Women requesting Caesareans: ethical implications in light of the new National Institute for Clinical Excellence guidelines

At what cost?

The International Federation of Gynaecology and Obstetrics (FIGO) committee for the Ethical Aspects of Human Reproduction argued that CSMR was unethical, as there was insufficient evidence to suggest a net benefit to health. ³ Although it was written over 10 years ago, the statement still resonates today. Holding the idea that CSMR should not ‘dent the resources of public health care’ ⁴ one could infer that the total risks associated with CSMR outweigh the benefits and it is therefore unjustifiable in a health-care service that strives to allocate resources with a net benefit.

NICE also recently assessed how worthy CSMR was of National Health Service (NHS) money. ⁴ In an analysis that considered both the costs of birth and ‘downstream’ costs, it found that planned vaginal delivery was approximately £800 cheaper than a maternal request CS. It would follow that the NHS could save £5.6 million for every percentage point reduction in CSs. However, it then goes on to say that because other outcomes such as urinary incontinence were not considered, the findings were less certain. It concludes that on balance the model does not provide strong evidence to refuse a woman's request for CS on cost-effectiveness grounds. This fact, along with the knowledge that denying a request could cause psychological harm in a woman with a fear of childbirth, leads to the conclusion that she should be supported in her wish for CS, and the NHS will fund it.

This utilitarian approach to CSMR is, however, problematic. What constitutes a net benefit to health in the case of labour and delivery is not a straightforward question to answer. Obstetric practice is a balancing act, an attempt to reduce the morbidity and mortality associated with delivery for both mother and baby. For the baby this means minimizing the chances of death, disability and injury during delivery. And for the mother it means minimizing damage to the pelvic floor and perineum and providing psychological stability by allowing her the birth she wants. The psychological relief that could accompany a planned CS for a woman suffering from a morbid fear of vaginal delivery - tokophobia - surely contributes to a better outcome, for she will be less distressed in the lead up to delivery and may be able to enjoy the process of giving birth, rather than fearing it. This may contribute to a better overall birth experience, making it easier for her to bond with her baby and increasing the chances of successful breastfeeding. Obstetricians, in caring for their patients as a whole, are therefore obliged to consider both the mental and physical health of the mother, again making utilitarian analyses of ‘net benefit’ difficult to apply in practice.

Utilitarian analyses of CSMR are also hampered by a lack of evidence. There is very little unbiased data comparing healthy women who opt for vaginal delivery or elective CS. ⁵ As such, both obstetricians and patients cannot always assess the likely consequences of their respective decisions.

In fact, it seems improbable that there will ever be a randomized controlled trial carried out where women are allocated either to vaginal delivery or elective CS. Currently we rely on observational data: comparisons of women who have elective CS for medical reasons, and those who have vaginal deliveries. These are obviously confounded by the medical problems existing in the women before CS, so it is impossible to compare the outcomes. This situation is unusual: typically when the safety of a medical procedure is unknown, trials are set up to at least limit access to the procedure, so as to monitor those who undergo it and follow-up the outcomes. After a fixed amount of time, it might be decided that the procedure is indeed safe, even if it is impossible to compare it with a control. However, in the case of CSMR, it remains widely practised with no clear evidence. One can only assume this is because of the inference that if CS carries minimally increased risks for women with medical indications, the risks for otherwise healthy women must surely be lower and are thus acceptable.

NICE rules

In contrast to these broad utilitarian approaches, the NICE guidelines on CSMR focus on rule-governed approaches to specific clinical scenarios. The previous criteria published by NICE in April 2004 offered a list of indications for planned elective CS. ⁶ This list included major or minor placenta praevia, pregnant women who are HIV-positive and those with a twin pregnancy where the first twin is not cephalic. These indications provided a basic set of rules for obstetricians to follow when considering a planned elective CS. However, these rules should not be read as a set of moral imperatives, such as the Hippocratic Oath. Instead, these guidelines emphasize the need to consider the specific consequences of obstetric interventions in light of certain conditions: for example, HIV. As such, they are best read in terms of rule consequentialism, rather than deontology.

