Wednesday, 13 August 2008

Living Wills

In this posting I want to explore a familiar theme of this blog respect for autonomy. In previous postings I have stressed the importance of respect for autonomy. However I will now argue in the specific context of living wills respect for a patient’s autonomy is usually misplaced and that as a result the circumstances in which a patient’s living will should be respected are extremely limited.

A legal will states how someone’s goods are to be distributed after her death. A living will states someone’s preferences as to how she wishes be treated if she ceases to able to give fully informed consent for herself. The terms ‘living will’ and ‘last directive’ have the same meaning. In this posting I will use the term ‘living will’ as I believe this term best reflects our intuitive ideas. In the U.K. the validity of living wills is recognised by The Mental Health Act of 2005 and The Department of Health’s ‘Reference Guide to Consent for Examination or Treatment’ page 10. A living will is only valid if it refers to the treatment proposed and the will maker was competent at the time she made her will. If a patient is competent this usually means she meets the requirements of the so called C test (Re C, [1994] 1 All ER 819). However in practice there seems to be no way to assess whether a will maker actually meet the standards laid down by the C test at the time she made her will. When considering a patient’s living will it seems she would be regarded as competent when she made her ‘living will’ provided she was autonomous at the time.

It does not automatically follow from the fact that, we should respect someone’s current autonomous decisions, that we should also respect her past autonomous decisions. Consider the following scenario. Suppose someone is autonomous and signs and gets witnessed a document specifying how her vote should be cast in any future general election if she suffers from dementia at the time. It might be assumed for the sake of argument that this person has always supported only one party. At the date of a future general election such a document would be meaningless even if the person involved actually did suffer from dementia. However such a document seems in some ways to be analogous to a living will. This example suggests that perhaps we need not respect all living wills in all circumstances. I will consider whether we should respect a patient’s will in two sets of circumstances.

  1. When a patient has permanently lost her capacity for autonomy. In these circumstances I will argue there are no reasons to respect her living will.
  2. When a non-autonomous patient is expected to regain her capacity to make autonomous decisions but has permanently lost her capacity to implement these decisions. In these circumstances I will argue her living will need only be respected when the treatment or non-treatment specified in her will is irreversible.


Consider an elderly patient with dementia who catches pneumonia. Let it be assumed that this patient’s dementia means she has permanently lost her capacity to make or implement autonomous decisions. Clearly such a patient cannot give valid informed consent. Let it be further assumed her doctors believe provided she is placed on a ventilator for a short time she will recover and be able to live in the state she was in immediately prior to catching pneumonia. Let it be still further assumed that prior to her becoming demented this patient made a living will specifying that in these precise circumstances she was not to be placed on a ventilator in order to prolong her life. Lastly let it be assumed that this patient, prior to catching pneumonia, enjoyed simple pleasures such as sitting in the garden even if prior to becoming demented she hated gardens and enjoyed mountaineering. Under current legislation and medical guidelines it would seem such a patient should not be placed on the ventilator even though she had a reasonable quality of life.

I wish to question whether respecting the living will of such a patient is really connected to respecting her autonomy in circumstances when she will never be autonomous again? In what follows accepting or respecting someone’s decisions refers to decisions that do not harm others unless stated otherwise. Let it be accepted the reason why we respect someone’s living will is in order to respect her informed consent decision. Let it be further accepted the reason why we should respect someone’s informed consent decision is in order to respect her autonomy. I have previously argued that making autonomous decisions does not require any great intellect and that autonomous decisions need not of necessity be good decisions, see my posting of 01/07/08. It follows before a doctor thinks about respecting a living will she must first be totally sure the patient is non-autonomous. However let it be assumed our patient is clearly non-autonomous. Does respecting this patient’s living will respect her autonomy? Clearly it does not respect any autonomy she now possesses because she is non-autonomous. Does then respecting her living will respect her previous status as an autonomous person? The answer to this second question is yes. However the answer to this second question suggests a third question. If it is accepted that we should respect someone’s status as an autonomous person does this mean we should also respect her previous status as an autonomous person? In order to answer this third question we should consider a fourth question; why do we respect someone’s status as an autonomous person?

