Pressure and judgement within a dichotomous landscape of infant feeding: a grounded theory study to explore why breastfeeding women do not access peer support provision

Abstract

Lack of support is reported as a key reason for early breastfeeding cessation. While breastfeeding peer support (BPS) is a recommended intervention to increase breastfeeding rates, a number of studies identify that engagement with BPS is problematic. Due to paucity of research in this area, this study explores why breastfeeding women do not access BPS in South‐West England. Utilising a constructionist grounded theory approach, 33 participants (women (n = 13), health professionals (n = 6) and peer supporters (n = 14)) participated in a semi‐structured interview (n = 22) or focus group (n = 11). Analysis involved open coding, constant comparisons and focussed coding.

One core category and three main themes explicating non‐access were identified. The core category concerns women's experiences of pressure and judgement around their feeding decisions within a dichotomous landscape of infant feeding language and support. Theme one, ‘place and space of support’, describes the contrast between perceived pressure to breastfeed and a lack of adequate and appropriate support. Theme two, ‘one way or no way’, outlines the rules‐based approach to breastfeeding adopted by some health professionals and how women avoided BPS due to anticipating a similar approach. Theme three, ‘it must be me’, concerns how lack of embodied insights could lead to ‘breastfeeding failure’ identities. A background of dichotomised language, pressure and moral judgement, combined with the organisation of post‐natal care and the model of breastfeeding adopted by health professionals, may inhibit women's access to BPS. A socio‐cultural model of breastfeeding support providing clear messages regarding the value and purpose of BPS should be adopted.

Introduction

Observational evidence suggests that poorer health outcomes for both mothers and babies are linked to formula feeding when compared with breastfeeding (Ip et al. 2007). The Global Strategy for Infant and Young Child Feeding (World Health Organisation (WHO) 2003) recommends that babies are exclusively breastfed up to the first 6 months of life, with continued breastfeeding up to ‘two years and beyond’. However, no countries who are members of the Organisation for Economic Cooperation and Development meet these recommendations (Organisation for Economic Cooperation and Development, 2009). Increasing exclusive breastfeeding rates in the first 6 months of life to at least 50% is one of the six global nutrition targets for 2025 (WHO/UNICEF 2014). While socio‐economic variation in breastfeeding rates is well reported (McAndrew et al. 2012), a lack of suitable support is identified as a key reason for breastfeeding cessation (Schmied et al. 2010; Hoddinott et al. 2012). In the UK, it is estimated that moderate increases in breastfeeding could lead to a saving in treatment costs of £17 million per annum in relation to four acute diseases in infants: gastrointestinal disease, respiratory disease, otitis media and necrotising enterocolitis (UNICEF 2012a).

Breastfeeding peer support (BPS) is an advocated intervention to increase breastfeeding rates (WHO 2003; WHO/UNICEF 2014; DH 2004; NICE 2005, 2008). Peer support may be defined as:

The provision of emotional, appraisal, and informational assistance by a created social network member who possesses experiential knowledge of a specific behaviour or stressor and similar characteristics as the target population (Dennis 2003, p. 329).

While qualitative research highlights the value of peer support in breastfeeding continuation and maternal well‐being (Thomson et al. 2012), trials of BPS interventions in the UK and other developed countries have been found to be ineffective in increasing breastfeeding rates (Jolly et al. 2012). However, authors such as Hoddinott et al. (2011) and Thomson & Trickey (2013) call attention, among other issues, to the heterogeneous and reductionist trial designs, implementation difficulties including the influence of contextual factors, the lack of underpinning theory concerning possible mechanisms of efficacy and a lack of high‐quality evidence.

In a recent UK national survey, 69% of breastfeeding women were given contact details of voluntary organisations or community groups that support women with breastfeeding (e.g. BPS provision) at discharge from the maternity hospital. However, only approximately a quarter sought support from these sources (McAndrew et al. 2012). Several UK and international trials also note that access to BPS is problematic (Jolly et al. 2012). For example a UK trial conducted by Graffy et al. (2004) reported that 38% of women in their intervention group received no post‐natal BPS. Non‐access to support was also reported in almost half of the intervention group in a trial of BPS for young mothers in the United States undertaken by Di Meglio et al. (2010). These insights thereby add to the difficulties in interpreting existing intervention data as it impossible to determine whether a lack of effect was due to the intervention or a lack of uptake. This therefore poses problems in determining the relative importance of non‐access in relation to the efficacy of BPS interventions.

There is some evidence suggesting that the quality of the peer–professional relationship is important in facilitating access to BPS. Raine (2003) attributed variability in referral rates by health professionals to BPS to an initial lack of acceptance of peer support provision, although it was anticipated that this reluctance would decrease as the intervention became embedded. However, others have noted that some health professionals did not want lay people involved in the care of women (Muirhead et al. 2006), particularly if there were concerns of women feeling pressurised to breastfeed (Thomson et al. 2015a,). Furthermore, while Bronner et al. (2001) considered that good relationships directly facilitated effective peer support via increased access, Kaunonen et al. (2012) suggests that these relationships require ongoing work and investment.

