When choice becomes limited: Women’s experiences of delay in labour

Abstract

Choice and patient involvement in decision-making are strong aspirations of contemporary healthcare. One of the most striking areas in which this is played out is maternity care where recent policy has focused on choice and supporting normal birth. However, birth is sometimes not straightforward and unanticipated complications can rapidly reduce choice. We draw on the accounts of women who experienced delay during labour with their first child. This occurs when progress is slow, and syntocinon is administered to strengthen and regulate contractions. Once delay has been recognized, the clinical circumstances limit choice. Drawing on Mol’s work on the logics of choice and care, we explore how, although often upsetting, women accepted that their choices and plans were no longer feasible. The majority were happy to defer to professionals who they regarded as having the necessary technical expertise, while some adopted a more traditional medical model and actively rejected involvement in decision-making altogether. Only a minority wanted to continue active involvement in decision-making, although the extent to which the possibility existed for them to do so was questionable. Women appeared to accept that their ideals of choice and involvement had to be abandoned, and that clinical circumstances legitimately changed events.

Introduction

Choice and patient involvement in decision-making are strong aspirations of contemporary healthcare (Greener, 2009; Le Grand, 1997) – embodied within policy discourse and codes of conduct for professionals (Department of Health, 2004a, 2010; General Medical Council, 2006). The move towards delivering person-centred care (Gerteis et al., 1993) has been endorsed through policy and professional statements emphasising its centrality to the delivery of good care. An increased emphasis on paying attention to and displaying suitable respect for patients’ values and preferences is framed as an important guard against the dangers of paternalism and autocratic practice on the part of professionals (Corrigan, 2003). Indeed, it has been argued that patients are now being primed for action rather than passivity (Armstrong, 2014).

However, attempts to translate these principles into practice reveal several potential problems. This article is primarily concerned with two and the potential interaction between them. First, the extent to which individual patients really welcome choice and actively wish to engage in making decisions about who provides their care, where and using what treatments or interventions is not clear (Fotaki et al., 2005), and there is some evidence that the continuing asymmetry within clinician–patient interactions is co-constructed between the two parties rather than being a simple matter of professional dominance (Pilnick and Dingwall, 2011). Policies emphasising choice tend to position patients in particular ways and in relation to other actors, most obviously professionals (Greener, 2009). In doing so, they can serve to offer up preferred identities for patients (McDonald et al., 2007), with choice often being argued to be based on a rational consumer model that may not always be well-suited to the healthcare context. As a result, many have argued that (at least some) choice models seem to ask rather a lot of patients, and it is not always clear that they want to take on this work (Clarke et al., 2006; Fotaki et al., 2005; Greener, 2009).

Second, decisions about which course of treatment or which intervention to pursue take place in a range of clinical circumstances. While some are undoubtedly amenable to patients being involved in decisions about their care (e.g. conditions for which different treatment options exist, but each carries different benefits and risks), there are others in which, it can be argued, patient involvement in decision-making is either not possible or not desirable. Indeed, some would argue that seeking to involve patients in these contexts is inappropriate and does not give sufficient credit to professional knowledge or training. Emergency situations raise important challenges to involvement in decision-making because the options available, as well as the time that would be needed to discuss them, can be extremely limited. Contemporary health empowerment discourses have been critiqued for their tendency to ignore or obscure the complex forms of dependence that characterise many healthcare experiences and situations (Henwood et al., 2011), and research has shown that even relatively well-established processes such as the taking of consent can be challenging in these circumstances (Dixon-Woods et al., 2006; Habiba et al., 2004).

