(with whom Lebel J, Fish J, Rothstein J & Cromwell J joined, delivered the judgment of the court)
This case presents us with a tragic yet increasingly common conflict. A patient is unconscious. He is on life support – support that may keep him alive for a very long time, given the resources of modern medicine. His physicians, who see no prospect of recovery and only a long progression of complications as his body deteriorates, wish to withdraw life support. His wife, believing that he would wish to be kept alive, opposes withdrawal of life support. How should the impasse be resolved?
In the past, disputes between next of kin and physicians over consent regarding life support and other forms of medical treatment for incapable patients were resolved through the courts, under the common law. However, in Ontario, the Health Care Consent Act, 1996, S.O. 1996, c. 2, Sch. A (“HCCA”), provides a statutory scheme for resolving such disputes. Under the HCCA, a designated substitute decision-maker – often a close family member – has the right to determine whether life support can be withdrawn in the first instance. In making that decision, she must act in accordance with the provisions of the HCCA, which aim at protecting patients’ autonomy and medical interests. In the event of disagreement, it is open to the attending physician to challenge the substitute decision-maker’s decision on the ground that it is not in accordance with the HCCA, by applying to the Consent and Capacity Board (“Board”). The HCCA empowers the Board to make the final decision on the issue of consent for incapable patients.
The appellant physicians in this case take the position that the HCCA does not apply because consent is not required for withdrawal of life support that does not provide any medical benefit to the patient. The courts below rejected that contention, as would I. It follows that the appeal should be dismissed. Where a substitute decision-maker does not consent to the withdrawal of life support, the physicians’ remedy is an application to the Board.
This case turns on statutory interpretation – what the HCCA provides. It is not a case about who, in the absence of a statute, should have the ultimate say in whether to withhold or withdraw life-sustaining treatment. Nor does the case require us to resolve the philosophical debate over whether a next-of-kin’s decision should trump the physicians’ interest in not being forced to provide non-beneficial treatment and the public interest in not funding treatment deemed of little or no value. The Ontario legislature has addressed the conflicting interests and arguments that arise in cases such as this in the HCCA. The Court’s task is simply to determine what the statute requires. I note that the parties did not address resource implications or Charter issues in this appeal.
B. The Events
In October 2010, Mr. Hassan Rasouli underwent surgery at Sunnybrook Health Sciences Centre (the “Hospital”) to remove a benign brain tumour. Following the procedure, Mr. Rasouli developed an infection that caused severe and diffuse brain damage. As a result, Mr. Rasouli has been unconscious since October 16, 2010, and is being kept alive by mechanical ventilation, connected to a tube surgically inserted into his trachea, and artificial nutrition and hydration, delivered through a tube inserted into his stomach. Without these life-sustaining measures, it is expected that Mr. Rasouli would pass away.
The physicians responsible for Mr. Rasouli’s care, including the appellants, Dr. Cuthbertson and Dr. Rubenfeld, formed the opinion that Mr. Rasouli was in a persistent vegetative state, that all appropriate treatments for his condition had been exhausted, and that there was no realistic hope for his medical recovery. In the opinion of the physicians, continuing to provide life support would not provide any medical benefit to Mr. Rasouli and may cause harm. They seek to remove his life support and to provide palliative care until his expected death.
The physicians informed Mr. Rasouli’s wife, Ms. Parichehr Salasel, who is also his litigation guardian and substitute decision-maker under the HCCA, of Mr. Rasouli’s diagnosis and their proposed course of action. She would not agree. Ms. Salasel and her family wish to keep Mr. Rasouli alive. Ms. Salasel does not accept that Mr. Rasouli is in a state of permanent and irreversible unconsciousness and believes that, as a devout Shia Muslim, he would wish to be kept alive. She contends that new evidence on Mr. Rasouli’s neurological function indicates an increased level of consciousness.
In the face of Ms. Salasel’s disagreement, the Hospital arranged for a second opinion from a neurologist who had not been involved in Mr. Rasouli’s care. The neurologist concurred with the original diagnosis and assessment. The Hospital also contacted another facility to see whether Mr. Rasouli could be treated elsewhere, but that facility was not prepared to admit Mr. Rasouli. In addition, the physicians offered Ms. Salasel the opportunity to independently obtain an opinion from another neurologist, which she chose not to do.
Faced with an impasse, the physicians agreed to postpone their plans to withdraw life support until Ms. Salasel could apply to the Ontario Superior Court of Justice for an order restraining the physicians from withdrawing Mr. Rasouli from life support, and directing that any challenge to her refusal of consent be made to the Board. The physicians cross-applied for a declaration that Mr. Rasouli is in a permanent vegetative state, that consent is not required to withdraw life support where such treatment is futile and that the Board has no jurisdiction to decide these issues.
C. Court Decisions
The Ontario Superior Court of Justice, per Himel J., granted Ms. Salasel’s application for an order that life support could not be removed without her consent, and that any challenge to her refusal to consent must be brought before the Board: 2011 ONSC 1500, 105 O.R. (3d) 761.
The Ontario Court of Appeal upheld this order. It held that withdrawal of life support and administration of end-of-life palliative care were integrally linked and should be viewed as a “treatment package”: 2011 ONCA 482, 107 O.R. (3d) 9, at para. 52. Since consent to the administration of palliative care was clearly required under the HCCA, it should also be required for the “treatment package” of withdrawal of life support and administration of palliative care.
In January 2012, before the hearing of the appeal in this Court, assessments by two neurologists resulted in a change to Mr. Rasouli’s diagnosis from permanent vegetative state to minimally conscious state. As a result, the appellant physicians took the view that further investigations were required to determine whether Mr. Rasouli may be capable of any communication, which could bear on their assessment of whether life support should be continued. Ms. Salasel, on the other hand, brought a motion to quash the appeal given the change in diagnosis. The motion was dismissed by this Court in May 2012.
In November 2012, both parties brought motions to adduce new evidence on Mr. Rasouli’s neurological function. These motions were referred to the panel hearing the appeal to be determined at the hearing. In light of my conclusion that the substance of the dispute must be determined by the Board, I would dismiss the motions to adduce fresh evidence, without prejudice to the Board receiving any evidence it deems relevant.
This appeal raises two questions.
The first is whether the HCCA governs the issue of withdrawal of life support with the consequence that Ms. Salasel’s consent to withdrawal of life support is required and that her refusal can be challenged only before the Board.
Only if we conclude that the HCCA does not apply, do we reach the second question – whether at common law this Court should order that Mr. Rasouli’s life support can be removed without Ms. Salasel’s consent.
In enacting the HCCA, the Ontario legislature both codified and in important ways modified the common law of consent to medical treatment. It is therefore useful to begin by situating the statute within the common law legal landscape.
A. The Common Law Backdrop
At common law, medical caregivers must obtain a patient’s consent to the administration of medical treatment: Reibl v Hughes,  2 S.C.R. 880; Hopp v Lepp,  2 S.C.R. 192. The physician cannot override the patient’s wishes to be free from treatment, even if he believes that treatment is in the vital interests of the patient. The patient’s consent must be given voluntarily and must be informed, which requires physicians to ensure the patient understands the nature of the procedure, its risks and benefits, and the availability of alternative treatments before making a decision about a course of treatment. The requirement for informed consent is rooted in the concepts of an individual’s right to bodily integrity and respect for patient autonomy: see Fleming v Reid (1991), 4 O.R. (3d) 74 (C.A.).
The common law of consent to medical treatment works well for patients who have the capacity to decide on consent to treatment, in the sense of being able to understand the nature, purpose, and consequences of the proposed treatment. The patient’s autonomy interest – the right to decide what happens to one’s body and one’s life – has historically been viewed as trumping all other interests, including what physicians may think is in the patient’s best interests.
However, the traditional common law approach to medical treatment is more problematic when a patient is incapable of appreciating the nature, purpose, and consequences of the proposed treatment. As explained in Malette v Shulman (1990), 72 O.R. (2d) 417 (C.A.), at pp. 423-24, the common law doctrine of informed consent “presupposes the patient’s capacity to make a subjective treatment decision based on her understanding of the necessary medical facts provided by the doctor and on her assessment of her own personal circumstances”. When such capacity is lacking, the patient is not in a position to exercise his autonomy by consenting to or refusing medical treatment.
If a patient is incapable, disputes over consent to treatment at common law are resolved in the courts. The focus shifts from the patient’s autonomy interest, which is compromised or extinguished, to whether receiving treatment is in the best interests of the patient. In emergency situations, where treatment is necessary to save the life or preserve the health of an incapable patient, treatment may be provided without consent: Malette, at p. 424. In non-emergency situations, treatment may be authorized by a court, acting under its parens patriae jurisdiction, or in the case of an incapable minor, by the child’s parents or legal guardian. See e.g. E. (Mrs.) v Eve,  2 S.C.R. 388; B. (R.) v Children’s Aid Society of Metropolitan Toronto,  1 S.C.R. 315, at para. 83; Re S.D.,  3 W.W.R. 618 (B.C. S.C.), at p. 629.
B. The Statutory Scheme
Many provinces found the common law regime for the treatment of incapable patients unsatisfactory and devised new approaches through legislation. In 1996 the Ontario legislature passed the HCCA, which provides a statutory framework governing consent to treatment for capable and incapable patients. Similar legislation has been adopted in other provinces. See The Health Care Directives Act, C.C.S.M. c. H27; Health Care (Consent) and Care Facility (Admission) Act, R.S.B.C. 1996, c. 181; Care Consent Act, S.Y. 2003, c. 21, Sch. B; Civil Code of Québec, S.Q. 1991, c. 64, articles 11-25; Consent to Treatment and Health Care Directives Act, R.S.P.E.I. 1988, c. C-17.2.
Each of these statutes provides a framework for resolving the difficult issues surrounding treatment of patients who lack capacity to decide for themselves: Starson v Swayze, 2003 SCC 32,  1 S.C.R. 722. Generally speaking, the statutes give effect to the patient’s autonomy interest insofar as possible. If the patient’s autonomy is compromised by lack of capacity, they seek to balance it against considerations related to the best interests of the patient. Finally, some statutes provide for resolution of disputes by specialized tribunals instead of the courts. The HCCA does all these things.
The purposes of the Act are:
The HCCA starts from the general premise that medical treatment cannot be administered without consent: s. 10(1). Building on this premise, the HCCA goes on to provide a detailed scheme governing consent to treatment for incapable patients. It provides that a substitute decision-maker must consent to treatment of an incapable patient: ss. 10(1)(b) and 20. The statute sets out a clear hierarchy designating who will serve as substitute decision-maker: s. 20(1). This will often be a close family member of the patient, furthering the statutory objective of ensuring “a significant role for supportive family members when a person lacks the capacity to make a decision about a treatment”: s. 1(e).
The substitute decision-maker does not have a free hand to grant or refuse consent at will. She must respect prior applicable wishes of the patient expressed while the patient was capable: s. 21(1). If there are no such wishes, the substitute decision-maker must decide based on the best interests of the patient, taking into consideration a series of mandatory factors relating to the medical condition, well-being, values, and wishes of the patient: s. 21(2).
