Part 1: the case of ‘M’

First then the case that started me off along this track.7 In 2003 Ms M, a 43-year-old woman, went into a coma as a result of a devastating viral brain infection. She was subsequently diagnosed as being in a vegetative state and 4 years later reassessed to be in a ‘minimally conscious state’. After intensive but unsuccessful efforts in a neurorehabilitation centre of excellence to enhance her ability to communicate, her very loving relatives applied, in 2011, to the Court of Protection for permission to withdraw her artificial nutrition and hydration. Her family argued that M, who had been a very independent person, would have hated to go on being kept alive in her current and incurable MCS of total dependence on others, doubly incontinent, suffering from apparently painful contractures of her arms and legs and manifesting hypersensitivity to being handled during her care.

Following a very thorough hearing of evidence on both sides the judge ruled that it was in M's best interests to continue to be kept alive by providing artificial nutrition and hydration. The judge ‘unreservedly’ accepted the testimony of M's nearest and dearest that she would definitely not have wished to be kept alive any longer in such a state. But he ruled that it would be wrong to give her prior views, when she was competent and independent, much weight when assessing her best interests now. Her reported prior views did not specifically address the situation she was now in, she might in any case by now have changed her mind, and in the absence of a valid and applicable advance decision to refuse artificial nutrition and hydration, although the law required him to consider her previous informally expressed views he did not give them much weight—‘it would in my judgement be wrong’. What was to be given substantial weight in assessing M's best interests, explained the judge, was ‘the sanctity of human life’. In addition the judge reiterated that under a rule of the Court of Protection, PD9E, any decision to withhold or withdraw artificial nutrition and hydration (ANH) from a person in vegetative state or MCS ‘must be referred to the court’ (ref. 7, para 257). In the meantime such patients must continue to be provided with ANH.

That judgement led me to do three things. I wrote the BMJ editorial5 to which I've referred and I began to draft my own formal Advance Decision (see online supplementary file 1), to refuse any life-prolonging treatment whatsoever after 3 months of very probably irreversible legal incapacity to make decisions about my treatment.

Third I delivered a Gresham College Lecture6 in which I enlarged on my BMJ editorial and responded to seven counterarguments to my own position (which for reasons of space I do not repeat in this paper—but the paper and lecture are easily available on the internet).

Part 2: the judgement in the case of Aintree v James

Since I wrote my BMJ editorial a further and in my view brilliant judgement by the UK's Supreme Court, in the case of Aintree v James,8 followed by a Court of Protection judgement in the case of M and Mrs N,9 has considerably, but by no means entirely, ameliorated my concerns. Although the Supreme Court judgement in Aintree v James was not concerned with ANH/CANH (clinically assisted nutrition and hydration) it was from my perspective a breath of fresh air from the highest court in the land in relation to the legal interpretation of the MCA. First, it crucially reaffirmed that Act's emphasis on the importance to be given to a legally incapacitated person's prior wishes, views, beliefs and values, even if these have been only informally expressed, and even in cases involving life-prolonging treatment, when assessing that person's best interests: As the judgement explains, (para 11.1) the MCA's tests for ‘best interests’ contain “a strong element of ‘substituted judgement’ ”.

The judgement also states explicitly that the Act does not give legally incapacitated patients or their representatives any more powers to instruct doctors what treatment they must provide than those patients would have had if they were legally competent. Thus (Para 17) ‘This Act is concerned with enabling the court to do for the patient what he could do for himself if of full capacity, but it goes no further. On an application under this Act, therefore, the court has no greater powers than the patient would have if he were of full capacity. The judge [in the first court] said “A patient cannot order a doctor to give a particular form of treatment, although he may refuse it. The court's position is no different”’.

The Aintree v James judgement cites with apparent approval (para 17) a previous Appeal Court judgement confirming the General Medical Council's assertion that if a doctor has assessed that it was not in a patient's best interests to provide a particular treatment, including ANH/CANH which was the main subject of that case, then the doctor was not legally obliged to provide it even if the patient demanded it.