Under the April 2004 guidance, a maternal request for CS was ‘not on its own an indication for CS’. Obstetricians faced with a maternal request for CS were instead advised as follows: •

Explore and discuss specific reasons;

These guidelines rightly suggested that it is imperative that the woman understands the risks and benefits of such a procedure if she is to agree to it; she must give informed consent. The last point makes it clear that an obstetrician could refuse to perform the Caesarean, but that they were bound to refer the woman to another obstetrician for a second opinion. (This is not to say the second obstetrician would necessarily agree.) A woman intent on having a CS may have been given another chance, but this was not a guarantee.

While acknowledging some of the complexities of the decision-making process, the April 2004 NICE guidelines still only provided obstetricians with a set of rules to be followed in the face of a maternal request. A number of problems arise from such an approach.

It is widely acknowledged that obstetric practice varies significantly from country to country, around the world and within Europe. This is the case for CSMR. ⁵ Differences in attitude do not seem to be based on scientific evidence; instead cultural factors, legal liability and variations in perinatal care all seem to play a role. ⁵ In the study reported by Habiba et al. ⁵ , obstetricians’ compliance with a hypothetical woman's request for CS simply because it was ‘her choice’ varied between 15% in Spain and 19% in France, to 75% in Germany and 79% in the UK. These stark differences cannot be explained quickly or simply, but would require an in-depth analysis of the medical culture among patients and medical staff alike. For our purposes here, I will focus on practice in the UK alone.

A rate of 79% compliance among a sample of obstetricians surveyed in the UK is high relative to other European countries. What reason might an obstetrician have for accepting a woman's request for CS when there are no medical indications and the surgery itself will put her at risk? Some might say that the woman's choice itself is enough to agree to a CS if that is what she wants. Valuing the autonomy of the patient, a doctor might argue that the woman's right to choose how she delivers her baby is paramount, and they will not be the one to tell her that she has to deliver vaginally. A woman suffering with tokophobia presents another reason to agree, but there is arguably a role for psychotherapy and counselling here. Previous CS is a persuasive reason for UK obstetricians, with 98% of those surveyed agreeing to a woman's request with this factor in mind. ⁵ Other reasons include previous traumatic vaginal delivery, previous intrapartum death or the first child being disabled. Obstetricians in Britain seem to value the woman's choice highly, as they will often agree to CS out of respect for the woman's wishes.

Rule-based approaches to CSMR, such as the April 2004 NICE guidelines, do a good job of directing obstetricians to engage with patients’ concerns, but they offer little guidance on mediating actual decisions in light of the range of reasons obstetricians invoke for both accepting and declining maternal requests.

Who requests a CS and why?

To understand this problem in more detail, it is helpful to consider the reasons women themselves request CS, even when they know about the increased morbidity and mortality. The reasons may be psychological: an intense fear of anyone examining the vagina due to sexual assault in the past; they may be socio-cultural: a culturally held belief that vaginal birth is unpleasant and deforming; they may simply be for reasons of convenience and the desire to plan the date of delivery. ⁷ Often they may be due to traumatic experience in a previous delivery, and the wish to avoid this happening again. Some of the reasons will be deeply held and immovable, whereas others may be open to change, especially if they are based upon misinformation.

It is estimated that 6-10% of women in the UK suffer from tokophobia. ⁵ The fear appears to be a combination of fear of pelvic floor injury, fear of a traumatic emergency CS, fear of losing the baby and fear of being left alone in labour. ⁵ These fears are understandable and it is likely that all pregnant women experience them to some extent. But in women with tokophobia, the thought of labour and vaginal delivery causes such stress and worry that they are unable to contemplate a trial of labour and insist upon CS. The management of women with this level of tokophobia is to refer them for psychological counselling, where their fears can be explored, in the hope of reducing the fear associated with childbirth so that they might attempt a vaginal delivery. The management of a woman who returns from counselling with her fear intact presents a problem to the obstetrician, who must now decide whether to grant her the CS she wishes. Most doctors would argue that her mental health is one aspect of her health care that the obstetrician is duty bound to protect, and that attempting vaginal delivery would be detrimental to her mental state. ⁸ Others, including the outspoken critic of obstetric intervention Marsden Wagner, are not so sympathetic: ‘Vaginal birth is the consequence of being pregnant, a state for which the woman and her sexual partner must take responsibility, not the medical profession’. ⁹

In the previous NICE guidelines, the obstetrician was permitted to decline a request for CSMR. This suggests implicitly that some obstetricians may have reservations about operating on perfectly healthy women, when CS is not indicated. They may have ethical concerns about subjecting her to the risks of surgery merely because she wants it. Again, the rule-based approach offered by NICE cannot help mediate these tensions.

