Ontario Court of Appeal considers whether consent is needed to remove patient from life support

Auteur: Rahool P. Agarwal Publication | Juillet 2011

Pour l'instant, cette publication n'est disponible qu'en anglais

End-of-life decisions are not only emotionally overwhelming for the family members involved, but also raise a host of complicated legal issues. This complexity is illustrated by the recent Court of Appeal decision in Rasouli v Sunnybrook Health Sciences Centre,1 which addresses whether a physician must obtain the consent of a substitute decision-maker to withdraw life-sustaining measures from a patient where there is no chance of medical recovery.


Mr. Rasouli is a patient at Sunnybrook Hospital in Toronto. Following surgery for the removal of a brain tumour, he developed serious complications and was placed on mechanical life support.

His doctors concluded that Mr. Rasouli had no chance of recovery and ongoing mechanical intervention would provide no medical benefit. They recommended to Mr. Rasouli’s wife, who was also his substitute decision-maker, that life support be discontinued and he be placed on palliative care until his death. Mr. Rasouli’s wife, however, did not agree with this recommendation and applied to court for an order prohibiting Mr. Rasouli’s doctors from removing his life support without her consent. The doctors and the hospital, in response, applied for a declaration that they did not need the consent of Ms. Rasouli to withdraw her husband’s life support and to proceed as recommended.

The central legal question for the court was whether the withdrawal of life support constitutes “treatment,” and therefore requires consent. Under the Health Care Consent Act, 19962 (the Act), all “treatment” requires the patient’s consent or, if the patient is incapable, the consent of a substitute decision-maker. If there is a dispute between the attending physician and the substitute decision-maker, the Act provides that the physician may apply to the Consent and Capacity Board for a determination as to whether the substitute decision-maker’s decision is in the best interests of the patient.

The application judge held that the physicians’ proposal did constitute treatment and ordered that the dispute between the doctors and Mr. Rasouli’s wife be referred to the Consent and Capacity Board.3 The doctors appealed this decision to the Court of Appeal.

The Court of Appeal decision

At the Court of Appeal, the doctors argued that withdrawing treatment that is of no medical value does not constitute “treatment” and therefore does not require consent. To hold otherwise would have serious implications for the doctor-patient relationship, as patients would have the right to insist on treatment their doctors believe to be medically inappropriate or ineffective.

The Court of Appeal recognized the merit of this argument, saying it had “difficulty accepting that the legislature intended to include within the definition of treatment measures that attending physicians consider to be of no medical value and therefore worthless.” Despite this, the court ultimately decided the appeal without determining whether withdrawing medically futile treatment requires consent.

The court found that the doctors’ “treatment package” consisted of two parts: (i) removing life support and (ii) placing the patient on palliative care. In the court’s view, the two parts were integrally linked and could not be separated: “The removal of the ventilator is a necessary precondition to the administration of palliative care and palliative care is a necessary response to removal of the ventilator.” Palliative care is expressly included within the definition of “treatment” in the Act and thus requires consent. Accordingly, where palliative care is recommended as an adjunct to the removal of a ventilator, consent is required for the entire “treatment package.”  

On the basis of this approach, the court held that Mr. Rasouli’s doctors could not implement their proposal without the consent of Mr. Rasouli’s wife. Accordingly, the decision of the application judge was affirmed.


There are several issues raised by the decision that warrant further discussion.

First, the Court of Appeal’s “treatment package” approach – that is, consent is required whenever the termination of existing life support measures is to be immediately followed by palliative care – does not address the doctors’ concern about the physician-patient relationship. Palliative care is frequently the logical next step after removing life support. From a physician’s perspective, these are the situations where the ability to exercise his or her professional discretion without interference will likely be most critical.

Second, the decision leaves physicians with little practical guidance. The court apparently supports the notion that removing medically futile treatment does not require consent. They point to the example of withdrawing ineffective chemotherapy from a terminally ill cancer patient as an instance where no consent would be required, because the “treatment package” does not include immediate palliative care. But the court does not provide any further analysis on this issue. As a result, physicians may be left guessing as to whether their treatment proposal is closer to the chemotherapy example, where consent would not be required, or the situation in Rasouli.

Third, absent from the court’s decision is any acknowledgment that applying to the Consent and Capacity Board is not always desirable. End-of-life situations are emotionally charged and highly stressful. Engaging the adversarial board process could create or exacerbate tensions between the family and the doctor. A physician may reasonably choose not to trigger the board process simply to avoid a difficult and uncomfortable situation for the physician, the hospital staff and the patient’s family. Accordingly, while the board process may provide a fair and reasonable solution in theory, in practice it may not be a preferable option.

Fourth, because the court viewed the Act as providing a complete answer to the parties’ dispute, it did not review the Canadian common law regarding consent and the withdrawal of life-sustaining treatment. The common law is important because, in any case where treatment proposals fall outside of the Act’s definition of “treatment,” the common law would govern. Consent issues moreover arise across Canada, and the common law is a patchwork of differing approaches from Canada’s various jurisdictions.

Finally, the Court of Appeal did not address the respondent’s argument that doctors’ treatment decisions could be subject to review under the  Canadian Charter of Rights and Freedoms. The lower court considered the issue and concluded that the Charter did not apply, but its analysis was limited. Perhaps most glaring was its failure to fully consider the growing body of Supreme Court jurisprudence that supports the Charter’s application to health care delivery.4 As a result, whether the Charter applies to the actions and decisions of doctors has not yet been adequately analyzed. The issue would certainly benefit from further judicial consideration, whether in this case at the Supreme Court of Canada or in another similar case.

The author wishes to thank Christopher Guerreiro, summer student, for his help in preparing this legal update.


1 Rasouli v Sunnybrook Health Sciences Centre, 2011 ONCA 482.

2 Health Care Consent Act, 1996, SO 1996, c 2, Sch A, ss 1, 2, 10, 21.

3 Rasouli (Litigation guardian of) v Sunnybrook Health Sciences Centre, 2011 ONSC 1500.

4 See e.g., Eldridge v British Columbia (Attorney General), [1997] 3 SCR 624; Chaoulli v Quebec (Attorney General), [2005] 1 SCR 791.