The Aintree v James judgement also and crucially makes clear, in para 22, that ‘the focus is on whether it is in the patient's best interests to give the treatment, rather than on whether it is in his best interests to withhold or withdraw it. If the treatment is not in his best interests, the court will not be able to give its consent on his behalf and it will follow that it will be lawful to withhold or withdraw it. Indeed, it will follow that it will not be lawful to give it. It also follows that (provided of course that they have acted reasonably and without negligence) the clinical team will not be in breach of any duty towards the patient if they withhold or withdraw it’ (Emphasis added).

Finally and again to me reassuringly, the Aintree v James judgement also specifies that (Conclusions 1) ‘there is nothing in this judgement which is inconsistent with the sensible advice given by the General Medical Council in their guidance on Treatment and care towards the end of life: good practice in decision making’.

Thus I, as a non-lawyer, interpret the Supreme Court's judgement in Aintree v James as making it clear that doctors when assessing a legally incapacitated patient's best interests must assess wherever possible and as ‘an element of substituted judgement’ what that patient's views and values when competent would have been; and that the MCA does not give the Court of Protection any powers to require doctors to provide treatments that the doctors, following the Act's and the Act's Code of Practice's procedures for assessment of best interests, consider not to be in the best interests of their patients.

Part 3: the judgement in the M and Mrs N case

A more recent Court of Protection case, M and Mrs N8 encouragingly follows the Supreme Court's advice about the importance of an incapacitated patient's prior views and values, even when these have not been written down. Mrs N was treated by the Court as being in an MCS (there was disagreement among the medical expert witnesses as to whether she was in MCS or permanent vegetative state (PVS)) resulting from gradual deterioration over many years of her multiple sclerosis. Mrs N was previously a fiercely independent person who had vigorously resisted coming to terms with her increasing disabilities. However by the time of the Court hearing in 2015 she was either minimally conscious or not conscious at all, quadriplegic with generalised flexor spasticity, and affected by epileptic fits, chronic sacral pressure ulcers and recurrent chest infections. She was receiving ANH through a surgically implanted percutaneous endoscopic gastrostomy tube and her daughter M, as the judge explained: ‘saw in her no recognition of the outside world, no interaction with any stimulus and absolutely no quality of life whatsoever’ and asked the Court to agree that it was not in her mother's best interests to continue to be treated with ANH/CANH.

Placing great emphasis on the need to give full weight to the patient's prior views and values despite the absence of a legally valid and applicable advance decision to refuse life prolonging treatment, the judge in a very extensive judgment agreed with M that it was not in Mrs N's best interests to continue ANH/CANH. He explained [sections 59–61]:

‘The family has each, in their different ways, been able to create a clear and compelling impression of who Mrs. N is and what her values were. They have, if I may say so, given her a clear voice in this courtroom.

60. I am left with little doubt that Mrs. N would have been appalled to contemplate the early pain, increasing dependency and remorseless degeneration that has now characterised her life for so long. I have no difficulty in accepting the family's view that she would not wish to continue as she is. More than that, she would have wished to have discontinued her treatment some considerable time ago’. And the judge concluded:

‘….to force nutrition and hydration upon her is to fail to respect the person she is and the code by which she has lived her life’.

As well as in its emphasis on Mrs N's prior views wishes and feelings and the beliefs and values and other factors that would be likely to influence her decision about treatment had she retained legal capacity this judgment is also consistent with the Supreme Court's emphasis in Aintree v James that the important issue to be assessed is not whether a treatment should be withdrawn but whether it is in a patient's best interests and therefore lawful to provide it at all. Thus [section 68]

‘Baroness Hale in Aintree University Hospital Trust v James [2014] AC 591 also took care to ensure that the question was properly formulated not as the ‘withholding of treatment’ but focused instead on the patient's best interest.’

So should my worries about excessive judicial concern for sanctity of life be allayed by these two cases? I would say yes, were it not for the continuing apparent judicial conflict addressed in the next section.

