The following is a summary from information in the public domain: Hassan Rasouli underwent surgery for a benign brain tumour in 2010, but a postoperative infection resulted in severe brain damage and minimal consciousness. At the time the Supreme Court of Canada released its decision, he remained at Sunnybrook Health Sciences Centre dependent on life support (e.g., mechanical ventilation, artificial nutrition and hydration). His physicians had concluded that there was no realistic hope for recovery and that ongoing life support was not appropriate. They proposed to withdraw mechanical ventilation and provide palliative care and further advised Mr. Rasouli’s wife and substitute decision-maker, Ms. Salasel, that they would not attempt resuscitation in the event of cardiac arrest. Ms. Salasel disagreed with this plan.
The decisions of the courts
In this case, the substitute decision-maker applied to the Ontario Superior Court of Justice to prevent removal of life support from Hassan Rasouli. The judge found that consent was required to withdraw life support, given the definition of “treatment” in Ontario’s Health Care Consent Act.1 An injunction was not needed, and the matter could be referred to the province’s Consent and Capacity Board established under the Health Care Consent Act.2 The physicians appealed to the Ontario Court of Appeal, which decided that in cases where palliative care will be offered after withdrawal of life support and death is imminent, the palliative care treatment and withdrawal are part of a treatment plan, and consent is required.3 However, the Court noted that this decision did not mean that consent is otherwise required for the withdrawal or withholding of treatment or that a right is created to demand treatment of no medical benefit. The imminence of death was a key distinction.
The physicians appealed to the Supreme Court of Canada, arguing that consent is not required for the withholding or withdrawal of treatments outside the standard of care, whether or not other treatments are administered; that imminence of death should not determine the need for consent; and that requiring physicians to provide nonbeneficial treatment that may cause harm places them in breach of their legal and professional duties.4,5
A majority of the Supreme Court dismissed the appeal in a 5–2 decision (Cuthbertson v. Rasouli, hereinafter referred to as the Supreme Court’s decision) focused on a statutory interpretation of the Health Care Consent Act. In the Act, “treatment” is defined as “… anything done for a therapeutic, preventive … or other health-related purpose … includ[ing] a plan of treatment. … ” “Plan of treatment” is “the administration … of various treatments … and may, in addition, provide for the withholding or withdrawal of treatment. …” Consent is required for the “administration” of treatment.
The Supreme Court found that in the case of Hassan Rasouli, withdrawal of life support fell within the definition of “treatment” in the Health Care Consent Act and therefore that consent was required. The Supreme Court also found that the definition of “treatment” included treatments outside the standard of care and that withdrawal of treatment “may sometimes, although not always, constitute treatment.” The Supreme Court rejected the Ontario Court of Appeal’s finding that consent was required only if withdrawal of life support was part of a plan of treatment but did not specify when withdrawal would require consent (other than in cases like that of Mr. Rasouli). It noted that withdrawal of life support at the end of life may involve physical contact and is tied to administering palliative care, both of which normally require consent. It rejected the physicians’ argument regarding imminence of death, stating that “[b]y removing medical services that are keeping a patient alive, [withdrawal of life support] impacts patient autonomy in the most fundamental way.” It suggested that the physicians could apply to the Consent and Capacity Board.
Simply put, the Supreme Court ruled that the definition of treatment in the Health Care Consent Act is sufficiently broad that the need for consent cannot be determined by medical benefit or futility. Court decisions sometimes establish a test that can be applied to other cases, but the Supreme Court did not do so here. However, its reasoning suggests that the following factors may indicate the need for consent before withdrawal of treatment:
-
administration of other treatment
-
need for physical contact (which might constitute battery)
-
likelihood of death shortly after withdrawal
Implications for practice
If the Supreme Court decision in this case applies only to situations of intractable disagreement about continuing life support in similar cases in Ontario, then it will affect few patients. Physicians, patients and substitute decision-makers usually agree about care decisions. Intractable conflicts about withdrawal of life support are rare, as are applications to the Consent and Capacity Board.6
Yet futile care may be common,7,8 and the number of referrals to the Consent and Capacity Board for withdrawal of life support has increased recently.9,10 It is beyond the scope of this article to discuss the definition of “futile or nonbeneficial care,” but others have done so.8 We focus on the implications of the Supreme Court’s decision for cases in which the medical team feels that ongoing life support would be nonbeneficial, however the team arrived at that opinion. The Supreme Court’s decision may trigger more referrals to the Consent and Capacity Board or may increase the incidence of nonbeneficial care, if it is interpreted to grant a right to demand treatment and physicians accede to such demands to avoid the legal system. Here, we explore the potential consequences of the Supreme Court’s decision by addressing some practical questions.