To date, there are no published studies that specifically focus on the reasons why breastfeeding women do not access BPS. Insights into this issue are important as many women stop breastfeeding before they intended, and a lack of support is reported to be a key reason for early cessation. BPS interventions are part of current strategies for increasing breastfeeding rates via increased support, yet there is evidence of non‐engagement within trial data. Exploration of these issues from a professional‐peer‐woman perspective would help to identify the difficulties and barriers faced, as well as insights into how breastfeeding support should be provided. The aim of this study was to elicit the reasons for non‐access among women, health professionals and peer supporters within Cornwall, a specific geographical region in South‐West England. Key factors that might serve to facilitate increased access to BPS, as identified by study participants, are considered in the discussion.

Key messages.

• Reasons for breastfeeding women's non‐access to breastfeeding peer support have not been previously explored.

• Pressure, moral judgement and dichotomous language and practices impacted on breastfeeding women's access to peer support provision.

• Mechanistic constructions of breastfeeding, and the rules‐based approach adopted by some health professionals, led women to polarise themselves as either those who could or could not breastfeed.

• A socio‐cultural model of breastfeeding that provides clear messages regarding the value and purpose of peer support should be adopted.

Methods

Study context

Cornwall ranks 143rd out of 326 local authorities in terms of overall socio‐economic deprivation (Cornwall Council 2015). While breastfeeding initiation in Cornwall in 2012/2013 was 79.8%, some 5.9 points above the English national average (73.9%), continuation rates in Cornwall at 6–8 weeks were 46.7%, 0.5% below the English average of 47.2% (ChiMat 2015). Full UNICEF Baby‐Friendly status has been in place for all hospital, community and children's centre services in Cornwall since 2012 (UNICEF Baby Friendly 2012b). BPS training, developed by County Infant Feeding Co‐ordinators, is delivered by not‐for‐profit social enterprise ‘Real Baby Milk’. BPS services consist of weekly drop‐in groups (n = 33) at children's centres, run and ‘owned’ by peer supporters (n = ~120 at the time of the study). This BPS provision is the only additional breastfeeding support available to women aside from standard maternity and health visiting care or contact with national voluntary help lines. Maternity care at the time of the study consisted of women receiving a phone call from a midwife on day 1 or the day after discharge, when the contact venue for face‐to‐face contact could be discussed and agreed. On days 5 and 10, face‐to‐face contact would take place, usually in a clinic environment. Some areas also had maternity support assistants who could provide home visits for additional feeding support. Health visiting care comprised a pre‐birth visit, one home visit between days 11 and 14, another between weeks 12 and 20 and access to a health visiting team member via the phone or in a clinic environment. The BPS service estimates that around 70% of women who initiate breastfeeding in Cornwall do not access BPS.

Study design

Grounded Theory (GT) was developed in the 1960s as a way of developing theory about social processes. As an emergent method, it is regarded as a useful approach when studying under‐researched areas (Charmaz 2008). GT's positivist roots of assumed researcher objectivity and the ‘discovery’ of social processes are acknowledged by Charmaz (2006, p. 9), who argues that ‘we can use basic grounded theory guidelines with twenty‐first century methodological assumptions and approaches’. Charmaz's interpretivist theoretical perspective informs her constructionist GT methodology. This method is focused on the interactions between the researcher and participants and how theories are constructed from and between these interactions. The resulting theories are offered not as exact versions of the worlds in question but as rather co‐constructed interpretations (Charmaz 2006). Charmaz emphasises the importance of staying grounded in data when creating interpretations and suggests broad guidelines, rather than strict rules and procedures.

Reflexivity was central to this study. The first author has a nursing background, has breastfed three children and has been involved with BPS provision in the study area, both in a paid and voluntary capacity, for 10 years. The likely impact of personal experience was recognised, and participants were recruited from areas where she was unknown. The second author is from a psychology background, has personal experience of breastfeeding and has undertaken a number of research/evaluation studies into BPS provision. A reflective interview designed to identify prior values and assumptions was undertaken with the second author before data collection began. A reflexive journal was also kept throughout, and ideas were shared and discussed between the authors.

Ethics

Full ethical permission was obtained via the National Research Ethics Service system of proportional review (REC reference 13/LO/0775), with subsequent permission gained from the Cornwall NHS Research and Development Department (reference 2013.CFT.08), Cornwall Council ethics committee and the Built Environment, Sport and Health (BuSH) ethics subcommittee at the University of Central Lancashire (reference BuSH180).

