To what extent are women free to choose where to give birth? How discourses of risk, blame and responsibility influence birth place decisions

Abstract

Over the past 50 years, two things have changed for women giving birth in high-income nations; birth has become much safer, and now takes place in hospital rather than at home. The extent to which these phenomena are related is a source of ongoing debate, but concern about high intervention rates in hospitals, and financial pressures on health care systems, have led governments, clinicians and groups representing women to support a return to birth in ‘alternative’ settings such as midwife-led birth centres or at home, particularly for well women with healthy pregnancies. Despite this, most women still plan to give birth in high-technology hospital labour wards. In this article, we draw on a longitudinal narrative study of pregnant women at three maternity services in England between October 2009 and November 2010. Our findings indicate that for many women, hospital birth with access to medical care remained the default option. When women planned hospital birth, they often conceptualised birth as medically risky, and did not raise concerns about overuse of birth interventions; instead, these were considered an essential form of rescue from the uncertainties of birth. Those who planned birth in alternative settings also emphasised their intention, and obligation, to seek medical care if necessary. Using sociocultural theories of risk to focus our analysis, we argue that planning place of birth is mediated by cultural and historical associations between birth and safety, and further influenced by prominent contemporary narratives of risk, blame and the responsibility. We conclude that even with high-level support for ‘alternative’ settings for birth, these discourses constrain women’s decisions, and effectively limit opportunities for planning birth in settings other than hospital labour wards. Our contention is that a combination of cultural and social factors helps explain the continued high uptake of hospital obstetric unit birth, and that for this to change, birth in alternative settings would need to be positioned as a culturally normative and acceptable practice.

Keywords:

• pregnancy
• birth risk
• decision-making
• longitudinal
• narrative

Introduction

Planning where to give birth is arguably one of the most important decisions made during pregnancy, yet it is often taken for granted that birth takes place in hospital. Recent history shows that this remains the case despite almost 50 years of challenges to ‘technological’ birth by lay groups, feminist scholars and other maternity activists. These groups have long argued that hospital birth, with its reliance on surveillance medicine, high-technology equipment and adherence to protocol-led care, changes birth from a positive, life-affirming rite of passage to a dehumanised, mechanistic process. It is now well established that when healthy women with low-risk pregnancies give birth in traditional hospital labour wards (or obstetric units ‘OUs’ – see Box 1 for categorisation of the types of facilities available to women), they are more likely to experience interventions, surgical birth and their sequelae; what then accounts for the continued popularity of the ‘hospital’ birth model?

Drawing on empirical narrative research with pregnant women in England, in this article, we explore women’s experiences of deciding where to give birth. Our starting point is an acknowledgement that such decisions are taken under conditions of uncertainty, when the outcome is unknown. By drawing on sociocultural concepts of risk, we begin to tease apart the subtle impact brought to bear on these decisions by discourses of risk, blame and responsibility. We argue that beliefs and assumptions about birth risk are deeply ingrained, reflect varying perceptions of who is to blame if things go wrong and incorporate differing views of both nature and technology in relation to birth. We start the article with a discussion of existing literature in relation to birth place preference, we then review wider public debates about how the relationship between medical technology and nature influences constructions of birth as ‘safe’, ‘risky’ or ‘uncertain’, before presenting findings from our empirical research.

Risk, choice and place of birth

In England, there has been high-level policy support for choice of place of birth for two decades (Department of Health 1993, 2007), yet giving birth in settings other than a hospital OU (or ‘labour ward’) remains unusual. Over the past two decades, the proportion of home births in England has remained virtually static at around 2.5%, although national data (HCC 2008) shows that a further 5% of women now give birth in alongside or freestanding midwifery units (AMUs or FMUs – see Box 1).

Box 1. Terms used to denote different birth facilities or settings.

OU – Obstetric Unit (‘labour ward’): Provides 24 hour routine and emergency maternity care, with access to NICU; staffed by midwives, obstetricians and other specialists.

Ninety-three per cent of births in England take place in obstetric units.

AMU – Alongside Midwifery Unit (‘birth centre’ or ‘snoezelen room’) – Maternity unit staffed and led by midwives, co-located on site of an obstetric unit but organisationally separate. No epidural or surgical services.

Three per cent of births in England take place in AMUs.

FMU – Freestanding Midwifery Unit (‘birth centre’). Maternity unit on a separate, community site, not based in a tertiary hospital. Staffed by midwives, no epidural or surgical services.

Two per cent of births in England take place in FMUs.

Home birth – birth in woman’s own home, supported by a midwife (legal and provided free of charge through NHS care in the United Kingdom).

Around three per cent of births in England take place at home.

Source for definitions: Rowe (2011).Source for proportion of births in each setting: HCC (2008).

Previous studies of birth place decisions consistently show that hospital birth is associated with safety for many women (Houghton et al. 2008, Pitchforth et al. 2008, 2009). Birth place preferences are also thought to be influenced by socio-economic status (Nelson 1983, Davis-Floyd 1990, 1992, Zadoroznyj 1999), by access to private or publicly funded models of care (Liamputtong 2005, Murray and Elston 2005), by moral considerations (Viisainen 2000) and by cultural, religious and ethnic allegiances (Donner 2003).