The HCCA does not neglect the role of health practitioners in the treatment of incapable patients. First, where there is a prior wish by the patient, the attending physician may ask the Board to find that the wish is not applicable to the patient’s current circumstances (s. 35), or to permit a departure from the wish because the likely result of treatment has significantly improved since the wish was made: s. 36. Second, if the physician feels that a substitute decision-maker has not complied with the HCCA’s rules for giving or refusing consent to treatment, he may challenge the consent decision by application to the Board: s. 37. Such a challenge will generally focus on medical considerations within the s. 21(2) best interests analysis. The physician’s views of what will medically benefit the patient are obviously critical to the Board’s determination of the patient’s best interests. However, the HCCA gives the Board final responsibility to decide disputes over consent to treatment for incapable patients, based on an objective assessment of whether the substitute decision-maker complied with the requirements of the HCCA.
In summary, the HCCA contemplates disputes between physicians and substitute decision-makers over the care of incapable patients, and provides for their resolution by the Board, an independent, quasi-judicial body with specialized jurisdiction over matters of consent to medical treatment.
C. “Treatment”: Measures That Serve a Health-Related Purpose
The HCCA requires consent to all measures that constitute “treatment”. Section 10(1) of the Act provides:
“Treatment”, in turn, is broadly defined as care given for a health-related purpose. Section 2(1) provides:
.... anything that is done for a therapeutic, preventive, palliative, diagnostic, cosmetic or other health-related purpose, and includes a course of treatment, plan of treatment or community treatment plan ....
The issue raised in this case thus comes down to the interpretation of “treatment” and “health-related” purpose under s. 2(1) of HCCA.
The basic rule of statutory interpretation is that “the words of an Act are to be read in their entire context, in their grammatical and ordinary sense harmoniously with the scheme of the Act, the object of the Act, and the intention of Parliament”: R. Sullivan, Sullivan on the Construction of Statutes (5th ed. 2008), at p. 1. Every statute “shall be given such fair, large and liberal interpretation as best ensures the attainment of its objects”: Legislation Act, 2006¸ S.O. 2006, C. 21, Sch. F, s. 64(1).
There is no dispute between the parties that, in general, the provision of life support constitutes treatment under the HCCA and therefore requires consent. The question is whether withdrawal of life support constitutes treatment on the facts of this case. The physicians argue that it does not. They raise three arguments: (1) life support that is not “medically indicated” is not “treatment” under the HCCA; (2) in any case, withdrawal of treatment does not itself constitute “treatment” under the HCCA; and (3) that requiring consent for withdrawal of life support will place them in an untenable ethical position. I will consider each argument in turn.
D. The Argument That “Treatment” Is Confined to What Is of Medical Benefit to the Patient
The physicians argue that treatment under the HCCA is limited to what the attending physician or caregiver deems to be of medical benefit to the patient (in other words, what is medically indicated). Mr. Rasouli’s physicians have concluded that life support no longer offers a medical benefit, despite keeping him alive, given his unconscious state and the extreme unlikelihood of his recovery. It follows, they argue, that the provision of life support to Mr. Rasouli has ceased to be treatment under the HCCA, obviating the need for consent to its withdrawal.
The difficulty with the physicians’ argument is that it substitutes a physician-made criterion for treatment (medical benefit) for the criterion specified in the HCCA for consent (health-related purpose). These concepts sound similar, but they are in reality different.
The concept of “medical benefit” is a clinical term used by physicians to determine whether a given procedure should be offered to a patient. This clinical term has legal implications for the physician’s standard of care. If a treatment would be of medical benefit to the patient in this sense, the physician may be required to offer that treatment in order to comply with his standard of care. Whether a given treatment offers a medical benefit requires a contextual assessment of the patient’s circumstances, including the patient’s condition and prognosis, the expected result of treatment for that patient, and any risks of treatment for that patient: A.F., at para. 44.
The concept of “health-related purpose”, by contrast, is a legal term used in the HCCA to set limits on when actions taken by health practitioners will require consent under the statute. Treatment is “anything that is done” for one of the enumerated purposes (therapeutic, preventive, palliative, diagnostic and cosmetic) or “other health-related purpose”. Under the HCCA, only acts undertaken for a health-related purpose constitute treatment, and therefore require consent. The concept of health-related purpose in the HCCA does not interfere with a physician’s professional assessment of whether a procedure offers a medical benefit. Its only function is to determine when the actions of health care practitioners require patient consent.
The issue here is not the correctness of the physicians’ professional opinion that sustaining life in Mr. Rasouli’s situation confers no medical benefit. In fact, their opinion appears to reflect a widely accepted view in the medical community. The issue at this stage of the argument is whether maintaining Mr. Rasouli’s life serves a health-related purpose within the meaning of the HCCA.
The wording of the HCCA does not limit “health-related purpose” to what the attending physician considers to medically benefit the patient. The HCCA does not use the terms “medical benefit” or “medically indicated”. The legislature could easily have taken this approach but instead chose to define “treatment” more broadly with a wide-ranging and non-exhaustive list of health-related purposes.
The words of the HCCA on their face cover provision of life support that is effective in keeping the patient alive and forestalling death. Life support arguably falls within “therapeutic” and “preventive” purposes, listed in the definition of “treatment” in s. 2(1).
The New Oxford Dictionary of English (1998) defines “therapeutic” as “relating to the healing of disease”, but also as “having a good effect on the body or mind” (p. 1922). Maintaining life support for Mr. Rasouli does not serve the purpose of “healing of disease”. However, it can be argued that maintaining life support has a “good effect on the body”, in the sense of keeping it alive.
The same dictionary defines “preventive” as describing a medicine or other treatment “designed to stop disease or ill health from occurring” or “designed to keep something undesirable such as illness, harm, or accidents from occurring” (p. 1469). If death is considered harmful or a manifestation of ill health, then life support serves a preventive purpose so long as it is effective in preventing death.
Inclusion of life support in “treatment” is also generally supported by the objects of the HCCA. It provides consistency with respect to consent, protects autonomy through the requirement of consent, and provides a meaningful role in the consent process for family members. An interpretation of “treatment” that is confined to what the medical caregiver considers to be of medical benefit to the patient would give these statutory purposes short shrift. The legislature cannot have intended such a crabbed interpretation of “treatment”.
Reading the words of the statute in their ordinary sense and in their context, and having regard to the objects and scheme of the Act, I cannot accept the physicians’ argument that “treatment” and “health-related purpose” are confined to procedures that are of medical benefit in the view of the patient’s medical caregivers.
E. The Argument That “Treatment” Does not Extend to Withdrawal of Treatment
The physicians argue that withdrawal of life support does not constitute “treatment” under the HCCA, because it is not treatment but withdrawal of treatment. They argue that the Act distinguishes between administering a particular type of care, which is “treatment” requiring consent, and removing that care, which is not “treatment” and does not require consent. Consequently, they argue, withdrawal of Mr. Rasouli’s life support does not require Ms. Salasel’s consent. I conclude that this argument cannot succeed, essentially because withdrawal of life support involves – indeed may be viewed as consisting of – a series of acts that serve health-related purposes, and because the critical interests at stake where withdrawal of life support is concerned go to the heart of the purposes of the HCCA.
On its face, the definition of “treatment” in s. 2(1) appears broad enough to include “withdrawal of treatment”. The opening words of the definition could not be more expansive: “.... anything that is done” for one of the enumerated health-related purposes or other health-related purpose is included in “treatment”.
The breadth of the concept of “treatment” is reinforced by the express exclusions from this term under the HCCA. For example, the assessment or examination of a person, the taking of a person’s health history, and the communication of an assessment or diagnosis are all excluded from the definition of “treatment”: s. 2(1). That the legislature felt it necessary to specify that such actions are not included within the definition of treatment strengthens the view that “treatment” was intended to have a very broad meaning.
The regulatory powers conferred by the HCCA further support this conclusion. The statute provides machinery for the scope of the term “treatment” to be narrowed by regulation, but not to be enlarged. The Lieutenant Governor in Council may pass regulations prescribing actions that do not constitute treatment, but cannot prescribe actions that constitute treatment: ss. 2(1) and 85(1)(f). The delegated authority to narrow, but not to enlarge, the definition of “treatment” suggests that the legislature intended the overall concept of treatment to be broadly construed.
Given the breadth of the definition of “treatment” articulated in the HCCA, it seems on first impression that withdrawal of treatment could fall within this term. Withdrawal or discontinuance of a given treatment clearly may be something done for a therapeutic, preventive, palliative, or other health-related purpose.
The scheme of the HCCA suggests that the legislature contemplated that withdrawal of treatment requires consent in some cases. One form of treatment identified under the HCCA is a “plan of treatment”, which is a defined term under the statute: s. 2(1). A physician may obtain consent for a plan of treatment that provides for various treatments and may provide for the withholding or withdrawal of treatment: ss. 2(1) and 13. Section 29(3) then states that if a treatment is withheld or withdrawn in accordance with a plan of treatment that the physician believes reasonably and in good faith was consented to, the physician is not liable for withholding or withdrawing the treatment. This provision would serve no purpose if consent were not required for the withholding or withdrawal of treatment in some circumstances.
The objects of the HCCA also support the view that “treatment” may include withdrawal of treatment. The values of autonomy – critical where life is at stake – and providing a meaningful role for family members support regarding withdrawal of life support as “treatment” requiring consent. These values must be balanced against that of ensuring appropriate care for incapable patients. The HCCA aims to strike the right balance among these values. Its purposes would be ill served by an interpretation that holds withdrawal of life support cannot constitute “treatment” under the Act.
Nor can I accept that the HCCA does not encompass withdrawal of life support because the legislature’s intention was simply to codify the common law. Nothing in the HCCA suggests that it is merely a codification of the common law. While the HCCA builds on the common law, its consent requirement is in some ways broader and in other ways narrower than the common law, based as it is on the detailed definition of “treatment” in s. 2(1): B. F. Hoffman, The Law of Consent to Treatment in Ontario, (2nd ed. 1997), at pp. 7, 10 and 163. For anything done which is outside the definition of “treatment”, the common law applies: s. 8(2). For anything done which is inside the definition of “treatment”, the HCCA applies. The HCCA goes on to provide an administrative scheme that is entirely independent of the common law.
Even if the HCCA merely codified the common law, this would not answer the question of whether consent is required for withdrawal of life support. The common law is not at all settled on this issue. While the common law has traditionally viewed consent as giving patients the right to refuse medical care rather than to refuse its withdrawal, courts have struggled with the applicability of this paradigm in the end-of-life context and have reached divergent conclusions: see. e.g., Golubchuk v Salvation Army Grace General Hospital, 2008 MBQB 49, 227 Man. R. (2d) 274, at paras. 22 and 25; Sweiss v Alberta Health Services, 2009 ABQB 691, 483 A.R. 340, at para. 48; Children’s Aid Society of Ottawa-Carleton v C. (M.) (2008), 301 D.L.R. (4th) 194 (Ont. S.C.J.), at paras. 33-34.
The arguments just reviewed undermine the view that withdrawal of life support is not “treatment” requiring consent. The physicians, however, advance two counter-arguments.
The first is textual. It is argued that by expressly including withdrawal of treatment in “plan of treatment” the legislature indicated that it did not intend withdrawal of treatment to be treatment requiring consent, unless the withdrawal is part of a plan of treatment.