Recruitment and data collection

Purposive and theoretical sampling methods were used to recruit three groups of participants: mothers who had initiated breastfeeding and continued for 5 days or longer and had not accessed BPS (n = 13); health professionals who signpost mothers to BPS (n = 6); and peer supporters who provide BPS (n = 14). Mothers were given study information sheets and reply slips and recruited either by health visitors (n = 4) or at children's centre baby groups (n = 9). Peer supporters and health professionals were recruited via covering letters and information sheets sent to their work addresses.

In order to engage with broad theoretical insights from the beginning, initial sampling aimed to recruit participants with a wide range of backgrounds, ages and breastfeeding histories (Table 1). Interview schedules were adhered throughout. However, during later interviews, particular areas of theoretical interest formed a focus. For example, health professional interactions were explored in greater depth with participants recruited later in the study, in order to illuminate theoretical ideas about the manner by which health professionals discuss breastfeeding. Rather than seeking population representativeness, the sampling strategy aimed to exhaust theoretical ideas associated with non‐access. Overall, 33 participants took part in either a face‐to‐face (n = 16) or telephone (n = 6) interview or a focus group discussion (n = 11).

Table 1.

Socio‐demographic profile and interview type for women participants

Women	n = 13
Age, years (n)	• 20–24 (3)
	• 25–29 (4)
	• 30–34 (3)
	• 35–39 (2)
	• 40+ (1)
	Mean: 29.5 years. SD: 6.839 years.
Range of educational level (n)	• No formal qualifications (1)
	• GCSE/NVQ level 2 (2)
	• A levels/NVQ level 3 (3)
	• Diploma level (2)
	• Degree level (5)
Parity range (n)	• 1 child (8)
	• 2 children (4)
	• 5 children (1)
Age of baby at time of interview, months (n)	• 0–2 (3)
	• Over 2–6 (5)
	• Over 6–12 (4)
	• Over 12 (1)
LSOA of postcode range, % (n)	• Most deprived 0–10 (1)
	• Most deprived 20–30 (1)
	• Most deprived 30–40 (5)
	• Most deprived 40–50 (5)
	• Least deprived 30–40 (1)
Infant feeding method at time of interview (n)	• Exclusive breastfeeding (4)
	• Exclusive formula feeding (3)
	• Mixed feeding of breastmilk and formula via EBM or direct breastfeeding (2)
	• Breastfeeding plus solid food (2)
	• Formula feeding plus solid food (2)
Interview type (n)	• Face to face (10)
	• Telephone (3)

LSOA, lower super output area; GCSE, General Certificate of Secondary Education; NVQ, National Vocational Qualification.

All participants were offered the opportunity to have the main themes of the study sent to them and to take part in a second ‘member check’ interview. Following analysis, the main themes were sent to all participants who had requested them. Seven participants (two women, four peer supporters and one health professional) opted to take part in a second telephone interview and were in broad agreement with the main themes identified.

The socio‐demographic profile of women (Table 1) and the job roles and interview types for participating health professionals and peer supporters (Table 2) are presented below.

Table 2.

Job role and interview type for health professional and peer supporter participants

Health professionals (n = 6)	• Health visitor (n = 1)
	• Community nursery nurse (n = 2)
	• Student health visitor (n = 1)
	• Health visitor and practice teacher (n = 1)
	• Assistant family health practitioner (n = 1)
Interview type (peer supporters, n = 14)	• Face to face (n)
	Health professionals (4)
	Peer supporters (2)
	• Telephone (n)
	Health professionals (2)
	Peer supporters (1)
	• Group interview (n)
	Peer supporters (11)

Two semi‐structured interview schedules were developed, one for service users and one for health professionals and peer supporters. Both schedules covered participants' awareness and perceptions of BPS, barriers to and facilitators of access to the service, and recommendations for service development. For women, their infant feeding experiences and support needs were also explored, while health professionals and peer supporters discussed women's referral to the service. All participants were asked to sign a consent form (face‐to‐face interviews) or provide verbal consent (telephone interviews) prior to data collection.

Data collection and analysis were undertaken concurrently, with field notes written immediately following interviews. All interviews took between 25–90 min to complete, were audio recorded, transcribed verbatim by the first author and uploaded onto qualitative data analysis software (MAXQDA) for analysis purposes.

Data analysis

Analysis was initiated as soon as data collection commenced using Charmaz's broad analytical guidelines. First, transcripts were read multiple times, and open coding was used to categorise the text. Second, constant comparisons were undertaken. Comparisons were drawn between codes or events in the data, and written memos of such were recorded. These comparisons and subsequent questioning of the data were undertaken on an iterative basis until theoretical ideas emerged. Third, focussed coding took place when prominent codes that occurred frequently or seemed important were identified and compared against other sections of data (Charmaz 2006). The codes were then grouped to form categories. Links between tentative categories were made, and diagrams were drawn to represent these links. A sample of the transcripts was read by the second author. The theoretical ideas and diagrams were shared and discussed between the authors until consensual validation was obtained. During this process, one category emerged that held multiple links to others and acted as a reference point for other categories. This subsequently became the core category. Recruitment and concurrent analysis continued until no new theoretical ideas arose.