Part 4: an apparent conflict—the legal requirement to prolong life unless the Court of Protection permits otherwise

It seems clear to this non-lawyer, and to the social scientists Professors Jenny and Celia Kitzinger1 ,2 and to the academic lawyer Professor Simon Halliday,3 that there is an obvious conflict between elements in the Code of Practice to the MCA and elements in the rules of the Court of Protection on the one hand, and the judgement of the Supreme Court in Aintree v James on the other hand. The Code of Practice and the Supreme Court Rules seem clearly to require doctors to provide a particular type of life-prolonging medical treatment to certain categories of patients even when those doctors, having followed the instructions of the MCA and its code of practice concerning assessment of best interests, believe that the treatment is no longer in the best interests of particular patients. The Supreme Court on the other hand reaffirms in Aintree v James an earlier judgement that if a doctor concludes that a treatment is ‘not clinically indicated he is not required (ie, he is under no legal obligation) to provide it’.

Thus the Code of Practice states that proposals to withdraw or withhold ANH/CANH from patients in PVS must be referred to the Court of Protection11 and the rules of the Court of Protection in PD9E,10 reiterated and confirmed in the M judgement,7 require proposals to withdraw or withhold ANH/CANH from patients in PVS or MCS to be referred to the court. PD9E is part of the Rules of the Court of Protection, and according to a legal analysis3 these Rules have the force of Statute, since the MCA empowered the Court of Protection to make its own rules of practice. (I thank an anonymous JME referee for introducing me to this scholarly paper). PD9E concerns applications to the Court of Protection that involve serious medical treatment and under the heading ‘Matters which should be brought to the court’ (emphasis added) subsection 5 states “Cases involving any of the following decisions should be regarded as serious medical treatment for the purpose of the Rules and this practice direction, and should be brought to the court: (a) decisions about the proposed withholding or withdrawal of artificial nutrition and hydration from a person in a permanent vegetative state or a minimally conscious state”. By extending the requirement to patients in MCS PD9E extends (and as I shall argue in section 5 entirely logically extends) to a further category of patients the legal requirement specified in the code of practice of the MCA that patients in PVS must be referred to the Court if it is proposed to withhold or withdraw ANH/CANH.

Given that the rules of the Court of Protection state clearly, in PD9E, that any proposed withholding or withdrawing of artificial nutrition and hydration from patients in PVS or MCS should be brought to the Court it seems indisputable to this non-lawyer that doctors are legally required to provide ANH/CANH to patients in MCS or PVS until and unless the Court of Protection authorises withdrawal in particular cases, even when doctors, having followed the instructions of the MCA and its code of practice concerning assessment of best interests, believe that the treatment is no longer in the best interests of particular patients. And it seems equally clear that this requirement is in plain contradiction to the assertions by the Supreme Court in Aintree v James, quoted in part 2 above, that the Court of Protection is not empowered to require a doctor to provide a treatment that the doctor considers not to be clinically indicated, and in the case of an incapacitated patient, not to be in the patient's best interests and therefore unlawful to provide.

Informal discussion has indicated that some lawyers reject these claims that the MCA Code of Practice and PD9E create a legal obligation on doctors to refer to the court before withdrawing ANH/CANH—instead, they merely advise that it is good practice to do so. My admittedly summary response is that while this distinction may signify a difference to lawyers, I am pretty confident that for doctors, health authorities and medical practice in general it is a distinction without a difference. If a Statute's Code of Practice, backed up by the Court of Protection's Rules and a specific unreversed judgement in that Court, tell doctors that it's ‘good practice’ to refer to the Court before withdrawing ANH/CANH in certain sorts of case then doctors and health authorities—and probably most patients' families, friends, loved ones and representatives too—will understand this to be a legal requirement, whether it is a direct legal obligation or merely legally recommended as good practice by a Statute's Code of Practice, by the rules of the Court of Protection, and by a particular judgement of that Court!

However if it is not a legal requirement, direct or indirect, the Court of Protection should surely make this pellucidly clear in its forthcoming revision of its rules. The revised rules could simply state that if a doctor, supported by his or her health authority were to withdraw ANH/ CANH from a patient in PVS or MCS on the grounds that the doctor, having carried out the assessment of the patient's best interests as stipulated in the MCA 2005 and its Code of Practice and having decided that it was no longer in the best interests of a particular patient to continue ANH/CANH, and where those who knew the patient and others caring for the patient agreed that provision of ANH/CANH was no longer in the patient's best interests, that doctor would not be committing any sort of legal offence or bad practice or failure of a duty of care under the MCA. In particular the doctor would not be in Contempt of that Court contrary to the suggestion in reference 3 at p.11.

