationship. Social and cultural discourses are inevitably implicated in the reflective recount of subjective experience. The findings from this meta‐ethnographic synthesis have highlighted the contribution of sociocultural discourses to the sense of disillusionment, and failure which many women express. It is also evident that the words and language health professionals and other support people use may be contributing to the lack of confidence, and sense of guilt and failure which some new mothers report. Health professional language and practices have the potential to enhance maternal self esteem, to transform societal discourses and to contribute to improvements in the experience of breastfeeding for women. Further research into the nature of health professional discourses during breastfeeding support, and the impact of these upon women, is necessary to gain a deeper of understanding of the micro and macro determinants of breastfeeding experience.

Source of funding

This work has been funded by an Australian Research Council Linkage grant for the project: ‘Establishing Breastfeeding: an analysis of the language and practices used by midwives and lactation consultants when interacting with new mothers’.

Conflicts of interest

No conflicts of interest have been declared.