However, it is difficult to draw inferences of legislative intent on point from these provisions. As pointed out by the courts below, measures must be “treatment” to be included within a “plan of treatment”, making the necessary process of inference circular. Moreover, it is unclear what the legislature meant by “plan of treatment”. A plan of treatment entails obtaining consent to all elements of the plan: s. 13. But the HCCA does not clarify whether a plan of treatment is fixed and must be fully specified in advance, or whether it permits flexible alteration in response to changes in the patient’s situation – an understanding that might extend to a case such as this. Whatever the correct response to these questions, the point is simply that it is not clear that the legislature intended only withdrawals of treatment that are part of a “plan of treatment” to be “treatment” under the HCCA.
Moreover, common sense suggests that the legislature cannot have intended withdrawal of life support to require consent only in the context of a plan of treatment. This would place the issue of consent at the sole discretion of physicians. A plan of treatment is simply a way in which physicians may choose to group and present various treatments to the patient for the purpose of obtaining consent. Allowing physicians to unilaterally determine whether consent is required in any given case cuts against patient autonomy and the statutory objective of providing consent rules that apply consistently in all settings: s. 1(a).
The second argument against regarding treatment as including withdrawal of life support is that it could lead to deeply undesirable results. If consent is required for withdrawal of life support, patients could arguably compel the continuation of any treatment, regardless of its medical implications. The legislature cannot have intended this. Common sense suggests that many withdrawals of treatment – for example, refusal to renew a prescription for a drug that may harm a patient – must be excluded from the definition of “treatment” under the Act.
The difficulty with this argument is that it treats everything that can be termed a withdrawal of treatment – from refusal to refill a prescription to ending life support – as equivalent for purposes of consent under the HCCA. A more nuanced view that withdrawal of treatment may sometimes, although not always, constitute “treatment”, better fits the provisions of the HCCA and the realities of medical care.
At a minimum, if the processes involved in withdrawal of care are health-related, they do not cease to be treatment merely because one labels them cumulatively as “withdrawal of treatment”. This applies to withdrawal of life support, as described in this case. The reality is that in Mr. Rasouli’s situation, the distinction between “treatment” and “withdrawal of treatment” is impossible to maintain. The withdrawal consists of a number of medical interventions, most if not all done for health-related purposes. Viewed globally, a series of distinct acts may be viewed as “withdrawal” of treatment. But viewed individually, each act may be seen as having a health-related purpose, and hence constitute “treatment” requiring consent.
The precise elements of withdrawal of life support will vary from case to case, but the substance of what is being done is the same: the provision of life-sustaining treatment is brought to an end and appropriate care is provided to ease suffering and prevent indignity at the end of life. Typically, the steps taken in withdrawal of life support are either physically required to effect the process of dying or directed to minimizing distress and discomfort as the dying process occurs. Providing services to address these problems serves health-related purposes within the meaning of s. 2(1) of the HCCA.
Many of the acts involved in withdrawal of life support entail physical interference with the patient’s body. The reality is that while “withdrawal” sounds like purely negative conduct, it typically involves physically touching or performing procedures upon the patient’s body. This is borne out by the case law of the Consent and Capacity Board: see E.J.G. (Re), 2007 CanLII 44704; G. (Re), 2009 CanLII 25289; and also Golubchuk, at para. 23.
Under the HCCA, as at common law, physical interference requires consent. The right to be free from unwanted physical interference goes to the heart of the law of consent to medical treatment. As described by the Ontario Court of Appeal in Malette, at p. 423:
The right of a person to control his or her own body is a concept that has long been recognized at common law. The tort of battery has traditionally protected the interest in bodily security from unwanted physical interference. Basically, any intentional nonconsensual touching which is harmful or offensive to a person’s reasonable sense of dignity is actionable.... Thus, as a matter of common law, a medical intervention in which a doctor touches the body of a patient would constitute a battery if the patient did not consent to the intervention.
Furthermore, withdrawal of life support, on the evidence here, entails the provision of palliative care. The case law of the Board suggests that this will generally be the case. Physicians regularly administer palliative care upon the removal of life support: see A.K. (Re), 2011 CanLII 82907, at p. 21; G. (Re).
The Court of Appeal held that withdrawal of life support and the administration of end-of-life palliative care are integrally linked. It reasoned that removal of mechanical ventilation is a necessary precondition to end-of-life palliative care and end-of-life palliative care is a necessary response to removal of the ventilator. Since the administration of palliative care clearly requires consent, the court concluded that consent should be required for the entire “treatment package”, including the removal of the ventilator (at paras. 50-52).
The Court of Appeal’s “treatment package” approach has been criticized on the basis that removal of life support will not invariably trigger the need for palliative care, and that the administration of palliative care may actually precede the decision to remove life support.
While the Court of Appeal’s assertion that removal of life support will always lead to the administration of end-of-life palliative care may be too broad, the evidence shows that palliative care will be administered in the process of withdrawal of life support in cases like Mr. Rasouli’s. It may be impossible to predict precisely how much distress the patient will suffer in the dying process, and hence what palliative care will be required. But what seems clear is that palliative care will inevitably be administered in Ontario hospitals as part of the process of withdrawing life support in cases like Mr. Rasouli’s. The simple fact is that appropriate medical care at the end of life, including palliative care, is closely tied to the withdrawal of life support.
In summary, withdrawal of life support aims at the health-related purpose of preventing suffering and indignity at the end of life, often entails physical interference with the patient’s body, and is closely associated with the provision of palliative care. Withdrawal of life support is inextricably bound up with care that serves health-related purposes and is tied to the objects of the Act. By removing medical services that are keeping a patient alive, withdrawal of life support impacts patient autonomy in the most fundamental way. The physicians’ attempt to exclude withdrawal of life support from the definition of “treatment” under s. 2(1) of the HCCA cannot succeed.
The practice of the Board, although not determinative, reinforces the conclusion that treatment under s. 2(1) includes withdrawal of life support. Whether implicit or explicit, a specialized tribunal’s interpretation of its home statute constitutes persuasive authority: Sullivan, at p. 621; P.-A. Côté, in collaboration with S. Beaulac and M. Devinat, The Interpretation of Legislation in Canada (4th ed. 2011), at pp. 584-85. The Board has regularly exercised its jurisdiction in cases where physicians proposed to withdraw life support, consistent with the view that withdrawal of life support constitutes “treatment” under theHCCA: see A.K. (Re); E.J.G. (Re); G. (Re). Courts on review have endorsed this interpretation: see Scardoni v Hawryluck (2004), 69 O.R. (3d) 700 (S.C.J.).
These considerations lead me to conclude that “treatment” in the HCCA should be understood as extending to withdrawal of life support in the situation at issue here and as that process is described in these proceedings. This case does not stand for the proposition that consent is required under the HCCA for withdrawals of other medical services or in other medical contexts.
F. The Argument That Requiring Consent for Withdrawal of Life Support Will Place Physicians in an Untenable Ethical Situation
A final argument raised by the physicians is that they may be placed in an untenable ethical situation if consent is required for withdrawal of life support. They could effectively be compelled to continue providing life support, even where they consider it to provide no medical benefit to, or even to harm, the patient. This could place physicians in breach of their legal and professional obligations to act in the best interests of the patient.
Legally, a physician cannot be faulted for following the direction of the Board, any more than he could be faulted for abiding by a judge’s direction at common law not to withdraw life support. Implicit in the physicians’ request that a judge resolve the present dispute is acceptance that if a judge orders that life support cannot be withdrawn, they must comply. Their legal position under the HCCA is no different.
However, a physician may feel that his legal obligation not to withdraw life support is in tension with his professional or personal ethics. Such tensions are inherent to medical practice. Indeed, the law of consent to medical treatment evolved through cases in which the patient did not wish to be treated, but the physician felt a professional obligation to treat: see Malette, at p. 420; Fleming, at pp. 85-86. The law is now clear that treatment cannot be administered without consent, irrespective of the ethical imperative that physicians may feel. Similarly, a physician’s duty of care may require that treatment not be withdrawn despite the physician’s ethical objections to its administration: see R. (Burke) v General Medical Council,  EWCA Civ 1003,  3 W.L.R. 1132, at para. 34. If the present case were resolved as my colleague Justice Karakatsanis proposes, the physicians may still be required not to withdraw life support based on their common law or fiduciary duties. Their ethical position under the HCCA is no different.
The HCCA’s scheme for dispute resolution offers several avenues through which a clash with a physician’s ethical compunctions may be averted. If the substitute decision-maker refuses consent to withdrawal of life support based on a patient’s prior wish, the physician may seek direction from the Board on whether the wish is applicable to the patient’s current circumstances (s. 35) or request permission for the substitute decision-maker to depart from the wish: s. 36. Where there is no applicable prior wish, the substitute decision-maker must act in the best interests of the patient. If the physician feels that the substitute decision-maker has not done so, he can challenge the decision before the Board: s. 37. In each of these types of proceedings, the physician’s submissions on the patient’s condition, the nature of the proposal to withdraw life support, and what will medically benefit the patient will be highly relevant to the Board’s analysis.
Wherever one tries to draw the line, it is inevitable that physicians will face ethical conflicts regarding the withdrawal of life support. No legal principle can avoid every ethical dilemma. What may be needed is a practical solution that enables physicians to comply with the law and to satisfy their professional and personal ethics. In this case, for example, the physicians explored the possibility of transferring Mr. Rasouli to a different Toronto hospital. Alternate staffing arrangements within Mr. Rasouli’s present hospital could also be considered. Finally, other physicians qualified to undertake Mr. Rasouli’s care may not hold an ethical objection to continuing the administration of life support. Such practical solutions could go far in averting any ethical conflict.
While the end-of-life context poses difficult ethical dilemmas for physicians, this does not alter the conclusion that withdrawal of life support constitutes treatment requiring consent under the HCCA.
G. Resolving Disagreements Over Withdrawal of Life Support
Having rejected the physicians’ arguments, it follows that the consent regime imposed by the HCCA applies in this case. I earlier outlined that regime. At this point, it may be useful to discuss in greater depth the role of the substitute decision-maker, health practitioners and the Board in cases like this.
To recap, theHCCA is a carefully tailored statute. It deals with patients capable of consent and patients who no longer have the power to consent. It seeks to maintain the value of patient autonomy – the right to decide for oneself – insofar as this is possible. This is reflected in the consent-based structure of the Act. If the patient is capable, she has the right to consent or refuse consent to medical treatment: s. 10(1)(a). If the patient is incapable, the HCCA transfers the right of consent to a substitute decision-maker, often next of kin (s. 10(1)(b)), who is required to act in accordance with the patient’s declared applicable wishes or failing that, the patient’s best interests: s. 21. Finally, it provides that a physician may challenge a substitute decision-maker’s consent decision by application to the Board: ss. 35 to 37. The physician may make submissions to the Board regarding the medical condition and interests of the patient. If the Board finds that the substitute decision-maker did not comply with the HCCA, it may overrule the substitute decision-maker and substitute its own opinion in accordance with the statute: s. 37(3). To be clear, this means that, even in life-ending situations, the Board may require that consent to withdrawal of life support be granted.