NICE rights

The new NICE guidelines that appeared in November 2011 ² contain the following recommendations concerning CSMR: •

1.2.9.5. For women requesting a CS, if after discussion and offer of support (including perinatal mental health support for women with anxiety about childbirth), a vaginal delivery is still not an acceptable option, offer a planned CS.

Compared with the previous guidelines, where the obstetrician either agreed or disagreed with the request for CS once a woman had been fully informed and possibly attended counselling, the algorithm has been tweaked. Now, if the counselling does not work, so to speak, the obstetrician is guided to ‘offer a planned CS’. This apparent order to offer CS is followed by a caveat, which is sensitive to the range of views on CSMR within the obstetric profession: if the obstetrician is unwilling to perform the CS then they are not obliged to. They must simply refer to someone who will. In the discussion following the initial recommendations in October 2011, the Guideline Development Group stated that it:

This statement introduces the language of deontology into the new NICE guidelines: rights and duties now play a much greater role in framing the rule-based guidance. This produces a conflict. The statement suggests that an individual obstetrician has the right to decline the request, but that as a wider body, obstetricians have a duty to comply. Similarly, it appears to increase women's power, granting a ‘right to express a preference for a CS’. The guideline presumes that if the first obstetrician refuses to perform the CS, the second one will agree. However, with the new obstetricians’ rights, any obstetrician could theoretically refuse, so the right of the woman depends on the way the obstetrician makes use of his/her rights. For a given woman and her obstetrician, the rights cannot be exercised simultaneously.

The press was quick to comment on the new recommendations, with The Sunday Times running a front page article: ‘AH women get right to Caesarean section’ on 30 October 2011. However, a statement from the Royal College of Obstetricians and Gynaecologists on 31 October declared this was misleading:

This statement emphasizes that women do not have an ‘automatic right’ to CS. The analysis above of the NICE guideline explains why this is the case. The woman's right is to express a preference but the management of this preference depends on the way in which the obstetrician exercises his or her own rights.

The new NICE guidelines therefore marry the rule-based guidance of previous guidelines with a more explicit deontology: rights and duties now play a much more prominent role. However, the rights of obstetricians and patients are also placed in conflict. This is not necessarily a bad thing; the new NICE guidelines will encourage obstetricians to engage more regularly with maternal requests for Caesareans. However, the tensions which follow cannot be mediated by NICE's rule and right-based advice alone. In addition, obstetricians need guidance on the kind of personal characteristics needed in order to approach maternal requests in an ethical manner. Here, recent developments in the application of virtue ethics to medicine should prove instructive.

The virtuous obstetrician

A virtue-based approach to medical ethics provides two core advantages over rule- and right-based guidance alone. First, virtues allow doctors to make good ethical judgements in the application of rules. Second, virtues ensure doctors conduct themselves in an ethical manner which extends beyond that which is prescribed by rules, rights and duties. Virtues, therefore, should be understood to supplement rather than replace deontology and consequentialism. ¹¹

One virtue in particular will be central for obstetricians approaching maternal requests for Caesareans in light of the new NICE guidelines: that virtue is temperance. In the context of medicine, it is understood as the ability to protect against under-treatment while also avoiding ‘inappropriate overtreatment’. ¹²

Temperance will aid in resolving the tension between the rights of obstetricians and patients. Should an individual obstetrician exercise their right to refuse a request for CS, they must be confident that their action will not lead to under-treatment. In such cases, obstetricians must take into account the possible psychological consequences of both CS and natural delivery discussed earlier in this paper. Furthermore, it is precisely a failure of temperance which permeates the thinking of certain critics of obstetric practice, such as Marsden Wagner:

Demonizing obstetrics as a patriarchal force intent on intervening in every birth whether it be normal or not, detracts from the question of how women see their bodies and how they want to give birth. It is good to congratulate women when they deliver at home using next to no pain relief, but for those women who require intervention, not conforming to this ‘natural’ model can feel like a failure. Whether women in labour should shun the medics or embrace them remains a decision that can only be made by the individual women themselves. The temperate obstetrician, with an eye to balancing under-treatment against over-treatment, manages rather than obfuscates patient choice.