Scholars have also argued that women associate different risks with different settings for birth (for example, Davis-Floyd 1990, 1992, Abel and Kearns 1991, Devries et al. 2001, Bryant et al. 2007). Women who prefer home birth are usually described as being concerned about risks imposed on ‘natural’ birth by birth in a hospital, and by separation from their families (e.g. Abel and Kearns 1991, Davis-Floyd 1992, Edwards 2005, Cheyney 2008). In contrast, women who prefer hospital birth are thought to be concerned about the risks and uncertainties of ‘natural’ birth, preferring to ‘stay in control’ and seeing hospital medical technology as a means of reducing risk by securing a clinically ‘clean’ and ‘safe’ birth, with access to anaesthetic pain relief during labour (Davis-Floyd 1994, Bryant et al. 2007). The assumption that women’s birth place decisions are polarised between preference for either ‘natural’ or ‘medical’ birth has rested largely unchallenged in the sociological literature since the early 1970s, and this dichotomy increasingly fails to capture the nuances of women’s experiences, or the breadth of contextual influences upon their decisions (Coxon 2012). The polarisation of ‘natural’ vs. ‘technological’ perspectives of birth also reflects a debate central to risk theory identified by Zinn (2008, p. 439), which is the attempt to contrast the relative merits of ‘objective’ or ‘rational’ knowledge (associated with an ‘expert’-led technological knowledge) with ‘non-rational strategies’ informed by personal experiences, beliefs and socially mediated knowledge such as folk-wisdom or ‘old wives’ tales’. Zinn (2008) argues that in practice, individual decision-making relies on strategies which are ‘in between’ these extremes, leading to decisions which not only draw upon objective knowledge, but are also infused with intuition, trust and emotion.

In recent decades, ‘home-like’ settings in maternity hospitals have been provided in many high-income countries. ‘Alongside midwife-led units’ or ‘birth centres’ provide low-technology midwife-led care to carefully screened healthy women whose pregnancies are categorised as ‘low-risk’ (Hodnett et al. 2010), with the additional ‘safety net’ of easy access to high-technology care, should this be required. Such units might be considered an attempt by hospital institutions to address exactly this need to manage the uncertainty of birth with an ‘in-between’ setting, both humane and home-like but with rapid access to high-tech facilities. In England, the provision of AMUs in maternity hospitals has expanded rapidly in recent years (Redshaw et al. 2011), and these increasingly appear to be the preferred solution to the ‘problem’ of providing women with both choice and safe, high-quality, individualised maternity care.

Birth place decisions in the United Kingdom: is birth safe or risky?

Childbirth in the United Kingdom is increasingly considered to be ‘safe’, at least relative to low-income countries without health infrastructure or access to trained and qualified birth attendants. However, a combination of higher maternal age, an ethnically diverse population, rising caesarean rates and increasing levels of obesity and chronic disease, such as diabetes, mean that more pregnancies are medically complex, and demand for tertiary-level, critical and intensive care in maternity services is increasing (RCOG 2013).

Despite these public health challenges, women and babies in England are demonstrably safer during birth than was once the case, yet birth is still understood and, to some extent, experienced as ‘risky’ or potentially so. In the context of welfare, Taylor-Gooby (2000, p. 4) described anxiety about risk where evidence suggests improved health, wealth and security as ‘timid prosperity’, occurring where ‘concern [about risk] is pervasive although vulnerability is concentrated among the weakest groups in an increasingly unequal society’. In relation to birth, this seems to be the case; women who do not have universal access to high-quality maternity care bear much of the burden of poor outcomes, but those in more affluent settings remain highly sensitised to the presumed risks of birth. In common with Giddens (1991) and Beck (1992), Taylor-Gooby considers this to be evidence of risk reflexivity, which occurs as a consequence of well-educated, affluent populations developing ‘more critical attitudes to received wisdom and to professional expertise’ (2000, p. 10).

Despite the critical and reflexive responses to expert notions of risk that these theories anticipate, there remains an enduring public acceptance that births ought to take place in hospital OUs, particularly for first babies (Topliss 1970, Cunningham 1993, Barber et al. 2006). Contemporary discourses of parenthood, examples of which are found in media debates on parenting, advertising targeted at parents, parenting magazines and advice provided by professionals involved in supporting parents and parents-to-be, ensure parents are made aware that they must be seen to be responsible, effective ‘risk managers’ in relation to birth, upbringing and infant feeding (Green 1999, Lee 2008, Scamell 2011, Scamell and Alaszewski 2012). Parents-to-be, particularly pregnant women, are also required to demonstrate moral, physical and emotional ‘fitness’ for pregnancy and beyond (Marshall and Woollett 2000, Mansfield 2008, Nash 2011). Green (1999, p. 29) has also charted increasing societal unwillingness to describe or accept deaths as ‘accidental’, and argues that deaths once considered as misfortunes have been re-categorised as an outcome of mismanaged risk, something that ‘should never have happened in the first place’ (2007, p. 35). That this was the case in relation to place of birth by the turn of the 1960s can be inferred from the following (anonymous) editorial comment in the Lancet: ‘Good antenatal care and safe delivery in a hospital fully equipped and staffed to deal with any emergency can prevent family tragedies once thought to be unavoidable’ (Lancet 1963, p. 1208, emphasis added).

The view that birth in hospital is essential to prevent death is the product of a powerful historical discourse, and one which has been reiterated by sources perceived as authoritative, with international reach and influence.