Under the HCCA, the substitute decision-maker does not have carte blanche to give or refuse consent. He or she must comply with the requirements of s. 21 of the Act, which contemplates two situations. The first is where the substitute decision-maker knows of a prior expressed wish by the patient which is applicable to the circumstances. The second is where there is no such wish, in which case the substitute decision-maker “shall act in the incapable person’s best interests”.
(1) Prior Expressed Wishes
If the substitute decision-maker knows of a prior wish regarding treatment that the patient expressed when capable and over 16 years old, and that is applicable in the circumstances, the wish must be followed: s. 21(1). This reflects the patient’s autonomy interest, insofar as it is possible.
While the HCCA gives primacy to the prior wishes of the patient, such wishes are only binding if they are applicable to the patient’s current circumstances. This qualification is no mere technicality. As the Ontario Court of Appeal held in Conway v Jacques (2002), 59 O.R. (3d) 737, at para. 31:
.... prior capable wishes are not to be applied mechanically or literally without regard to relevant changes in circumstances. Even wishes expressed in categorical or absolute terms must be interpreted in light of the circumstances prevailing at the time the wish was expressed.
Needless to say, where an incapable patient has expressed a prior wish that life support not be withdrawn, the intended meaning and scope of the wish must be carefully considered: see Fleming, at p. 94. The question is whether, when the wish was expressed, the patient intended its application in the circumstances that the patient now faces: see Conway, at para. 33; Scardoni, at para. 74. Changes in the patient’s condition, prognosis, and treatment options may all bear on the applicability of a prior wish: Conway, at paras. 37-38. For example, had Mr. Rasouli expressed a prior wish regarding life support, his substitute decision-maker would have to consider whether, when the wish was expressed, Mr. Rasouli intended the wish to apply if he were in a permanent vegetative state, with recovery extremely improbable according to medical evidence, and facing the health complications associated with long-term provision of life support.
A prior wish need not identify every possible future development in order to be applicable: Scardoni, at para. 74; K.M.S. (Re), 2007 CanLII 29956 (Ont. C.C.B.). However, a wish that is unclear, vague, or lacks precision may be held inapplicable to the circumstances. On this basis, the Board has found there were no prior wishes relating to life support applicable to the existing circumstances in numerous cases: D.D. (Re), 2013 CanLII 18799; P. (D), Re, 2010 CarswellOnt 7848; E.B. (Re), 2006 CanLII 46624; G. (Re); E. (Re), 2009 CanLII 28625; H.J. (Re), 2003 CanLII 49837. I have been unable to locate any case in which there was a prior expressed wish opposing withdrawal of life support that was held to be applicable and therefore binding in the circumstances.
If it is unclear whether a prior wish is applicable, the substitute decision-maker or physician may seek directions from the Board: s. 35. Alternately, if the substitute decision-maker acts on a prior wish that the physician believes is not applicable, the physician may challenge the consent decision before the Board: s. 37. The physician’s submissions on the patient’s condition, prognosis, and any adverse effects of maintaining life support will be relevant to the Board’s assessment of applicability.
In addition, either the substitute decision-maker or physician may apply to the Board for permission to depart from prior wishes to refuse treatment: s. 36. The Board may grant permission where it is satisfied that the incapable person, if capable, would probably give consent because of improvement in the likely result of the treatment since the wish was expressed: s. 36(3).
I note that the HCCA also provides that the substitute decision-maker is not required to comply with an expressed prior wish if “it is impossible to comply with the wish”: s. 21(1)2. This is not raised on the facts of this appeal, and I consider it no further.
(2) The Best Interests of the Patient
If the substitute decision-maker is not aware of an expressed prior wish of the patient or if the wish is not applicable to the circumstances, the substitute decision-maker must make her consent decision based on the best interests of the patient, according to the criteria set out in s. 21(2). These criteria include the medical implications of treatment for the patient, the patient’s well-being, the patient’s values, and any prior expressed wishes that were not binding on the substitute decision-maker. This legislative articulation of the best interests of the patient aims at advancing the values that underpin the HCCA: enhancing patient autonomy and ensuring appropriate medical treatment.
The substitute decision-maker is not at liberty to ignore any of the factors within the best interests analysis, or substitute her own view as to what is in the best interests of the patient. She must take an objective view of the matter, having regard to all the factors set out, and decide accordingly. This is clear from the mandatory wording of the opening portion of s. 21(2): the decision-maker “shall take into consideration” the listed factors. The need for an objective inquiry based on the listed factors is reinforced by s. 37, which allows the decision of the substitute decision-maker to be challenged by the attending physician and set aside by the Board, if the decision-maker did not comply with s. 21. The intent of the statute is to obtain a decision that, viewed objectively, is in the best interests of the incapable person.
The first consideration under s. 21(2), heavily relied on by Ms. Salasel in this case, concerns the values and beliefs of the incapable person. Section 21(2)(a) provides that the substitute decision-maker must consider the values and beliefs that the incapable person held when capable and that the substitute decision-maker believes that the incapable person would still act on if capable. Here, Ms. Salasel argues that sustaining life as long as possible accords with the religious beliefs of Mr. Rasouli, and that as a result he would not have consented to the removal of life support.
The second consideration relates to known wishes of the incapable person that were not binding on the substitute decision-maker under s. 21(1)1. For example, wishes expressed when a person was under the age of 16 or when incapable do not bind a substitute decision-maker, but must be taken into consideration in this stage of the best interests analysis.
Third, in addition to considering the values and beliefs of the patient and any relevant wishes, s. 21(2)(c) requires that the substitute decision-maker consider four factors that relate to the impact of the treatment on the patient’s condition, well-being, and health. This stage of the best interests analysis focuses on the medical implications of the proposed treatment for the patient. The attending physician’s view of what would medically benefit the patient must be taken into account.
The first factor asks whether receiving the treatment is likely to improve the patient’s condition or well-being, prevent deterioration of the person’s condition or well-being, or reduce the extent or rate of the deterioration of the person’s condition or well-being: s. 21(2)(c)1. In this case, the inquiry must determine whether removing life support would improve, prevent deterioration of, or reduce the extent or rate of deterioration of, Mr. Rasouli’s condition or well-being. The physicians argue that artificially prolonging Mr. Rasouli’s life will lead to health complications such as bedsores, respiratory infections, and organ failure – a scenario that can be avoided if life support is removed. On the other hand, Ms. Salasel argues that new evidence and evaluation suggest that Mr. Rasouli’s condition may improve in the future, militating against removal of life support.
The second factor requires the substitute decision-maker to consider whether, in the absence of the proposed treatment, the incapable person’s condition or well-being is likely to improve, remain the same or deteriorate: s. 21(2)(c)2. In this case, the inquiry is into the likely medical outcomes for Mr. Rasouli if life support is not withdrawn. The decision-maker must cast her mind into the future and ask what the patient’s condition will be in one year, five years, or ten years.
The third factor requires the substitute decision-maker to consider risks of harm associated with the treatment and weigh whether the benefits from the treatment will outweigh those risks: s. 21(2)(c)3. This factor is particularly important in cases where the substitute decision-maker must decide whether to go ahead with a risky procedure, like high-risk surgery, that while offering some hope, could worsen the patient’s situation. In this case, the substitute decision-maker must consider the benefits of removing life support, such as avoidance of protracted physical deterioration from bedsores, infections and organ deterioration ultimately leading to death, against the risks, which quite plainly are the hastening of death and the loss of whatever chance of recovery Mr. Rasouli has according to medical evidence.
The fourth factor requires the substitute decision-maker to consider alternative courses of treatment – whether less intrusive or restrictive treatment would be as beneficial as the treatment proposed: s. 21(2)(c)4. In a case such as this, the question is whether maintaining life support would be less intrusive or restrictive than its withdrawal, and if so, whether maintaining life support would be more beneficial to the patient than withdrawal.
As I see it, this review of s. 21(2) reveals that although a patient’s beliefs and prior expressed wishes are mandatory considerations, there is no doubt that the medical implications of a proposed treatment will bear significant weight in the analysis.
Where physicians and substitute decision-makers disagree about whether withdrawal of life support would be in the best interests of the patient, the HCCA provides the procedure for resolving this conflict. Under s. 37, the health care practitioner may apply to the Board to have the decision of the substitute decision-maker set aside on the ground that it is not in the best interests of the incapable person, having regard to the factors set out in s. 21(2) of the Act. This is an important avenue of recourse for physicians who believe that life support can no longer be ethically administered because it is not in the best interests of the patient to do so. The Board must duly consider the physician’s professional opinion and submissions on what would be of medical benefit to the patient.
If the Board agrees that the substitute decision-maker did not act in the best interests of the patient, it may substitute its own opinion for that of the substitute decision-maker: s. 37(3). Alternatively, if the Board concludes that the substitute decision-maker did act in the best interests of the patient, it can affirm the decision of the substitute decision-maker. In making these determinations, the Board must objectively apply the same criteria that substitute decision-makers are required to consider under s. 21. The Board is well placed to make a determination of whether treatment is in the best interests of the patient, in light of the statutory objectives of enhancing patient autonomy and ensuring appropriate medical care. This was observed by the Ontario Court of Appeal in M. (A.) v Benes (1999), 46 O.R. (3d) 271 [para. 46]:
A case will come before the Board only when the health practitioner disagrees with the S.D.M.’s application of the best interests test under s. 21(2). The Board will then have before it two parties who disagree about the application of s. 21: the S.D.M., who may have better knowledge than the health practitioner about the incapable person’s values, beliefs and non-binding wishes; and the health practitioner, who is the expert on the likely medical outcomes of the proposed treatment. The disagreement between the S.D.M. and the health practitioner potentially creates tension and the Act recognizes this by providing for a neutral expert board to resolve the disagreement. Indeed, after hearing submissions from all parties, the Board is likely better placed than either the S.D.M. or the health practitioner to decide what is in the incapable person’s best interests.
The Board must apply a standard of correctness in reviewing the decision of the substitute decision-maker: Benes, at para. 36; Scardoni, at para. 36. The wording of s. 37, which provides for full representation and gives the Board the right to substitute its decision for that of the substitute decision-maker, indicates that the Board must consider the matter de novo. The critical nature of the interests at stake support the Board’s obligation to review the decision of the substitute decision-maker on a correctness standard.
The legislature has given the Board the final responsibility to decide these matters. This is not to say that the courts have no role to play. Board decisions are subject to judicial review. This mechanism for court oversight ensures that the Board acts within its mandate and in accordance with the Constitution.
Over the past 17 years, the Board has developed a strong track record in handling precisely the issue raised in this case.
In some cases, the Board has upheld the decisions of substitute decision-makers to refuse withdrawal of life support as being in the best interests of the patient: D.W. (Re), 2011 CanLII 18217; S.S. (Re), 2011 CanLII 5000; P. (D.), Re. In others, it has reversed the decision of the substitute decision-maker and required consent to be given for the withdrawal of life support: A.K.; E.J.G.; N., (Re), 2009 CarswellOnt 4748. The particular facts of each case determine whether withdrawal of life support is in the best interests of the patient.
