Article Text


Ethics of refusing parental requests to withhold or withdraw treatment from their premature baby
  1. R J Boyle1,
  2. R Salter1,
  3. M W Arnander2
  1. 1Department of Paediatrics, Hillingdon Hospital, London, UK
  2. 2Department of Medicine, St Thomas Hospital, London, UK
  1. Correspondence to:
 Dr R Boyle
 Paediatric Registrar, Department of Paediatrics, Pield Heath Road, Hillingdon Hospital, Uxbridge, Middlesex UB8 3NN, UK;


In the United Kingdom women have access to termination of pregnancy for maternal reasons until 24 weeks’ completed gestation, but it is accepted practice for children born at or beyond 25 weeks’ gestation to be treated according to the child’s perceived best interests even if this is not in accordance with parental wishes. The authors present a case drawn from clinical practice which highlights the discomfort that parents may feel about such an abrupt change in their rights over their child, and argue that parents should have greater autonomy over treatment decisions regarding their prematurely born children.

Statistics from

Since the Abortion Act in 1967 women in the UK and increasingly in other countries have had widespread access to legal abortions.1,2 In the UK 98% of abortions take place to safeguard the woman’s physical or mental health, and such abortions are permitted up to 24 weeks completed gestation.3 This contrasts with practice regarding children born at this stage—guided in the UK by the Children Act—which emphasises that the welfare of the child should be the primary concern when making decisions surrounding their care.4 Practice among paediatricians, supported since 1989 by the Children Act, has been to put the child’s perceived best interests above their parents’ wishes in cases of conflict. For example if parents wish to withdraw treatment from their baby and the medical team disagree then most practitioners in the UK will overrule the parents.5 This abrupt change in attitude toward the fetus in utero and the child ex utero at similar gestations may lead to conflict. We present a case from our practice which brings into focus this potential for conflict, and highlights some of the ethical issues surrounding the treatment of children born very prematurely.


A 31 year old woman in her second pregnancy presented at 23+2 weeks’ gestation with spontaneous rupture of membranes. She was counselled that the timing of labour and delivery were uncertain, but the risk of delivery before 26 weeks’ gestation was high. She was advised that the child had a 10% chance of survival to discharge from hospital if born at 23 weeks’ gestation, and a 50% risk of disability if they did survive.6 Increased survival rates were given for the child if born at 24 or 25 weeks’ gestation, but similar rates of disability. This information was based on recent UK data shown in table 1. She was advised that although figures for 26 weeks’ gestation births are less complete, the disability rate at this gestation is likely to be considerably less than 50%.7,8 The medical team explained that although a child born in very poor condition or at 23–24 weeks’ gestation might not be resuscitated (depending in part on parental wishes), they would feel obliged to resuscitate and support a child born at 25 weeks’ gestation or later in reasonable condition. The woman and her husband were concerned at the prospect of having a disabled child—she had some experience of caring for disabled children in her work as a teacher, and through having a disabled niece. They were disturbed that at 25 weeks’ gestation they might not have the opportunity to withhold resuscitation despite a 50% risk that the child would die before leaving hospital, and a 50% risk of disability in the child if they did survive. For this reason they elected to terminate the pregnancy.

Table 1

 Outcome of extreme prematurity in the United Kingdom and Ireland, March to December 19956


The outcome of pregnancies with rupture of membranes before 26 weeks’ gestation has been the subject of a number of reviews.9–13 These estimate the mean latency before delivery as 10–21 days, with delivery occurring within seven days of rupture of membranes in up to 50% of cases. In some cases delivery may not occur for 10 weeks or more. Although the overall risk of premature delivery, death, and disability is high with rupture of membranes at 23 weeks’ gestation, those who do deliver at or beyond 26 weeks’ gestation have a reduced risk of poor outcome.7,8 If the couple in the case described were assured they would have the option of not resuscitating a child born at 25 weeks’ then they might have continued this pregnancy, in the hope of going beyond 25 weeks’ gestation and delivering a child with a higher chance of survival and a lower risk of disability. In the light of this case we shall review law and practice in relation to refusing parental requests to withhold treatment from their 25 week gestation baby, and discuss whether such a denial of parental wishes is justified.