As well as being important in their own rights, the potential interconnections between these two critiques of choice warrant consideration. Perhaps the most obvious connection is the contrast between patients who may say they want choice in the abstract but then find this an unwelcome burden, or simply inappropriate, when actually ill or injured (Schwartz, 2004). In exploring this tension, we turn to the work of Annemarie Mol (2008), who argues that, while they can sometimes be complementary, patient choice and good care are much more often at odds with each other. For Mol, the ideal of patient choice carries with it a whole set of assumptions (a ‘logic’) that acts as a mode of organising and interacting, of understanding and of distinguishing between good and bad outcomes. She argues that the ‘logic of choice’ assumes professionals limit themselves to presenting facts which the patient assesses in order to make his or her choice of desired outcome, and the professional then uses appropriate techniques to deliver this. However, deciding to do something is rarely enough to actually achieve it, and central to Mol’s critique is the idea that the ‘logic of choice’ unhelpfully focuses attention on discrete end products. This is unhelpful because, she argues, care is better understood as an interactive and often open-ended process that is shaped and re-shaped depending on its results. What is or is not achievable in any particular care context cannot always be known or clearly set out in advance, but rather is contextual and changeable; in care, time twists and turns and there is no crucial moment when all facts are known. By focusing on ‘end products’, the logic of choice oversimplifies the relationship between means and ends. It is for these reasons that Mol argues the ‘logic of choice’ is in tension with the ‘logic of care’, and that the latter is preferable in a great many situations. She concludes her work by calling for further explorations of how these two logics ‘interfere’ with each other in specific healthcare contexts:

That the logic of choice and the logic of care are so profoundly different begs the question as to what happens when these two modes of thinking and acting get mixed together – as they do in real life. The possible interferences are many. (Mol, 2008: 96)

This is precisely what this article seeks to explore.

Choice in maternity care

We take as our focus maternity care, a setting in which competing discourses about the most appropriate way to care for and support labouring women are well established (Walsh, 2010) and different options are (at least in theory) open to women (Miller and Shriver, 2012). While we acknowledge there are some important differences between maternity care and the diabetes care context within which Mol’s work developed, we believe it is appropriate to use her work here. While childbirth itself does not, of course, equate with illness, the women in our study had all experienced a particular complication (explained below) which, under current guidance, is managed medically.

Within the United Kingdom, a range of organisations are influential in shaping maternity care, including both professional bodies (such as the Royal College of Obstetricians and Gynaecologists and the Royal College of Midwives) and consumer groups (such as the National Childbirth Trust), with the National Institute for Health and Care Excellence (NICE) responsible for producing evidence-based guidance. Recent UK policy and practice in this area has focused on choice (Department of Health, 2004b, 2007; NICE, 2007, 2014; Royal College of Midwives, 2012), for example, over place of birth (the four options in the UK context being home birth, a freestanding midwife-led unit, a midwife-led unit alongside a hospital or a hospital obstetric unit or ‘labour ward’), and women are invited to develop a ‘birth plan’ (Kitzinger, 1992; Lothian, 2006) during pregnancy in order to record their preferences. The emphasis is therefore increasingly placed on women engaging in a process of information-seeking about the choices available and subsequently making decisions that best fit their preferences. This can be illustrated in the UK context by reference to the ‘Pregnancy Planner’ – a National Health Service (NHS)-provided online resource for pregnant women which explains a birth plan as follows:

A birth plan is a record of what you would like to happen during your labour and after the birth. You don’t have to create a birth plan but if you would like one your midwife will be able to help. Discussing a birth plan with your midwife will give you the chance to ask questions and find out more about what happens in labour. It also gives your midwife the chance to get to know you better and understand your feelings and priorities. (NHS Choices Website, 2015)

In principle, this would seem a good idea as there is evidence that the involvement in decision-making can improve women’s birth experiences and lead to better physical and emotional outcomes (Hodnett et al., 2010), and also that place of birth can impact women’s birth experiences (Overgaard et al., 2012). Pregnant women may vary enormously in their preferences – some women highly value easy and quick access to medical technology and welcome interventions such as an epidural for pain relief, others prefer to approach labour more ‘naturally’ and prefer to have little or no pain relief or other medical intervention (Lupton and Schmied, 2013).