Bringing its expertise to the issue, the Board’s decisions may be expected to bring consistency and certainty to the application of the statute, thereby providing essential guidance to both substitute decision-makers and health care providers in this difficult area of the law.
H. The Dissenting Reasons
I have had the benefit of reading Justice Karakatsanis’ reasons. Respectfully, I cannot agree with her approach in this case.
First, I do not agree that the important role the HCCA accords to prior wishes indicates that the legislature did not intend the statute to apply to the withdrawal of life support (reasons of Justice Karakatsanis, at paras. 140-50). When it comes to the life and death matter of withdrawal of life support, there is every reason to think that the legislature intended a patient’s applicable prior wishes to be respected. Moreover, as discussed, a prior wish will only be binding if it is applicable to the patient’s current circumstances. Vagueness in a prior wish or changes in the patient’s condition, prognosis, or treatment options may mean that the prior wish is inapplicable. Where prior wishes are inapplicable, the best interests analysis governs.
Second, my colleague’s proposal that consent is required under the HCCA only where withdrawal of life support is part of a “plan of treatment” under the Act merely pushes the analysis back one step to the thorny problem of defining “plan of treatment” (paras. 154-55). Moreover, since the HCCA gives physicians discretion over whether withdrawal of life support will be presented as part of a plan of treatment, the ultimate issue of whether consent is required might well turn on the physician’s choice of what to call the procedure, risking arbitrariness. (As the record shows, withdrawal of life support is in practice often but not always proposed as a “plan of treatment”: see P. (D.), Re, at para. 16.
Third, if the HCCA does not apply, the fact that the common law tort of battery may prohibit the withdrawal of life support without consent in some cases raises further difficulties. Justice Karakatsanis leaves open the possibility that the law of battery may be engaged if the mechanics of withdrawal of life support are invasive (paras. 162-63). However, it is unclear what rules for substitute consent would govern in these circumstances, and on what legal basis the physician could proceed with withdrawal of life support.
I cannot agree that, on the record before us, withdrawal of life support from Mr. Rasouli would not be invasive (reasons of Justice Karakatsanis, at para. 163). Mechanical ventilation is delivered to Mr. Rasouli through a tube surgically inserted into his throat. His hospital chart indicates that withdrawal of life support would entail “extubation” (A.R., vol. II, at p. 50), which is the “[r]emoval of a tube from an organ, structure, or orifice”: Stedman’s Medical Dictionary (28th ed. 2006), at p. 688. On the record before us, the possibility of an invasive medical procedure to withdraw life support cannot be excluded.
Fourth, it can be questioned whether the legal tools my colleague relies on to provide judicial oversight of physician’s decisions to withdraw life support are sufficient for the task. At this point in the development of the standard of care and fiduciary duty, they do not appear to support the broad-based common law/equity review that my colleague envisions (paras. 190-201).
There has been no trial on the standard of care in this case, so we can only speculate as to its content in situations like Mr. Rasouli’s. Whatever its content, the standard of care does not hold physicians to a standard of perfection but, rather, only to one of reasonable care: E.I. Picard and G.B. Robertson, Legal Liability of Doctors and Hospitals in Canada (4th ed. 2007), at p. 225; Crits v Sylvester (1956), 1 D.L.R (2d) 502 (Ont. C.A.), at p. 508, aff’d  S.C.R. 991. As such, standard of care is a blunt instrument for reviewing physician decisions to withdraw life support, and not one that supports the broad-based review contemplated by my colleague.
As for fiduciary duty, once again the waters into which my colleague ventures are untested. While this Court has recognized that the doctor-patient relationship is fiduciary in nature, it has never reviewed physicians’ good-faith treatment decisions on the basis of fiduciary duty: McInerney v MacDonald,  2 S.C.R. 138, at p. 149; Norberg v Wynrib,  2 S.C.R. 226, at p. 274. My colleague’s approach thus contemplates a substantial expansion of the role of fiduciary duty in regulating the doctor-patient relationship.
Finally, how the standard of care and fiduciary duty relate to one another is unclear on the proposed approach. Which analysis comes first – standard of care or fiduciary duty? And in case of conflict, which prevails?
Fifth, if the HCCA does not apply, it is unclear who will protect the interests of the incapable patient when a physician determines that life support should be withdrawn. The proposed approach contemplates the involvement of a “substitute decision-maker”, but does not state who this person would be or under what legal authority he would act (reasons of Justice Karakatsanis, at paras. 202 and 204). The HCCA creates the legal designation of “substitute decision-maker”, and sets out a hierarchy for identifying this individual. However, this designation is unavailable if the HCCA does not apply to the withdrawal of life support.
Sixth, I am concerned about the practical impact of the proposed approach on patients, families, and physicians. It may heighten the vulnerability of incapable patients, since the legal burden will be on family or friends to initiate court proceedings to prevent the withdrawal of life support, rather than on physicians to obtain consent before acting. The implications of this shift are particularly troubling where the incapable patient lacks a network of family and friends with the financial resources to fund legal action, which could entail a trial on the medical standard of care. Furthermore, it is unclear on the proposed approach whether the physician could withdraw life support without judicial oversight if no legal action is taken by the patient’s family or friends.
One of the legislature’s primary motivations in enacting the HCCA was to simplify the law governing the treatment of incapable patients. The HCCA sets out clear rules requiring consent before treatment can occur, identifying who can consent for an incapable patient, stating the criteria on which consent must be granted or refused, and creating a specialized body to settle disputes. The legal framework of theHCCA has been used to resolve end-of-life disputes in Ontario for 17 years. I would be reluctant to close off access to this established regime and cast these matters back into the courts.
I conclude that the following steps apply under the HCCA in a case such as this, where the substitute decision-maker and the medical health care providers disagree on whether life support should be discontinued.
The health practitioner determines whether in his view continuance of life support is medically indicated for the patient;
If the health practitioner determines that continuance of life support is no longer medically indicated for the patient, he advises the patient’s substitute decision-maker and seeks her consent to withdraw the treatment;
The substitute decision-maker gives or refuses consent in accordance with the applicable prior wishes of the incapable person, or in the absence of such wishes on the basis of the best interests of the patient, having regard to the specified factors in s. 21(2) of the HCCA;
If the substitute decision-maker consents, the health practitioner withdraws life support;
If the substitute decision-maker refuses consent to withdrawal of life support, the health practitioner may challenge the substitute decision-maker’s refusal by applying to the Consent and Capacity Board: s. 37;
If the Board finds that the refusal to provide consent to the withdrawal of life support was not in accordance with the requirements of the HCCA, it may substitute its own decision for that of the substitute decision-maker, and permit withdrawal of life support.
Applying the HCCA in the manner just discussed, we arrive at the following conclusions.
The appellant physicians, having determined that in their view Mr. Rasouli should be removed from life support, were obliged to seek Ms. Salasel’s consent to the withdrawal. Since Mr. Rasouli had not expressed a wish within the meaning of s. 21(1)1, Ms. Salasel was required to determine whether removal of life support was in Mr. Rasouli’s best interests, having regard to the factors set out in s. 21(2) of the Act.
If the appellant physicians do not agree that maintaining life support for Mr. Rasouli is in his best interests, their recourse is to apply to the Board for a determination as provided by s. 37(1) of the HCCA.
When the application is brought, it will be for the Board to determine whether Ms. Salasel’s refusal to provide consent to the withdrawal of life support was in Mr. Rasouli’s best interests, within the meaning of s. 21(2) of the HCCA. If the Board is of the opinion it was not, it may substitute its decision for that of Ms. Salasel, and clear the way for removal of Mr. Rasouli’s life support.
It follows that I would dismiss the appeal. I would also dismiss the motions to adduce fresh evidence on the appeal to this Court, without prejudice to the Board receiving any evidence it deems relevant on the hearing before it.
This being a matter of public interest, I would not award costs.
(with whom Abella J joined)
This appeal addresses the roles of the doctor, the incapable patient’s substitute decision-maker, Ontario’s Consent and Capacity Board and the courts, in the decision to withdraw or withhold life support.
Unlike the Chief Justice, I conclude that the common law, and not the Health Care Consent Act, 1996, S.O. 1996, c. 2, Sch. A (the HCCA or the Act), governs when doctors and substitute decision-makers disagree regarding the proposed withdrawal of an incapable patient’s life support. Thus, the court, and not the Consent and Capacity Board, is the appropriate forum for resolving any disputes between the doctors and the incapable patient’s substitute decision-maker.
In my view, the HCCA does not apply to the withdrawal of treatment. The HCCA codifies the deeply rooted common law right to refuse treatment, no matter the medical consequences. It does not, however, give patients, or their substitute decision-makers, the right to insist on the continuation of a treatment that is futile, harmful, or contrary to professional medical standards of care.
In reviewing whether a physician is acting within the professional standard of care at common law, the court should determine whether the life support has any chance of being medically effective and whether withdrawal of the treatment is in the best interests of the patient. This necessarily includes consideration of the patient’s wishes, values and beliefs, in addition to the broad mental and physical implications for the patient’s condition and well-being, all within the framework of the governing legal principles.
Accordingly, I would allow the appeal and remit this matter to the Ontario Superior Court of Justice, so that it may make the necessary findings of fact and to determine whether the withdrawal of life support is in accordance with the standard of care and the best interests of the patient.
The patient in this case suffered brain damage that, according to his doctors, left him in a persistent vegetative state. For almost three years, he has been kept alive by life support, including mechanical ventilation and artificial nutrition. It is his medical team’s opinion that there is no realistic hope for his recovery. In their view, the provision of life support offers him no medical benefit and may, in fact, cause harm. As such, they wish to withdraw life support.
The patient’s wife, who is his substitute decision-maker, does not want life support withdrawn. She has filed more recent medical assessments as fresh evidence of a change in the patient’s diagnosis to a minimally conscious state. Further, she believes that her husband’s religious beliefs dictate that he would wish to be kept alive in these circumstances. Accordingly, she brought an application to restrain the patient’s doctors from withdrawing life support.
The application judge concluded that, under the HCCA, the physician’s decision to remove life support requires the consent of the patient or the patient’s substitute decision-maker. As a result, the doctors are required to apply to the Consent and Capacity Board if they wish to challenge the substitute decision-maker’s decision to refuse to consent to the withdrawal of life support (2011 ONSC 1500, 105 O.R. (3d) 761).
 The Court of Appeal dismissed the doctors’ appeal (2011 ONCA 482, 107 O.R. (3d) 9). While the court was prepared to accept for present purposes “that the Act does not require doctors to obtain consent from a patient or substitute decision-maker to withhold or withdraw ‘treatment’ that they view as medically ineffective or inappropriate” (para. 46), it concluded that the withdrawal of life support was inextricably bound with palliative care as a “treatment package” and therefore fell within the definition of “treatment” in the Act. In these specific circumstances, the Court of Appeal held that consent was required.
For the reasons that follow, I conclude that the Ontario legislature did not intend the HCCA to require patient consent for the withdrawal of medical treatment.