UK legislation

In the UK legal decisions regarding withdrawing or withholding care from children are underpinned by the Children Act, which makes the best interests of the child the overriding consideration, and the Human Rights Act which emphasises the universal right to life.4,14 These Acts of Parliament support the maintenance of life in children unless it is clearly against their interests to do so, but the regularity with which treatment withdrawal cases reach the Courts attests to the difficulty of determining what is in the child’s best interests. Moreover the issue of whether the best interests of a very premature neonate are viewed with the same weight in these decisions as those of an older child is not clear. Perhaps the best known Court case in which parents wished to withhold treatment from their child against medical advice was that of Baby Alexandra, a child with Down’s syndrome in need of life saving surgery.15 The Court of Appeal upheld the physicians’ request for surgery against parental wishes, concluding that the infant’s life was not “demonstrably so awful” that she should be allowed to die. Many would agree with this stance that children should be given the opportunity to live unless their life is clearly not worth living.

A contrasting case is that of Child T, who had failed liver surgery for biliary atresia at age 3 weeks and was offered a liver transplant. Parental refusal of this life saving treatment was upheld by the Court of Appeal as in the “child’s best interests”.16 At the time of this decision, liver transplant surgery in children had a 10 year survival rate of 76%, with just 29% of these survivors needing a further liver transplant during those 10 years.17 These outcome figures might be viewed as comparable with those of a child born at 25 weeks’ gestation, and it is conceivable that the same court would allow parents to withhold treatment from a 25 week neonate against medical advice.

United States law

In the US parental autonomy over the upbringing and nurturing of their child has been interpreted as constitutionally safeguarded.18–20 However this principle may be overridden if the health and safety of the child are in jeopardy.21,22 As in the UK, the legal position is unclear when the best interests of the child are not evident, and where treatment is only marginally beneficial parental autonomy to withdraw treatment may be upheld.23,24


Professional guidelines

UK practice guidelines do not easily apply to the resuscitation of a child born at 25 weeks’ gestation, but they do suggest that in cases of dispute between physician and parent treatment should generally be continued until the issue is resolved, if necessary by the Courts.25,26 Canadian and US guidelines suggest that whereas non-resuscitation of children at 23 or 24 weeks’ gestation might be appropriate depending on parental wishes and prognostic factors, children of 25 weeks’ gestation or greater should be resuscitated in all cases in the absence of fatal anomalies.27–29

Surveys of practice

In UK hospitals many nurses and doctors feel that the ultimate responsibility of deciding to withdraw treatment from a premature baby is too great a burden for parents to bear.5,30 In practice, treatment withdrawal decisions are usually made by the medical team, not only in the UK but in the majority of European countries.31,32 More specifically, most European practitioners (with the exception of those in the Netherlands) would resuscitate a baby born at 24 weeks’ gestation even in the face of parental opposition.33


The law in the US and UK does not provide specific guidance as to when paediatricians should accede to parents’ wishes regarding their child’s treatment, but surveys suggest that neonatal practice in the UK and Europe is patriarchal. Although parents may be involved, it is often clinicians who ultimately decide when treatment might be withdrawn, and this is also common practice when deciding whether to resuscitate children born very prematurely. In many centres those born at 25 and even 24 weeks’ gestation in good condition are routinely resuscitated whether their parents wish it or not. Below we discuss three important ethical questions which arise from this practice.


If the best interests of the child are to be given greater weight than parental wishes then this relies on the child being a person and having the rights of a person. The acquisition of this “personhood” or moral status by a developing human being is widely debated in the context of abortion and the use of human embryos for scientific research.34,35 Personhood (and with it the protection of one’s interests by society) is believed by some to be fully acquired at conception, and by others to be acquired gradually or stepwise through pregnancy and childhood. Advocates of the latter approach may invoke criteria such as viability, birth, or self awareness as prerequisites for personhood.34 In practice viability and birth are the most widely used and we discuss these below.

A viability criterion of personhood

Obstetric and neonatal practice in the UK and US rests on the premise that when a fetus becomes viable they acquire the status of a person.36 The major criticism of this viability criterion is that the viability of a fetus depends very much on the available technology of the society into which they are born, rather than on any intrinsic worth that they may have.37 If 25 week gestation babies are to be resuscitated against their parents’ wishes because they are viable then we might imagine a future in which neonatology has advanced such that fetuses 6 weeks old can be reliably salvaged with little morbidity. To many it would seem absurd for the fetuses of women who suffered an early miscarriage or ectopic pregnancy to be resuscitated and given full intensive care against their wishes.

