In favour of withdrawal

Some pointed to the lack of benefit for Vincent Lambert in continuing existence in a state of unconsciousness. Because he lacked any conscious interests, it was not in his best interest to keep him alive.11 If treatment did not confer any positive benefit to Vincent, it may be regarded as futile, or pointless.12 French legislation on patient’s rights and end-of-life issues specifically direct doctors not to continue treatment with ‘unreasonable obstinacy’,3 this includes treatment ‘n’ont d’autre effet que le seul maintien artificiel de la vie’ (having no effect other than to maintain life artificially).13

In Vincent Lambert’s case, his wife maintained that while he did not have a written advance directive, he had repeatedly expressed his wish not to be kept alive in a highly dependent state. The Conseil d’Etat ruling had noted

It is apparent …from the testimony of Mrs Rachel Lambert, that she and her husband, both nurses, had often discussed their respective professional experiences… and that Mr Lambert had on several such occasions clearly voiced the wish not to be kept alive artificially if he were to find himself in a highly dependent state. The tenor of those remarks, reported by Mrs Rachel Lambert in precise detail and with the corresponding dates, was confirmed by one of Mr Lambert’s brothers6

If this is correct, continued treatment violated his prior autonomy, and seemed to frustrate Lambert’s right to have a say in his medical care and his right not to have his life prolonged in a state that he would have regarded as deeply undesirable. It may be thought to deprive Lambert of his right to die with dignity,14 15 and to discriminate against him since, were he not so profoundly disabled, he would have been able to refuse treatment.

Against withdrawal

However, it is not clear that artificial nutrition and hydration was ‘futile’ for Vincent Lambert.16 Such treatment had manifestly succeeded in the goals of providing nourishment and in keeping him alive over the last 11 years. It may be that some judge the quality of his life to be too low to warrant keeping him alive, but others, notably Lambert’s parents, disagreed. Whether it is ‘reasonably’ or ‘unreasonably’ obstinate to sustain life in this state is an ethical, not a scientific, question. While some, as noted above, would have found his continued treatment undignified, dignity is a deeply contested concept in bioethics.17 18 Others argued that Lambert’s essential human dignity remained despite his profound disability, and that his life should have been protected ‘until its natural end’.19

The UNCRPD was due to hear a petition relating to the Lambert case. This committee might have had concern for the rights of the severely disabled to receive life-prolonging medical treatment. The decision to withhold treatment from Lambert was directly related to the extent of his brain injury. If he were less severely disabled, artificial nutrition and hydration would potentially bring benefits other than to maintain life artificially; there would not be a question of ‘unreasonable obstinacy’. As a consequence, the UNCRPD might have felt that the Conseil d’Etat decision represented discriminatory denial of healthcare on the basis of disability (and was thus arguably contrary to Article 25 of the UN Convention on the Rights of Persons with Disabilities).20

However, there is a potentially competing right. Article 12 of the UN Convention emphasises the importance of advance planning, and of respecting the wishes and preferences of disabled persons (including their prior expressed preferences).21 If the decision, by Lambert’s doctors, was based on an accurate assessment of his individual situation, of his best interests and his wishes, it arguably did not in any way represent unjust discrimination. Furthermore, if it is accepted that Vincent had ‘no interests’ because he was permanently unconscious (as was concluded in the case of Tony Bland),11 it is not clear that he could be discriminated against.

Dissensus

It can be useful to debate such questions, yet the long history of similar cases points clearly to the ongoing challenge of reaching a common view on these issues. Quite simply, we will never all agree on what should happen .

Our societies are increasingly diverse. In societies like France, the UK, the USA and many others, we have to accept and tolerate that people have a range of different values. That acceptance and tolerance means that we should allow people to live their lives based on their own ethical views and values (as long as they do not harm others). It is perfectly acceptable for people to express their views about situations like Vincent Lambert. However, it would not be acceptable to impose those views onto Vincent’s life.