A. The Health Care Consent Act, 1996
The HCCA codifies and builds upon the common law of consent in Ontario. Similar legislation exists in several provinces across Canada. The Act provides that a medical practitioner who proposes a treatment cannot administer it without the consent of the patient, or if the patient is incapable of consenting, the consent of the substitute decision-maker (s. 10(1)). It does not say that a physician who proposes to withdraw treatment must obtain the consent of the patient or substitute decision-maker. Further, “treatment” is defined as “anything that is done for a .... health-related purpose” (s. 2(1)). The definition does not say that “treatment” includes the withdrawal or withholding of treatment.
The legislative history, similarly, discloses no intention to create a right for a patient to insist on treatment that a physician considers medically futile. Rather, the Act is designed to provide for findings of incapacity and an orderly and principled regime for substitute decision-makers (see H. Young, “Why Withdrawing Life-Sustaining Treatment Should Not Require ‘Rasouli Consent’ (2012), 6:2 M.J.L.H. 54, at p. 66).
(1) Codifying Patient Autonomy
One of the HCCA’s purposes is to require that medical professionals adhere to the treatment wishes expressed by a person while capable and over the age of 16 (s. 1(c)(iii)). The known express wishes of a patient to refuse treatment must prevail, notwithstanding the medical consequences (s. 10(1)). As with the common law, this respects the autonomy of a person to refuse a proposed treatment – no matter the reason – even if the treatment is medically necessary (see Reibl v Hughes,  2 S.C.R. 880; Hopp v Lepp,  2 S.C.R. 192). Thus, for example, an adult patient has the right to refuse a blood transfusion even if, as a result, death is inevitable. When it comes to refusing treatment, personal autonomy is paramount.
However, the converse is not true. As discussed below, there is no clear right under the Act or at common law for a patient to insist on a particular treatment if the doctor is not prepared to provide or continue to provide it. The HCCA reflects the consensus at common law, and does not require that a patient’s wishes prevail. When the issue is the withdrawal of treatment that is no longer medically effective or is even harmful, a patient’s choice alone is not an appropriate paradigm. A patient’s autonomy must be balanced against broader interests, including the nature of her condition, the implications of continuing the treatment, the professional obligations of her physicians, and the impact on the broader health care system. This reflection of the common law is evident from the purposes, provisions, and scheme of the Act.
(2) The Purposes, Provisions, and Scheme of the Act
The purposes of the HCCA under s. 1 include:
The Act sets out the framework for choosing a substitute decision-maker, determines the principles and process by which treatment decisions are made for incapable patients, and provides a limited mechanism to resolve disputes that may arise between the decision-maker and medical practitioners.
For the reasons that follow, I am of the view that a withdrawal of treatment does not fall within the meaning of “treatment” under the Act. Further, given the role of known wishes under the Act, I am satisfied that the legislature did not intend for the Act to extend to such decisions, and has left such matters to be determined before the courts, according to the common law.
(a) The Role of Known Wishes in the Scheme of the Act
My conclusion that the HCCA does not require patient consent for withdrawal or withholding treatment is informed by the scheme of the Act.
(i) Principles for Giving or Refusing Consent
The scheme of the Act ensures that when treatment is proposed, doctors, substitute decision-makers and the Board, are all bound by the patient’s known wishes, if clear and applicable. This is true for all treatment; there are no special provisions for end-of-life scenarios.
Where the wishes of the patient are not known, the Act provides a broad test for determining whether it is in the best interests of the incapable patient to consent to or refuse treatment (s. 21(2)). It requires that the incapable patient’s values, beliefs, and wishes be considered along with the consequences of a treatment on the patient’s medical condition and broader well-being of the patient going forward.
The best interests test has no role, however, when the express and clear wishes of the patient are known and applicable. The provisions of the Act, read together, ensure that if the express wishes of the patient are known they must be followed.
When the Act is engaged, s. 21(1)1 requires a substitute decision-maker to follow the express wishes of the patient if those wishes are known. Only if the wishes are not known can the substitute decision-maker act in the best interests of the patient:
(ii) Limited Role of the Consent and Capacity Board
The Act permits applications to the Consent and Capacity Board with respect to the consent of incapable patients under three different provisions. (These applications to the Board are not available to resolve disputes between a doctor and his capable patient; any such disputes must be resolved through the courts.) These provisions permit the substitute decision-maker or the health practitioner to seek clarification or direction from the Board. They do not, however, give the Board the authority to override any clear and applicable known wishes.
Section 35 provides for an application to the Board (by the substitute decision-maker or the health practitioner) for directions if there is need for clarity about the wishes of the incapable person. However, in giving directions the Board shall apply s. 21, which in turn requires consent be given or refused in accordance with any known wish.
Section 36 permits a substitute decision-maker (or health practitioner) who seeks the consent for treatment, despite known wishes to refuse the treatment, to apply to the Board. However, the Board may override the known wishes only if the patient would have likely consented because “the likely result of the treatment is significantly better than would have been anticipated in comparable circumstances at the time the wish was expressed” (s. 36(3)).
Finally, a health practitioner can apply to the Board under s. 37 to determine whether the substitute decision-maker complied with the requirements for providing consent under s. 21. Once again, pursuant to s. 21, the best interests test applies only if there are no clear known wishes. There is no provision that gives the Board authority to permit a substitute decision-maker to ignore known wishes and to determine and act on the best interests of the patient if the patient, over the age of 16 years, expressed a clear and applicable wish.
Thus, where there are known wishes, the only available applications to the Consent and Capacity Board are (1) to obtain clarification of those wishes if unclear; or (2) to consider whether the wishes would likely have been different because of a change in the effectiveness of the treatment; or (3) to ensure the known wishes are being respected.
A clearly expressed and still applicable wish to refuse treatment must therefore be followed by the substitute decision-maker, the physicians, and, crucially, by the Consent and Capacity Board. In this context, if the legislation was intended to go beyond the common law right to refuse medical treatment – to provide a patient or their substitute decision-maker with the right to insist on the continuation of treatment – it would have done so in clear terms. It provides no such right.
(b) “Treatment” Does Not Include Withdrawal of Treatment
In Ontario, s. 10 of the HCCA requires physicians to secure consent to treatment:
While ss. 10 and 21(1) give a central role to consent in the context of treatment, the Act also specifically provides that it does not affect the law relating to giving or refusing consent to anything not within the definition of “treatment” (s. 8(2)).
Section 2 of the HCCA provides the scope of the procedures for which consent must be secured under s. 10(1) in the following definition of “treatment”:
“treatment” means anything that is done for a therapeutic, preventive, palliative, diagnostic, cosmetic or other health-related purpose, and includes a course of treatment, plan of treatment ....
and of “plan of treatment”, which is defined as follows:
“plan of treatment” means a plan that,
On its face, the definitions of “treatment” or “plan of treatment” in s. 2 do not require a doctor to obtain consent to withdraw treatment, except where it is specifically provided as a component of a “plan of treatment”. The word “treatment” in the Act is associated with something that is proposed by a health practitioner (s. 10) and done for a health-related purpose (s. 2). Although withdrawal of life support is something that is proposed by a health practitioner, it is not done for a health-related purpose. Rather, it is done for the purpose of discontinuing treatment. Stated differently, withdrawing life support brings that treatment to an end. The definition of “treatment” does not say that it includes discontinuing treatment.
In my view, the reference to withholding or withdrawing treatment, in the definition of a “plan of treatment”, does not provide support for interpreting “treatment” generally to include its withholding or withdrawal. Rather, this specific reference to withholding or withdrawal of treatment in the definition of “plan of treatment” makes it clear that the framers of the Act differentiated between treatment and the withholding or withdrawal of treatment. They could have specifically included it in the definition of “treatment”, if they wished to do so. Instead, withholding or withdrawal of treatment is included only as an optional additional element in the context of an overall plan that is focused on providing “various treatments or courses of treatment” to deal with a health problem. While the definition of a “plan of treatment” may create some arbitrariness by including some withdrawals of treatment and not others, the Act has addressed only those withdrawals of treatment that are bound up with an overall plan that provides for more than one treatment or course of treatment.
The protection of liability provisions further support the conclusion that the HCCA only contemplates a requirement of consent for the withholding or withdrawal of treatment that is included in a “plan of treatment”. A physician is protected when there is an apparently valid consent to “treatment” or apparently valid refusal of “treatment” (s. 29(1) and (2)). When a physician obtains an apparently valid consent to a “plan of treatment”, no liability can ensue for the withholding or withdrawal of the treatment, as set out in the plan of treatment (s. 29(3)). There is no similar general provision to protect a physician from liability if a patient consents to the withdrawal of “treatment” that is not part of a “plan of treatment”.
(3) The HCAA Does Not Govern Consent to Withdrawal of Treatment
In light of the purpose, text and scheme of the Act, I conclude that the definition of “treatment” does not include the withdrawal of treatment. As the Court of Appeal recognized, “if the legislature intended that consent was required to the withholding or withdrawal of life support measures that are considered to be medically ineffective or inappropriate”, clearer language to that effect would have been used in the statute (para. 41). The reasonable conclusion is that the HCCA does not alter the common law of consent by creating an entitlement to treatment.
The Chief Justice, however, concludes that withdrawal of treatment may sometimes, although not always, constitute “treatment” as defined in the Act. She notes that the definition of “treatment” does not include the withdrawal of treatment in every case. For example, a doctor would not require consent for the withdrawal of prescription medication that was not medically indicated. To require consent in that scenario would allow patients to compel the continuation of any treatment despite any of the attendant medical implications and would be absurd.
The Chief Justice concludes that the withdrawal of life support is included in the definition of “treatment” in s. 2 and consent is required under s. 10(1) because the processes involved in the withdrawal of life support are a series of distinct acts – most serving health-related purposes – and the critical interests at stake go to the heart of the purposes of the Act. In particular, the steps taken in withdrawing life support are physically required to effect the process of dying (or are directed to minimizing distress and discomfort as the dying process occurs), and serve health-related purposes. She concludes that such steps are often (but not always) bound up with palliative care treatment. They often (but not always) involve physical interference with the body of the patient.
I take a different view. I am not satisfied that, under the Act, there is a coherent basis upon which to conclude that some withdrawals of treatment require consent and others do not. In my view, whether withdrawal of treatment is considered to be treatment should not depend upon the process involved in the withdrawal of the treatment. Nor should it depend upon whether it is withdrawn in conjunction with other treatment to ease distress or discomfort, or upon how invasive the particular process is. Finally, it should not depend upon how a treatment has been administered in a particular case, whether orally, by injection or intravenously. These distinctions provide little clarity about when the Act will be engaged.
There is no doubt that, under the Act, the provision of treatment to provide palliative care requires consent. However, while the decision to withdraw life support may also lead to a decision regarding palliative care, these decisions are not always bound together. The record shows that the relationship between the two, as a matter of implementation, will depend upon the specific circumstances of the case. The requirement for consent for withdrawal should not rest on whether palliative care is required or has preceded the withdrawal of life support. They are separate issues. There is nothing in the Act to support tying these separate issues together as a “treatment package”.
The withdrawal of life support does not necessarily engage the common law tort of battery. Discontinuing life support need not require physical touching of the patient. Stopping a ventilation machine or discontinuing the provision of sustenance to the patient are both activities undertaken without any physical interaction with the patient.
