Who decides on a Dnacpr?

There is considerable confusion over the status of ‘do not attempt resuscitation’ (DNAR) orders. David Tracey's claim against Addenbrookes hospital, in which he claims that the Trust unlawfully inserted a DNAR order into his wife's notes, without consulting with her or him, has brought the issue into the public eye.1 A DNAR order is a statement in the patient's notes to the effect that cardiopulmonary resuscitation (CPR) should not be attempted. This article analyses the legal status of such orders, and outlines some ethical and practical corollaries.

‘DNAR order’ is not a legal term: it refers to two types of direction which are legally very different. The first (which we will call ‘type 1’) is a clinical decision for a patient who lacks capacity: a decision that should be made in the patient's best interests at the time of cardiac arrest. The second (‘type 2’) is an advance decision to refuse resuscitation made by the patient when capacitous and binding on clinicians if valid, and if applicable to the circumstances.

At the time that a clinician might be considering CPR, the patient will not be mentally capacitous: he will be unconscious. The legal position in relation to the treatment of incapacitous patients is governed by the Mental Capacity Act 2005 (MCA), and is straightforward. Clinicians must assess the patient's best interests, and act in accordance with that assessment.2

How does the law understand best interests? The MCA provides no definition, but sets out (in s. 4) a list of issues which must be considered by the clinician in making the assessment and coming to a decision. Broadly speaking these issues fall into two categories: first, the wishes and values that the patient had when capacitous before the arrest; and, second, the patient's current wishes and feelings. The relevant wording of the Act is as follows:

(6) He [the relevant clinician] must consider, so far as is reasonably ascertainable—

(a) the person's past and present wishes and feelings (and, in particular, any relevant written statement made by him when he had capacity),

(b) the beliefs and values that would be likely to influence his decision if he had capacity, and

(c) the other factors that he would be likely to consider if he were able to do so.3

The general tenor of the Act is to give as much decision-making power to an incapacitous patient as possible. This generally means that clinicians should act in accordance with what the patient would wish if she had fully understood and considered the situation she is now in. Ascertaining this is notoriously difficult. The decision must be informed by all the information available about the patient, including her known beliefs and values. Although the required best interests determination is an objective one, it is crucially informed by the clinician's assessment of the patient's likely wishes – wishes that are themselves crucially tinged with subjectivity. If, for instance, a clinician believes that resuscitation, if successful, will leave the patient in a condition so bad that life itself is contrary to the patient's best interests, the clinician can only act on that belief if he concludes that the patient, if fully appraised of the relevant information, would have come to the same conclusion.

There will often be great doubt about where a patient's best interests lie. If there is doubt about whether continued life is in his best interests, legally the safest course is to presume that it is. There is ample authority for this: see, for instance, Re J (a Minor) (Wardship: Medical Treatment),4 and most recently, Re M.5

Crucial to understanding the legal status of this ‘type 1’ DNAR order is that the best interests determination that must be carried out is of the best interests at the time of the cardiac arrest. One needs only note the tense in which s. 4(1) of the MCA is couched: ‘(1) In determining for the purposes of this Act what is in a person's best interests….’ (emphasis added). This sounds trite, but often clinicians, when considering DNAR orders, forget it and instead regard a DNAR order made some time previously, in different clinical circumstances, as a legally acceptable best interests determination at the time of cardiac arrest. But it cannot be. The only lawful best interests determination will be one made at the time that it becomes relevant, and that, in the case of a decision about CPR, will be when the arrest happens.

Of course it makes great sense for clinicians to sit down and determine, before the dramatic circumstances arise which might call for a decision about CPR, where the patient's best interests are likely to lie. There is nothing at all unlawful about that. Indeed, if the patient has capacity at the time a DNAR order is being considered, this is a good time to ascertain the patient's relevant beliefs and values. The MCA states that: [the relevant clinician] must, so far as is reasonably practicable, permit and encourage the person to participate, or to improve his ability to participate, as fully as possible in any act done for him and any decision.6

This presumably includes encouraging the patient to participate in decisions about resuscitation. If the patient lacks capacity at the time that a DNAR ‘order’ is being considered, the Act imposes a further duty of inquiry on clinicians:

He [the clinician] must take into account, if it is practicable and appropriate to consult them, the views of—

(a) anyone named by the person as someone to be consulted on the matter in question or on matters of that kind,

(b) anyone engaged in caring for the person or interested in his welfare,

(c) any donee of a lasting power of attorney granted by the person, and

(d) any deputy appointed for the person by the court,

(e) as to what would be in the person's best interests and, in particular, as to the matters mentioned in subsection (6) [i.e. the best interests considerations set out above].7

For a ‘type 1’ DNAR order however – that is one based on a best interests assessment – the order, made in advance of any cardiac arrest, should not, from a legal perspective, be perceived to, or actually, crystallize into a definite, non-negotiable decision. Instead, such a DNAR order is an intelligent foreshadowing of what the legally relevant decision is likely ultimately to be.