Temperance will also be needed to protect against reckless over-treatment. Under the new NICE guidelines it would be possible for an obstetrician to grant a CS to every woman who requested one. However, despite adhering to the rules, this would not guarantee ethical practice. There is a point at which over-treatment becomes morally inappropriate even if it is technically permitted by NICE.

But what exactly would constitute inappropriate over-treatment? As discussed earlier in this paper, a woman with obvious psychological distress surrounding thoughts of vaginal delivery may benefit from an elective CS. However, in the absence of any psychological factors, there are at least two ways in which CSMR can be considered inappropriate. First, CSMR brings with it increased risks to both mother and baby. For the mother, CS constitutes major abdominal surgery with a significant recovery time and risks of infection, bleeding and numerous other possible complications. For the baby, there is an increased incidence of early respiratory problems. There also remains the chance that, even with elective CS scheduled for 39 weeks, the woman may go into labour spontaneously before then. The elective Caesarean would then become an emergency Caesarean, bringing with it all the additional associated risks including aspiration and increased maternal mortality.

Second, CSMR may constitute inappropriate over-treatment in the sense of wasted resources. The case of a young middle-class couple highlights this issue. Encountered in an antenatal clinic at a district general hospital in the South East of England, both young and healthy, the couple was in their first pregnancy. With three previous miscarriages, they had conceived again spontaneously and there had been no complications in the pregnancy. The woman was at 34 weeks gestation and they were requesting a CS. When the consultant obstetrician explored the reasons, it appeared that the woman was not scared of childbirth; she did not profess fear of the pain of labour, she simply did not want to labour. She was healthy otherwise and apart from her prior history of miscarriages, had no other obstetric history. She wanted a predictable delivery, as she considered herself high risk. This woman and her husband worked at the hospital. Clearly they felt they had a right to a CS; they had just heard the news of the new NICE guidelines. But the obstetrician was unwilling to agree to a CS. She referred the couple to another obstetrician in the hospital, who - she told me later - would ‘possibly give in’. The only reason for agreeing to a CS in this case would be out of respect for the woman's choice, considering that she wanted a predictable delivery, possibly out of anxiety that something might go wrong during labour. Although this couple may have been able to pay for the operation themselves privately, or at least contribute towards the cost, the new guidelines supported their case for CS, with the NHS paying the price. Granting this woman a CS could therefore constitute inappropriate over-treatment in terms of wasted resources.

Conclusion

CSMR is a complicated debate within obstetrics. It forces one to explore the ethics of surgical procedures that lack medical indication, and what possible psychological benefits they can offer. Analysis of the changes to the NICE guidelines on CSMR extracted their exact meaning. The new NICE guideline will reinforce tensions between obstetricians and their patients: the rule-based approach found in the previous guidelines is now supplemented by the language of deontology. This is not necessarily a bad thing if it ensures obstetricians more readily engage with their patients’ concerns. However, the new ‘rights’ gained by women and obstetricians are at odds with one another and will remain mutually exclusive in a clinical setting. In making this important decision, who should have the final say - the woman or the doctor? If one has power over the other, the mutual teamwork so necessary to a happy, successful delivery is threatened. A virtue-based approach to CSMR is therefore needed to mediate such tensions. This should supplement the rule- and right-based approach found within the NICE guidelines. While a range of common virtues, such as compassion, are often applied rather broadly to medical practice, this paper argued that temperance in particular will be central for dealing with the specific clinical scenario of CSMR. How the profession can encourage the cultivation of temperance remains a subject for future research. To date, studies have concentrated narrowly on the reduction of medicalization rather than the cultivation of virtue. Nonetheless, developing attitudes similar to those identified by Johanson et al., ¹³ such as pride in a low CS rate, may prove relevant. It is also impossible not to consider the cost implications of the guidelines to the NHS. With the new guidelines, obstetricians will face new challenges from women who believe they have a right to CS. Although this may not technically be correct, their case has certainly been strengthened. There is a possibility that more women will begin to request CS and this may cost the NHS more than it currently spends. Within the context of NHS cutbacks and national austerity measures, many may question why this recommendation has been made now. But it also increases the amount of choice women have in their deliveries. It will create a world in which we do not allow just natural choices; we allow choices that span the spectrum of medicalization, all the way to elective Caesarean. Temperance will be central in navigating such a world in an appropriately ethical manner.