Vaginal

In Australia, the maternal mortality ratio was 8.5 deaths per 100 000 women in 2016.19 Leading causes of maternal death are pre-eclampsia, obstetric haemorrhage, sepsis and cardiovascular problems.20 A vaginal birth or a trial of labour can range from a birth requiring no external or medicative assistance to the use of interventions such as forceps or vacuum extraction and the use of analgesic medications. Further despite all efforts, a caesarean section can still be required following a trial of labour.

The risk of perineal tearing, and its associated complications, is an obvious concern for pregnant women. Approximately, 3% of all women who give birth vaginally in Australia suffer a third or fourth-degree perineum tear,21 22 for first-time mothers, this rate is increased to represent 5%22 of all vaginal births. A third-degree perineal tear involves an injury to the anal sphincter, whereas a fourth-degree tear also involves damage to the anal mucosa (or lining). Without repair, the tears can result in long-term consequences such as ‘continued perineal pain, faecal incontinence, painful sexual intercourse, reduced quality of life and depression’.22 The risk of perineal tears increases to 7.3% with instrument-assisted vaginal births. There has been a rise in reported rates of severe perineal trauma, but this may reflect an increase in diagnosis as opposed to changes in clinical care.23

One risk of pregnancy and perineal tearing is the potential to increase the chance of urinary incontinence shortly following birth and long term. A study found that at 12 years following childbirth, 52.7% of women reported urinary incontinence at some point, 23% reported weekly occasions of incontinence and 5.4% reported daily occurrences.24 Similarly, childbirth is associated with faecal incontinence. Women suffering from faecal incontinence report having a reduced quality of life, utilised pads, plugs or constipating medicine as a result. Incontinence is greatly reduced in women who have exclusively had caesarean births, indicating the procedure has a protective element.24

Vaginal birth increases a woman’s risk of pelvic organ prolapse to 29% as compared with 5% for women who have only had caesarean births.25 This risk increases further with the use of forceps in vaginal delivery.25 An obstetric fistula is a less common risk of vaginal birth in Australia, associated with obstructed labour occurring in two in 100 000 women.25

Caesarean section

Caesarean births are not without complication, involving a ‘major, open-abdominal procedure, often performed in an emergency setting’.26 As with any surgery, risks include blood loss, wound infection, deep vein thrombosis, pulmonary embolus, damage to adjacent organs necessitating corrective surgery, slower recovery and risks associated with anaesthesia such as vomiting, nausea, postdural puncture headache and low blood pressure.27

RANZOG guidelines provide that elective caesareans have a 7% risk of complications, compared with 16.3% for emergency caesareans and 12.9% for instrumental vaginal deliveries.5 Postoperative sepsis or puerperal fever is an important risk, at a rate of 9.7% for emergency procedures and 6.8% for planned caesareans.26 Caesarean delivery is associated with a postpartum haemorrhage rate of 5%, but this rate drops to 1.1% for planned caesarean sections.26 Elective caesareans are also associated with a reduction in breastfeeding rates initially postbirth, but not once feeding has commenced.28

Relevantly for future or subsequent births, caesareans are associated with an increased placenta accreta—where the placenta does not come away from the muscle of the womb after birth—which compounds each subsequent pregnancy. It may be associated with significant maternal mortality and morbidity, including massive haemorrhage requiring emergency hysterectomy.5 Women who have a caesarean delivery for their first pregnancy are also less likely to have any subsequent pregnancies.29 Caesarean delivery is commonly associated with the development of abdominal adhesions, which are correlated with their own morbidities, such as chronic pelvic pain, secondary infertility and bowel obstruction. The incidence of abdominal adhesions increases with each caesarean procedure and also increases the risk that subsequent surgical procedures will be required.30

Neonatal risks

According to the RANZOG Guidelines, there is a lack of reliable and accurate data on the effect of all birthing options, limiting understanding of the respective risks.5 A Western Australian study discovered rates of hypoxic ischaemic encephalopathy (a complication where inadequate blood flow reaches the baby’s brain) were significantly lower with elective caesarean sections than alternative modes of birth.5 31 There is the potential for a planned early caesarean delivery to be preventive for injuries to the baby, as cerebral palsy and Erb’s palsy are ‘unequivocally greater after vaginal birth’ with rates reported at up to 0.3%.5