Findings

Overall, one core category and three key themes were identified in the data set. The core category concerned pressure and judgement within a dichotomous landscape of infant feeding. Pressure and judgement operated as the social, personal and cultural backdrop to many women's infant feeding decisions and experiences. Women sensed pressure (from professionals, media and social networks) to breastfeed and moral judgement around their feeding decisions. It was felt that women were made to feel ‘guilty and bad’ if they chose not to breastfeed and felt like a ‘failure’ if breastfeeding difficulties arose. Dichotomous discourses and practices were also prominent across all the themes that explained non‐access. Discourses around infant feeding frequently employed dichotomised language. For example ‘can’/‘can't’, ‘success’/‘failure’ and ‘you either breastfed or bottle‐feed’ appeared to signify that breastfeeding was often thought of as an ‘all or nothing’ practice. Dichotomies in terms of how support was offered and provided were also apparent. Theme one, ‘place and space of support’, relates to the tension in the early post‐natal period between promoting breastfeeding and a lack of appropriate forms of support. Theme two, ‘one way or no way’, relates to the sense of there being only one correct way to breastfeed. A ‘rules‐based’ model of breastfeeding support was employed by some health professionals and refers to the mechanistic manner by which breasts and breastfeeding were often constructed. The final theme, ‘it must be me’, concerns how health professionals and women's lack of insight into the value and purpose of embodied breastfeeding knowledge can lead to non‐access and to women forming ‘breastfeeding failure’ identities. It is important to reflect that, while the aim of this study was to explore reasons for non‐access to BPS among breastfeeding women, this study also highlighted operational and practice‐based issues as to why women discontinue breastfeeding early, thereby rendering BPS an unviable option. These issues are now discussed in‐depth, contextualised by participant quotes.

Theme one: place, space and timing of support

This theme illustrates that while some women felt pressurised to breastfeed their infants, the contrasting realities of inadequate or inappropriate early support may lead to it being ‘too late to support’ and to early breastfeeding discontinuation. Whereas, ‘It's not what you need’ explains the inappropriateness of the group nature of BPS during the early post‐natal period.

Too late to support

Despite the ‘breast is best’ rhetoric, some women reported minimal breastfeeding support in the hospital:

The girls […] at the hospital definitely didn't make sure that I was feeding properly. (Belinda, mother)

Women also frequently reported how the post‐natal contact came ‘too late’. As reflected in the wider literature (e.g. Graffy & Taylor 2005; Hoddinott et al. 2012), participants considered that practical help ‘earlier on’ was crucial in order for ‘successful’ breastfeeding to be established:

There needs to be more preparation and more emphasis on trying to, if people are going to breastfeed, there needs to be more emphasis on the time when it's going to be optimum for them to try. (Kim, trainee health visitor)

The lack of early support directly impacted on access to BPS due to many women discontinuing breastfeeding before they felt able to get ‘out and about’:

I think if you're not getting it [early support from a health professional] properly, […] women aren't carrying on long enough to the point where they can get out of the house. (Jacky, peer supporter)

‘It's not what you need’

During the post‐natal period, women perceived themselves and were often perceived to be ‘vulnerable’ due to recovering from the birth and coping with the demands of caring for a newborn baby:

I was trying to feed him, […] he was so hungry and I just couldn't do it, I was just like a mess. (Dolly, mother)

The prospect of accessing a group at this sensitive time and when they had ‘no confidence already’ to breastfeed was described as ‘nerve wracking’. Many women referred to how they would have benefited from knowing someone at the group, as their social anxiety about ‘walking in through the door’ into an unknown and unfamiliar environment was a key barrier. While many women identified their need for support and were aware of its availability, the group environment was often not considered appropriate:

I didn't want to go to somewhere where there are lots of ladies breastfeeding, I didn't want to sit there and have somebody show me how to breastfeed in a room, I wanted to be at home where I usually am going to be doing it, and be shown […] different ways to lay like in the bed to feed him […] which I wasn't shown, […] the breastfeeding support group is there, but yeah it's not what you need, not when they're that young. (Belinda, mother)

Peer supporter and health professional participants mentioned practical barriers to access such as group timings and transportation issues more frequently than women. These issues, however, were perceived to be ‘add on’ barriers and were often discussed after more central concerns had been expressed.