On the other hand if the doctor would be committing an offence or bad practice or failing in a duty of care under the MCA the revised rules of the Court of Protection should explain what the offence, bad practice or failure of duty would be and what the likely sanction would be. Doctors, patients and their loved ones and carers need to know the answers to these questions. At present, let me add, it seems perfectly clear to me—but as a non-lawyer—that as the M judgement states (ref. 7, para 257) the Court of Protection through its PD9E does instruct doctors, directly or indirectly, that they must continue to provide a particular life-prolonging treatment to two categories of patients, even when the doctors, having followed the instructions of the MCA and its code of practice concerning assessment of best interests, believe that the treatment is no longer in the best interests of a particular patient. Furthermore this seems to me to be equally clearly in plain contradiction to the assertions by the Supreme Court in Aintree v James, quoted in part 2 above, that the Court of Protection is not empowered to require a doctor to provide a treatment that the doctor considers not to be clinically indicated, and in the case of an incapacitated patient, not to be in the patient's best interests. Indeed I understand, informally, that there even seems to be legal doubt as to whether a valid and applicable advance decision to refuse life-prolonging treatment including ANH/CANH (such as my own in online supplementary file 1), would be legally binding on a doctor and thus sufficient to override PD9E section 5(a), if the patient has already been diagnosed as being in MCS or PVS. (I am trusting in my own case that since my own ADRLPT requires all life-prolonging treatment to be withheld and withdrawn after 3 months of legal incapacity if I am medically judged very unlikely to recover legal capacity, there will not have been enough time for me to be formally diagnosed as being in MCS or PVS—but by then I and my doctors may face the further problem described in the next section.)

Part 5: logical implications of the requirement to treat unless the court permits otherwise

If I and others are right, as argued in section 4, that there is a clear legal requirement, whether direct or indirect, for doctors to provide a particular life-prolonging treatment, ANH/CANH, to all patients in PVS and MCS, until and unless the Court of Protection agrees that withdrawing the treatment is in a particular patient's best interests, it is important to ask why it is there. Given the name of the Court it seems obvious that the main purpose is to protect legally incapacitated patients in PVS and MCS from possibly mistaken decisions by doctors to withhold or withdraw a particular life-prolonging treatment. But as I have argued previously, briefly5 and then extensively,6 if legally incapacitated patients who have no consciousness or minimal consciousness must be legally protected in this way then logically all other legally incapacitated patients, whatever their degree of consciousness, should have the same legal protection. And since the obligation to refer to the court specifies both withholding and withdrawing of ANH/CANH the logical implication of these rules is that whenever any incapacitated patient's life might be prolonged by ANH/CANH doctors should provide such treatment until and unless the Court of Protection rules that it is not in a particular patient's best interests to do so. There must be many thousands of incapacitated patients who die each year in oncology services, palliative care services, accident and emergency services, intensive and critical care services, care of the elderly services, and indeed in old age care homes, whose lives could be prolonged by ANH/CANH but from whom such treatment is withheld because doctors decide that such treatment is not in those patients' best interests.

To pursue the logical implications further, why should provision of the life-prolonging treatment of ANH/CANH be legally singled out for special protection? It certainly seems clear that any available life-prolonging treatment falls under the definition of serious medical treatment provided in the Court of Protection's PD9E. Thus the withholding or withdrawing of any life-prolonging treatment ‘would be likely to involve serious consequences for P’ where ‘serious consequences’ are defined by the Court of Protection as those which ‘could have a serious impact on P’ including ……..‘potentially major consequences for P’ and or ‘a serious impact on P's future life choices’.10 If the life-prolonging serious medical treatment of ANH/CANH must not be withheld or withdrawn without Court of Protection approval then surely any other available treatment that might prolong the life of a legally incapacitated patient should similarly be provided until and unless the Court of Protection decides that it is not in the particular patient's best interests. Once again it is probably safe to assert that each year there must be thousands of legally incapacitated patients from whom potentially life-prolonging treatments other than ANH/CANH are withheld or withdrawn because doctors consider that these life-prolonging treatments would not benefit those particular patients. Such treatments include fluid and electrolyte replacement therapy, respirators, dialysis treatments of various sorts, a wide range of pharmaceuticals, potentially life-prolonging surgical operations, pacemakers and cardiopulmonary resuscitation procedures. To give but one example, the Intensive Care Society told the Court of Appeal in the case of Burke v GMC in 200513 that some 50 000 patients a year (presumably in the UK) are admitted to intensive care units of whom some 30% die while still in hospital, ‘most of whom die because treatment is withdrawn or limited’. The logic of PD9E, if implemented, would require each proposed withholding or withdrawing of an available life-prolonging treatment from a legally incapacitated patient to be referred to the Court of Protection, and meanwhile to be provided to the patient even though the doctor believes that the treatment is not in the patient's best interests.