The Guidelines provide that ‘caesarean delivery without labour is associated with an increased risk of neonatal respiratory complications’ (rapid breathing).5

While RANZOG guidelines state that the risk of fetal injury is low with caesarean delivery, there is still the possibility of injuries such as skull fractures, intracranial haemorrhage, brachial plexus palsy or cervical spine, spinal cord and/or vertebral artery injury.32 Complications and injury can still arise in cases of a very large fetus, but the incidence is much less than with vaginal birth.32 There is also evidence that caesarean births without labour (not necessarily as a result of maternal request) increase short-term neonatal respiratory problems (including transient tachypnea) and childhood infections until 5 years of age.5 6 33

Autonomy

A primary concern when considering medical decision-making and appropriate information sharing is the need to ensure that patient autonomy is respected. Proponents of ‘natural birth’ provide examples of paternalistic medical interventions and experiences of women who have had traumatic births in which they have not had their autonomy or rights respected. They warn of the exploitation of statistical and risk information as a tool to influence women’s decisions about their care44 and fear a slippery slope, whereby pregnancy is further pathologised, creating additional fear, pressure, conflict and loss of control among women.45 This leads to a blanket suspicion of any and all medical interventions in childbirth, and a preference for ‘natural’ deliveries (ie, vaginal, in a non-medical setting, often with no pharmaceutical pain relief or assistive technologies). Outside the natural birth movement, authorities such as WHO also seek to reduce the rates of caesarean sections, suggesting they should be performed only 10%–15% of the time.46 47 There is concern that, where caesarean section rates exceed this it is driven by practitioner rather than patient preferences, perhaps due to the convenience of being able to schedule births or a desire to mitigate risks.48 Thus, autonomy may be used as an argument against policies that might increase caesarean section rates.

Ethically, there is a difference between having a right to refuse a certain intervention, and having a right to demand some treatment.49 50 Previously, a lack of evidence showing a net benefit of caesarean sections spoke against a right to demand such a procedure, yet a more holistic consideration of how women might reasonably weigh the outcomes of different modes of birth for both themselves and their future child means the question of net benefit is not so clear.13

We suggest it is implausible that a concern for autonomy should direct us to withhold risk information regarding trial of labour (both relating to the woman and the fetus) from pregnant women.18 Given the risks described earlier, a rational woman might well determine that she is willing to accept the harms associated with surgery (based on personal experience or impersonal data) in order to avoid the harms associated with vaginal delivery, such as third-degree or fourth-degree tears, urinary or faecal incontinence and the risk of extreme acute pain. Alternatively, she may be prepared to sacrifice her own interests to avoid even a small risk to her baby. Such altruistic preferences should be respected. There is an excellent case for considering childbirth a preference-sensitive decision: the risks and benefits of different modes of delivery are not decisive in terms of which will best promote the interests of the mother and baby, and it will depend on each individual’s values and preferences.51 It is perhaps instructive that obstetricians and other healthcare professionals seem to prefer caesarean births relative to the general population.52–54

Furthermore, medicalised birth is not inherently incompatible with women’s autonomy.18 Women may choose a medicalised birth and prefer their healthcare team to take control over their delivery, meaning the medical management of birth can be experienced as empowering for some women.45 The use of terms such as ‘normal’, ‘natural’, ‘right’ or ‘wrong’ to describe modes of delivery or procedures is also not supportive of beneficial power and a pregnant woman’s ability to maintain control over her perception of the world.55

In accordance with standards of shared decision-making, the respective risks associated with different modes of birth should be explicitly discussed with the pregnant woman such that she can plan the birth she prefers. Moreover, the current legal landscape, as defined by Rogers v Whittaker, requires that informed consent is given for any medical interventions that occur during assisted delivery, including communicating information about risks likely to be material to the mother. As already argued, prior discussion of the harms and benefits of different modes of birth would ensure women are enabled to make informed choices regarding their birth options and can consent to any interventions likely to be encountered (recognising that consent will need to be resought and may be removed at the time any interventions occur).

