Some of the key features of the judgment are as follows:

1. DNR, standards of care and consent

The judgment turns on consent, and the consent issue is decided by reference to the standard of care.

Non-beneficial treatments are outside of the standard of care.

In this case CPR was determined, by the physicians, to be almost certain not to bring a benefit to the patient. Sunnybrook policy and CPSO policy at the time referred to three categories of potential benefit to patients of CPR:

Patient is likely to benefit

Patient is unlikely or uncertain to benefit – prognosis is poor or uncertain and the likelihood of adverse consequences is high

Patient almost certainly will not benefit… either because the underlying illness or disease makes recovery or improvement virtually unprecedented, or because the person will be unable to experience any permanent benefit.

So:

1. It is a physician’s judgment that a patient will almost certainly not benefit (or indeed be harmed) by a resuscitation attempt
2. A treatment that is almost certainly not going to benefit the patient is outside of the standard of care.
3. There is no obligation to offer a treatment that is outside of the standard of care. (Is there an obligation not to offer?)
4. If there is no offer of treatment there is nothing to consent to.. so consent is not required.
5. Therefore a physician should write a DNR order in these cases and should not seek consent
6. The physician should inform the patient/SDM
7. If there are objections there should be a dispute resolution process
8. If the patient arrests while the dispute resolution process is unfolding the physician at the bedside must make a clinical judgment as to the appropriateness of the resuscitation attempt.

2. DNR, withdrawing life sustaining treatment and the issue of touching

All of the steps to 6 above were followed in the Rasouli case.  Cuthbertson v. Rasouli, 2013 SCC 53, [2013] 3 S.C.R.

1 The physicians determined that Hassan Rasouli would not benefit from continued life support.

2, 3, 4 apply as above.

5 The withdrawal order was written without consent being sought

6 The SDM was informed.

She objected all the way to the Supreme Court of Canada and she won.  What is the difference?

Judge Cavanagh argued that the crucial difference is that the withdrawal of life support requires a touching or interference with the patient’s body.

Para 198 … In Rasouli, the court focused on the reality of the situation, and examined the “processes” involved in the withdrawal of care for Mr. Rasouli. McLachlin C.J.C. noted at para. 62-63 that many of the processes involved in withdrawal of life support entailed physical interference with the patient’s body and, under the HCCA, as at common law, physical interference requires consent.

Whereas the decision to write – and the writing of a DNR order are processes that do not involve any interference or touching of the patient’s body.

3. Consent, treatment and “little or no risk of harm to the person.”

What follows are quotations from the judgment, the paragraph numbers are in parenthesis.

[245] The definition of “treatment” in the HCCA includes a number of specific exclusions. One of these exclusions is found in clause (g) of the definition in s. 2(1) of the HCCA: “a treatment that in the circumstances poses little or no risk of harm to the person”.

[246] I have concluded that Dr. Livingstone and Dr. Chapman met the required standard of care when they concluded following their assessments of Mr. DeGuerre on September 22, 2008 that the administration of CPR would almost certainly not be of benefit to him and would only cause him harm. It follows that the defendants’ medical decision not to offer CPR as a treatment option and to write the DNR order, and Dr. Chapman’s decision at Mr. DeGuerre’s bedside not to offer or administer CPR, posed little or no risk of harm to Mr. DeGuerre. To the contrary, these decisions and the actions taken to implement them saved Mr. DeGuerre from harm.  (Emphasis added.)

[247] If I had concluded that Dr. Livingstone’s and Dr. Chapman’s acts in writing the DNR order to give effect to their medical decision and Dr. Chapman’s actions in deciding not to offer CPR as a treatment option at Mr. DeGuerre’s bedside would otherwise fall within the definition of the term “treatment” in the HCCA, I would have concluded that these things are described in clause (g) of the definition of “treatment” and are excluded from the meaning of “treatment”.

Comment

I think this is a novel use of the “little or no risk of harm to the patient” exclusion from the definition of treatment.

First, harm is a contested concept.  Is “harm” subjective (determined by, e.g. the patient) or objective – there is some scientific notion of harm that transcends an individual’s own conceptions? In this quotation from HCCA there seems to be an objective notion of harm being used.  But the rest of the Act is autonomy all the way.  When HCCA Section 21.2 talks about “best interests” (a concept closely related to “harm”) it says that when a substitute decision maker tries to determine a person’s best interests that must be done in “consideration of the person’s values and beliefs.”  There is no objective notion of best interests at play here, but rather a subjective – patient/person-centred understanding of the concept.

Second, it is not clear how an action that causes little or no risk of harm being exempt from the definition of “treatment” and therefore not needing consent would fare if the action involved a touching or interference with a person.  Does the consent required for a touching trump the exemption for an act that causes little or no risk of harm?

4. Language, clarity and practice

Judge Cavanagh examined a number of CCB cases that the plaintiff argued supported the contention that DNR needs consent.