Such a development would require a vast increase in referrals to the Court of Protection and in the meantime massive diversion of medical resources towards providing life-prolonging treatments for legally incompetent patients which doctors considered not to be provided in the patients' best interests and thus on the face of it to be provided unlawfully (since the MCA, as emphasised in Aintree v James, requires that any and all treatments for mentally incapacitated patients must only be provided if the provider reasonably believes that the treatment serves the incapacitated person's best interests). Furthermore unless governments were prepared to provide considerable additional funding for National Health Service provision of these treatments there would have to be a diversion of available funds away from treatments that doctors and patients do consider to be beneficial, in order to provide treatments that doctors and patients and their families do not consider to be beneficial in specific cases—but which would be legally required until and unless the Court of Protection agreed in particular cases that the treatments were not beneficial. As Richard Cookson, the health economist coauthor of ref. 3 argues, ‘In the context of a cash-limited NHS operating within a fixed budget, investment in one form of medical intervention comes at the cost of another’. At what he considers to be a conservative estimate, each referral of a patient in PVS to the Court of Protection costs an additional £122000.3 If all available life-prolonging treatments had to be provided to all legally incapacitated patients until and unless the Court of Protection authorised withdrawal the economic costs to the NHS would clearly be enormous. And as Cookson and his coauthors point out ‘Costs to the NHS are, ultimately, costs to individual patients in terms of foregone health care and the loss of health benefits from that foregone health care’.3

Part 6: two alternatives and a recommendation

It is important for patients and potential patients, their families and friends, and their doctors to know what the law states about the provision of life-prolonging treatments including ANH/CANH to patients who are legally incapacitated from making their own treatment decisions. I think I have demonstrated that at present there are contradictory elements in UK law, with the Supreme Court stating in Aintree v James that the MCA does not give the Court of Protection powers to order doctors to provide particular treatments—‘A patient cannot order a doctor to give a particular form of treatment, although he may refuse it. The court's position is no different.’. In apparently clear contradiction to this the Court of Protection's PD9E directly or indirectly requires doctors to provide ANH/CANH to patients in PVS and MCS, even when the doctors supported by a patient's family and by other carers believe such treatment to be no longer in a patient's best interests, until and unless the Court of Protection has ruled, in each individual case, that it agrees with the doctor.

It is important to note the development of this legal obligation. It did not originate in an Act of Parliament but in the legal case of Anthony Bland, finally determined by the House of Lords.13 In that judgement the practice of applying to the Court for approval of proposed withdrawal of life-prolonging ANH treatment in cases of PVS was judged desirable, as Lord Keith put it, ‘at least for the time being and until a body of experience and practice has been built up which might obviate the need for application in every case’ (ref. 13, p. 4). This potentially temporary element of good medical practice was strengthened in the Code of Practice of the MCA which states: ‘8.19 The case law requirement to seek a declaration in cases involving the withholding or withdrawing of artificial nutrition and hydration to people in a permanent vegetative state is unaffected by the Act and as a matter of practice, these cases should be put to the Court of Protection’.11 The Rules of the Court of Protection in PD9E further extend (entirely logically, as argued in part 5), the requirement to refer to that court cases of MCS in addition to cases of PVS.9 Thus from a recommendation in a particular legal judgement concerning possibly temporary good medical practice the Court of Protection is potentially developing a role for itself as an executive and supervisory arm of the medical profession. For as argued above the Court is logically, though not yet legally, committed to requiring all decisions to withhold or withdraw available life-prolonging treatments from all legally incapacitated patients to be referred to the Court meanwhile requiring doctors to provide those treatments even in uncontested cases where the doctors, supported by patients' relatives and carers, believe that it is not in the best interests of particular patients to do so.