Sometimes, in the face of a lack of consensus, it can be tempting to avoid making decisions. However, in situations like this, there is no neutral alternative. Continuing treatment is, itself, a decision.ii

The best ethical response to reasonable disagreement in cases like that of Vincent Lambert is to make decisions based on his values and his wishes. If Vincent himself would not have wished to remain alive, then the wishes of other family members, or politicians, religious leaders or the UNCRPD are not relevant. Treatment should be withdrawn.

If Vincent would have wished to remain alive, then artificial nutrition and hydration should potentially be continued. If there is disagreement about what the patient’s wishes were or would be, there is a need for an impartial process such as a court to weigh up the available evidence. In Lambert’s case, the Conseil d’Etat and the European Convention on Human Rights had carefully weighed up such evidence and found that it did not support continuation of treatment.6

Drawing on ethical dissensus, we should accept reasonable disagreement and provide treatment that respects the wishes of the patient. We recommend erring on the side of providing options that are of possible benefit or that are most likely to respect the prior autonomy of the patient.

In favour of withdrawal

There is an argument which is rarely made in relation to such cases (relevant exceptions are the studies by Kondziella et al and Daniels9 22). Even if Lambert had wished to continue to be treated, there is a reason not to do this: distributive justice. People do not have the right to demand limited community or medical resources such as artificial nutrition and hydration, nursing and hospital care. Such resources must be allocated according to reasonable objective values.

No matter how much a person wants to be kept alive in an unconscious state, we must ask: is this a fair and just use of community resources?

In the UK, it is estimated that the nursing home care for patients in a VS costs £90 000–120 000 per year.23 Wade has pointed out that this value is considerably in excess of the £20 000–30 000 cost/quality-adjusted life years (QALY) threshold commonly used for deliberations about providing novel therapies in the National Health Service (NHS).24 As an example, the NHS has recently declined to fund the novel therapy for cystic fibrosis lumacaftor/ivacaftor.25 This drug costs a similar amount per year per patient (£105 000) as the nursing home care for patients in VS; however, the benefit for patients with cystic fibrosis appears considerably greater (an additional 2.9 years of life/2.4 QALY on average).26 Although it may be considered controversial to assess whether nursing home care for VS is cost-effective, it does not seem controversial or even contestable that funding lumacaftor/ivacaftor would be more cost-effective.iii

From information available in the public domain, it does not appear that Lambert has other major health conditions (ie, affecting other organs than his brain). However, imagine that he developed renal failure and dialysis were judged to be consistent with his prior wishes. In contrast with artificial nutrition and hydration, there is widespread acceptance that renal dialysis should not be maintained for patients in VS. This is accepted, even in countries that are normally loathe to accept rationing of treatment. An American Society of Nephrology shared decision-making guideline indicates that dialysis should be withheld or withdrawn from ‘patients with irreversible, profound neurological impairment such that they lack signs of thought, sensation, purposeful behaviour and awareness of self and environment’.27 28 (Of relevance, the estimated annual cost of dialysis is approximately £30 000,29 one-third the cost of long-term nursing home care.) In the case of Tony Bland, artificial nutrition and hydration were classified as ‘medical’ treatments.11 This suggests that they should be treated in the same way as dialysis—and potentially withheld from patients in VS.

Against withdrawal

Others would argue that it would be wrong to withdraw artificial nutrition and hydration from patients in a VS, if they would have wanted such treatment.

There is no question that the nursing and daily care of patients with severe disorders of consciousness is expensive. The cost of treatment is obviously higher than the cost-effectiveness limits that are sometimes used to decide about treatment.

However, many people do not accept that artificial nutrition and hydration are ‘treatment’,30 and would strongly oppose the idea that nursing home care should be regarded as optional. Some hold that provision of food and water is a basic human right and there is a moral obligation to provide it where it is possible to do so.

