By Charles Foster
From the time of the decision of the House of Lords in Airedale NHS Trust v Bland (1993) until the decision of the Supreme Court in An NHS Trust v Y (2018) (which I will refer to here as ‘Y”) it had been understood that the withdrawal of life-sustaining treatment (typically clinically assisted nutrition and hydration – ‘CANH’) from patients in a vegetative state should be endorsed by the court. Over the years, this practice had been extended to cover such withdrawals in Minimally Conscious States too.
In Y, the Supreme Court held that there was no requirement for court review or endorsement. Why?
There were broadly two reasons:
First, the process of taking a case to court might itself be intrinsically undesirable. Lady Black commented, at [121]:
‘As King LJ observed in In re Briggs, quite apart from the pressure that court cases place on the overstretched resources of NHS trusts, they add greatly to the strain on families facing acutely distressing decisions. In a case where all the proper procedures have been observed and there is no doubt about what is in the best interests of the patient, there is much to be said for enabling the family and the patient to spend their last days together without the burden and distraction, and possibly expense, of court proceedings. In addition, I do not disagree with Peter Jackson J’s observation that there is a risk that the need to go to court might deflect clinicians and families from making true best interests decisions and might lead in some cases to inappropriate treatment continuing by default. Equally, it is not inconceivable that it might, as the BMA suggest, generate a reluctance, in some cases, to start CANH because of the procedures attending its withdrawal.’
And second, if there was agreement between the relevant clinicians and those who had a legitimate interest in the patient’s welfare, and if the provisions of the Mental Capacity Act 2005 had been complied with and the relevant professional guidelines followed, no reference to the court was necessary to safeguard the patient against the (uncontroversially real) risks of, for instance, a questionable diagnosis or prognosis, and a consequentially wrong determination of best interests.
Much of the argument before the Supreme Court centred on the adequacy of the professional guidelines. They were, it was contended, an exhaustive statement of good practice. If followed, no one, including the court, could possibly have a legitimate concern. The Supreme Court agreed:
‘The documentation supplied to us shows that the difficulty that there is in assessing the patient and in evaluating his or her best interests is well recognised. The process is the subject of proper professional guidance, covering vitally important matters such as the involvement in the decision-making process of a doctor with specialist knowledge of prolonged disorders of consciousness, and the obtaining of a second opinion from a senior independent clinician with no prior involvement in the patient’s care. The second opinion, as contemplated in the guidance …. is, in my view, a crucial part of the scrutiny that is essential for decisions of this sort, and the guidance sets parameters which should ensure that it is an effective check, in that the clinician who provides the second opinion must (so far as reasonably practical in the circumstances of the case) be external to the organisation caring for the patient, and is expected to carry out his or her own examination of the patient, consider and evaluate the medical records, review information about the patient’s best interests, and make his or her own judgement as to whether the decision to withdraw (or not to start) CANH is in the best interests of the patient. Thus the interests of patients and their families are safeguarded, as far as possible, against errors in diagnosis and evaluation, premature decisions, and local variations in practice. If, at the end of the medical process, it is apparent that the way forward is finely balanced, or there is a difference of medical opinion, or a lack of agreement to a proposed course of action from those with an interest in the patient’s welfare, a court application can and should be made….’ (At [124] – [125])
The court concluded that:
‘If the provisions of the MCA 2005 are followed and the relevant guidance observed, and if there is agreement upon what is in the best interests of the patient, the patient may be treated in accordance with that agreement without application to the court….’ At [126]
But the ‘if’ is terribly important. Are the relevant procedures diligently followed? Was it in fact safe to dispense with court monitoring? A recent article by Gray, Pickering and Sturman in Clinical Medicine suggests not.
Freedom of Information (FOI) requests were sent to 342 hospital trusts and clinical commissioning groups in England – representing 95% of providers. The response rate was 88.8%.
They were asked if, in their trust/CCG, any formal register was kept of deaths resulting from the withdrawal of CANH that occurred under the provisions of the Mental Capacity Act, and to which the BMA/RCP guidelines of 2018 applied? There were supplementary questions about whether, if such a register was kept, there was any independent internal or external audit made of such deaths, and the degree of compliance with the guidelines, the number of such deaths in 2018 and 2019, and the proportion of cases where the checklist in the guidelines had been used and fully completed.
Not one respondent kept a register of CANH withdrawal cases. The supplementary questions were therefore otiose.
An FOI request to the Care Quality Commission (CQC) asked if the trusts, practices or other organisations inspected kept a register of the type detailed above, and if so, whether the CQC performed an audit of the deaths and of compliance with the guidelines, and if any such audits were performed, what they showed about the number of cases and the use of the checklist.
The CQC responded that they held no information about these matters, and that ‘No register of deaths occurring as a result of withdrawal of [CANH] is required to be kept by organisations registered with CQC.’