Of course one way of resolving the apparent contradiction between the judgement in Aintree v James and these requirements of the Code of Practice and of the Court of Protection is for the law to make clear that indeed all legally incapacitated patients must be given all and any available life-prolonging treatment until and unless the Court of Protection agrees in individual cases that this is not in the particular patient's best interests. This would extend the protection against mistaken medical assessments currently afforded to patients in PVS and MCS to all legally incapacitated patients. I have argued that pursuing this solution would lead both to the compulsory overriding of what many people would, when competent, have wished for themselves and to a highly undesirable skewing of healthcare towards non-beneficial life-prolonging treatment.5 ,6 It would also require an enormous increase in resources for the Court of Protection, as it turned itself into an effective executive and supervisory arm of the medical profession, to enable it to cope efficiently with the vastly increased workload that would result from pursuing this option.

Far preferable it seems to me is to pursue the reasoning of the Supreme Court in Aintree v James as outlined in Part 2 above and therefore to modify those aspects of the Rules of the Court of Protection and of the Code of Practice of the MCA that contradict that reasoning. The amended Rules could make clear that, following the judgement of the Supreme Court in Aintree v James, referrals to the Court of Protection concerning withholding or withdrawing of medical treatments, including life-prolonging and life-sustaining treatments, are not legally obligatory and that the Court does not and cannot instruct doctors what treatments they must provide—including in cases of PVS and MCS. However doctors could be advised that it would be good practice to refer particularly difficult decisions about withholding or withdrawing life-prolonging treatments of which the most important are cases when there is disagreement, or continuing uncertainty, about an individual patient's best interests that can't be resolved by discussion, provision of a second opinion, discussion with a clinical ethics committee or mediation. The Rules might explicitly remind doctors that they are of course expected to make their serious medical treatment decisions for legally incapacitated patients about withholding or withdrawing life-prolonging treatments, in accordance with the procedures for determining a patient's best interests specified in the MCA and its Code of Practice—but excluding the parts of that Code requiring referral to the Court of Protection. The Court of Protection could recommend to Parliament that the Code of Practice should be revised to accord with the Supreme Court's judgement in Aintree v James, and state that meanwhile the Code should be interpreted in accordance with that judgement and in accordance with the revised PD9E of the Court of Protection. Such modifications would of course require detailed legal drafting and consultation but the underlying purpose should be to bring PD9E and the Code of Practice into line with the enlightened and presumably legally overarching interpretation of the MCA manifested in the Supreme Court's judgement in Aintree v James. Such essentially simple amendments would, I believe, cut through what is potentially an ever more complex Gordian knot.

Part 7: some reflections

My paper is based on several presuppositions. The first is that there can be an important philosophical, ethical and legal distinction to be made between deliberately killing a patient and allowing that patient to die. Not all philosophers agree with that presupposition but it accords with most medical legal and common sense views. In the context of withdrawing life-prolonging treatment, as Lord Goff pointed out in the Bland case, ‘the question is not whether it is in the best interests of the patient that he should die. The question is whether it is in the best interests of the patient that his life should be prolonged by the continuance of this form of medical treatment or care’ (ref. 13, p. 14). Corresponding to that distinction between killing and allowing to die are two aspects of the widely accepted moral and legal right to life; a negative right not to be deliberately killed (a right that may be lost by aggressors) with the corresponding duty of others not deliberately to kill people (again not necessarily applicable to aggressors); and the much more restricted positive right to be kept alive, with its corresponding duty on at least some others, and especially doctors, to keep people alive. In relation to this second positive duty my second presupposition is that while medical practice always starts with a presumption in favour of prolonging life, medical prolongation of life is not an end in itself—it is not based on the ‘sanctity of life’—but is instead properly understood as a means to an end—the morally enormously important beneficial end of prolonging lives that the people living those lives would wish to have prolonged.