Sociocultural theories of risk, blame and normality

The belief that scientific or technological advances create dangers as well as solutions is also part of the ‘risk society’ thesis (Giddens 1991, Beck 1992). Beck, in particular, argues that reflexive anxiety about risk is a consequence of global technological advances, because the risks these carry are beyond the control of government or the democratic will of individuals or nations. A decade earlier, Douglas and Wildavsky (1982) had also tried to tackle the paradoxical increase in risk anxiety which seemed to accompany improved levels of security and well-being. In Risk and Culture, Douglas and Wildavsky (1982) examined why it might be that an increasingly affluent US population, apparently benefitting from cheap nuclear energy and improved living standards, developed an increasing aversion to the technological and environmental risks of nuclear power. They concluded that ‘technological’ risks, rather than necessarily being ‘objectively real’ or ‘measurable’, are better understood as having been ‘selected’ from an array of potential hazards, to support and validate a particular political perspective (for example, being pro-nature, and opposed to pollution, or other impacts of scientific or technological activity on the environment). A further insight, drawn from Douglas’ earlier anthropological studies (Douglas 1966, 1970), was that the opportunity to allocate blame is central to the selection of risks: ‘Blameworthiness takes over at the point where the line of normality is drawn. Each culture rests upon its own ideas of what ought to be normal or natural. If a death is held to be normal, no-one is blamed’ (Douglas and Wildavsky 1982, p. 35).

In this article, we examine the extent to which these sociocultural approaches to risk might enhance our understanding of contemporary birth place decisions. As we have argued, for ‘natural birth’ proponents, birth at home or in a non-OU setting minimises ‘unnecessary intervention’, and provides the best opportunity for labour and birth to be a fulfilling, life-enhancing experience, to assist with the transition to parenthood and relationship formation between mother and child (Kitzinger 2005, Fahy et al. 2008) and increasingly, though to a lesser extent, father and child (Miller 2011). Detractors argue that whilst birth without obstetricians to hand may be safe for many, it is unsafe for a few, or at least associated with untenable pain, and difficulty in predicting who will benefit and who will suffer makes an OU the ‘rational’, common-sense answer. The increasing costs of insuring obstetric and maternity care, particularly in relation to lifelong claims for damage and disability in infants, provide a further strand to this complex situation. As well as potential medical and obstetric risks, birth carries litigation and reputational risks for professionals. This leads to ‘litigation–based practice’ (Dahlen and Homer 2013, p. 168), where birth risks are managed through adherence to (and sponsoring women’s compliance with) ‘active management’ protocols and procedures, to reduce professional and organisational exposure to medico-legal risk.

Birth may then be constructed as both safe and potentially risky, for women, babies and clinicians, and choosing home, or birth in a midwife-led setting over a traditional OU birth might either increase or decrease perceptions of different types of risk, depending on the perspective adopted. Comparative epidemiology and public health evidence shows that some birth risk is objectively present, although rare, but qualitative studies reveal that women and partners’ risk perceptions are often subjectively distorted beyond the level of ‘actual’ threat (Houghton et al. 2008). It is likely that the heightened perception of risk goes some way towards explaining women’s birth place preferences, but this alone fails to account for why some women come to adopt strong opposition to birth in hospital OUs, whilst others positively opt for this. The research described in this paper arose from a need to understand better what accounts for birth place preferences, particularly the continued uptake of hospital OUs, in a context where alternative settings, whilst not always immediate or accessible, are freely provided and actively promoted, yet remain under-used.

Method

Much research into this subject has relied on single interviews, often conducted retrospectively, after birth has taken place. This has the effect of producing results based on women’s sense-making of their birth experiences, sometimes many years after the event. This research used a prospective, longitudinal narrative interview design in order to provide a contemporaneous account of women’s views and beliefs about place of birth, and to identify any changes which occurred over the course of pregnancy. We used a narrative methodology to prompt unguided accounts, through the use of a single ‘narrative-eliciting’ question at the beginning of each interview, which invited participants to discuss where they planned to give birth, and any events or experiences which contributed to this. In shorter ‘end of pregnancy’ interviews, we asked women to comment on their experiences since the first interview, and explored whether their views or birth place intentions had altered in the interim.

We decided to use the narrative method as a means of inviting accounts without ‘framing’ the study in terms of risk or safety, which, as Henwood et al. (2010) note, can mean that researchers’ a priori assumptions about risk form the basis of the investigation, rather than allowing new or different perspectives to emerge. Our approach was informed by Wengraf’s (2001) biographical-narrative interpretative methodology, and we adopted Riessman’s analytical methods for exploring how accounts are ordered and structured (Riessman 2008), and assessing the moral positioning of the self as a protagonist within the account (Bury 2001, Ehrich 2003). Ethical approval was granted by a National Health Service (NHS) research ethics committee [09/H0808/45].

Fieldwork was undertaken between October 2009 and November 2010 at three maternity services in England. The sites included inner city and semi-rural locations with different geographies and transfer time scenarios. Birth place options varied between the sites, but women had access to home birth (all sites), FMUs (one site), AMUs (two sites) and OUs (all sites). The proportion of births taking place in AMU, FMU or at home in these sites ranged from 6.7% to 18.5%, and was near or above the average for England (7%) at the time the fieldwork was undertaken (HCC 2008).