A birth criterion of personhood

The legal criterion for the acquisition of personhood in the UK and many other countries is birth. In Israel the distinction around birth is particularly clear: here abortion is permitted at any time during pregnancy on maternal grounds, yet neonatal policy is for all viable babies to be given full resuscitation and treatment.38 Where birth is used as the sole criterion of personhood, women in premature labour may quite frequently feel forced into aborting a wanted pregnancy through doctors’ aggressive resuscitation policy, in cases similar to that described above. And unless one views killing and withholding treatment as morally equivalent (a view which some do hold) then such cases argue against a reliance on birth for the acquisition of personhood.39,40

The gradual acquisition of personhood

Some ethicists reject the view that personhood is acquired suddenly at a given moment in development and argue that personhood develops in a continuous fashion from conception onwards.41–43 If a child born at 25 weeks’ gestation is less a person than an older child, then their perceived best interests may not overrule parental autonomy in decisions surrounding their care, and practice which applies to withholding treatment from older children would not be automatically transferred to the child born very prematurely.

Whether the best interests of a very premature child should supersede the wishes of their parents clearly depends on one’s view in the widely debated area of the development of moral status; but whichever philosophical stance one takes, the reliance of current practice on viability or birth to define personhood is imperfect.


If we assume that the 25 week newborn child’s interests do supersede their parents’ interests regarding their fate, it is still not clear that resuscitation is beneficial for the child. The best available UK data suggest that children born at 25 weeks’ gestation have a 40–50% chance of surviving their initial hospital stay—a stay which lasts three months, involves regular potentially painful procedures, reduced maternal contact, and prolonged periods of starvation. At age 30 months 50% of survivors have a significant disability.6 In later childhood there is an increased risk of behaviour problems and more minor motor and learning difficulties, and in the context of resuscitation against parental wishes these children may have either reluctant or foster parents.44–47 Although many would feel a 25% chance of life without disability is in the child’s best interests, it is at least arguable that it may not always be so. And where there is significant medical uncertainty about the benefits of a given course of action for the child, then parental autonomy should not be subverted.29


Where resources for medical treatment are tightly rationed as in public health systems, resuscitation and intensive care treatment of very premature children with a high chance of early death or disability may not be a justifiable use of such resources.48 This may particularly be the case where such treatment is against parental wishes, and the child places an economic burden on both the social and health services of a society. Indeed this “economic justice” criterion is cited by both the Danish Council of Ethics and US ethicists in their recommendations for the management of extremely premature babies.29,49


Published data suggest that children born at 26 weeks’ gestation have an improved outcome compared with 25 week births, and estimates of their survival to discharge range from 54% to over 90%.7,8,50–52 The long term outcome for such children is less certain, but we would suggest that in countries with a high standard of neonatal care it is in the best interests of a 26 week gestation baby born in good condition to be resuscitated. Considerations of the child’s moral status at 26 weeks’ gestation and appropriate resource allocation may in some cultural and financial contexts justify withholding treatment from such children at parents’ request.


In an era of fast improving neonatal care, the ethical management of fetuses or children at the threshold of viability is increasingly complex. Those born before 24 weeks’ gestation would have inevitably died in the past but can now be salvaged, yet may also be legally aborted, bringing into focus the debate over their moral status. In the UK women can sanction the killing of their child in utero at 23+6 gestation, but a day later they may not be permitted to let the child die at birth without medical intervention. If we agree that the fetus in utero is not morally distinct from the baby ex utero, then current practice rests on a premise that the fetus suddenly becomes a person at 24 weeks’ gestation; a view which has no physiological or philosophical basis. In the context of current practice regarding the resuscitation of children born very prematurely, termination of pregnancy for maternal indications at close to 24 weeks’ gestation may seem inappropriate. However it is equally important to offer couples greater flexibility in resuscitation decisions at 24–25 weeks’ gestation. Children born at this gestation should not be resuscitated against their parents’ wishes. Although the “personhood” of such children is widely debated, whatever their moral status it is not unequivocally in their interests to be resuscitated. To do so against the wishes of their parents may be both an unnecessary subversion of parental autonomy and an inappropriate use of healthcare resources.


We are grateful to Julian Savulescu, Professor of Applied Ethics at Oxford University, and to John Wyatt, Professor of Neonatal Medicine at University College London for their helpful comments on the manuscript.


View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.