A hypothetical adult case of desired experimental treatment

In a case like that of Colin Giles, there would be a major focus on his views about treatment.³ Why? An adult’s wishes might conflict with what would actually be best for him. In some situations, the patient’s views might be idiosyncratic, based on religion or other supernatural beliefs or unreasonable. (Imagine that Colin’s decision to choose this particular experimental treatment was because he liked the name of the drug, or because he had a seen a picture of the doctor offering the treatment and thought that she looked attractive). Perhaps we feel that Colin has made an unwise choice, and that it would be much better to accept his mortality and not prolong his dying in intensive care? He should, you might believe, die with dignity. Nevertheless, as long as Colin had capacity at the time that he indicated his wishes, the usual principle is that doctors should try to respect his previous wishes. So, in this imagined case, if there is good evidence that the adult Colin had wanted to be kept alive on a ventilator, if he would have wanted to pursue experimental treatment overseas, despite the apparently slim chance of benefit (and, from our discussion in Chapter 4, as long as this would be within the boundaries of distributive justice and consistent with fair allocation of limited health care resources) there is a strong argument that treatment should be provided or continued. It seems far less likely that the hypothetical case of Colin Giles would reach the court, and if it had reached the court, it seems more likely that the decision would have ultimately been in favour of providing treatment.

The ethical difference between the hypothetical Colin Giles and the real Charlie Gard case is in two separate ways. First, when thinking about decisions in medical ethics for adults, we give significant weight to patient autonomy. ‘Autonomy’ refers to the value of self-determination, or a right to make decisions about one’s own life. For medicine, it is about respecting patients’ wishes about medical treatment. Autonomy is often understood as freedom from interference – it is a negative right. Because of that, there is a very strong belief that patients’ refusal of treatment must be respected. We would allow an adult to refuse a blood transfusion, or to decline cancer treatment, even if those treatments would be life-saving and with few side effects.

There is less significance of autonomy in patient requests for treatment.⁴ Autonomy doesn’t mean that patients have a positive right to whatever treatment they choose. However, even if we don’t think that autonomy gives patients a right to demand treatment nevertheless respecting patient autonomy gives us a separate (and potentially strong) reason to provide treatment if the patient desires it and there aren’t other good reasons against doing so.

Second, if we are trying to determine what would be in a patient’s best interests, their values and wishes are important. It is difficult to know how to weigh up the risks and benefits of experimental treatment. But a previously expressed desire for this treatment (despite uncertainties and risks) would be relevant to determining what would be best for Colin Giles. While this obviously overlaps with the argument based on autonomy, there is a more nuanced relationship between someone’s wishes and best interests. If Colin’s desire for treatment is part of long-standing values that were important to him – that gives us extra reason to think that treatment would be in his own best interests. (If his expressed desire had been more transient, or less obviously related to values that were central to him, we might feel that this was less relevant to his interests).

For example, a person might request that mechanical ventilation be continued because he believes that we should not intervene in God’s will. Such beliefs are arguably irrational, even in religious terms.⁵ However, they are frequently respected. It is also important not to discriminate against secular non-religious beliefs. If a person requests ongoing ventilation because she does not believe in an afterlife, and believes that the Singularity is coming and she will be able to upload into an artificial intelligence, that belief ought to be accorded the same respect.⁶

In the real Charlie Gard case, however, we could not give any weight to the autonomy of the patient, to Charlie’s wishes. He was never in a position to understand his medical condition or to develop values relevant to the decision. His parents, though, clearly did have values and views about his treatment.

What significance should be placed on the autonomy or the wishes of parents in cases of disputed treatment for children?

Parental autonomy

One view might be that parental autonomy is just like autonomy for adult patients. We allow patients to make decisions about medical treatment because it is their life and their body. As long as they aren’t harming anyone else, why should it be anyone else’s business whether they decide to have a particular medical treatment? Some might suggest that parents have a right to decide about treatment because it is their child and their family.

But the autonomy of parents is very obviously different from the autonomy of patients to make decisions for themselves. While adult patients are generally thought to have an absolute right to refuse medical treatment for themselves, we don’t usually think that parents can refuse all medical treatment for their children.