Two important practical points arise from this analysis. First: the person who legally makes the relevant best interests determination is not the person who, perhaps several weeks before the arrest, put the DNAR order in the notes. It is instead the person who attends the arrest (who may be the nurse who decides not to call the arrest team), and who has the responsibility for deciding whether or not, at that time, it is in the patient's best interests to attempt CPR. And second, the notes and information available to the person attending the arrest should be such that she can rapidly make the best interests assessment. These notes should summarize the grounds on which it has been decided that the best interests of the patient lie in not being resuscitated, were he to arrest, so that the person attending the arrest can rapidly decide whether or not any changes in the circumstances materially affect the best interests assessment tentatively and pre-emptively made earlier.

It is worth noting that a clinician cannot be required, by an advance decision or otherwise, to do anything that the clinician thinks is not clinically indicated. While there is always an obligation to provide 'basic care' (such as non-medically administered food and hydration),8 the sophisticated procedures involved in attempts to resuscitate plainly do not fall within that class.

In contrast with the type of DNAR order so far considered, which is centred on a best interests assessment, a type 2 order is an advance decision, made by the patient herself, to refuse resuscitation. It has been long established in English law that a patient with capacity can refuse any treatment, including life-saving treatment. It would be battery for a clinician to impose a treatment on a capacitous adult without that person's consent. The MCA, for the first time in statute in English law, extends that power of refusal of treatment to patients through an advance decision. Such a decision (often called an advance directive or living will) is one made at a time when a patient has the relevant decision-making capacity but to take effect in the future at a time when the patient lacks such capacity. Thus a patient with capacity can state that, were she to have a cardiac arrest (and at that time to lack capacity), that she refuses consent for an attempted resuscitation. If a clinician subsequently attempts resuscitation under those circumstances, she has prima facie committed a battery. From the legal point of view the situation is little different from imposing a treatment on a capacitous patient against the patient's will. But for the advance decision to refuse resuscitation to be binding on clinicians it has to be valid and applicable. ‘Valid’ presumably means that at the time the patient refused, he had capacity, knew the information relevant to the decision, that his wishes have been accurately recorded and (in the case of the refusal of potentially life-sustaining treatment) that the record complies with certain formalities (see below). ‘Applicable’ presumably means that the clinical circumstances (including the risks and benefits of the available treatment) that obtain at the time of the arrest include those that the patient had in mind when he made the decision, that there are no grounds to believe that the patient changed his mind between the decision and the arrest (or, perhaps, in some circumstances, grounds for believing that he did not change his mind), and that the patient does indeed lack capacity at the time of the arrest (which would of course be the case).

The person attending the arrest in the case of a ‘type 2 DNAR order’ has to make a very different rapid assessment to that appropriate for a type 1 order. Instead of making a best interests assessment she has to assess whether the order is valid and applicable. If it is, then resuscitation cannot (legally) be attempted (even if to attempt it would be in the ‘objective’ best interests of the patient). It would be the responsibility of the clinician who recorded the DNAR order to ensure that the order was valid at the time it was made and to specify clearly to what it applied: refusal of attempted resuscitation in the event of cardiac arrest, in the situation we are considering. The person attending the arrest has to make her own determination of the validity and applicability of the order (of course relying insofar as appropriate – which will of course usually be heavily – on the person who recorded the original decision). For a type 2 DNAR – that is an advance decision refusing resuscitation made under the MCA – it must be made in writing, signed by the patient, and witnessed.9 The notes available to the person attending the arrest must make it clear that the DNAR is of this type – that it is an advance decision – so that the person does not wrongly consider that her duty is to make a (rapid) best interests assessment.