We recruited a diverse sample (Patton 2002) in order to explore the extent to which accounts of birth place risk and safety varied amongst respondents with different parity, birth experiences, socio-economic, cultural and ethnic backgrounds. Initial recruitment yielded a relatively homogenous sample, but following ethical approval of a substantial amendment to the original protocol, store vouchers for £20 were provided to participants in recognition of the time contributed to the research. The final sample included 41 women recruited at antenatal clinics or through interpreters (birth partners were also recruited, but this paper presents only findings from interviews with pregnant women). Participants were aged 19–42, from various ethnic backgrounds, five of whom required an interpreter to participate. Interviewees held a range of qualifications, and employment status varied. Just under half were expecting their first baby (44%), reflecting the national average (ONS 2009). Married, cohabiting and single women took part. Overall, respondents were slightly older and held higher qualifications than is the case in the maternal population (when compared with national and regional data sets) but there was diversity in terms of employment, spoken English, education and ethnicity.

Most initial interviews took place between 12 and 24 weeks of pregnancy at women’s homes, and a small number were conducted by telephone, where this was the woman’s preference. Interviews lasted between 30 and 75 minutes, but were usually about an hour long. Short follow-up telephone interviews were carried out in the last month of pregnancy. All interviews were conducted by the same researcher [Kirstie Coxon]. Altogether, 82 interviews were undertaken, with no loss to follow-up.

Interviews were digitally recorded and transcribed, with the exception of three participants who preferred not to be audio-recorded; these interviews were documented in contemporaneous field notes. Interview transcripts and field notes were entered into N-Vivo 8 (N-Vivo 8 QSR International Pty Ltd.) and initially analysed using thematic narrative analysis (Riessman 2008). This approach uses a coding process similar to grounded theory method, but with some key differences; in narrative thematic analyses, context is retained, text is interpreted in relation to each woman’s overall ‘story’, and interviewer prompts and questions, or other interruptions, are retained and considered part of the co-constructed interview discourse.

Antenatal data analyses were also ‘future-blind’, meaning that they were undertaken before the outcome of the pregnancy and actual place of birth were known. In narrative analysis, the ‘outcome’ of the story is understood to change how the events are interpreted by the teller and by the listener, so ‘future-blind’ analysis helped identify women’s pre-birth beliefs, concerns and aspirations before these were altered by the course of events. Disconfirming or theoretically ambiguous data were discussed with co-authors [Jane Sandall and Naomi Fulop], peer researchers and service user groups to clarify and make sense of the issues that these raised.

Findings

In this article, we draw on data from the interviews to examine how women made sense of birth risk and the uncertainties they faced during pregnancy, and related these factors to their planned place of birth. From a clinical perspective, three quarters of participants (30 out of 41) had healthy pregnancies, making them eligible for birth in a full range of possible settings (home birth, FMU, AMU or OU). The remainder (11 out of 41) had a risk factor which meant they either required further assessment, or would routinely be advised to give birth in a hospital OU according to existing English National Institute of Health and Care Excellence (NICE) guidance (NCCWCH 2007). The reason for recruiting women with both clinically healthy pregnancies and with risk factors present was to explore the ‘fit’ between messages received from health professionals or others about the riskiness or otherwise of their pregnancies and women’s own expectations of birth risk.

As others (Lee et al. 2012) have also found, women’s perspectives could differ markedly from clinical risk assessments. Some felt they could safely give birth at home despite having clinical risk factors which meant that hospital OU birth would be recommended. This viewpoint was however relatively unusual; it was far more often the case that women were healthy and had ‘low-risk’ pregnancies, but could not countenance giving birth anywhere other than hospital OUs. There was also a clear difference in experience between women expecting their first babies, and those who had given birth before; in first pregnancies, women were more open to the idea of birth in different settings, but those expecting second or subsequent babies usually planned hospital OU births. For this reason, findings are presented separately for each group.

Where extracts from interviews are presented in this paper, pseudonyms are used and broad contextual information is provided, although identifying details have been removed. In interviewee quotes, pauses of up to three seconds are denoted by an ellipse within square brackets [...]. Non-bracketed ellipses indicate that some text has been removed, usually to shorten the quote, but the sequence of the narrative is unaltered.

Prioritising ‘medical’ risks of birth: planning birth in a hospital obstetric unit

When women preferred to give birth in hospital OUs, their accounts often presented birth as medically risky or subject to danger. Most women in the sample (25 out of 41) planned to give birth in hospital OUs by the final month of pregnancy. Of these, 11 had risk factors which meant that OU would be recommended for birth, but 14 had healthy pregnancies, and it is on this group that we focus here, given that they could all have chosen home birth (because this is available in all the areas where recruitment took place) and most had access to either AMU or FMUs too. Seven were expecting their first baby, and seven were planning their second or subsequent birth.