Moreover, decisions about allocation of resources are not made solely on the basis of maximising benefit. Other values are important to our community, and those values need to be balanced against benefit.31 32 For example, giving weight to fairness may mean departing from pure cost-effectiveness thresholds.31 In the case of patients with VS, fairness or equality would potentially favour providing treatment. After all, patients with less severe neurological impairment would unquestionably receive artificial nutrition and hydration if this were desired. It is deeply unfortunate and unfair for someone to sustain brain injury that leaves them permanently unconscious. It is arguably doubly unfair if they are then denied treatment (which they would have desired) because of the severity of their injury.33

Moreover, when society makes decisions about allocating treatment, it must assess whether there is a basic minimum level that should be provided to all patients. Some treatments may need to be rationed (ventilators, cancer drugs, organs for transplantation), but other treatments are provided uniformly where they offer some possible benefit (pain relief, basic nursing care). One justification for setting minimum standards, of this sort, is that this reflects a prioritarian concern for the worst off.34

Societies may differ in which treatments they regard as part of a basic level of healthcare provided to any and all patients, and which will be provided selectively. In the UK, for example, it would be generally judged to be appropriate not to admit to intensive care a patient in VS, nor to provide renal dialysis, nor to list for organ transplantation. However, for the present, at least, artificial nutrition and hydration and nursing home care are not subjected to rationing. That is, indeed, why it has never been subject to the sort of cost-effectiveness analysis that lumacaftor/ivacaftor has been.

We have suggested above that the UNCRPD should not classify as discrimination a decision to withhold treatment based on a patient’s wishes (and his interests). However, the committee may view differently an argument based on cost-effectiveness and disability. If that were the basis for stopping treatment in Lambert’s case, the UNCRPD could potentially have concluded that this breached his Article 25 right to health, and Article 2 right to life.

On the other hand, unless there is a reason to treat artificial nutrition and hydration differently from other treatments, it is difficult to see how this would not also apply to decisions to withhold dialysis, organ transplantation or intensive care admission.

Dissensus

There is reasonable disagreement about how to allocate healthcare resources like long-term nursing home care. However, unlike questions of individual benefit/harm, this cannot be resolved by deferring to the preferences of the patient (or surrogate). Some people now demand intensive care even in the face of meeting criteria for death.35 And the end point of dementia, including Alzheimer’s disease, is unconsciousness. There will be large numbers of people whose lives could be prolonged who are unconscious.36 In the dissensus approach, we proposed that treatments that are probably of overall benefit and probably fall within the affordability limits of the healthcare system should be provided.10 There is then a question of what this means for continued treatment for patients in VS.

Our aim here is not to resolve the question of whether nursing home care and artificial nutrition and hydration should be withheld from patients in VS on the basis of limited resources. Instead, our aim is to point out that it is that question which needs to be addressed and resolved at a community level.

The focus of debate about treatment for patients with severe disorders of consciousness needs to shift away from the question of whether it is permissible to allow them to die—instead to the question of whether it is permissible to keep them alive indefinitely.iv If they choose to use their own funds for this purpose, there is no obvious distributive justice objection.v However, when they request the use of limited community resources, then such a decision becomes a communal one.

There is clearly going to remain disagreement about allocation of resources. The dissensus approach does not settle how such disagreement should be resolved. A democratic process of deliberation will necessarily tend to favour the values and views of the majority. However, deliberators should take into account the range of values and views held within society. There may not be a single right way to balance competing values in the face of reasonable disagreement. One option may be to settle on a fair process for allocation of resources.22

A community like France or the UK might decide that it will keep the unconscious alive with medical treatment like artificial nutrition and hydration, if they so desire. Well-resourced healthcare systems may be able to afford to provide this as a basic minimum level of healthcare. If they do, it may mean that nursing home care and community nursing should receive more funding that they do at present, at the expense of hospitals and pharmaceutical budgets (and, inevitably, reducing treatment availability for other patients).

However, that decision should be made in an open, transparent fully informed and rational way, with awareness of the opportunity cost that it involves. One estimate suggests that there are between 4000 and 16 000 patients with VS in nursing homes in the UK.37 At the same time, there are approximately 10 000 patients in the UK with cystic fibrosis.38

We will not all agree the right way to allocate limited resources, but given the large numbers of affected patients, and ongoing examples of conflict, these uncomfortable questions must be confronted.