The offer of choice in maternity care is typically based on the assumption that women are ‘low risk’, that there are options available and that the risk is comparable between the choices. However, birth is often not straightforward and unanticipated complications can rapidly reduce the scope for choice and possibly meaningful involvement in decision-making (Malacrida and Boulton, 2014). The best laid plans may not be achievable in practice, and choice suggests an element of equipoise of outcome that may not always reflect reality. For example, it is common for women who begin their labours in midwife-led units to be transferred to obstetric units, especially for first pregnancies (Rowe et al., 2012). Transfers may take place for clinical reasons (such as concerns for mother and/or baby) but can also happen if women decide to pursue an intervention not available in a midwife-led unit, such as an epidural for pain relief.

Delay during labour

The particular clinical focus of this article is delay during labour, in which contractions are either not frequent and/or strong enough for labour to progress. Once in established labour (regular painful contractions and progressive cervical dilation from 4 cm), assessment of progress includes cervical dilation. Delay is suspected if dilatation of less than 2 cm in 4 hours occurs and confirmed if progress of less than 1 cm is found 2 hours later. To facilitate progress during this 2-hour period, the woman will be encouraged to mobilise, consider hydration (e.g. a sports drink) and discuss appropriate and effective pain relief. If her membranes are still intact, artificial rupture will be advised.

If delay is confirmed, transfer to obstetric-led care takes place (if not already the case), and the use of syntocinon (a synthetic form of the hormone oxytocin) is recommended to increase the strength and frequency of contractions (NICE, 2007, 2014). The safety of mother and baby is routinely assessed by more intense monitoring by the midwife and obstetrician, and this normally includes support and effective pain relief, monitoring of the strength and frequency of contractions, the woman’s observations and fluid balance. Electronic foetal heart monitoring is routinely offered to detect signs of foetal hypoxia, should they occur. Progress is re-assessed after 4 hours of syntocinon and a decision made about birth.

Having delay during labour confirmed therefore has several implications for how the woman’s labour and birth progress. First, women who have chosen to begin their labour elsewhere (i.e. at home or in either type of midwife-led unit) will be transferred to an obstetric unit. Second, the need to ensure adequate pain relief means the majority of women in this situation have an epidural inserted (Kenyon et al., 2013). Given that preferences about pain relief are commonly occurring features of birth plans (Pennell et al., 2011), recommendations to have an epidural are likely often not to fit well with what women had planned, but may nevertheless be welcomed given that their labour is now prolonged and they are likely to be extremely tired. Third, while evidence suggests that, in its current regimens, the use of syntocinon can shorten labour by about 2 hours, it also shows it will not ultimately change the mode of birth, that is, women who would have had a caesarean will still ultimately do so (NICE, 2007). High-dose regimens have not been fully evaluated, but may reduce the likelihood of a caesarean and increase the likelihood of spontaneous vaginal birth (Mori et al., 2011; Wei et al., 2010).

While we acknowledge the important debates around the systemic or structural context and how this may be shaping care practices within the birth setting (e.g. the potential medicalisation of what many would regard as a natural process and the shifting patterns of professional involvement in birth), these are not the central focus of this article. Similarly, while we acknowledge those who argue that delay in labour may be more complex than a ‘mechanical’ physiological problem (Dempsey, 2013; Downe, 2010), it is not our intention in this article to consider the appropriateness of the medical interventions these women experienced.

Relatively little is known about the experiences of women who become delayed in labour, and the evidence that does exist presents a mixed picture. For example, two UK-based studies using questionnaires found that intervention for delay was not necessarily viewed as negative by women (Blanch et al., 1998; Lavender et al., 1999), while two more recent small-scale interview studies from Scandinavia have suggested that experiencing delay can be problematic and lead women to need particular support from health professionals (Kjaergaard et al., 2007; Nystedt et al., 2006).

In this article, we focus on how women experience having delay in labour confirmed, and the subsequent interventions that follow from this. In particular, we explore how women understand, and come to terms with, their labours not progressing as they would have wished, how they experience a reduction in choice about birth options and what this reveals about the relative importance of choice compared to other outcomes.