Accounts from women expecting their first babies

When women planned to have their first baby in hospital OUs, it was often because they feared something might go wrong, or had received advice from close family to give birth in hospital. Kath described a discussion in her midwife ‘booking’ appointment in early pregnancy as follows:

I was really surprised actually because […] I’m 38 and … I’m not the world’s fittest person … and I was convinced that I was going to be some high-risk kind of, oh my gosh, you can’t step out of [a hospital labour ward]. So I was quite surprised when I came out low-risk, and [the midwife] said ‘Do you want a home birth?’ (Kath, expecting first baby, healthy pregnancy, planned OU birth)

Kath’s concerns related to overall age and general fitness, and these respond to existing discourse of being ‘fit’ to give birth (Mansfield 2008, Nash 2011). Other women were sceptical about the likelihood of an out-of-hospital birth being safe for themselves or their babies because of their family histories. Laura had a healthy pregnancy and was expecting her first baby, but thought her relatives’ experiences of birth might affect her own labour:

In deciding where to have the baby, I guess I was pretty determined I’d have it in hospital [OU]. Both my sister and my mother had problems during birth, I was born by emergency caesarean and my sister had an emergency caesarean with her first child, and then an elective caesarean for her second, so it made sense given the experiences of people close to me that I’d like to be somewhere with good medical care on hand, if something goes wrong. (Laura, expecting first baby, healthy pregnancy, planned OU birth)

Donna’s general medical practitioner (GP) advised her to avoid going to her local FMU with her first baby, and his concerns chimed with her own risk perceptions:

I said, ‘Look, don’t worry, I’m not going anywhere there’s no doctors.’ And he said, ‘Yes, I’m just saying, you know, because you know the chances are … It’s a 40 minute journey [referring to transfer to OU during labour]. Do you want to risk that?’ No! [Laughs] But yeah, that’s all he really said, but he was right…I’m not risking that, I’m not risking the baby’s life or my life. So … you know, it’s just eliminating all the risks as much as you possibly can. (Donna, first baby, healthy pregnancy, planned OU birth)

In their narratives, Kath and Laura stopped short of making the ‘risks’ of birth explicit; instead, they alluded to the need for a hospital labour ward and access to emergency caesareans. Donna, on the other hand, was clear that she wouldn’t risk ‘the baby’s life or my life’ by giving birth in a place without doctors. Referring to the ‘worst-case scenario’ in this way heightened the dramatic effect of the statement, and communicated Donna’s sense of anxiety about birth. In each of these accounts, ‘hospital’ was used as a short-hand to refer to obstetric medical care and a sense of protection from risk; ‘alternatives’ were not mentioned, and women did not anticipate having water births or similar ‘natural birth’ experiences in hospital OUs. The hospital itself was associated with safety and reassurance; later in her interview, Kath described a major teaching hospital in the following terms: ‘But yeah, it is that kind of, yeah. I’ve fallen for the propaganda, it’s that lovely big sign outside, which says “comfort”. Yeah, it feels really comforting’.

Accounts from women expecting their second or subsequent baby

Women expecting their second or subsequent babies often have had previous normal births themselves, but were still influenced by the experiences of others close to them. Patsy’s account illustrated this well, and she linked her decision to give birth in hospital OU, despite liking the idea of the facilities in FMU or AMU settings, to a sibling’s experience of stillbirth.

My brother had … well his wife had a baby at home and the baby died … and I think that affects … that sort of affects the family for a long time, you know, anyone in the family who was involved with that or remembers that, you can’t, [home birth is] just a no-no for us. (Patsy, multiparous, healthy pregnancy and straightforward obstetric history, planned OU birth)

Later in the interview, Patsy mentioned that her mother also had a child born with a disability at home. The clustering of infant death or disability within families may reflect socio-economic inequities or genetic conditions, but the impact is passed down through generations, and to plan a home birth could be construed as insensitive to affected family members, and implicitly disrespectful towards shared family memories.

For women born in countries other than the United Kingdom, family practices were also important, and these reflected the national context of birth in both developing and high-income countries. Alexandria’s family lived in Africa and although she had a healthy pregnancy and straightforward obstetric history, she was insistent that she should go to hospital OU to give birth.

Well basically I think it’s just what I’ve known. That’s what my mother did, all my family members have had their babies in the hospital, and I just think it’s a safer place, in case of any kind of emergency it’s better to be at the hospital where there’s a doctor close by.

…I think it’s culture as well, for me. It’s just where we’re from. You only have a baby at home if you can’t make it to the hospital, and even when that happens people are […] almost ashamed to say. They’ll still say, ‘Oh yeah, we went to the hospital.’

Preference for hospital OU was also found amongst families from European countries where hospital OU is the only birth option available. Hannah described a conversation with her birth partner, who had asked whether she wanted to give birth at home:

I just thought, what? What a strange thing to ask. Because for me, being [from a Nordic country], when you want to have babies, you go to the hospital, just like if you want to have an operation you go to hospital! (Hannah, multiparous, healthy pregnancy, planned OU birth)

Amongst women who had previously given birth, the sense that birth ought to take place in hospital emerged through tacit knowledge or received wisdom, which had been absorbed through family life and inter-generational life histories, but was also representative of national cultural discourses and normative practice. These inherited ideas and traditions might feasibly be brought into question by the events of women’s own previous labours, or by different approaches in the new, host country, but it seemed to be more often the case that beliefs learned through upbringing and family practices endured. Women expecting their first babies did not have past labour to draw upon, and amongst this group, perceived medical-obstetric risks of birth were an important basis for preferring hospital birth, but these rationalities supplemented an existing canon of expectations in relation to place of birth.