Theme two: one way or no way

This theme refers to dichotomies in relation to how support was provided and subsequently internalised by women. In ‘These are the rules’, the manner by which some health professionals employed ‘rules’ in explaining how breastfeeding ought to be performed and how this provided the sense of there being only one correct way to breastfeed is described. The ways in which some women provided mechanistic constructions of breasts and breastfeeding are discussed in ‘If it works, it works’. The ‘telling and advising’ communication style of breastfeeding support delivered by some health professionals and the detrimental impact this had on women's perceptions of and subsequent access to BPS is also explored.

‘These are the rules’

Peer supporters and women reported that many health professionals employed a functional, theoretical paradigm of infant feeding whereby breastfeeding correctly was a matter of following ‘the rules’ and adhering to guidelines. Women and peer supporter participants recalled ‘rules’ in relation to a wide range of breastfeeding‐related practices (e.g. the necessity for exclusive breastfeeding, demand feeding, breastfeeding rather than expressing and breastfeeding until 6 months of age). Some women perceived that for these health professionals, there was only one right way to breastfeed:

You've got midwives and things like that who have to follow certain rules, […] some of the nursery nursing teams are very strict and “these are the rules, and you follow these”, and they don't move very much. (Gail, mother)

There was no notion that what might be ‘right’ for one mother might not be ‘right’ for another or that ‘the answer’ might need to be personalised or adapted. This led women to place themselves either as rule ‘followers’ or rule ‘breakers’. While a number of women described similar experiences, Dana's case formed an illuminating example. Dana's baby wanted to feed for ‘three hours at a time’. She wondered whether he was ‘feeding for the whole time he's on there’ and how to manage this whilst simultaneously caring for her 2‐year‐old child. Although her health professional did not observe a breastfeed, discuss active feeding or follow‐up at a later point, Dana was advised to ‘tell’ her body to ‘produce more milk’ by letting her baby feed for as long as he wanted. The rule communicated was ‘if he's still latched on properly and it doesn't hurt, then let him’. This advice relied wholly on physiological knowledge, with no discussion, adaptation or application to the social situation at hand. Dana could not continue breastfeeding in line with this advice and decided to break the ‘exclusivity’ rule and give her baby formula. She explained that, in the end, ‘you just give up’ and follow ‘what I think is best [gestures towards heart]’.

‘If it works, it works’

Several women constructed their bodies and breasts as machines and accepted that with breastfeeding, ‘if it works, it works’. Like a machine, the women's descriptions appeared to suggest that when the ‘on’ switch is pressed, either success or failure ensues. In this way, breastfeeding was not seen in terms of a continuum or a process, rather a dichotomy: ‘Some people can do it, some people can't do it, some babies take to it, some babies don't’ (Esme, mother). Some peer supporters and health professionals also recognised this construction:

I feel like it's you try and you fail, or you try and you succeed and it's easy, there's no kind of middle ground. (Laura, peer supporter)

Some women discussed antenatal education in terms of how breastfeeding was presented theoretically, with its emphasis on functionality. Like the women's mechanistic constructions of breasts and breastfeeding, on occasion, this could sound like a mechanical sequence culminating in milk entering the baby:

They show you the theory of you know how it should happen, um, you know, you hold your baby like this and they do this, and that and this happens and,[…] so it was very, I don't, I don't know, it's very, um, text book. (Esme, mother)

Breastfeeding was constructed to be about the body, often ‘portrayed to be really simple to do’ and without ‘the grit’ of reality being addressed.

‘Telling and advising’: health professional's communication style

Women frequently recounted how professionals would ‘tell’ and ‘advise’ how to perform and adopt the functional and theoretically informed rules of breastfeeding. This was explained by Belinda:

It was more, ‘you've got to do this’ and ‘you've got to’ the words used […] weren't like helping, it was more telling me what to do. (Belinda, mother)

One peer supporter reflected:

You spend a lot of your early mothering experiences being told what to do by lots of different people. (Pippa, peer supporter)

For a number of women in this study, their ‘failure’ to follow the ‘right way’ often led them to ‘give up hope’ and to discontinue breastfeeding. For other mothers, it was the anticipation of a similar approach from the peer supporters, together with concerns of judgement due to non‐compliance with ‘the rules’, i.e. mixed feeding, that prohibited their access to BPS:

I felt like um every professional I'd spoken to, the nurse, midwife, doctor, GP, anybody at the hospital, they were very “these are the rules” you know “you should breastfeed until he's six months old, exclusively you shouldn't start food till then, and breastfeeding's best and” […] the people I met were very, like pushy to do things like as the book said, and I was a little bit afraid of you know, afraid's not the right word, but, of being judged, if I couldn't do it, […] in my mind I was afraid that they [peer supporters] were going to judge me and make me feel bad for perhaps finding it difficult and not being able to do it. (Esme, mother)

Theme three: ‘It must be me’

This theme concerns women's experiences of embodied and theoretical knowledge of breastfeeding. The seeming lack of awareness of the possible value of experiential as compared with theoretical knowledge by health professionals is outlined in their ‘sales pitch’ of BPS. For women, the divergence between their theoretical and embodied breastfeeding knowledge and lack of vicarious insights could lead to feeling that ‘there's something wrong with me’ and subsequent non‐access to BPS due to feeling ‘not like everyone else’ at the breastfeeding groups.