Neonatal well-being

The mother is not the only agent affected by the risks inherent in childbirth. It is well recognised ‘in both medical ethics and law that patients with capacity can refuse treatment, even if doing so is against their interests’.56 Despite this well-established principle, the moral permissibility of forced medical care on a pregnant woman when the unborn fetus is at risk is subjected to continued debate.56 57 As discussed, different modes of delivery are associated with different risks to the child as well, and it is appropriate to consider their interests when weighing different birth options. Occasionally, a mother’s preference regarding childbirth may conflict with the interests of her child, as can be seen in ethical debates concerning home birthing or the moral permissibility of forced caesareans.56 58 In such circumstances, where the birthing choice carries an additional risk of life-long disability for the child, an ethical challenge arises in the form of balancing the pregnant woman’s autonomy and right to self-determination, and the fetus’s interests in not being harmed. On one hand, ensuring a woman’s autonomy, which on pregnancy, she does not become a means to an end or a restrained ‘ideal baby-making machine’.59 However, women will typically be highly motivated to ensure that minimal harm comes to their child during delivery.60 Assuming that the pregnant woman intends on being responsible for the child’s care following birth and, therefore, has a secondary benefit in the health of the fetus, it also cannot be assumed that the two interests are mutually exclusive.61 Thus, ensuring that women are informed of the ‘material risks’ associated with vaginal and caesarean births should support them to make decisions in the interests of their child as well as themselves.

Requiring consent is conceptually confused

A principal objection might be that planned vaginal delivery just is not the sort of thing to which one can consent. Or rather, it is not the sort of thing to which one can refuse consent: it is a natural consequence of pregnancy, not an optional medical intervention.62 Furthermore, there are concerns that women are pressured into medical interventions during delivery and that childbirth has become routinely medicalised.63 Encouraging women to view vaginal birth not as the default mode of delivery but as one alternative, complete with attendant risks, might be seen as a further shift towards medicalised childbirth.

While it is true that there is something odd about adopting processes of consent for a planned vaginal birth, we feel that the potential benefits could outweigh the conceptual drawbacks. First, as discussed, it would provide an important (and much needed) opportunity for pregnant women to discuss the potential risks (and benefits) of different modes of childbirth. This would be at a point in time prior to labour when they are able to consider their preferences regarding different risks and discuss these with their healthcare team, as opposed to rushed decisions at particular times.18 This may enable some women to make decisions ahead of labour, which mean they experience a childbirth that fits better with their preferences (eg, seeking a caesarean section) and avoids outcomes they particularly wish to avoid.

Second, the mere ‘inevitability’ of childbirth does not render consent procedures irrelevant. As described by Manson and Neil, providing informed consent constitutes a waiver, permitting healthcare professionals to take actions that would ordinarily be a violation of one’s bodily integrity.39 Act of giving informed consent changes the meaning of those actions from violations and injury to remedy and care. As Kingma argues, there are characteristics of maternity care that ‘increase the need for explicit attention to, and respect for both autonomy and rigorous informed consent processes’.61 Giving women the opportunity to affirm their willingness to undergo interventions associated with vaginal delivery is, thus, a valuable process of permission-giving.

One particular concern is that the language of ‘normal’ and ‘natural’ birth encourages pregnant women to assume that opting for a vaginal delivery is very likely to result in a birth with no medical interventions. As discussed, interventions such as vaginal exams, instrument deliveries, episiotomies and emergency caesareans are relatively common. In choosing a trial of labour, therefore, the pregnant woman is effectively (though perhaps not knowingly) choosing to place herself in a position where she is likely to be asked to consent to any number of these interventions. It may be very difficult (not to mention unwise) to refuse instrumental delivery or emergency caesarean once one is partway through a difficult labour. Hence, the importance of being given the opportunity to consider, when forming a birth plan, which risks one is willing to be exposed to, and which interventions one is willing to undergo.

Risk information is distressing

A second objection might be that highlighting the risks associated with (vaginal) childbirth could cause additional distress. A German study from 2003 suggested that women labelled as ‘at risk’ in pregnancy may suffer psychologically.64 Medical professionals’ understanding of risk is claimed to be based on specialised training and knowledge, in conjunction with personal experiences and value. In contrast, pregnant women’s is considered ‘contextual, individualised and embedded in their social environment and individual lives’.64 Adopting a medical risk-based model of childbirth may emphasise clinical risks and neglect the positive aspects of childbirth—for instance, as a spiritual and empowering experience for mothers.