Conceptualising the ‘risks’ of hospital obstetric unit birth

Regardless of whether or not they had given birth before, women who planned to give birth in hospital OU discussed their need to determine which NHS hospital was most likely to provide them with the best care. The context of media is relevant here; health care is frequently subject to debate in the UK mass media, and headlines reflect the national characteristic of passionate support for universal NHS health care and equally passionate assertions that the population is being failed by an underfunded or overly bureaucratic health system. Although NHS heath care is free, and only a tiny proportion (0.5%) of women have private obstetric or midwifery care, private in vitro fertilisation (IVF) clinics often provide services to the NHS, and so women may encounter private provision at this time. Serena had experienced assisted conception, and implicitly felt that a more private-like setting might have more chance of success:

I mean I must admit when I went to [NHS obstetric hospital 1] to do the [fertility] programme, the offices were very kind of [...] very basic, quite [...] claustrophobic, you know, whereas [at NHS obstetric hospital 2] it was a new centre that they had, and even just the images on the wall, it had a feeling of like this is as if it’s a private [...] you’re walking into some kind of private unit. And not to say that [NHS obstetric hospital 1] couldn’t have [achieved a successful pregnancy], but it gave you that feeling of not sitting down a corridor squashed together, everyone can hear what you’re saying, it was very much [...] the way they presented themselves and the way the new unit, it felt like a new unit that they would have the better facilities. (Serena, first baby, healthy pregnancy, planned OU birth)

The need to avoid being ‘squashed together’ in ‘busy’ units conveys one of the perceived risks of NHS care. Hospitals were sometimes envisaged to be crowded, or chaotic, and there was a sense that women needed to compete for attention, and might even be overlooked:

I just know how easily mistakes are made, and it worries me, that I’m putting my faith and the life of my child in a [...] in an overrun crowded hospital full of people giving birth [...] yes, does make me a bit nervous. But [...] we’ll see. (Annette, first baby, healthy pregnancy, planned OU birth)

Others were also concerned about exposure to other people’s physical or social conditions:

You’re more at risk because you’re with people, you’re surrounded by … however many, 20 beds or something, and … you know, another 40 couples that are having their babies and you don’t know their health. And so disease, not disease but you know, things do spread. (Serena, expecting first baby, healthy pregnancy, planned OU birth)

In these accounts, OUs in hospital are privileged as sites of safety and protection because emergency care is available if ‘something goes wrong’, but this needs to be balanced against some potential risks of being in hospital. These include exposure to overcrowding, errors and physical or social ‘pollution’, a term used here in the sense described by Douglas and Wildavsky as ‘a contagious state, harmful, caused by outside intervention’ which ‘carries the idea of a moral defect’ (1982, pp. 36–37).

Although women were worried that maternity care might be anonymous, or error-prone, what is absent from these accounts is concern about high rates of intervention in maternity care. Inch (1982, p. 244) and others have conceptualised the ‘cascade of intervention’, a scenario where one birth intervention (such as induction, or augmentation of labour) almost inevitably leads to another, incrementally increasing the likelihood of an instrumental or surgical birth at each stage and reducing the chances that the woman will have a ‘natural’ birth. Surgical births and their clinical sequelae (haemorrhage, pain, infection) are then linked with further deleterious effects such as difficulties breastfeeding or ‘bonding’ with babies, and slow recovery from the medical-surgical birth experience. Given that this discourse of intervention as ‘risky’ to natural birth has been successfully expounded for 30 years and more, and is part of antenatal teaching, it might be expected that women would consider this to be problematic. However, this rarely featured in the narrative accounts of women who planned to give birth in hospital OUs. Instead, they positioned surgical interventions such as caesarean section as life-saving moments of rescue, to be adopted quickly if the clinical situation suggested they may be required. Laura’s comment during an ‘end of pregnancy’ interview conveyed her sense of needing to ‘try’ for a ‘natural’ birth, but with recourse to surgery being part of that intention:

We just won’t let labour go on that long – if things aren’t progressing, we’ll ask for a caesarean section...I’ve done everything I can to be fit and healthy. I’ve done NCT [National Childbirth Trust¹] – very helpful apart from they’re a bit mad, and you have to take it with a pinch of salt. They are very pro- ‘active birth’ and anti-drug. (Laura, first baby, healthy pregnancy, planned OU birth; sister and mother had emergency caesareans)

In this extract, Laura discusses having worked hard to stay fit and healthy, and attending NCT birth preparation classes, both of which indicate self-presentation as a responsible, ‘fit’ parent-to-be. In England, most maternity hospitals offer birth preparation classes, but these are sometimes over-subscribed and women may opt for NCT classes, for which there is usually a charge. Not all women attend birth preparation classes, and a relatively small proportion access NCT classes. Laura’s reflection on what she perceived as the NCT ‘ethos’, and her willingness to request a caesarean section, suggests that to some extent, she wished to distance herself from, or at least to challenge the authenticity of, a ‘pro-“active birth” and anti-drug’ ethos, allowing her to retain the option of technological care during birth.

Prioritising the ‘iatrogenic’ risks of birth interventions and preference for AMU, FMU or home birth

So far, we have discussed findings related to views and perspectives of women who planned to give birth in hospital OU, and this group included women expecting their first or subsequent babies, with and without clinical ‘risk factors’. The rest of the sample (16 out of 41) planned to give birth either at home, in FMUs or in AMUs. As mentioned earlier, women expecting their first baby made up most of this group (10 out of 16), and almost all (15 out of 16) who wanted to give birth in a non-hospital setting had healthy pregnancies and straightforward obstetric histories.