When do physicians require consent for withdrawal of life support?
Since the Supreme Court’s decision, consent is required in Ontario for withdrawal of mechanical ventilation if death is anticipated as a result and palliative care will be provided. Beyond this, the decision “does not stand for the proposition that consent is required under the [Health Care Consent Act] for withdrawals of other medical services or in other medical contexts.”5 The Supreme Court suggested that the need for consent to withdrawal of life support was in part derived from administration of additional therapy, need for physical contact and the anticipation of death shortly after withdrawal of life support. Such factors imply that consent may not be required to withdraw a different life-sustaining therapy, such as intermittent hemodialysis, so long as there is no additional therapy or physical contact proposed, and the patient is not expected to die shortly thereafter.
Do physicians need consent to stop cardiopulmonary resuscitation?
On the basis of the Supreme Court’s decision, some have asked whether consent may be required to stop cardiopulmonary resuscitation (CPR). After all, an individual could express a prior wish regarding duration or conduct of CPR, or a substitute decision-maker at the bedside could insist that resuscitation be continued for longer than is deemed appropriate. In addition, cessation of failed CPR is typically followed by an immediate declaration of death.
We suggest that the Supreme Court’s decision does not apply to cessation of CPR. First, this procedure is performed on an imminently dying (or dead) patient, which differs from the patient in stable condition considered by the Supreme Court. Second, although several treatments may be provided during CPR (e.g., chest compressions, medications), cessation of failed CPR is not usually followed by other treatment. Third, cessation of CPR does not involve new physical contact — quite the opposite. Ultimately, the decision to stop CPR is a clinical one and must be in keeping with the standard of care (i.e., that of a reasonable medical practitioner considering all the circumstances). As with other important clinical decisions, the rationale should be clear and documented.
Is consent required to withhold life support that physicians believe to be nonbeneficial?
The Supreme Court of Canada clearly indicated that it was not resolving the issue of “who, in the absence of a statute, should have the ultimate say in whether to withhold or withdraw life-sustaining treatment.” Although the Supreme Court specified that its decision applied to withdrawal of life support, it raised together the issues of withholding and withdrawal of life support, which suggests that there may be instances in which consent is required for withholding of life support or life-sustaining therapy.
Based on the reasoning of the Supreme Court, many withholding decisions may not require consent, such as withholding chemotherapy (where there is no additional treatment, physical contact or proximate death). In contrast, deactivation of an implantable defibrillator would involve physical contact and therefore would require consent even if defibrillation is not expected to benefit the patient.
CPR is a special consideration because it is a form of life-sustaining therapy that can be initiated by nonphysicians, and it is familiar to the public. It is also provided by default in many settings; in the absence of an individual’s medical history, any given person (or patient) may expect to undergo CPR in the event of cardiac arrest.
This does not mean that CPR should be provided to all patients. Where physicians have adequately assessed a patient and determined that such resuscitation would not be beneficial, they should not offer it.
On the basis of the Supreme Court’s decision, we would argue that physicians should not require consent to withhold nonbeneficial CPR. There is no contact or additional treatment required, and death would not be expected as an immediate result of writing a do-not-resuscitate order. This would be true regardless of whether the existing plan of care (i.e., to receive CPR) was established by default or as the result of a previous discussion with the patient or substitute decision-maker. Circumstances may change during the course of an admission, and physicians need to be able to make appropriate medical decisions rather than follow a prior treatment plan that has become obsolete. A similar argument applies to withholding nonbeneficial life-sustaining therapy (e.g., mechanical ventilation, inotropic support), although this would be a separate consideration given that some patients could benefit from mechanical ventilation even if they would not benefit from CPR. However, the reverse is not true: a patient who would not benefit from life-sustaining therapy would also not benefit from CPR. Thus, if an intensive care unit (ICU) consultant determines that a patient would not benefit from life-sustaining therapy, then the admitting physician should write a do-not-resuscitate order.