On this record, the withdrawal of life support – the stopping of the respirator or the provision of sustenance itself – would not be invasive. The withdrawal of treatment may in some cases, however, entail some physical interference so that it can be undertaken more comfortably, or to respect the patient’s dignity. In this case, it may be that extubation would be undertaken once the tube no longer serves a purpose in the provision of life support. To the extent such measures constitute palliative care, consent is required.
In conclusion, the HCCA was not intended to cover the withdrawal of treatment. The Act is not intended to provide a comprehensive scheme. Indeed, it specifically provides that it does not affect the law relating to giving or refusing consent to anything not within the definition of “treatment” (s. 8(2)). A “plan of treatment” is described as something proposed or “developed” by the health care team for health-related purposes (s. 2(1)). The Act itself does not refer to the withdrawal of treatment, except as a specific component of a larger plan for the provision of treatments or courses of treatment.
The Act is designed to give effect to the principle of patient autonomy – a principle with deep roots in our common law – that permits a patient to refuse medical treatment, no matter the consequences. However, neither the HCCA nor the common law permits a patient to dictate treatment; as I discuss below, there is no common law consensus that a doctor requires the consent of the patient to withhold or withdraw treatment. Neither the words nor the scheme of the Act contemplate a patient’s right to stop a doctor from withdrawing treatment that is no longer medically effective or is even harmful. Indeed, such an extension of patient autonomy to permit a patient to insist on the continuation of treatment that is medically futile would have a detrimental impact on the standard of care and legal, ethical, and professional duties in the practice of medicine.
The role of patient autonomy must be balanced with the physician’s role, expertise, and advice. As well, there are a myriad of important interests, such as the integrity of our health care system, at stake. But the doctor’s ability to challenge the decision of the substitute decision-maker under the Act is limited in scope, particularly where the patient has made a prior wish with respect to treatment. The Board in such circumstances has no authority under the Act to override express clear and relevant wishes of the patient.
For all these reasons, I conclude that the legislature did not intend that the Act would apply to the withholding or withdrawal of life-sustaining treatment.
B. The Common Law
In my view, the common law does not entitle a patient to insist upon continuation of treatment; it does not require a patient’s consent to the withholding or withdrawal of treatment that the physician considers has no chance of being medically effective and that is no longer consistent with the professional standard of care. For the reasons that follow, I conclude that such consent is not required at common law, even in the context of withholding or withdrawal of life support. I note that rights pursuant to the Canadian Charter of Rights and Freedoms have not been raised or argued in this appeal.
A patient’s wishes, ideals, and values are important considerations in end-of-life decisions in an institutional setting. However, the continuation of life is not an absolute value. The ultimate decision whether to withdraw life-sustaining treatment must respect the medical or physical consequences of withdrawal or continuation of life support, and also the personal autonomy, bodily integrity, and human dignity of the patient. A doctor cannot be required to act outside of his standard of care and contrary to his professional duties.
In addition to the obligation to perform their medical duties in accordance with the standard of care, however, doctors have fiduciary obligations to their patients (see McInerney v MacDonald,  2 S.C.R. 138). La Forest J. described the fiduciary duty in McInerney: “As part of the relationship of trust and confidence, the physician must act in the best interests of the patient” (p. 154).
In my view, these obligations should require doctors to undertake a certain process for resolving such important questions, similar to the decisions that families and substitute decision-makers must make in the end-of-life setting. For one, the doctor should include a role for the family or substitute decision-maker. The doctor’s obligations should include, for example, providing notice and a thorough and accommodating process for determining the condition and best interests of the patient. When the medical team determines an appropriate course of action and the patient or their substitute decision-maker disagrees, doctors should also explore alternative institutions willing to continue the treatment.
In light of the duties that the doctor owes to her patient in the end-of-life setting, if the family objects to the physicians’ and institution’s final assessment, the court will review the circumstances to ensure that the doctor’s decision to withdraw life support accords with the required standard of care and that the doctor has discharged her fiduciary obligations to act in the best interests of the patient.
(2) Common Law Jurisprudence
The right to refuse treatment is well entrenched in the common law. However, the reverse is not true. I know of no, nor have I been directed to any, Canadian decision holding that consent is a necessary condition for the withholding or withdrawal of treatment generally. In my view, there is no general common law right or entitlement to treatment that a doctor considers medically ineffective or contrary to the professional standard of care.
The withdrawal of life support, however, involves stark emotional responses, competing values and difficult choices. Increasingly, medical advances permit institutions to use extraordinary measures to prevent patients from dying. In Airedale N.H.S. Trust v Bland,  A.C. 789 (H.L.), at p. 868, Lord Goff of Chieveley stated that the Court’s task is not to determine “whether it is in the best interests of the patient that he should die. The question is whether it is in the best interests of the patient that his life should be prolonged by the continuance of this form of medical treatment or care.” As the application judge Himel J. noted, in her careful and thorough review of the common law (at paras. 53-83), the issue of whether consent is required for the withdrawal of life support treatment has come before the courts in recent years.
 A number of common threads emerge from the jurisprudence.
(a) Consent Is Not Required to Withdraw or Withhold Treatment That Is Medically Ineffective
First, even in end-of-life situations, I have not been directed to any Canadian decision ordering that a physician obtain consent to withhold or withdraw treatment that is not medically effective. When faced with the question of whether or not consent is required for the withdrawal or withholding of life support, some courts have reviewed a physician’s decision to withhold life-sustaining treatment on the basis of whether or not it was in the best interests of the patient. See Sweiss v Alberta Health Services, 2009 ABQB 691, 483 A.R. 340; I.H.V., Re, 2008 ABQB 250, 449 A.R. 211.
Even in those cases in which the court has intervened to prevent doctors from unilaterally withdrawing or withholding treatment, the courts did not conclude that consent was required. Rather, in those cases, the courts ordered an injunction pending trial. In Sawatzky v Riverview Health Centre Inc. (1998), 132 Man. R. (2d) 222 (Q.B.), the court granted an interim injunction removing a “do not resuscitate” order pending trial, as the case involved factual questions as to the patient’s status and raised the question of whether the Charter or The Human Rights Code, C.C.S.M. c. H175, prevented a doctor from unilaterally imposing a “do not resuscitate” order.
In Golubchuk v Salvation Army Grace General Hospital, 2008 MBQB 49, 227 Man. R. (2d) 274, the court observed at paras. 18-23 that there would be some situations in which withdrawal of life support may necessitate some touching of the patient, including for the administration of drugs for pain. The requirement of consent, therefore, would seem to be predicated upon whether it was necessary to touch the patient in order to withdraw life support or to make him more comfortable by administering palliative care. This is similar to the “treatment package” reasoning used by the Court of Appeal in this case and is subject to the same objections, outlined above.
In another set of cases, an injunction was ordered for a period of time in order to allow the patient’s representatives to procure a second opinion or an ethics committee report before the doctors would be able to act unilaterally (see Sweiss, at paras. 67-68; Jin v Calgary Health Region, 2007 ABQB 593, 428 A.R. 161, at paras. 40-42).
However, other courts, including the only other appellate decision in Canada prior to the Court of Appeal’s decision in this case (Child and Family Services of Central Manitoba v R.L. (1997), 123 Man. R. (2d) 135 (C.A.)), have explicitly concluded that consent is not required and that it is not appropriate for a court to interfere with medical doctors acting unilaterally and professionally in the best interests of a patient. See also I.H.V., Re; Rotaru v Vancouver General Hospital Intensive Care Unit, 2008 BCSC 318 (CanLII).
In R.L., the Manitoba Court of Appeal concluded that consent was not required to place a “Do Not Resuscitate” order on a patient’s chart. The patient was an 11-month old infant in a persistent vegetative state. Doctors agreed that the patient would not regain any form of consciousness and, like in this case, continuing life support would not improve the patient’s condition. The patient’s parents objected to the order. The Court of Appeal stated, at para. 17:
.... neither consent nor a court order in lieu is required for a medical doctor to issue a non-resuscitation direction where, in his or her judgment, the patient is in an irreversible vegetative state. Whether or not such a direction should be issued is a judgment call for the doctor to make having regard to the patient’s history and condition and the doctor’s evaluation of the hopelessness of the case. The wishes of the patient’s family or guardians should be taken into account, but neither their consent nor the approval of a court is required.
This reasoning in R.L. reflects the well-established approach in the United Kingdom. U.K. courts generally agree that consent is not required to withdraw life support. Holding that physicians were not obligated to adopt a course of treatment that, in their view, was not in the patient’s best interests, the Court of Appeal considered that a court order would be “an abuse of power as directly or indirectly requiring the practitioner to act contrary to the fundamental duty which he owes to his patient” (Re J (a minor) (wardship: medical treatment),  4 All E.R. 614, at p. 622; see also Re R (a minor) (wardship: medical treatment),  4 All E.R. 177; Bland).
In Bland, the leading case in the U.K., the House of Lords held that health care providers would not be criminally or civilly liable for withdrawing treatment from a patient in a persistent vegetative state, where, in the physicians’ view, there was no possibility that he would regain consciousness and that continuing life support was not in the patient’s best interests. The House of Lords concluded that the withdrawal of life support was not illegal without a court order [p. 875, per Lord Lowry]:
.... in the absence of an application, the doctor who proposes the cessation of life-supporting care and treatment on the ground that their continuance would not be in the patient’s best interests will have reached that conclusion himself and will be judge in his own cause unless and until his chosen course of action is challenged in criminal or civil proceedings.
In the United States, some state legislatures have addressed the withholding and withdrawal of life support directly. While U.S. courts have shown deference to patient wishes, they have declined to address the issue of whether a patient has the right to insist on life support (In Re: The Conservatorship of Helga M. Wanglie, No. PX-91-283 (Minn. Dist. Ct. (Prob. Ct. Div) 1991), reported in (1991), 7 Issues L. & Med. 369); and whether consent was required to withdraw life sustaining treatment that has no benefit to the patient (In the Matter of Baby “K”, 16 F.3d 590 (4th Cir. 1994)).
Thus, courts throughout Canada, the U.K. and the U.S. have been reluctant to require a doctor to provide or continue life support treatment that was found to be outside the professional medical standard of care. As Hilary Young points out (at p. 63), neither the origins of consent as a defence to battery, nor the more recent development of the doctrine of informed consent in negligence, ground a legal entitlement to life-sustaining treatment outside the standard of care.
 In my view, even in end-of-life situations, there is no common law right to insist on medical treatment that the doctor and the institution consider medically futile, harmful, and outside professional standards. Consent is not required to withdraw life-sustaining treatment in such circumstances. Patients cannot force doctors to act in violation of the standard of care.
(b) Doctor/Institutional Decisions Are Subject to Judicial Oversight
The second thread that runs through the jurisprudence is the court’s supervisory role in adjudicating end-of-life decisions; such decisions are not entirely within the discretion of doctors (see, for example, Sawatzky; Jin; Golubchuk; Rotaru).
Even in Bland, where the House of Lords held that at common law doctors do not require consent to withdraw life support treatment, the court suggested that, at least for a time, it would be desirable that physicians receive court approval before ending life support treatments (p. 859, per Lord Keith of Kinkel).
Typically, the courts have become engaged in end-of-life decision-making when a patient’s family has sought an injunction to stop the institution from withholding or withdrawing life-sustaining treatment. Most often, the analysis centres on the factual record and whether the treatment is futile or medically ineffective. In addition, courts have also looked broadly to the best interests of the patient.