As an example, in February 2017, a couple in Michigan refused medical treatment for their jaundiced newborn baby, Abigail. A midwife had visited and told the parents that the baby needed urgent treatment. However, Abigail’s parents believed that she would be healed by prayer. The baby died two days later from complications of newborn jaundice (a medical condition that is readily treatable).⁷ In cases like this, we don’t think that parental autonomy gives parents a right to refuse treatment. There have to be some limits.

“Where, as in this case, the parents spend a great deal of time with their child, their views may have particular value because they know the patient and how he reacts so well… Their own wishes, however understandable in human terms, are wholly irrelevant to consideration of the objective best interests of the child save to the extent in any given case that they may illuminate the quality and value to the child of the child/parent relationship.”⁸

In the case of MB, an infant with severe muscular weakness (spinal muscular atrophy), the judge saw value in parents’ perspective on treatment. However, he distinguished between the views of parents and the wishes of parents. Is there reason to distinguish between parents’ views and wishes in this way? How are they relevant to the child’s best interests?

Knowledge of child

One reason to take parents’ views into account is because of their knowledge of the child. If parents refuse antibiotics because they know that their child has previously had a serious allergic reaction to those same antibiotics – it would be vital to pay attention to their concerns. In an older child, parents will often be able to provide evidence about the day to day life of the child, what they like or dislike, and how much their life is affected by illness or impairment. In Chapter 3, we discussed the significance for decisions of the burdens of treatment or illness, as well as the objective benefits of life-prolonging treatment. The views of parents may be highly relevant to an understanding of what life is really like for the child and hence whether it would be best to keep them alive or to allow them to die. However, sometimes parents’ assessment may differ from that of others caring for the child and it may not be clear whether parents’ views are overly positive (because of an understandable strong desire that their child is improving), or whether professionals are overly negative. In the Gard case, Charlie’s parents had spent hours at his bedside. They believed that he was aware of his surroundings and responding to them. In contrast, the nurses and doctors caring for Charlie in intensive care felt that he did not respond. A judgement may need to be made about which evidence should be believed. In the Gard case, the judge appeared to be more persuaded by the medical evidence.

Effect on child

Parents’ views might, at least in some circumstances influence whether or not treatment would be in a child’s best interests. In chapter 3, we mentioned the case of Y, an eight year-old girl who had a long-standing spinal muscular atrophy. In that case, the judge indicated the importance for Y’s quality of life of her family’s care and devotion. She had: “a loving and supportive family who were acutely concerned with her wellbeing and are diligent in seeking to ensure that everything that can be done for her is done”. One of the medical consultants praised the “care, love, devotion and expertise which Y’s family lavished on her in supporting her through these technically demanding treatments for many years. Without their loving care and attention, Y would undoubtedly have died many years ago.”⁹ However, even with the most devoted parents in the world, there are limits to how much parents are able to influence the wellbeing of a child. Where a child’s condition causes them to suffer, or prevents them from appreciating or enjoying benefits from continuing life, the scales may still be tipped against treatment.

The judge in MB dismissed the ‘wishes’ of parents (as ‘wholly irrelevant’), but that might be too hasty. There are two reasons to perhaps take these into account.

Uncertainty

We have described in the previous chapters some of the challenges in determining whether treatment would work, how likely it is that it would work, and how to weigh up the risks and benefits of treatment. In many situations, it isn’t clear that there is a single right answer. There may be different reasonable views about whether to continue life-support or discontinue life-support, whether to provide experimental treatment or not.

Where there is genuine moral uncertainty (uncertainty about what is the right thing to do), we should in general allow parents to make decisions about treatment for a child. Why? Here, perhaps, is where parental autonomy and parental wishes are relevant – parents have a right to decide within reasonable bounds about how to raise and care for their children. After the child, they will be the ones who are likely to be most affected by the decision. That especially includes decisions where there might be a range of different views about what to do. It would exclude situations that are clear-cut. In the case of the jaundiced baby Abigail, there is not any real uncertainty about what the right thing would have been. There is not likely to be reasonable disagreement about what her parents should have done. They should have taken her to a doctor.

In others cases, there is more uncertainty. Whether we prefer one course of action or another will depend on our values. Sometimes, faced with disagreement, the temptation may be to stay with the status quo – for example, to continue current treatment, or not to embark upon a requested procedure/treatment. However, that “non-decision” is still a decision. It still involves a value judgment (about the current situation versus an alternative). We are morally responsible for the outcome because we could have acted otherwise. If there is real uncertainty about the right course (and as long as parents appear to be competent to decide) it would be unfair for courts or doctors to impose their own values on a family.