Even if these decisions do not require consent, they should always be communicated to the patient or substitute decision-maker whenever reasonably possible.
Does the Supreme Court’s decision affect “time-limited trials” of life support?
A time-limited trial of life support may help to determine reversibility in an instance of acute deterioration, without any commitment to protracted life support. For some, the Supreme Court’s decision in Mr. Rasouli’s case may discourage trials in borderline situations, on the basis that consent to withdrawal of life support may subsequently be withheld in the event that the trial fails, resulting in open-ended, nonbeneficial, potentially harmful life support. We suggest that trials should continue to be offered when appropriate, and that physicians document the rationale and discussions leading up to the consent. In the event of disagreement following trial failure, such documentation may help in subsequent dialogue or may assist the Consent and Capacity Board and others if a decision of the substitute decision-maker is reviewed.
When should physicians ask the Consent and Capacity Board to review a substitute decision-maker’s decision?
In Ontario, a substitute decision-maker must comply with a prior, applicable and capable wish (i.e., a request made by a competent individual 16 years of age or older that is applicable to the circumstances).1 In the absence of a prior, applicable and capable wish, the decision must be made in the best interests of the patient, determined on the basis of criteria in the Health Care Consent Act (e.g., the patient’s values, beliefs and wishes; factors related to the proposed treatment; alternatives; and associated benefits and harms).
If refusal of consent is based on a prior, applicable and capable wish, physicians may ask the Consent and Capacity Board to determine whether the wish is actually applicable to the circumstances, or the Board may grant the substitute decision-maker permission to depart from the wish in certain situations.1 However, if the prior, applicable and capable wish is truly applicable, and it is supported by the substitute decision-maker, the Consent and Capacity Board cannot override it. If refusal is based on a substitute decision-maker’s assessment of best interests, physicians can apply to the Consent and Capacity Board to review the assessment of best interests. In either case, medical opinion will be important to the Board. It has ruled in favour of and against withdrawal of life support, and either party can appeal decisions of the Consent and Capacity Board to the Ontario Superior Court of Justice.9,10
If a physician feels that an inappropriate decision was made by a substitute decision-maker, the Consent and Capacity Board or a court, he or she has a professional obligation to have that decision reviewed. The Consent and Capacity Board has no jurisdiction to hear cases involving competent patients; disagreement in such circumstances can be referred to the courts. For Canadian jurisdictions without an administrative tribunal like the Consent and Capacity Board (all except Ontario and the Yukon), physicians should apply to the courts if a decision was, in their opinion, inappropriately made, given the law of their province or territory.
What are the elements of valid consent?
Consent is required before administering treatment (except in some circumstances, such as an emergency). Consent is a process, not an event, and it can be withdrawn. The Health Care Consent Act requires that it be provided by a capable patient or an authorized substitute decision-maker. The decision-maker must be informed about what a reasonable person would need to know in the circumstances, as well as other information requested. This includes information about the benefits, material risks and alternative options, as well as the likely consequences of not receiving the treatment. In the context of proposed withdrawal of life support, it must be clear that continuation of life-sustaining therapy is an option. Finally, consent must be voluntary and not result from misrepresentation, fraud, undue influence or coercion.
Consent for withdrawal of life support will include an explanation of the process involved, the likelihood of death (i.e., anticipated time-frame and circumstances) and the strategies to manage symptoms. The means of imparting this information will depend on those involved in the discussion and the context; the terminology may be qualitative or more explicit. However, all relevant aspects must be understood by the substitute decision-maker.
How explicit does consent need to be?
Valid consent may be explicit (e.g., a signed consent form or saying words such as “I agree”) or implied (e.g., conveyed via body language), according to the Health Care Consent Act and the common law.11 There should be good evidence of consent and the communications that led to it, in case consent is subsequently challenged, but a consent form is neither necessary nor adequate. Physicians should focus on the process of consent. Forcing individuals to sign a consent form or give formal explicit verbal consent may cause psychological trauma and potentially increase the incidence of posttraumatic stress disorder.12 Implied (but clear) consent may be preferable.