Methods

Women were recruited as part of a pilot study comparing high- and low-dose syntocinon for delay in labour, led by SK (Kenyon et al., 2013). Women were only eligible for the study when they had delay confirmed and had already opted to receive syntocinon (usual care constituted the low-dose arm of the study). The pilot study was based in three English maternity units and recruited 94 women over a period of 7 months between November 2010 and May 2011. All women who took part in the pilot study were invited to take part in this interview study. Women were invited to interview 2 weeks after birth (alongside the receipt of other pilot study–related follow-up such as questionnaires). Women received a material including an invitation letter, a Participant Information Leaflet and an interview reply slip, which they returned if they were willing to be interviewed. Reply slips were returned to the pilot study office and forwarded to NA, who led the qualitative element. She recorded women’s characteristics using data supplied by the pilot study office (study site, mode of birth and pain relief) and un-blinded to reveal allocation to study arm (low or high dose). Women were contacted, given the opportunity to ask any questions, and an interview time and venue were arranged. Written consent was taken at the interview itself. The interviewer (see Acknowledgements) remained blinded to study arm allocation throughout.

We had planned to sample purposively to include women from the low- and high-dose arms; women who had had a caesarean section, instrumental or spontaneous vaginal birth; and women who did or did not have an epidural during labour. However, a relatively low response rate meant that all women who agreed to be interviewed were followed up. In total, 19 women responded and 18 were interviewed (we lost contact with the remaining woman). Although we were unable to sample purposively in these circumstances, the final sample was diverse. The sample characteristics are shown in Table 1.

Table 1.

Interview sample characteristics.

High or standard dose arm	Standard: 8 High: 10
Mode of birth	Spontaneous vaginal birth: 2 Obstetric birth (either caesarean section or instrumental delivery): 16
Epidural	Yes: 12 No: 2 Spinal for birth: 4

A semi-structured topic guide with broad areas was developed from a literature review, discussions within the project team and input from our consumer representative (see Acknowledgements). While this was used to guide the interviews, the emphasis was on encouraging women to discuss their own perspectives freely.

All interviews were audio-recorded and transcribed verbatim, with permission from participants. The purpose of the interviews was to explore women’s understandings of the study and the information-giving and consent-taking processes adopted (not reported here; see Kenyon et al., 2013), as well as their views and experiences of labour and birth.

Data from the interviews were analysed using the constant comparative method (Glaser and Strauss, 1967), assisted by NVivo 8 software. Transcripts were read in detail and open codes were initially applied line-by-line to the data. The open codes were then incrementally grouped into organising categories or themes. These categories were modified and checked constantly and further open codes were incorporated as analysis proceeded. The categories and their specifications (the coding scheme) were then programmed into the software. The coding scheme was used to process the data set systematically by assigning each section of text to a category, according to the category specifications.

The pilot study, including this interview study element, was reviewed and given a favourable opinion by the Leicester, Northampton and Rutland 1 NHS Research Ethics Committee (reference 10/H0406/30).

Findings

Choices had been made, but had to be revisited

While there are, of course, some circumstances in which women’s options about labour and birth may be limited by clinical factors, the exclusion criteria for recruitment to the pilot study meant that only women who had nothing in their medical history to make them ‘high risk’ (such as gestational diabetes, existing maternal or foetal disease or concern, previous uterine surgery or vaginal bleeding in the pregnancy of clinical significance) were eligible for recruitment (Kenyon et al., 2013). This meant that all options were initially open for these women, and that they were therefore ideally placed to engage in the kind of information-seeking and decision-making processes embodied within the current policy discourse around maternity care.

The majority of women interviewed did indeed report having planned how they wanted their labour and birth to be to some degree, including where they wanted to give birth and what kind of pain relief they would like to use. The emphasis was most commonly on wanting ‘as natural a birth as possible’:

I went to the birthing centre. I tend to worry quite a lot about, erm, clinical aspects, I don’t like needles and I don’t like that sort of environment. It worries me a bit and makes me a bit anxious. (Participant 1)

I really didn’t want to have anything to try and combat the pain … it’s just I wanted to try and have as natural a birth as possible without pain relief, rather than intervention like that. (Participant 11)