Interests of parents and other family members

Finally, one overlapping reason for allowing parents to make a decision (where there is uncertainty) is because these decisions have profound effects on the parents themselves and other members of the family. Where a decision would lead to a child surviving, that will often result in a substantial burden of care for other family members. Where the decision would lead to the child dying, in most cases it is the parents and family who will carry the greatest emotional burden of the loss of the child. It is they who will have to live with decision.

Stepping outside of the intensive care unit and the hospital, if we think about other serious decisions that parents make for their children, it is usual for parents to take into account their own interests as well as those of other members of the family. Every day, parents make decisions that affect the interests of their children. They decide what sort of food to provide, how to travel, how fast to drive, which school to send their children to, which activities to involve them in, whether to move house, which job to apply for. All of those decisions involve trade-offs. It would be possible for parents who have a single child to make every decision based only on what (they feel) would be best for their child. However, it isn’t clear that is what we should expect of parents. Some might think that a perfect parent would pay attention to what would be best for their child, and be prepared to make sacrifices for their child. However, it isn’t even clear that that is the ideal for parenthood. Surely, we wouldn’t expect parents to sacrifice all of their own interests and wellbeing as soon as they have a child? In any case, for parents with more than one child, that supposed ideal becomes impossible. Decisions that are the best for one child become less than the best for other children. Parents should be paying attention to the wellbeing of all members of the family.

We shouldn’t overstate this. If we pay attention to what is at stake for parents and other family members that does not mean that we should ignore what is at stake for the child, or allow the child to be harmed for the sake of other family members. If there is a serious clash between the interests of the child and of the parents, we should favour the child. Along these lines, the UN convention on the rights of the child states that “In all actions concerning children…the best interests of the child shall be a primary consideration”.¹⁰ Rather, the idea is that we shouldn’t ignore that the child is part of a family, and that these decisions affect other people than just the child. Returning to the balancing metaphor that we discussed in chapter 3, if there is a situation where the benefits and burdens are closely balanced, and it isn’t clear which way the scales are leaning – the interests and wishes of parents may tip the balance.

The Zone of Parental Discretion

There are a number of reasons that parents’ views and wishes are ethically relevant to decisions about medical treatment for children. If we take those reasons into account, that suggests that whether or not treatment is provided for a child with a given health condition will vary from family to family. Some families may strongly desire treatment, feel that it would be best for their child, and it will be provided. Other families may feel, in the same circumstance, that that treatment is not the best thing for their child and treatment will be withheld or will be stopped.

Situations where treatment might or might not be provided depending on the family’s wishes represent what Australian ethicist Lynn GIllam and her colleagues call the “zone of parental discretion”.¹¹ ‘Discretion’ here is the idea that parents may decide one way or the other. As she describes it – this is the “ethically and legally protected space where parents may legitimately make decisions for their children, even if the decisions are … not absolutely the best for them.”¹²

The idea that parents would be consulted about decisions, that they would be an integral part of the decision-making process for children is hardly controversial. It is, in fact, what happens all the time in the medical care of children all around the world. That includes the most serious and fraught decisions that are ever made – about providing or stopping life-support for seriously ill children. It is widely accepted that the right approach to these decisions in most cases involves sharing the decision between professionals and parents.¹³ However, if we think that there is a ‘space’ where parents’ views are important and their wishes will be respected, there must also be an area beyond where parents’ wishes are not going to determine whether or not a child has treatment. The more weight that we give to parental interests or autonomy, the wider the range of cases where parents will decide. While there might be agreement internationally that there is a ‘zone of parental discretion’, there is not as much agreement about how wide this zone is, or where the boundaries of the zone lie. In the Charlie Gard case, the UK courts and doctors clearly felt that Charlie’s situation was not one that fell within the zone of discretion. However, other commentators, including ethicists and doctors from the US felt differently – that this was a decision that parents should be allowed to make, even if we disagree with it.¹⁴ When the European Court of Human Rights considered the Charlie Gard appeal, they assessed whether the balance between the competing interests of child and parents had been struck “within the margin of appreciation afforded to States in such matters”.¹⁵ This highlights that the court expected and accepted that different states might draw the boundaries of parental discretion in different places.