(i) Decisions to Withdraw Life Support Must Be in Accordance With the Standard of Care
In my view, Canadian courts should assess whether the decision to withdraw life support accords with the physician’s standard of care and her fiduciary duty, as well as, considerations of patient autonomy and human dignity. In any review, the doctor’s medical diagnosis and view of the implications of continued treatment feature prominently. The wishes, values, and beliefs of the patient should be considered; however, they cannot be determinative. A doctor cannot be required to act contrary to her standard of care.
The common law protects the interests of Canadians in the medical realm – whether doctor or patient – by requiring physicians to act (1) in accordance with the conduct of a prudent practitioner of the same experience and standing in her field, including a duty to obtain informed consent (ter Neuzen v Korn,  3 S.C.R. 674; Reibl, at pp. 899-900) and (2) in the best interests of their patients (Norberg v Wynrib,  2 S.C.R. 226, at pp. 270-72). Typically, decisions to provide or to withdraw treatment are made on the basis of medical benefit to the patient. This approach will likely satisfy the standard of care and advance the patient’s best interests where the patient’s medical condition is the primary concern.
(ii) Fiduciary Duties Play a Role in the Withdrawal of Life Support
The fiduciary duty is a broad and evolving set of obligations inhering in some elements of the doctor-patient relationship. La Forest J. held in McInerney [p. 149]:
In characterizing the physician-patient relationship as “fiduciary”, I would not wish it to be thought that a fixed set of rules and principles apply in all circumstances or to all obligations arising out of the doctor-patient relationship. As I noted in Canson Enterprises Ltd. v Boughton & Co.,  3 S.C.R. 534, not all fiduciary relationships and not all fiduciary obligations are the same; these are shaped by the demands of the situation.
In other words, the fiduciary obligation is not a closed category; it maintains a flexibility to apply in a variety of situations and it is meant to be available if the relationship of trust required to ground the duty is present. Medical decisions in the end-of-life context are unique and challenging, and will give rise to obligations under the fiduciary duty that may not apply to other medical decisions. Given the early state of the jurisprudence in this domain, the exact contours of the obligation remain to be defined and explored in future cases.
Generally, in many typical doctor-patient relationships, the fiduciary obligation and the standard of care will likely overlap or resemble one another. It seems to me, however, that in the end-of-life scenario where ongoing life support is futile, the foundation and ambit of a doctor’s fiduciary duty would be a useful and appropriate conceptual paradigm to supplement the standard of care and address the broader best interests of the patient. In such difficult circumstances, in my view, the ambit and operation of the fiduciary and standard of care duties tend to diverge. As the Chief Justice observed in Norberg: “The foundation and ambit of the fiduciary obligation are conceptually distinct from the foundation and ambit of contract and tort. Sometimes the doctrines may overlap in their application, but that does not destroy their conceptual and functional uniqueness” (p. 272). The fiduciary may ensure that additional processes are undertaken to ensure that the patient’s best interests are respected, while the standard of care requires that the correct medical decisions and operations are undertaken according to medical standards.
(iii) Factors in the Decision-Making Process
As with all medical decisions, health care practitioners must consider the medical effectiveness of the course of action, which involves weighing the course of action’s risks and benefits, as well as its implications for the condition and the well-being of the patient.
The prospect of imminent death, however, elevates the significance of other interests, such as religious beliefs and personal values. Due to the important interests involved in life-sustaining treatment, factors such as maintaining respect for autonomy and human dignity are particularly vital in this balancing process. However, these considerations cannot prevail if a doctor considers the treatment to be outside the standard of care due to its futility or harmful effects.
Artificial continuation of life will not always be in the best interests of the patient. While the sanctity of life is an important principle of our legal system, it is not absolute; it is subject to exceptions where notions of dignity must prevail (Rodriguez v British Columbia (Attorney General),  3 S.C.R. 519, at p. 605, per Sopinka J.). Further, the suggestion that life is an absolute value is contrary to medical and scientific notions of treatment. The Law Reform Commission of Canada noted in “Euthanasia, Aiding Suicide and Cessation of Treatment”, Working Paper 28 (1982), that [p. 59]
|the guiding principle for medical decision-making is not life in itself as an absolute value, but the patient’s overall welfare. In most instances, this welfare imposes the maintenance of life, but this is not always the case. It is not the case when the prolonging of life has become purely artificial. It is not the case when the maintenance of life can only be achieved by an undue prolongation of the patient’s agony. It is not the case when the maintenance of life results only in the infliction of additional suffering. In other words, it is not the case when treatment is diverted from its proper end and merely prolongs the dying process rather than life itself.|
A physician’s duties in the end-of-life context are informed by policy statements from the governing professional bodies, including in Ontario, the College of Physicians and Surgeons of Ontario (“CPO”) (Policy Statement #1-06, “Decision-making for the End of Life” (July 2006) (online)). These guidelines allow a physician to withhold or withdraw life support against the wishes of a patient or substitute decision-maker under certain circumstances, if the patient will almost certainly not benefit from it, although the issue of benefit must take into account the patient’s values (CPO Policy, at p. 5). Indeed, various policy statements of professional medical organizations adopt the position that physicians are not obliged to provide treatments that will almost certainly not benefit the patient, either because the patient’s condition is such that recovery or improvement is virtually unprecedented or because the patient will be unable to experience any permanent benefit from the treatment (CPO Policy, at pp. 4-5).
The CPO Policy, however, also stipulates that patients have the “right to receive life-sustaining treatments that may be of benefit to them and that take into account their goals, values and beliefs. When it is not clear whether treatment might be of benefit, the choice should be made on the side of providing life-sustaining treatment” (p. 5). The CPO Policy clearly indicates that, in end-of-life decisions, physicians are required to account for the “personal, cultural and religious values, goals, beliefs and practices” of their patients (p. 2).
Moreover, the 1995 report of the Special Senate Committee on Euthanasia and Assisted Suicide states that life-sustaining treatment should not be withheld against patient’s wishes unless it is “futile” (Of Life and Death: Report of the Special Senate Committee on Euthanasia and Assisted Suicide (1995), at p. 45). “Futile treatment”, defined by the report, must be understood as “treatment that in the opinion of the health care team will be completely ineffective” (p. 15). See also E. I. Picard and G. B. Robertson, Legal Liability of Doctors and Hospitals in Canada (4th ed. 2007), at pp. 345-46.
While the common law does not permit personal autonomy to be the overriding consideration for the withdrawal of life support, it has long recognized the role of values, beliefs, and the dignity of human life, including dying with dignity (see Rodriguez, at pp. 585 and 605). This respect is reflected in the medical profession’s and institution’s policy statements. Thus, a doctor must consider these factors in determining the patient’s best interests, in accordance with her professional and fiduciary responsibilities. For example, a patient’s wishes may require that life support be withdrawn. Alternatively, as discussed previously, the wish to continue life support indefinitely may result in deferring a decision pending further discussions with the family, receipt of further medical opinions, or exploration of other available treatment facilities. However, if a doctor is ultimately satisfied that treatment is futile, she may discontinue treatment notwithstanding the wishes of the patient or family, provided they have followed these consultative processes and considered the patient’s best interests.
(iv) End-of-Life Decisions Must Follow a Fair, Inclusive, and Accommodating Process
These broader considerations, reflected in the profession’s policies, are framed by a doctor’s fiduciary obligation – to act in the best interests of the patient. These obligations include broad duties to ensure the well-being of the patient, including the duty to consult the patient (or the patient’s substitute decision-maker) in arriving at a decision regarding what constitutes the patient’s best interests in the circumstances.
In keeping with these duties, the various policy statements illustrate a process of giving notice, of seeking further medical opinions if requested, and of making efforts to transfer care to another institution willing to continue administering treatment. Indeed, in this case, each of these avenues was made available to the respondent’s substitute decision-maker.
Where a family member, or a substitute decision-maker, disagrees with the medical practitioner’s decision to withdraw life support, she may apply to the court to challenge the physician’s decision. That is what the substitute decision-maker did in this case.
In this case, the application judge made no factual findings about the patient’s condition and effectiveness of any treatment, and the patient’s diagnosis has been subject to change. Accordingly, there are factual issues that remain to be determined and it is not within this Court’s ability to do so.
I would allow the appeal and remit the matter to the Superior Court of Justice, rather than the Consent and Capacity Board, so that the application judge can determine whether to issue an injunction or declaration in accordance with these reasons and the facts as found on the application. In light of the public importance of the questions raised in this appeal, the parties should bear their own costs.
 See The Health Care Directives Act, C.C.S.M. c. H27; Health Care (Consent) and Care Facility (Admission) Act, R.S.B.C. 1996, c. 181; Care Consent Act, S.Y. 2003, c. 21, Sch. B; Civil Code of Québec, S.Q. 1991, c. 64, articles 11-25; Consent to Treatment and Health Care Directives Act, R.S.P.E.I. 1988, c. C-17.2.
 At least 11 states, including: Arizona, Idaho, Indiana, Kentucky, Maryland, Minnesota, North Dakota, Oregon, Pennsylvania, South Dakota, and Wisconsin, permit requests for life-sustaining treatments in an advance directive. Other states, however, have enacted statutes that allow for the unilateral withdrawal or withholding of life-sustaining treatment by health care providers when it is deemed medically inappropriate. For example, Virginia’s Health Care Decisions Act, Va. Code Ann., § 54.1-2990 (2013), provides that physicians are not required to provide treatment which, in their opinion, is medically or ethically inappropriate. The Uniform Health-Care Decisions Act, 9 U.L.A. 83 (2011), adopted (at least in part) by Maine, New Mexico, Mississippi, and Delaware, provides that health care providers may refuse to comply with an advance directive that “requires medically ineffective health care or health care contrary to generally accepted health-care standards applicable to the health-care provider or institution” (§ 7(f)). Other states have struck more of a middle ground in their legislation. In Texas, the Tex. Health & Safety Code Ann., § 166.052 (Vernon 2012), creates a process for resolving disagreements between physicians and substitute decision-makers with respect to the withdrawal of life support.
 See also The College of Physicians & Surgeons of Manitoba, Statement No. 1602, “Withholding and Withdrawing Life-Sustaining Treatment” (2007)(online); Canadian Healthcare Association, Canadian Medical Association, Canadian Nurses Association, and Catholic Health Association of Canada, “Joint Statement on Resuscitative Interventions (Update 1995)”(online); Canadian Healthcare Association, Canadian Medical Association, Canadian Nurses Association, and Catholic Health Association of Canada, “Joint Statement on Preventing and Resolving Ethical Conflicts Involving Health Care Providers and Persons Receiving Care” (1999)(online); Manitoba Law Reform Commission, “Withholding or Withdrawing Life Sustaining Medical Treatment”, Report #109 (2003)(online).
Harry Underwood, Andrew McCutcheon and Erica J. Baron (instructed by McCarthy Tétrault, Toronto), for the appellants.
J. Gardner Hodder, Guillermo Schible and Stefan A. De Smit (instructed by Hodder Barristers, Toronto), for the respondent.
all rights reserved