‘The sales pitch’

Congruent with a techno‐medical construction of breastfeeding, many of the health professional participants did not appear to value breastfeeding groups as somewhere where women could learn about and benefit from other women's varied experiences. The messages recounted in health professional's ‘sales pitches’ were reflected in a quote provided by a community nursery nurse:

I just sort of say ‘are you aware of the group?’ and direct them to the page in the book which has got the information about groups, let them know actually how to get there if they need to and that's about it. (Clare, community nursery nurse)

Peer supporters were described by health professionals as people to go to for ‘advice and help’ if breastfeeding ‘problems’ were experienced and when health professionals were unavailable. On one occasion, a health professional who had been invited to attend a breastfeeding group reported:

They've [peer supporters] had breastfeeding training, it's not that I'm going to be doing any different, to what the peer supporters would. (Phillipa, health visitor)

Phillipa assumed it was only the theoretical knowledge imparted via training that peer supporters would use in their supportive interactions with women. Additionally, several health professionals seemed unclear about the purpose of peer support and explained that women would be directed to groups ‘for [their] support’ and how peer supporters were ‘supporting other mothers’ in a very general way. Mothers described how health professionals did not explain ‘how it [peer support] worked’. When Chrissy was asked about the main reasons for non‐access, she suggested:

The fact that we're not really told what it is, or what the point of it is, or how it differs to other baby groups, really I think, they could, ought to a tell pregnant women a little bit more about it. (Chrissy, mother)

‘There's something wrong with me’

As reflected in previous research (e.g. Thomson & Dykes 2011), many women referred to the contrast between their theoretical knowledge of breastfeeding and the ‘shock’ of ‘actually doing it’. One mother described how she:

Naively thought before I had him that it would all come naturally, and they know, babies should know what they're doing and, it should just happen. (Esme)

Several women also revealed their lack of vicarious insights into other women's infant feeding experiences; ‘[I] didn't really have a lot of experience of seeing anybody do it [breastfeed] before’.

Women and peer supporters spoke of how ‘if they [babies] don't take to it straight away’, it became easy to assume that ‘you're not doing it right’ or even that ‘there's something wrong with me’, when the experience did not match their expectations. The self‐blame in women's accounts is evident in Heidi's depreciating remarks about the ‘quality’ of her breast milk:

[My] milk wasn't thick enough […] when I expressed it, it was really watery, runny,[…] lots there but just not thick enough. (Heidi, mother)

Heidi knew ‘what the problem is’, i.e. the seemingly inadequate quality of her milk, and assumed that nothing could be done. Indeed, Heidi, like other mothers in this study, had no knowledge of cluster feeding, growth spurts, different breastfeeding trajectories or other women's varied experiences of breastfeeding.

Not like everyone else

Women's mechanistic impressions of infant feeding, reinforced by those of health professionals, and a lack of embodied insights often resulted in the perception that breastfeeding was ‘easy’ for peer supporters and for those mothers who accessed BPS groups:

When you hear the term peer supporter you're definitely going to be assuming that they, they've had no problems, […] I think that you just assume that they're just going to be pros at it and have had no issues. (Chrissy, mother)

The conflict between women's self‐perceptions of being ‘a failure’ doing ‘a crappy job’ and the imagined ‘successful’ breast‐feeders who accessed the groups was a key barrier to access: ‘I'm not one of those [successful breast‐feeders], so how can I go into that [group]?’ In this way, breastfeeding was not perceived to be a skill that is gradually learned. This reflection was echoed by some of the peer supporters who could empathise with the women's reticence in entering an environment where ‘everybody else is doing it’.

Discussion

In this study, we report on a dichotomised landscape of infant feeding that worked in several ways to impact upon access to BPS. Mechanistic constructions of breastfeeding and a rules‐based approach contributed towards women polarising themselves as those who could or could not breastfeed and did or did not comply with the ‘rules’ for ‘successful’ breastfeeding. Women who did not access BPS often lacked vicarious insights and were surprised at the disparity between their expectations and embodied experiences. They struggled to follow ‘the rules’ and perceived themselves to be ‘failing’, thereby identifying themselves as a mother who ‘can't’ breastfeed. Non‐identity with peer supporters and other breastfeeding women at groups arose as women assumed that group attendees were successful breast‐feeders who had complied with ‘the rules’. As the public group BPS environment contrasted with the personal, internalised nature of women's emerging self‐identities, this made it an inappropriate environment for support.