Where should the boundaries be drawn?

The harm threshold

In Gillam’s analysis of parental discretion, she described the limits as set by the ‘harm threshold’: parents should not be able to make a decision “if the child is likely to suffer significant harm from the decision”.¹⁶ This builds on previous accounts of decision-making for children by US bioethicists and paediatricians Doug Diekema and Lainie Friedman Ross.¹⁷ It also aligns with the much older harm principle articulated by 19^th century English philosopher John Stuart Mill “The only purpose for which power can be rightfully exercised over any member of a civilized community, against his will, is to prevent harm to others.”¹⁸ As noted in the first chapter, some of the legal argument in the Charlie Gard case was based on whether a ‘harm threshold’ should be the legal test for deciding whether to overrule parents who have chosen a treatment alternative. The court ruled that that wasn’t the current standard in English law around medical treatment (though the judges also noted that even if such a standard had applied, that it would not have changed their verdict). However, there is good reason to think that the harm threshold is what should be the basis for overruling parents. To turn the question around: if we think that a particular decision isn’t likely to cause significant harm to the child, why should doctors or the courts step in? What business is it of theirs?

Here is an example case that, to our knowledge has never been tested in a UK court (or any court), but we suggest provides strong support for the harm threshold.

Adam is 6 years old. Tragically, at the age of four he was involved in a car accident that killed his mother. He suffered severe brain damage and has been left in a persistent vegetative state. Adam’s clinical condition has not changed despite the best available medical care. He lives at home with his father, who, since the death of his wife has devoted all of his energy to caring for Adam.

Adam is dependent on artificial nutrition and hydration provided by a feeding tube (gastrostomy), but does not require any other form of invasive medical treatment to stay alive. His medical treatment is funded by a generous legal settlement that Adam’s father was awarded following the accident.

The medical consultant caring for Adam has come to the view that Adam will never recover, and that it is not in his best interests to keep him alive. He applies to the court for permission to withdraw artificial feeding. Adam’s father is utterly opposed to this plan. He accepts that Adam will likely never improve, but wishes to continue to care for him for Adam’s sake, and for the sake of his deceased wife.

Hypothetical case

It seems to us unlikely (but not impossible) that a legal case such as the one described above would arise. We suspect that few or no medical professionals in a case like this would be willing to go to court to seek withdrawal of artificial feeding against the wishes of a devoted parent. However, it also seems to us that on the basis of English law and precedent, that the legal verdict in this hypothetical case should be clear. If Adam is genuinely in a persistent vegetative state, then the courts have determined that he would have no interest in continuing medical treatment. His father’s wishes about treatment are not thought to be relevant to his interests. As continued artificial feeding is not in Adam’s interests, it would be unlawful to continue to provide it. Accordingly, it should be withdrawn.

But if that is a correct interpretation of English law, we suggest that English law is ethically flawed. It would be wrong to withdraw artificial feeding from Adam against his father’s objections. While it is plausible that Adam is not benefiting from continued artificial feeding, it is also not harming him.

Should English law be revised to adopt the harm threshold for disputes about medical treatment? One question is whether it is actually necessary. We have suggested a hypothetical case, where the current legal approach might judge treatment to not be in a child’s best interests, though it is not harmful. In practice, though, it seems likely that most, if not all cases that reach the court would already cross the harm threshold. As we suggested above, it seems highly unlikely that any health professional or NHS trust would embark on legal action to oppose parents unless they felt that what parents were requesting was harmful. We do not actually think it is likely that Adam’s case would arise. If that claim is correct, a change in the law would not lead to a change in court decisions. It might be valuable to clarify the ethical basis for overruling parents, however, it wouldn’t actually change decisions that are made. The Appeal Court judges in the Gard case suggested that (in their view), the same decision would have been reached in that case whether a harm or best interests test were applied.

Even if there is no harm to the child, there is one potential justification for not acquiescing with parental requests for treatment. In chapter 4, we discussed in some detail the possibility that providing a particular treatment option would cause indirect harm to other children. Distributive justice considerations may preclude the state continuing to pay for such futile medical treatment. (That wouldn’t apply in the case of Adam, since providing artificial nutrition would not deprive any other patients of medical treatment).