Is the Supreme Court’s decision relevant across Canada?
The Supreme Court of Canada went to great lengths to restrict the impact of its decision. It made it clear that its decision was a statutory interpretation of the Health Care Consent Act as a whole, and not the common law or any other legislation. Although the Supreme Court focused on some key definitions in the Health Care Consent Act (and these definitions are similar to those found in legislation in some other provinces), it considered many other sections in reaching its interpretation of that statute. The only other Canadian jurisdiction with legislation substantially similar is the Yukon Territory, which also establishes a tribunal like the Consent and Capacity Board. In other provinces and territories, disputes are resolved in the courts, and even those with similar definitions have other key differences in their legislation. Other jurisdictions should be cautious about applying the rationale used in this case.
Comment
The Supreme Court of Canada did not discuss the societal impact of using limited ICU resources for patients with no reasonable prospect for recovery. Such cases are rare,13 but may become more common given Canada’s growing and aging population, high ICU occupancy and increasing ability to prolong life. Futile care may therefore result in increased ICU costs and decreased ICU availability.
A summary of suggestions arising from these considerations is presented in Box 1. Other issues relevant to this case are discussed in Appendix 1 (available at www.cmaj.ca/lookup/suppl/doi:10.1503/cmaj.140054/-/DC1).
Suggestions
-
Whether or not consent is required in any scenario, physicians should communicate and document their proposed treatment and its rationale whenever reasonably possible.
-
Physicians in Ontario must obtain consent for withdrawal of life support in any case resembling that of Hassan Rasouli.
-
Consent for withdrawal of life support need not be explicit, but it must be clear and should be documented.
-
Intractable disagreements about withdrawal of life support should be resolved according to the appropriate provincial or territorial legal framework.
-
Health care organizations should have written policies concerning consent to treatment (and, if applicable, its withdrawal or withholding) and conflict resolution, consistent with medical standards of care and the law, and consistent within a jurisdiction.
-
Physicians should offer time-limited trials of life support where appropriate. If consent to withdraw life support is withheld after a failed trial, the rationale for the trial, as well as all relevant factors, should be reviewed and agreement sought.
-
Physicians should apply for a review of any decision by a substitute decision-maker that they believe was made on an improper basis, where reasonable attempts to reach agreement have failed.
-
Legislation and legal processes to resolve disputes about treatment should be reviewed to ensure that they protect the interests of patients, that they are effective and timely, and that they consider issues of social justice.
-
Provinces considering legislative changes relating to consent should make clear:
-
whether consent is required for withdrawing or withholding a treatment of no medical benefit; and
-
whether rules relating to consent should be applied consistently, or if there are criteria for determining the need for consent.
-
-
Physicians should seek legal advice in any case of intractable disagreement.
-
In Ontario, physicians must obtain consent to withdraw life support in cases like that of Hassan Rasouli, even where such treatment is thought to be nonbeneficial.
-
In Ontario, consent may or may not be required to withdraw other treatment or to withhold treatment.
-
If the decision is applied in other cases in Ontario, whether consent is required may be based on the need for immediate additional treatment, physical contact with the patient and anticipation of death shortly after a withdrawal of life-sustaining treatment.
-
Considering the rationale of the Supreme Court’s decision in the Rasouli case, as well as other legal and ethical considerations, the authors argue that consent is not required to withhold nonbeneficial cardiopulmonary resuscitation, but appropriate processes, including those related to communication, must be followed.
Footnotes
-
Competing interests: James Downar is a member of the Canadian Critical Care Society, which was an intervenor in the Supreme Court case discussed herein, and is co-chair of the Advisory Council of Physicians for Dying with Dignity, an advocacy group for choice at the end of life. No other competing interests declared.
-
This article has been peer reviewed.
-
Contributors: All authors participated in the conception, writing and revision of the article. All authors gave final approval of the version to be submitted and agree to act as guarantors of the work.
References
- Health Care Consent Act, S.O. 1996, c. 2, Sched. A.