As previously explained, having delay in labour confirmed necessitates a transfer to an obstetric unit if the woman has begun her labour in any other setting as this is the only place in which syntocinon can be administered. It is also likely that more pain relief will be advised, and the possibility of a caesarean section becomes more likely. As it became clear that the anticipated progress through labour was not being made, the possibility that plans would have to change was naturally upsetting and women commonly reflected in their interviews on the ‘ideal’ labour and birth that they felt they had lost:

I first went to the birthing centre as opposed to the delivery suite. That was my ideal labour, would have been deliver in the birthing centre … I probably went up to the ward about quarter past nine and then I didn’t really like it. (Participant 4)

[the doctor] was saying how do I feel about having an epidural [prior to syntocinon] and they went through the pros and cons of the epidural and I decided that I was going to have one and to be honest, I’d already thought all the way through that I was kind of against it and I did kind of want, I would have loved a water birth. (Participant 2)

There came a point at which choice receded

As explained above, the exclusion criteria for the pilot study meant that only ‘low-risk’ women were recruited. Before labour commenced, therefore, all options were open for these women, but, as time went on, the scope for choice diminished. Corporeal realities began to set the agenda. On the whole, although they often found it upsetting, women were accepting of the fact that the plans they had made were no longer feasible.

Fundamental to this widespread acceptance was a recognition on the part of women themselves that they were not making the anticipated progress through labour. In many cases, the women were acutely aware that their body was not doing what it needed to in order to progress the labour as their uterine contractions and cervical dilatation patterns were being regularly monitored by midwives. Even if concerns were not immediately conveyed to the woman, the midwife’s language or behaviour could suggest to them that things were not going as would be hoped (Scamell, 2011).

The cervix needs to reach full dilatation (10 cm) before the second stage of labour can begin and the baby be born – this is often referred to colloquially as reaching ‘the magic 10’. Women typically focused in on the number of centimetres of cervical dilation they had reached and were aware that they were not where they needed to be:

So it got to the stage where I was dilated at five centimetres but then that just stuck, nothing changed. (Participant 11)

We got to eight centimetres ok and then examined me again and part of the cervix wasn’t dilating further although most of it was but part of it wasn’t and so they said ‘ok, we’ll give it another couple of hours’, so we gave it another couple of hours and then examined again, it was still the same. (Participant 16)

This focus on talking in terms of the number of centimetres of cervical dilation reached appeared to serve as a form of objective measurement of their progress through the first stage of labour. The numbers reported to them by health professionals following examinations were accepted seemingly without question, and women could themselves appreciate that there was a discrepancy between where they were and where they needed to be. Repeat examinations that produced the same figures led them to accept that their progress had become ‘stuck’. It was at this point that women commonly began to draw distinctions between what they had hoped would happen during their labour and birth and what their situation actually was:

In my ideal world I would have just had a water birth but that wasn’t feasible so I was open to suggestions. (Participant 2)

Coping with diminishing choice

Having accepted that their labour and birth were not progressing as they would have wished, there were differences in the ways in which women talked about how they went about approaching their revised circumstances and the degree to which they wished to stay involved in decision-making. The majority of women interviewed were ultimately willing to cede control of decision-making about how their labour progressed to the health professionals caring for them. In many ways, this was similar to entering the sick role in that they appeared to accept that they no longer knew what needed to happen and were happy to defer to those they regarded as having the technical expertise required to manage the changed situation effectively and make decisions on their behalf (Parsons, 1975; Williams, 2005):

I have no knowledge of birth, I’m not a midwife, I have no knowledge, so I think I was very much open to suggestion and open to what they were saying to me and always felt that whatever they were saying was always going to be in mine or baby’s best interests anyway. (Participant 4)

There appeared to be some kind of ‘tipping point’ at which the situation changed from being about what their preferences would be from a range of possible options (as it had been for place of birth, type of pain relief, etc.) to one in which the possibilities were much reduced and were being driven by clinical necessity rather than patient preference. Importantly, women who adopted this approach accepted the legitimacy of this change as it was being presented to them – typically because they trusted the health professionals caring for them:

I did feel that I had control over what was going on, what was happening, up until the point where they said we’ve got to get baby out and, at that point, I just thought ‘well whatever these guys think they need to do now’. (Participant 17)

At that point I was happy really to trust whatever they were saying to me … I was fairly trusting in the people that were advising me or sort of telling me what the options were. (Participant 5)

This is not to say that these women were always necessarily particularly happy with what ultimately happened to them (e.g. having a caesarean), but they did accept the legitimacy of health professionals’ assessments of their situation. The fact that their baby may be at risk added an extra layer of complexity – there was such a lot at stake and there did not seem to be any other option. This can be seen in the two data extracts below, both from women who ultimately gave birth via caesarean section:

DID YOU THINK YOU FELT INVOLVED IN THE DECISIONS THAT WERE MADE THROUGHOUT LABOUR AND BIRTH?

Yes to some degree. You can only have so much, I mean I cannot refuse a c-section because [of] the risk of my life and my baby’s life. (Participant 8)

He [health professional] came in and basically started talking to me then about the possibility of maybe having to have a caesarean but they don’t know for definite. So at that point then I was, like, ‘no, no, I would do anything, don’t give me a caesarean, I really don’t want a caesarean. I have come this far, I really want to have her naturally, I don’t want to have anybody take me down to theatre’. He said ‘well you might not have an option’. (Participant 1)

Combined with their acceptance of the legitimacy of these professionals’ technical expertise, the women were also very aware that their resources were rapidly depleting – they were very tired, frequently in a great deal of pain, and often affected to some degree by the effects of pain relief. They had reached a point at which they were happy to let someone else take charge and manage the situation:

By this point I was exhausted and the birth plan had well and truly, could have been ripped up anyway. (Participant 19)

Some women went further and actively tried to reject any kind of involvement in making decisions about what to do. While the women discussed above were happy to defer ultimate decision-making to health professionals, this smaller group went beyond this and actively sought to detach or remove themselves from the situation completely. While they of course could not do this bodily, they talked about mentally or emotionally seeking to withdraw from the situation as a coping mechanism and just wait for it all to be over:

I realised there’s no way of getting away from this [situation] and just thinking ‘I just need to pretend I’m dead’, like just completely take myself out of the whole situation and just shut down to get through it. (Participant 9)

It sounds very strange but I almost wasn’t very interested, you know … I was happy to just let everybody else worry about what was happening. (Participant 16)

So deep seated was these women’s desire to absent themselves that, in these cases, even involvement at the level of being asked to sign consent forms (e.g. for a caesarean section) was experienced as an unwelcome intrusion. In common with work exploring the role of patient consent in emergency surgical situations (Habiba et al., 2004), the seeking of consent was experienced as problematic and interpreted as largely tokenistic or ritualistic rather than having any real meaning or significance:

I definitely wasn’t in the frame of mind you’d normally get someone to consent for something in. (Participant 16)

Only a small minority of women reported having been keen to stay involved in the decision-making process and being unhappy if they felt they were being left out. In the extract below, one such woman recounts how she challenged health professionals who, she felt, were leaving her out of important discussions:

They were talking [about a possible] caesarean and I did actually pipe up and say ‘you’re talking about me in a room, talk to me, if caesarean is what you’re considering, then I want to be part of that decision-making’ and at which point, the, she was a surgeon, I presume she was a consultant as well came back and said ‘I am sorry, force of habit’ and we discussed the options. (Participant 6)

In contrast to those women who were happy to defer to health professionals who they regarded as having the necessary technical expertise to manage the situation on their behalf, for this woman the principle of maintaining involvement and being meaningfully consulted on what might have to happen retained its importance. These few women are an interesting exception, and it is debatable whether they really were involved in making decisions in any meaningful way as, given the clinical circumstances, the options were really very limited.