If that is right, then the boundaries of the zone of parental discretion would be determined by these two separate conditions: either where a decision would cause harm to the child, or harm to others (ie through a violation of principles of distributive justice). Of course, if distributive justice is the reason for limiting treatment, it only provides a limit to State funded treatment. If Adam’s father was prepared to meet the costs of providing his artificial nutrition and hydration, there would be not barrier to its provision.

For the rest of this book, we are going to assume that this is the correct ethical basis for overruling parents’ decisions about treatment. However, we can’t stop there. We are going to need to do more to decide what counts as enough harm. For example, in Doug Diekema’s formulation of the harm threshold, he argues that state intervention is justified if parental refusal of treatment places a child at serious risk of serious preventable harm. But how high a risk is a “serious risk”? What type or level of harm is “serious harm”? There may be different answers to those questions in different parts of the world.

Zones of parental discretion and medical tourism

In early 2017, the French supreme court (the Conseil d’Etat) considered an appeal relating to the care of a brain damaged infant, Marwa Bouchenafa.¹⁹ In France, the Public Health code L1110-5 allows doctors to discontinue treatment without parental consent if they believe it would be of “no use, disproportionate, or only serving to artificially maintain life without other effect”.²⁰ Marwa had contracted a severe viral illness of the brain in late 2016, and was left with brain damage, and was dependent on a ventilator. Doctors planned to withdraw treatment, however, her parents were opposed. A local court, and then the supreme court found that there was sufficient uncertainty about Marwa’s outlook that the criteria were not met. In particular, the supreme court made a comment about the importance of parental wishes, particularly for young infants.

“à défaut de pouvoir rechercher quelle aurait été la volonté de la personne s’agissant d’un enfant de moins d’un an … l’avis de ses parents, qui s’opposent tous les deux à l’arrêt des traitements, revêt une importance particulière.”

“for an infant younger than a year old, in the absence of being able to ascertain what would have been her wishes … the views of her parents, who are both opposed to withdrawal of treatment, are particularly important.”²¹

This quote suggests that courts in France may give considerably more weight to the wishes of parents than UK courts. That would mean that the zone of parental discretion would potentially be wider in France, than in the UK. This probably applies to the US and Australia, where, at least anecdotally, parents are accorded significantly more discretion over medical decisions.

If there is international variation in where the boundaries are drawn for decisions, that could give rise to situations where parents are unable to find doctors prepared to provide treatment in one country, but are able to find some overseas. Perhaps they could embark on a special form of so-called ‘medical tourism’? That was, of course, one of the key issues in the Charlie Gard case. It was also raised in the case of Child A, an infant diagnosed with brain death, whose parents wished to take him to Saudi Arabia, and in the case of Ashya King (discussed in the previous chapter).

What should we do in situations like this? Should patients or parents be prevented from accessing treatment overseas?

You might (or might not) be persuaded in the case of experimental medical treatment for Charlie Gard or continued mechanical ventilation for Child A or provision of artificial feeding in the case of Adam. But here are a number of other possible cases:

1. Gender reassignment. A ten year old apparently identifies as transgender. His parents have arranged for him to travel to Australia for hormone treatment, and surgical gender reassignment.
2. Female genital cutting. A 2 year-old girl’s mother had a form of female genital cutting as a young child. Her parents have arranged for travel to Sudan where a local doctor is prepared to provide the procedure. This involves the surgical removal of her clitoris and oversewing of the vagina.
3. Ashley procedure. The parents of a profoundly cognitively impaired 7 year-old have identified a specialist in the US who is prepared to provide hormone treatment to reduce C’s growth, as well as surgery to remove developing breast tissue and her uterus. C’s parents hope that these treatments will make C easier to care for.²² Euthanasia. A 9 year-old boy with a terminal neurodegenerative disorder has progressive neurological decline and is no longer able to communicate. His parents wish to travel to Belgium to access euthanasia.

For medical tourism involving adults there is some reason to believe that the state should only intervene in an individual’s freedom in order to avoid harm to others (that would be the implication of Mill’s harm principle mentioned above). If an individual’s competent decision to receive medical treatment affects only themselves, why should a government prohibit someone travelling to obtain that treatment, or prosecute them subsequently?