Women in this study accepted moral responsibility for infant feeding, experienced pressure to breastfeed and anticipated judgement of their infant feeding decisions. These findings are similar to those of Larsen et al. (2008) and Murphy (2003). The theoretical ideas of Michel Foucault (1991), which Larson and Murphy employ to explain their findings, can also help to interpret the insights generated in the current study. In the 18th century, the family became an important area of medicalisation, meaning that moral responsibility for the welfare of children was imposed upon women (Foucault 1991). By means of subtle pressure and multiple small ‘technologies’ (e.g. breastfeeding ‘rules’), the state, acting via ‘disciplines’ (e.g. the ‘discipline’ of medicine), simultaneously increased the ‘utility, docility and obedience of the people’ (Foucault 1995, p. 137–8). ‘Disciplines’ operate through sciences that appear to be ‘the foundation for society’ (Foucault 1995, p. 223), and through them, standards of normality are established (Foucault 1995). Expert discourses that suggest that one course of action is healthy, and thereby ought to be undertaken, and other actions are unhealthy, and thus ought not to be undertaken, form an understated form of control (Murphy 2003). Everyone in society knows what ought to be done, and subtly, people become subjectified, self‐regulating citizens (Murphy 2003).

Women in this study adopted a mechanistic can/cannot conception of breastfeeding similar to that described by Dykes (2005a). The dichotomised discourses in infant feeding were also similar to those recounted in relation to women's interactions with health professionals in the Hoddinott et al. (2012) study. In addition to leading women towards a sense of ‘failure’, this language, rather than suggesting a continuum of breastfeeding experiences, drew women towards polarisation and to contrast themselves against those who are ‘successful’ and for whom breastfeeding is ‘easy’ (i.e. those who do attend group provision). Avoidance of those assumed to be successful can be viewed as behaviour associated with self‐regulation. Foucault (1991, p. 195) explains how in a society where ‘the disciplines’ (e.g. medicine) are dominant, ‘value giving’ normalising judgements impose standards that everyone strives to achieve. By avoiding others assumed to be reaching the prescribed ‘normal’ (i.e. who follow the rules and are successful breast‐feeders), participants in the current study, and similar to the findings of Murphy's (2003) study, acquiesced to the validity of the dominant discourse and internalised themselves as breastfeeding failures. Many participants in the current study referred to the value of ‘personal bridges’ for peer support access, such as knowing the peer supporter in attendance or arriving with a friend, as highlighted by others (Hoddinott et al. 2006; Thomson et al. 2012). These ‘bridges’ were considered to have the potential to mitigate negative feelings associated with differences in perceived identity and should form part of standard care.

The functional–theoretical ‘disciplines’‐based model of breastfeeding utilised by health professionals in the current study can be seen in terms of Foucault's (1991) subtle control. The technical–medical knowledge of such state agents of medicalisation is privileged and acts to exert power over women (Murphy 2003). Echoing the findings of Burns et al. (2010) and Murphy (2003), women in the current study talked about health professionals telling them what to do and that for health professionals there was one right way to breastfeed. Burns et al. (2010) and Bartlett (2002) recognise a shift in authority regarding breastfeeding in western societies from women's embodied knowledge to expert knowledge. In the current study, it appeared that only theoretical knowledge of breastfeeding was known about. Problems were often interpreted in strictly functional–theoretical terms and without the practicalities of everyday life being considered. Previous research has identified how women's anticipation of ‘being told’ what to do formed a barrier to their accessing health professional support (Hoddinott & Pill 1999). While Hoddinott et al. (2006, p. 144) report that some women were ‘put off’ peer support due to concerns that peer supporters might be ‘snooty’, in our study, women anticipated that peer supporters would adopt the same mechanistic, functional approach they had experienced from health professionals. Our findings also support those of Dykes (2006a) in that women appear to value a manner and model of breastfeeding support that moves away from the medical model towards a more socio‐cultural, woman‐centred model of infant feeding.

Coupled with the privileging of medical knowledge and forming a further reason for non‐access was the lack of clear messages about the purpose or value of peer support and how it might differ from health professional support. While authors such as Raine (2003) and Muirhead et al. (2006) identified variability and reticence among health professionals when referring women into BPS services, a lack of clarity regarding how BPS might work has not previously been described as a reason for non‐access. In the current study, health professionals rarely mentioned experiential knowledge as a reason why peer support may be valuable. These insights thereby indicate that when only theoretical knowledge concerned with function is legitimised, experiential knowledge has the potential to be rendered irrelevant. In the current study, the BPS service did not have a clear underpinning theoretical base that health professionals understood and could articulate. This finding supports those of Thomson & Trickey (2013) who highlight a lack of underpinning theory regarding peer support projects that makes the interpretation of trial results difficult. In the current study, the lack of underpinning theory impacted directly on the practical functioning of the BPS intervention, hence, future exploration of this issue among stakeholders would prove valuable.