Whatever one’s ethical stance on the withdrawal of CANH from patient in prolonged disorders of consciousness (PDOC), these responses should cause grave concern. There simply is no way of knowing whether, in the words of Lady Black in the Supreme Court, ‘the provisions of the MCA 2005 are followed and the relevant guidance observed….’ The presumption of compliance with the Act and with the guidance was at the root of the Supreme Court’s decision in Y. Everyone agreed in Y that safeguards were ethically and legally necessary. If they are not in fact in place there are some serious ethical and legal worries. The register anticipated by the freedom of information request, and the audits by the trusts/CCGs and the CQC based on the information that would be in such a register would not be administratively onerous. But even if they would be onerous, they are necessary to ensure that Y does not have the effect of elbowing the law out of decisions about CANH withdrawal in PDOC cases.
Article 2 of the European Convention on Human Rights contains the ‘right to life’. Article 2(1) provides that ‘Everyone’s right to life shall be protected by law. No one shall be deprived of his life intentionally save in the execution of a sentence of a court following his conviction of a crime for which this penalty is provided by law.’ Article 2 also gives to a ‘procedural’ obligation to ensure that there is adequate review of cases where lives are lost under the auspices of a state institution (as is the case for NHS patients).
There is a positive obligation to adopt appropriate measures for the protection of lives. There is, too, a duty to ensure the effective functioning of the protective regime. This includes a duty of supervision and enforcement.
The criminal law of course extends to NHS hospitals. So does the jurisdiction of the coroner. Those are both ways in which a state’s procedural obligations under Article 2 might be satisfied. But are they sufficient in PDOC withdrawal cases?
If a best interests determination under the Mental Capacity Act is made, as envisaged by the Supreme Court in Y, using the relevant professional guidelines diligently as a template for deciding where the patient’s best interests lie, the determination will be lawful. Since the Supreme Court in Y presumed that those guidelines would be used, it follows that if they are not, the best interests determination has not been reached in the way decreed by the country’s highest court, and (it would depend on the facts) may also be wrong in its result.
If a best interests determination is made wrongly, there is the possibility of personal liability on the part of the decision-maker. There is a statutory defence. Section 5 of the Mental Capacity Act provides:
(1) If a person (“D”) does an act in connection with the care or treatment of another person (“P”), the act is one to which this section applies if—
(a) before doing the act, D takes reasonable steps to establish whether P lacks capacity in relation to the matter in question, and
(b) when doing the act, D reasonably believes—
(i) that P lacks capacity in relation to the matter, and
(ii) that it will be in P’s best interests for the act to be done.
(2) D does not incur any liability in relation to the act that he would not have incurred if P—
(a) had had capacity to consent in relation to the matter, and
(b) had consented to D’s doing the act.
(3) Nothing in this section excludes a person’s civil liability for loss or damage, or his criminal liability, resulting from his negligence in doing the act.
(4) Nothing in this section affects the operation of sections 24 to 26 (advance decisions to refuse treatment).
It is open to doubt whether a clinician who had reached a decision about best interests without following the professional guidelines, as presumed by the Supreme Court, could ‘reasonably believe’ that it was in the patient’s best interests to (e.g) withdraw treatment – and so avail herself of the statutory defence under s. 5. The section refers to ‘acts’, and withdrawal is classified as an omission, but that, I suggest, does not alter the basic jurisprudence of the situation. It is true that it is the continued treatment of a patient that requires justification, but no one doubts that if continued treatment is the patient’s best interests, the state has (subject to arguments about resources which do not arise here), an obligation to continue to provide it – and also that failure to provide it might amount to a substantive breach of Article 2.
A ‘reasonable’ decision will no doubt be an Article 2 compliant decision. And – and this is the pont of raising section 5 here – vice versa. The section 5 defence relates to individual liability on the part of a decision-maker, but it is strongly arguable that part of the state’s responsibility is to monitor whether its agents in the discharge of its substantive Article 2 obligations are making their decisions in an Article 2-compliant way. Can it really delegate that responsibility to (e.g.) the General Medical Council?
If there is no way of knowing on a systemic level whether the Mental Capacity Act is being flouted by the Supreme Court’s implicit guidance being ignored, it seems arguable that the state is in breach both of its positive and procedural obligations under Article 2. It is surely no answer to say that a breach applies only to an individual case rather than to a large conglomeration of individual cases. The Bland rule – whereby PDOC cases would be reviewed – purportedly applied to all cases. It was replaced in Y by a declaration that all cases would be dealt with safely and lawfully if the guidelines were followed. Both those decisions were seen as articulating the state’s Article 2 obligations. If those guidelines are not in fact being followed, there will be many individual breaches of the Article each year. There is an obligation to monitor to see if the regulatory regime demanded by Article 2 is working. A register and the possibility of consequential audit may be an indispensable part of the regulatory regime.
Not knowing whether the law is being complied with might itself be unlawful.
A thorough and detailed piece of work, thank you for addressing this important issue.
Comments are closed.