In their accounts, these women referred to medical interventions during labour, and, as others have documented (Edwards 2005, Cheyney 2008), women who opted for non-OU births often intended to use non-pharmacological methods of pain relief, and to exert control over their birth, over the decisions made in labour, over the ambiance of the setting and presence of others in the birth environment. Marylin explained why she planned to give birth in an FMU as follows:

I just really don’t want to give birth in hospital [OU]. I don’t like the environment very much and I prefer it to be kind of more natural, and without intervention as much as possible. So hence why I prefer to go to the [FMU] because it’s kind of more natural and they kind of leave you to it, I don’t really want epidurals or anything like that, I just want to kind of keep active throughout and [...] and do it all that way. (Marylin, expecting first baby, planned FMU birth)

She also compared her own hopes with her friend’s induced labour, to explain why she wanted to avoid a ‘more forced on you’ birth:

…a close friend has given birth in hospital [OU] because she had to be induced, and the whole procedure [...] it just seems kind of more forced on you and more [...] scary, rather than just doing it at your own pace and dealing with it and the pain and everything that’s happening at that time yourself. So ideally I’d stay [home] as long as possible and then go to the FMU.

Alison also sought control over her birth environment, having seen some conflicting accounts of hospital OU environments, particularly postnatal wards, on internet discussion fora:

…you’ve got control over your environment, you can decide what position you’re in, whether you need something to eat or a bath or a scented candle or, you know, you might want none of those things, you might have time for none of those things… And being somewhere that is familiar and safe and happy and that is not intruded on by other people and their various dramas, positive or negative. And where you can control the cleanliness and the food and anything else, and you can go to your own bed afterwards and … yes. It just feels to me some … more comfortable. (Alison, expecting first baby, healthy pregnancy, planned home birth)

Hilary planned a home birth from the beginning of her pregnancy:

… I’ve kind of thought … that it’s worth aiming for the best-case [scenario] of being at home, and doing … I mean the things that I’ve either started doing or been thinking about doing have been pregnancy yoga, which is much more about kind of relaxation techniques and breathing than it is about, I think, flexibility or normal yoga as we would think of it. And I’ve been thinking about getting a doula [paid birthing companion] as well – again, because I’ve heard quite good things about them through internet and through my own research… (Hilary, expecting first baby, healthy pregnancy, planned home birth)

In these accounts, the importance of relaxing in a private environment with known carers, having ‘control’ during labour, using alternatives to pharmacological pain relief were all evident; the ‘risks’ posed by hospital to natural birth, particularly the interventive approach associated with hospital OUs, were central to women’s decisions to birth elsewhere. However, the medico-obstetric risks of birth, and concerns about safety, which were such an important feature of the narratives discussed earlier, were barely mentioned. Other ‘risks’ of hospital birth such as the busy environment and loss of privacy at a vulnerable time were also part of these narratives, but were cited as a reason for avoiding hospital OU, rather than something which has to be borne in order to secure the medical safety of specialist obstetric care.

Discussion

In this research, we sought not only to explore birth place decisions generally, but also to provide a contemporary account of preference for hospital OU care. We believe that our findings offer a more nuanced account of birth place preferences than that provided by the view that women’s heightened anxiety about birth risks is sufficient to explain OU preference. This has implications for continued use of a ‘polarisation’ argument, whereby ‘natural’ and ‘technological’ views of birth are seen as being in opposition to each other, and as providing an explanatory framework for birth place decisions. Rather than choosing one stance over another, it seemed that women’s responses were shaped by numerous discourses. These originated in family histories and childhood, incorporated national traditions of public health, and, for families who have moved to the United Kingdom in recent generations, the influence of home nation provider model of care seemed more powerful than the host nation’s approach. These factors contributed to the immediate cultural milieux of interviewees, which was influenced by close family birth stories, as well as national cultural identities.

Previous studies on this topic have also recognised that although the rationale for providing alternatives to hospital OU may be to enhance access to ‘natural’ or ‘low-technology’ birth, women in fact seek control over birth in diverse ways. Authors have argued that some women eschew technology and embrace a ‘natural’, social or holistic approach (Garcia et al. 1990, Davis-Floyd and Sargent 1997, Viisainen 2000, 2001, Edwards 2005), whilst others suggest that sense of control is enhanced through ‘technological’ birth and ‘choosing’ medical interventions such as epidural anaesthesia or caesarean section (see Davis-Floyd 1990, 1994, Donner 2003, Bryant et al. 2007). Existing interpretative research on birth place decisions also suggests that women planning non-OU births still speak of the ‘safety net’ that specialist or acute care provides (Cheyney 2008, Pitchforth et al. 2008), although whether this is because women are mindful of ‘medical-obstetric’ risks of birth when planning a ‘natural’ non-OU birth, or concerned with the need to present themselves as moral, responsible, rational and ‘fit’ parents-to-be (Marshall and Woollett 2000, Viisainen 2001, Nash 2011) remains uncertain.