Discussion

Choice and patient involvement in decision-making are strong aspirations of contemporary healthcare, but translating these somewhat abstract principles into practice is often far from straightforward, and it has been argued that politicians have stoked up choice as something which doctors and patients often do not recognise and/or cannot achieve (Greener, 2009; Newman and Vidler, 2006). In this article, we sought to explore how these ideals of choice and patient involvement in decision-making may be disrupted by unanticipated complications that can rapidly limit the extent to which either is likely to be achievable in practice. By drawing on the accounts of women who experienced delay during labour with their first child, we have examined how these women experienced the sometimes rapid reduction in the potential for them to exercise choice about how their labour and birth progressed. We situated our work in the context of Annemarie Mol’s (2008) work on the inherent tension she sees between the ‘logic of choice’ and the ‘logic of care’.

We have demonstrated how, although they often found it upsetting, women commonly accepted that the choices and plans they had made about how they wanted their labours and births to be were no longer feasible as their labours became ‘abnormal’. Women appeared to accept that the ideal of making choices that fitted with their values and preferences had to be abandoned, and that clinical circumstances legitimately changed events. The majority were willing to defer to clinical staff who they regarded as having the necessary technical expertise, while some women actively rejected any involvement in decision-making altogether appearing not even to want to be kept informed. Only a minority sought to continue an active role in decision-making, although it is not clear to what extent this was actually possible.

What is noticeable is that, for many women, the plans they had made were let go fairly easily – they ‘went out the window’ or were ‘ripped up’. These were women having their first child, so there is some recognition on their part that, with the benefit of hindsight, they had been ‘planning in the dark’ as they had little idea what labour and birth would be like and their ideal hopes and expectations were often very different to their actual experiences (Lally et al., 2008; Pennell et al., 2011). In situations such as this in which corporeal realities set the agenda, and the women accepted that the safety of either themselves or their baby was potentially at risk, ideals of patient choice and involvement in decision-making appeared to be readily abandoned and were sometimes completely inverted. The group of women who sought to detach themselves as fully as possible from the situation adopted a very passive patient role and could be understood as wanting to be ‘rescued’ by health professionals (Habiba et al., 2004). While ‘informed consent’ was sought for each intervention, women talked about these interventions as ‘not being an option’ – there did not seem to be a choice to be made anymore.

It is interesting to reflect on how and why women were apparently so able to reconcile themselves to their changed circumstances and be prepared to accept that the choices and plans they had made could not now be followed. As Mol (2008) has suggested, the ‘logic of care’ is characterised by fluidity and as an interactive and on-going process which is shaped and re-shaped depending on its results. What seems to be the case here is that the slow progress of their labours had given women time to adjust to, and accept, the fact that their birth would not be as they had planned or would have liked. This longitudinal aspect, combined with the apparent objectivity of the degree of cervical dilation measured and reported at each examination, perhaps meant that these women were more able to accept their changed situation than if it had been presented to them completely out of the blue and with no prior warning.

Having accepted the legitimacy of having to do things differently to how they may have liked or planned, and that corporeal realities were now setting the agenda, the priority for these women very firmly became the safe delivery of their baby, rather than their own preferences and choices about labour and birth. It seems, then, that the emphasis switches very clearly from what they may want to what their baby needs. What does this tell us about the relative importance of choice compared to other outcomes? It would appear that, in this context at least, choice becomes framed as associated with the mother and for her benefit, and that, while this may be nice to have where possible, it does not come above the safety of the baby. When the latter appears to be at risk, the former loses any significance. Mol (2008) has argued that one of the problems with the ‘logic of choice’ is that it focuses on discrete end products that, it assumes, are all deliverable by health professionals. Before their labours began, the majority of women we interviewed had made choices about how they wanted their labours and births to be, typically this meant a ‘natural’ process with minimal (ideally no) medical intervention. The ‘end product’ here was the birth itself. The unanticipated complications they experienced served to make them re-evaluate what the important ‘end product’ actually was, and they subsequently focused on a safe and healthy baby that must be achieved through any means necessary.