[233] The fact that in some of the decisions the Board used words to describe the physician’s view that the proposed treatment would not be in the patient’s best interests according to the principles in s. 21(2) of the HCCA, for example, that it would be “futile”, does not allow me to conclude that the physician had made a medical decision that the treatment would almost certainly not benefit the patient, and that the treatment was not offered. In each case, it appears that CPR or other life support treatment had been proposed by the physician and consent to the proposed treatment had been requested and given. In a number of these Board decisions, the physician had recommended that the substitute decision-maker refuse to consent to administration of CPR by consenting to a DNR order. The issue on these applications was whether the patient’s substitute decision-maker, in giving consent to CPR or other life support treatment, had acted in the incapable patient’s best interests in compliance with s. 21 of the HCCA.  (Emphasis added.)

The issue here is exactly how a situation is described and exactly what was being sought from the CCB.

I would argue that if a treatment is “futile” that means it is pointless, that it cannot achieve the effect for which it intended.  That it will not work.  I would argue that a futile treatment is a paradigm case of a treatment that is non-beneficial, a situation where, by definition, the patient almost certainly will not benefit.  According to Cavanagh therefore, such a treatment would be outside of the standard of care, and so need not (should not?) be proposed or offered – and so does not reach the threshold of requiring consent.

The cases only reach the CCB because a health care provider has sought consent, which has been refused, so someone must have offered a treatment in the first place.  The “error” here is in assuming that a DNR order is a treatment and therefore needs consent, the DNR is offered or proposed, the SDM refuses so off we go to the CCB.  But a DNR order is an order not to do something, rather than to do something.  So provided any particular DNR order meets the standard of care no treatment is being proposed so no consent is required.

The CCB in  Re (MW), 2008 CanLII 119685 determined:

In this case, [the substitute decision-maker] is asking the Board to make a determination about consent for a hypothetical treatment. No such treatment has been proposed by any physician treating [the patient]. The Board is asked to consider a treatment that is not and may never be at issue. This is simply not within the Board’s authority. It is not within the role of the Board to propose treatment or to determine the appropriateness of the scope of treatment. There is no role for the Board in this matter. (Page 10)

The question before the CCB in the other cases therefore concerns whether the SDM is appropriately refusing consent to a proposed treatment or intervention.  Under the HCCA the SDM must follow the patient’s previously expressed capable wishes – if possible, and where there are no relevant instructions, the decision is to made on the basis of the patient’s best interests (taking into consideration his or her values and beliefs.)

The crucial message here is that physicians should not offer futile or inappropriate treatments e.g. CPR and then seek to encourage a patient or SDM to refuse the inappropriately offered treatment – consent to a DNR order.  Far better not to offer such a treatment or intervention and explain why the intervention is not being proposed.

This is supported at:

[249] The interpretations which I have given to “treatment” and “plan of treatment” are consistent with the provisions of the HCCA as a whole and with the purposes of the HCCA as set out in s. 1 thereof. The HCCA does not provide that a physician is required to propose every conceivable treatment to a patient and allow the patient to choose which treatment or treatments to receive, and if consent is given to a treatment on the menu of choices, to administer the treatment unless directed otherwise on an application to the Board under s. 37(1). To the contrary, ss. 10 and 37 of the HCCA only apply where a health practitioner is proposing or has proposed a treatment. This interpretation is supported by the decision of McLachlin C.J.C. in Rasouli, who recognized that the standard of care applicable to physicians requires that they undertake a contextual assessment of the patient for the purpose of deciding whether a given treatment should or should not be proposed. To interpret the HCCA as the plaintiff asks would significantly diminish if not eliminate the proper role of a physician in exercising clinical judgment and making the necessary medical decision about whether administration of a given treatment would be medically inappropriate for his or her patient, and whether the treatment should be proposed to the patient.  (Emphasis added.)

5. Patient and family centred care.

The plaintiff in this case also asserted that she was entitled to damages because the physicians owed her a duty of care. (Paragraph 340)

She argued that as a close family member involved in the father’s care and as the SDM the physicians had a fiduciary duty to her.

This idea was rejected:

[370] I accept that recognition of such a duty of care owed by a physician to a family member or substitute decision-maker of an incapable patient would have the potential to put the physician in a conflict of interest because the wishes of the close family member or substitute decisionmaker may not align with the physician’s medical opinion of what is in the patient’s best interests. The imposition of a duty of care owed by a physician to a patient’s family members or a substitute decision-maker might influence the physician, in attempting to comply with competing and potentially conflicting duties of care, to act in ways in which he or she would not otherwise act and put the patient at risk of harm.

This highlights the tensions that are inherent in the language and practice of “patient/family centred care.”  The implication embedded in the phrase “patient and family centred care” is that somehow the patient’s and the family’s interests are necessarily aligned.  But, of course, that is not always so.  If the physician has a commitment to both patient and family there arises the possibility of a conflict of interest.  That conflict of interest could lead to a physician demonstrating less than total commitment to the interests of the patient.  In order to avoid that possibility the physician can only have a duty to the patient.

Robert Butcher November 2019