Using an analytic approach which incorporated ideas of sociocultural risk ‘selection’ led to a more detailed identification of individual birth risk perspectives. The narrative method was valuable in this endeavour, because it provided an opportunity to compare accounts, and consider which issues were neglected, as well as which were privileged or prioritised. The importance of safety from risk or harm was evident amongst narrative accounts of women who planned hospital OU birth, and this was often built on foundations already well established in family histories, so that the setting selected chimed best with what women already expected of birth, and of themselves during birth; over the course of time, cultural discourses had further attenuated women’s sensitivity to and awareness of the ‘medical’ risks of birth. The ‘problem’ of medical intervention was therefore not recognised as such – instead, interventions were cast as an essential ‘rescue’ from the hazards or uncertainties of birth. This sits well with work by Zinn (2008, p. 445) and others (Taylor-Gooby 2000, Brown et al. 2011), which suggests that individuals seek to establish trust in institutions as a strategy for managing uncertainty. Zinn argues that ‘stable institutions’, established on the basis of dependable rules and protocols, might be viewed as a source of trust and security (2008, p. 448). As Scamell and Alaszewski (2012) point out, childbirth can be considered a ‘fateful moment’ in the sense outlined by Giddens (1991); a time when ‘people might choose to have recourse to more traditional authorities’ (p. 142). When maternity hospitals are thought of as sites of medical safety, it makes sense that women anticipate that these are the ‘best’ place for birth, even when adverse media coverage and an inquiry-based, ‘name and shame’ culture might reasonably be expected to unsettle this view. This may also help to explain why, when women planned hospital OU birth, the ‘risks’ that they discussed were not that they might experience iatrogenic complications, but instead included propensity for error, overcrowding and exposure to others’ misfortunes, which threatened their view of hospital as a benevolent or stable institution.

The need to seek sanctuary during birth, coupled with a belief system that is sensitised to messages of riskiness and potential tragedy, might be sufficient to ensure women give birth in hospital OUs, but to this is added a strong societal expectation that parents take steps to actively present themselves as responsible and to be seen ‘to do the right thing’ in terms of self-education and preparation, which for many, includes planning to give birth in a hospital OU.

Whilst this was a prevailing perspective amongst interviewees, others had of course taken a different stance, and once again, risk selection helped to highlight some of the reasons for this. The principal feature of narratives which favoured non-OU birth ‘alternatives’ (home, FMU and AMU birth) was that the facilities and presentation of settings, and the extent to which one would be given privacy, dignity and individualised caregiver support, were considered very important. Here, the hospital OU was no longer an obvious safe haven, but instead characterised as a monolith, resistant to individualised needs and concerns, and where strict adherence to protocol was threatening to ‘normality’, rather than the basis for much-needed rescue. In these narratives, women did not convey an immediate sense that something might ‘go wrong’, but rather an intuitive belief that if they trusted themselves, their bodies and their caregivers, they would feel safe. Adding weight to Zinn’s (2008) argument that individual decisions lie somewhere in between rationality and an emotional or heuristic belief that ‘this’ is right for them, these accounts still paid homage to the need for a ‘safety net’, but this was less central, allowing social, emotional and relational factors to be a prime motivation for selecting a particular setting for birth.

Very few women, representing a minority within the smaller group who opted for non-OU birth, actively took issue with the idea that birth was potentially risky and argued a case that birth without medical oversight was safer, suggesting either that the ‘natural’ birth argument has declined in recent decades, or that the recent proliferation of discourses of risk, blame and responsibility renders such a perspective difficult to voice in contemporary times. This may be out of step with epidemiological evidence that birth is generally very safe (Birthplace in England Collaborative Group 2011), but it remains persistent, and reflects the discourses and uncertainties facing women. These observations raise further questions about whether women are, in practice, able to make decisions that conflict with their immediate cultural reference point, whether this originates in their ‘natal’ family and nationality or in the cultural public and professional narratives which continue to position birth as risky despite evidence of safety.

Conclusion

Douglas and Wildavsky’s (1982) assertion that some risks are politically ‘selected’ to support a given argument, whilst others are overlooked, is, to some extent, supported in this analysis. However, the term ‘risk selection’ perhaps suggests too strongly that a conscious decision has been made to focus on particular risks, and to allow this to underpin decision-making. Rather, it would seem that decisions are informed by impressions and understandings gleaned over time, which merge to form a sense of risk, and that individuals may not be immediately aware of the source of their beliefs, feelings and expectations. On the other hand, Douglas and Wildavsky’s (1982) observation that the ‘line of normality’ provides a basis for risk differentiation proves consistently valuable; if hospital OU birth is considered ‘normal’, then women who elect to birth without doctors present may be held to account if death or mishap occurs. It might then be expected that this view could change if home and other non-OU settings for birth became culturally positioned as ‘normal’, but given the historical basis of women’s beliefs, such a change is unlikely to be rapid or even to occur within a generation. Given the prevailing national and international discourses of birth as predominantly risky, we argue that marginalisation of ‘alternative’ settings is likely to persist. It also seems likely that expressing views or preferences which are at odds with the prevailing culture will remain difficult, even in high-income countries with good health infrastructure and high-level policy support for birth in settings other than OUs.

Acknowledgements

The views and opinions expressed therein are those of the authors and do not necessarily reflect those of the NIHR, NHS or the Department of Health. We are grateful to those who gave their time and participated in interviews, and to the NHS trusts which supported this research.

Funding

The research was supported by the National Institute of Health Research [NIHR/RTF/08/01/022].

Notes

1. The National Childbirth Trust is a UK charity which aims to support parents, through providing information, networking and education for pregnancy and beyond.