Medicine's ever-growing capability of prolonging life in states that people would not wish to have prolonged requires us all, especially our legislators and our judiciary, to distinguish carefully between doctors' legal obligation not intentionally to kill others or help them to kill themselves—the voluntary euthanasia and assisted suicide debate—and doctors' legal obligation to prolong life. The negative and positive aspects of the right to life are increasingly frequently and mistakenly conflated. Even if we agree that doctors should be legally prohibited, along with everyone else, from deliberately killing any of their patients or helping them to kill themselves—no matter how benevolent their intentions—and I do agree and this paper is entirely consistent with that agreement—we should be very careful not to identify this negative obligation within the ‘right to life’ with the more limited positive obligation that doctors have to sustain and prolong their patients' lives. That positive obligation must be tempered first and foremost by patients' own views when these can be discovered, but also by professional knowledge and skill in determining whether or not particular treatments to prolong a particular patient's life would actually benefit the patient, would do so without excessive harm, and would do so without involving unfair use of limited healthcare resources thus unjustly depriving other patients of beneficial healthcare. And of course doctors' positive obligation to prolong their patients' lives must also conform to morally acceptable laws.

We can all do our part towards clarifying the first of these elements by writing our own Advance Decisions (if we wish to refuse certain treatments) and Advance Statements indicating our values wishes and preferences and indicating treatments and interventions that we would like to have. These will help our doctors to make decisions about our healthcare once we no longer have legal capacity to make our own healthcare decisions. My own very detailed AD (online supplementary file 1) is available in the online version of this paper, both to solicit responses on ways to improve it, and in case people may wish to consult it when developing their own ADs. This may interest people who share my vigorous concern of avoiding life-prolonging treatment when irreversibly in a legally incapacitated state; but it may also interest those who have the diametrically opposite preferences; none of us is legally entitled to have demands for specific treatments met but our advance statements of treatment preferences must be taken into account when doctors assess our best interests once we have lost legal capacity to make our own decisions. Those who have strong wishes, for example, to be given life-prolonging ANH/CANH until they die can say so and are then very likely to have their wishes fulfilled, as the General Medical Council makes clear in its advice to doctors14 (advice specifically commended as ‘sensible’ in the conclusions of the Supreme Court's judgement in Aintree v James). I have added links to organisations that help people create their own advance decisions.ⁱ

But our judiciary too can and in my respectful view should also do its part by clarifying to what extent—if any—the law requires doctors to administer life-prolonging treatments to their legally incapacitated patients, even when those doctors, in accord with patients' relatives and carers, believe that providing such treatment is against their patients' best interests, until and unless judges have scrutinised the individual patients' cases themselves. In the current state of the law it seems clear that doctors must continue to treat their patients in PVS and MCS with one type of life-prolonging treatment—ANH/CANH—until a judge agrees that it is not in the individual patient's best interests to continue to do so. Furthermore, as argued in section 5 above, the logical implication of this legal requirement is that all legally incapacitated patients whatever their diagnoses, should be similarly protected by being treated with any available life-prolonging treatment until the Court of Protection rules otherwise in particular cases.

The MCA gives clear and admirably patient-supportive instructions about the legal requirement to act in the best interests of people who do not have legal capacity to make their own healthcare decisions and gives clear instructions about how to assess those best interests. I believe doctors, as a profession, can and should be trusted by the Court of Protection to follow those requirements. In this context let me commend the words of their fellow judge Lord Goff, again in the Bland case:The truth is that, in the course of their work, doctors frequently have to make decisions which may affect the continued survival of their patients, and are in reality far more experienced in matters of this kind than are the judges. It is nevertheless the function of the judges to state the legal principles upon which the lawfulness of the actions of doctors depend; but in the end the decisions to be made in individual cases must rest with the doctors themselves. In these circumstances, what is required is a sensitive understanding by both the judges and the doctors of each other's respective functions, and in particular a determination by the judges not merely to understand the problems facing the medical profession in cases of this kind, but also to regard their professional standards with respect. (ref. 13, p. 18)