We do not respect someone’s status as an autonomous person on a mere whim but because we believe autonomy has value. Autonomy has both instrumental and intrinsic value. However in the case of a patient who is non-autonomous it is hard to see how her previous status as an autonomous person could possibly have any instrumental value to her now. It follows in the context of living wills if we respect autonomy we do so because of the intrinsic value of someone’s autonomy in the past. I have previously argued the intrinsic value of someone’s autonomy depends upon the recognition of her as the kind of person capable of making her own decisions, see my posting of 23/03/08. It follows respecting the intrinsic value of someone’s autonomy means accepting her decisions that do not harm others. Now that we are clear about why we might respect autonomous decisions, in the context of living wills, we are in a position to tackle the question as to whether we should respect someone’s previous status as an autonomous person. Consider someone who has previously made an autonomous decision to prefer x to y. Let it be assumed she changes her mind and makes a further autonomous decision preferring y to x. Clearly we do not respect her autonomy if we fail to accept her current autonomous decision preferring y to x because we feel we should respect her previous autonomous decision preferring x to y. One reason why we fail to respect her autonomy in this case is that we fail to recognise that she is the kind of person capable of making her own decisions. The above suggests we can only respect the intrinsic value of someone’s autonomy at the time she exercises this autonomy. If my suggestion is correct then there are no reasons based on respect for the intrinsic value of autonomy as to why we should respect someone’s past autonomous decisions. It can then be concluded if respect for living wills is based on respect for the intrinsic value of autonomy there is no reason to respect someone’s living will in circumstances in which she has permanently lost her capacity to make autonomous decisions. It can be further concluded in these circumstances the patient’s doctor should make a decision on her behalf based on what is in the patient’s best interests after consulting her relatives.

I now wish to consider whether we should respect living wills in circumstances in which treatment, or lack of treatment, means the non-autonomous patient is expected to regain her capacity to make autonomous decisions whilst at the same time losing her capacity to implement her decisions. Once again I will use an example. The example I will use will be the case of Ms B (Re B, [2002] All ER 449). Ms B suffered from a cavernous haegmangioma. She recovered but the angioma recurred and she became tetraplegic and was placed on ventilator. Prior to being placed on the ventilator she made a living will stating she wished treatment to be withdrawn if she was suffering from a life threatening condition or permanent mental disability or unconsciousness. Ms B regained consciousness and wished to have her ventilator switched off. The hospital trust in question refused. Eventually The High Court agreed that Ms B had a right to have the ventilator switched off and awarded her a small amount in damages for trespass. I now only consider the question as to whether her doctors were correct in ignoring the wishes she expressed in her living will. I will argue the answer to this question is not as straightforward as it might appear.

Let it be assumed Ms B was fully autonomous when she made her living will. Let it also be assumed after being placed on the ventilator Ms B could again make autonomous decisions but that she could not implement these decisions. I believe whether Ms B’s doctors were correct in ignoring, the wishes expressed in her living will depend on whether the effects of her treatment could be reversed. I will argue in contexts in which the effects of a patient’s treatment can be reversed that even if some treatment is undertaken, which runs contrary to the wishes expressed the patient’s living will, the patient’s autonomy might still respected. When treatment is undertaken, which runs contrary to the wishes expressed the patient’s living will, her doctors cannot ask her if she is sure about these wishes. However once the patient becomes capable again of expressing her autonomous wishes her doctors may ask her if her views are unchanged. Moreover in this context if her views are unchanged her doctors can reverse the treatment. The above position is partly analogous to the position in many other fields. In these fields we can still respect someone’s autonomous decision even if we ask her to think again. Once she has thought again we respect her autonomy by simply accepting her decision regardless of whether she has changed her mind or not. How good this analogy is depends on what is meant by being able to reverse the treatment.

It might be assumed that in the case of Ms B the effects of her treatment could be reversed by simply switching off the ventilator. In practice this would have caused Ms B to suffer. Her ventilation should have been reduced slowly whilst at the same time giving her painkillers. It follows reversing some treatments might mean doctors having to employ their medical skills in assisting this reversal. It follows in practice treatment is only reversible if the following three conditions can be satisfied.

  1. It is possible to reverse the treatment and leave the patient in a similar condition to that she was in prior to the commencement of treatment. In most circumstances reversing treatment will mean stopping treatment.
  2. It is not illegal to reverse the treatment.
  3. Any doctors prepared override a patient’s wishes expressed in a valid living will and treat the patient must also be prepared reverse this treatment. This reversal of treatment might involve taking measures to ensure the reversal does not cause the patient unnecessary suffering once she regains the power to make autonomous decisions.

In the case of Ms B I believe her medical team acted correctly by placing her on a ventilator. However her medical team failed to respect her autonomy because they were not prepared to stop the treatment once she regained the power to make autonomous decisions. It can be concluded there is no reason, based on respect for autonomy for doctors to respect a patient’s living will, in circumstances when she is expected to regain her capacity to make autonomous decision provided the proposed treatment is reversible subject to the three conditions outlined above.

I have argued there are no good reasons based on respect for autonomy to respect the majority of living wills. However there are some circumstances in which living wills should be respected provided respect for living wills is based on respect for patient autonomy. Living wills should be respected in all cases in which a patient is expected to be able to regain the ability to make autonomous decisions and in which the effects of her treatment cannot be reversed. Accepting the above would mean non-autonomous patients who state in their living wills that in certain circumstances they should not be resuscitated and who once are expected to regain the capacity to make autonomous decisions again should have their wishes respected. Further examples might include limb amputation and perhaps, though I write this reluctantly, blood transfusion.


Engaging with Robots

  In an interesting paper Sven Nyholm considers some of the implications of controlling robots. I use the idea of control to ask a different...