It may, indeed, be distressing to be informed of the risks associated with a trial of labour, and to face the prospect of perineal tearing, incontinence and severe pain. Yet to allow the unpleasantness of being informed—truthfully—of one’s medical risks to justify withholding relevant information from people is dangerous. This was the justification provided by Dr McLellan in the Montgomery case: Dr McLellan was concerned that information regarding the size of the baby was causing Mrs Montgomery distress, and, hence, no scan was conducted at 38 weeks; the (high) risk of shoulder dystocia was considered likely to provoke a request for a caesarean section and, hence it was not discussed.

Compassion is a laudable quality in a medical professional, and a desire to avoid unnecessary suffering is understandable. This should not, however, extend to misleading a patient (including via omission) regarding the material risks she faces.

In the Montgomery case, it appears that Dr McLellan was aware of the materiality of the risks of shoulder dystocia to her patient, since she predicted that, if such risks were made known to Mrs Montgomery, she would have requested a caesarean section, and Dr McLellan judged a caesarean section to ‘not be in the maternal interests’. Yet a patient’s right to make decisions not regarded in their interests by a medical professional is protected by medical ethical guidelines and law. So long as patients meet the requirements of competence to decide on their treatment, they should be permitted to choose which treatments they consent to and which they refuse. This is only possible if patients are provided with the relevant information in order to make those decisions.

Earlier reviews and studies of the psychological consequences of caesarean deliveries noted ‘increased fear during labour, loss of positive self-esteem and body image, feelings of failure and self-blame and post-traumatic stress’.65 The description of vaginal birth as a ‘normal’ or ‘natural’ birth may contribute to women’s feelings of failure or disappointment in their birth experience, as by inference, their birth was thereby ‘abnormal’ or ‘unnatural’. Furthermore, a woman’s perception of her control or absence thereof, during the birth can also contribute to potential psychological consequences of a caesarean delivery.65 One suggested method of reducing this risk is to ‘provide women with information about caesarean delivery prenatally and help them develop realistic expectations about childbirth’.65 Providing more information to women may assist in the fear and anxiety they feel at the prospect of losing control and autonomy as they approach labour and birth.66 As such, it is at least plausible that additional information provision will not ultimately result in more, but rather less, distress.

Cost

Finally, cost considerations might speak against seeking consent from women for vaginal birth. The time needed to inform women of the risks of vaginal delivery would create additional costs to the healthcare system. Furthermore, the cost of a vaginal birth is estimated at $9089 in Australia compared with $14 086 for a caesarean delivery.67 However, cost estimates from Canada suggest caesarean delivery (prior to onset of labour) is comparable to vaginal delivery, and cost-saving relative to assisted vaginal delivery.68

Nonetheless, if seeking consent for planned vaginal delivery and routinely providing information regarding the risks of all modes of birth were to result in more caesarean sections, then this could plausibly make maternity care more expensive. It is not clear that pregnant women should be offered a medically unnecessary procedure when an acceptably safe, more resource efficient alternative is available. This is compounded by the likelihood that a caesarean section makes it more likely that any subsequent deliveries will be by caesarean section as well. Highlighting the risks of trial of labour by conducting informed consent procedures could increase MRCSs and lead to a slippery slope, whereby the majority of births are by caesarean section and vaginal delivery is a rarity.69

Any additional costs should be weighed against benefits that might be gained later in the birthing process or after birth. For practitioners and healthcare providers, increased elective caesareans may decrease insurance premiums53 and may assist time management and resourcing allocations, although this would need to be measured against the increased length of stay for mothers’ postsurgery. On the other hand, the decreased rates of permanent neonatal injuries and disabilities that could occur with higher rates of caesarean delivery may counterbalance the time cost incurred in the consent process. Once the cost of a potential stay for babies in the NICU (more frequent with vaginal birth),68 along with the long-term cost of prolapse, is calculated, the argument that caesarean deliveries cost more is not true.70 71

It is, therefore, unclear that routinely engaging in shared decision-making processes and seeking consent for different modes of childbirth would be additionally costly. Even if the absolute cost was to increase, it is unclear that it would, all things considered, be cost inefficient. Given the benefits of providing additional risk information (in terms of respecting women’s autonomy and facilitating better decision-making regarding their mode of delivery), any additional expense may be worthwhile in terms of the pay off.

Indeed, even if caesarean sections were more costly, and should not be funded within a public healthcare system, women should still be informed of the risks of planning a vaginal delivery and enabled to make their own decision whether to pursue alternatives privately.