There is a danger that all DNAR orders are understood by clinicians to have the effect of an advance decision even when they are type 1 orders, and have been made without any consultation at all with the patient, and with the patient having been denied the protection of the formalities demanded by s. 25(6). As we have emphasized, it is the person attending the cardiac arrest (who may well be the nurse who has to decide whether to call the resuscitation team at all) who has the legal responsibility for making the necessarily rapid assessment of the circumstances. Each of the two types of order requires a different type of assessment.

At the time of arrest, a decision must first be made whether or not to call the resuscitation team. This decision is usually made by the relevant nursing staff. The purpose behind making the (type 1) DNAR order is normally to ensure that the resuscitation team is not called. According to our analysis, such a decision (not to call the resuscitation team) must, legally, be made at the time of the arrest and on the basis of the patient's best interests as assessed at that time. It is sometimes thought that the nurse has to decide whether or not to disobey the DNAR order. This is a misreading of the law. There is no order to be obeyed or disobeyed. Rather, (for ‘type 1’ DNAR orders) it has been decided at some time that, should the patient arrest at that time, it would be in the patient's best interests not to be resuscitated, and, further, that this is likely to be the case until the next time at which the DNAR order is reviewed. So, when the best interests assessment is made at the moment of arrest, the law requires the nurse (say) to assess whether there has been any change relevant to the patient's best interests since the DNAR ‘order’ was made, or last explicitly reassessed and reconfirmed. There may, for instance, have been a change in the patient's condition or prognosis, or new evidence about the patient's previous views and values (for example from a relative or from the patient herself). In either case, one would hope that the DNAR order would have been reviewed.

Our analysis indicates that at the time of making the DNAR order, careful thought should be given to the reasons for the order and the changes that might affect it. These reasons should be succinctly and accessibly recorded, and available for the nurses (and for the resuscitation team if they are called), and should form an explicit part of the ‘handover’ information as nursing shifts change.

Clinicians will often be reluctant to broach the question of a DNAR order with a patient. We sympathize with that reluctance. There may be potent ethical reasons for not embarking on that conversation; and there may be ethical reasons for objecting to the MCA's approach to best interests and advance decisions,10 but it would be legally problematic for a clinician who, in the case of a patient who is capacitous at the time that a DNAR order is considered, does not seek to ascertain the patient's view about the proposed order. This is simply because it can hardly be said that the patient's view is not ‘reasonably ascertainable.’ It may not be comfortably ascertainable, but that is not the same thing.

The Resuscitation Council (UK), the British Medical Association, and the Royal College of Nursing issued, in October 2007, a joint statement relating to CPR. They wrote: ‘There is no ethical or legal requirement to discuss every possible eventuality with all patients, and if the risk of cardiorespiratory arrest is considered very low, it is not necessary to initiate discussion about CPR with the patient, or with those close to patients who lack capacity.’11 It is not clear from this whether the joint statement considers that a DNAR order would be inappropriate anyway where the risk of CPR was very low. If what is being said is that there are circumstances where a DNAR order is appropriate, but that there is no legal need to ascertain the patient's wishes direct from the patient if it is practicable to do so, we think that this advice is hard to square with the MCA.

CF and TH contributed equally to this paper

1. ‘Trust and health department deny acting illegally over “do not resuscitate” order'. BMJ 2011;343:d5659 [PubMed] [Google Scholar]

2. Mental Capacity Act 2005. Section 1(5) of the MCA provides that ‘(5)An act done, or decision made, under this Act for or on behalf of a person who lacks capacity must be done, or made, in his best interests.’

3. Mental Capacity Act 2005: Section 4(6)

4. [1991] 1 Fam. 33

5. [2011] EWHC 2443 (Fam)

6. Mental Capacity Act: Section 4(4)

7. Mental Capacity Act: Section 4(7)

8. R (Burke) v General Medical Council [2005] QB 424

9. Mental Capacity Act: see s. 25(6)

10. Hope T, Slowther A, Eccles J Best interests, dementia and the Mental Capacity Act (2005). J Med Ethics 2009;35:733–8 doi:10.1136/jme.2009.030783 [PubMed] [Google Scholar]

11. Decisions relating to cardio-pulmonary resuscitation: A joint statement from the British Medical Association, the Resuscitation Council (UK), and the Royal College of Nursing, October 2007: http://www.resus.org.uk/pages/dnar.pdf (last accessed 2 April 2011)