While the current socio-cultural context can be argued to make adopting an apparently passive patient role difficult (Lupton, 1997), the women in this study successfully managed the transition from being active choosers to relying on doctors’ judgement. While the contemporary discourses and policy statements about choice and patient involvement position patients and professionals as equal partners (Mol, 2008), it is clear that the vast majority of women in this study were very willing to adopt instead an asymmetrical relationship with those caring for them (Pilnick and Dingwall, 2011), preferring to place their trust in professionals to make choices and decisions in their best interests. Women talked very clearly about recognising and respecting the technical knowledge and expertise they regarded these professionals as possessing. They were prepared to put their trust in them with the expectation that whatever they did would be in the best interests of both them and their babies. Indeed, while discourses of choice may be argued to offer up preferred identities to patients (McDonald et al., 2007; Mol, 2008), in this context at least the discourse around motherhood was also a powerful force – a ‘good mother’ relies on the doctor’s expertise to keep her baby safe from harm (Miller and Shriver, 2012) rather than pursuing her preference for a vaginal birth when the doctor tells her that her baby is showing signs of distress and a caesarean section is needed.

If choice appears to be something of a ‘red herring’ in this context, then what can we learn from our data about what is important to women and how they can best be supported during this type of event? The importance of aspiring to offer choice and involvement in decision-making must be balanced with the need to keep other possible outcomes always in mind. As Mol (2008) has argued, the ‘logic of care’ places the fluidity and uncertainty of care centre-stage; the ‘logic of choice’, in contrast, fails to do so. While not so evident in our study, evidence from elsewhere shows that many women who have emergency caesarean sections have not thought about the possibility antenatally (Murphy et al., 2003). Knowing about, and being prepared for, what may potentially happen if things do not go to plan may be important in helping women cope in these circumstances.

This study has some important limitations. First, all of the mothers and babies in the pilot study, and therefore eligible for recruitment to the interview element, were healthy after birth and the views and experiences of others with poorer outcomes may be different. Second, the data on which this article is based are drawn from interviews conducted after the event and not on direct observation of these women and their encounters with health professionals during their labours and births. Third, due to the lower than anticipated response rate, the sample size is smaller than we would ideally have liked, although the sample is heterogeneous and includes women from the low- and high-dose arms of the pilot study; those who had a caesarean section, instrumental or spontaneous vaginal birth; and those who did or did not have an epidural during labour. Furthermore, analysis showed that the demographic characteristics of the interview sample did not differ significantly from the wider pilot study population.

Notwithstanding these limitations, this study makes an important contribution to understanding how women who have been offered, and engaged with, the aspirations of choice and involvement in decision-making cope with experiencing a clinical situation in which choices rapidly recede. Their acceptance of the changed circumstances as legitimately limiting choice, combined with the willingness of many to defer to the technical knowledge and expertise of professionals, clearly demonstrates that, in this context at least, the importance attached to choice diminishes rapidly in favour of other outcomes. To return to Mol’s (2008) call for more empirical work on the interferences that happen when the ‘logic of choice’ and the ‘logic of care’ get mixed together in real life, our work has shown that the latter can very easily displace the former and be accepted as the preferred and most appropriate mode by all concerned. Deborah Lupton (1997) has argued that in interactions with doctors, patients may pursue both consumerist and passive patient subject positions simultaneously and variously due to the complex and changeable nature of healthcare. Even those supportive of a consumerist model generally would, she argues, place their trust in doctors to make decisions on their behalf on some occasions. As we highlighted in our introduction to this article, contemporary health empowerment discourses tend to ignore or obscure the complex forms of dependence that characterise many healthcare experiences and situations (Henwood et al., 2011). Yet, dependency is a central feature of much illness experience and works against the full taking-up of a consumerist approach – patients cannot always be ideal-type consumers (Lupton, 1997). The challenging role for health professionals is to effectively bring about a smooth transition from the ‘logic of choice’ into the ‘logic of care’ – to do so in a person-centred way that affords people dignity, compassion and respect and offers them personalised care and support (Health Foundation, 2014). The accounts of women we interviewed for this study suggested that, in the vast majority of cases, they believed this had been achieved. We have already highlighted as a limitation of this work that it is based solely on the accounts of women shortly after the event and not on observations in real time. Future observational work of this kind would be very valuable in further understanding how health professionals working in such contexts can best manage this transition.