The finding that women need practical help in the early post‐natal period is supported by a wide body of research (e.g. Graffy & Taylor 2005; Hoddinott et al. 2012; Schmied et al. (2010); Thomson & Dykes 2011; Thomson et al. 2012; Thomson et al. 2015b), as is the variable quality of support provided by health professionals (e.g. Dykes 2005b; McInnes & Chambers 2008; Hoddinott et al. 2012; Thomson et al. 2015b). Participants in this study, and as reported by Dykes (2006b), emphasised the importance of place and space in relation to breastfeeding as a public or private activity. One of the key facilitators of access to BPS identified in the current study was that it be delivered by telephone or face to face in the women's own home. Emotional and physical vulnerability meant that women want support to come to them. The finding that a group environment was not appropriate in the early post‐natal period also supports the findings of Hoddinott et al. (2009) and emphasises the need, as reported in the NICE guidelines, of early and proactive support (NICE 2008).

The group environment can place breastfeeding in the sphere of the community, society and culture and can value the embodied knowledge of women. This positioning contrasts with the way many study participants situated breastfeeding and the medicalisation discourse that can explain many of the findings of this study. Hoddinott et al. (2012) highlight ‘pivotal points’ of support that arise from dissonance between idealism and reality and suggest a family‐centred discursive approach to anticipating them. Discussion between women and health professionals about the function of peer support and perceived barriers to access could facilitate deeper discussion concerning expectations and realities of breastfeeding, family context and support. In this way, although access to peer support can represent the confluence of two contrasting ways of thinking about the very nature of breastfeeding (i.e. breastfeeding as a socio‐cultural practice or breastfeeding as a technical–medical bodily function), it also represents an opportunity to explore this in the context of women's personal circumstances.

To our knowledge, this is the first study to specifically explore why women who initiate breastfeeding do not engage with BPS provision. As an original interpretation, this study has utilised a theoretically informed method to generate hypotheses for non‐engagement that are relevant to policy and practice. Incorporation of mothers', health professionals' and peer supporters' perspectives is a strength of the study, enabling consideration of diverse viewpoints. Women were also recruited across a broad range of ages and backgrounds. The participants' views may have been influenced by the data collection method employed. For instance, those who took part in a focus group may have felt unwilling to offer open and honest reflections. Telephone interviews have often been assumed to be an inferior research method when conducting qualitative interviews (Holt 2010) and are criticised in terms of quality due to an absence of interpersonal cues (Aquilino 1994); however, others would argue that this evidence is lacking and suggest that this mode of contact can enable more in‐depth information to be disclosed (Novick 2008). Rigour was considered from the outset and sought through reflective practices, member checking and consensual validation between the authors. Overall, however, the data set was limited and drawn from a restricted geographical area in which only one model of BPS was utilised. The findings interpret and give voice to the experiences and views of the participants at a particular time, place and context, hence may not be generalisable to others. Further qualitative research in other areas where differing models of BPS are in use could enable greater understanding of common or divergent influences.

Conclusions

Perceptions of pressure and judgement experienced within a dichotomised landscape of infant feeding prevented breastfeeding womens' access to peer support. Dichotomies in language, the structure of services and the manner of support were reported, with both mothers and providers often constructing breastfeeding as an ‘all or nothing’ endeavour. These dichotomies highlight how the medicalisation of infant feeding and the hegemony of technical and medical knowledge have undermined and de‐valued embodied insights, leading to early breastfeeding cessation and castigations of failure. Mothers and health professionals' lacked insight into the value of other women's embodied knowledge. Non‐identity with peer supporters and attending women arose as women assumed attendees had followed ‘the rules’ and were ‘successful’ breast‐feeders. This meant that BPS was not a viable option for many women. While post‐natal feeding support needs to be re‐evaluated to ensure it is acceptable and appropriate for women, these findings emphasise the need for a socio‐cultural model of breastfeeding, providing clear messages regarding the value and purpose of peer support.

Source of funding

There were no external funding sources for this research.

Conflicts of interest

The authors declare that they have no conflicts of interest.

Contributions

Study design is by LH and GT. Data collection and analysis was carried out by LH with support and guidance from GT. Manuscript was authored by LH with guidance and feedback from GT. The final manuscript was read and revised by both authors. Both authors meet all criteria for authorship.

Hunt, L. , and Thomson, G. (2017) Pressure and judgement within a dichotomous landscape of infant feeding: a grounded theory study to explore why breastfeeding women do not access peer support provision. Maternal & Child Nutrition, 13: e12279. doi: 10.1111/mcn.12279.