By Ramona Coelho, David Shannon, and Trudo Lemmens, December 1, 2025

Capacity is essential for making informed decisions in all aspects of a person’s life. It’s especially critical when decisions are serious and irreversible.

In the medical context, capacity means the ability to make one’s own, authentic health care decisions. Patients must understand their treatment options, appreciate the consequences of accepting or refusing specific treatments, and make voluntary, well-informed choices. If patients “choose” treatment but lack the capacity to do so, they may be exposed to serious harm, unwanted interventions, or decisions made on their behalf without real understanding.  Evidence shows that Canada’s euthanasia and assisted suicide regime, called medical assistance in dying (MAiD), often fails to uphold these requirements. Yet, MAiD happens in situations that – by their very nature – present the kinds of vulnerabilities that demand heightened scrutiny.

Canada’s MAiD regime is framed as resting on two pillars: physicians’ duty to diminish suffering and individuals’ autonomy to make health-care decisions. Although the first pillar emphasizes relief of suffering, the framework permits MAiD even when reasonable interventions remain available. In fact, scholars have noted that Canadian MAiD policy and practice often push relief of suffering to the background, placing even greater weight on autonomy. Moreover, some have also warned about relying too much on capacity assessments as a safeguard, considering their complex nature. For these reasons, we need even more rigorous scrutiny of standards and practices related to capacity and consent.

However, evidence from media investigations, parliamentary testimony, academic research, and government reviews reveals inconsistencies in how these safeguards are enforced. Clinicians and health policy experts have also warned that MAiD has proceeded for some patients even when capacity assessments did not meet professional standards or when qualified treating teams determined an individual lacked capacity. Additionally, concerns exist regarding failures to identify factors that undermine free and informed consent.

As members of Ontario’s MAiD Death Review Committee (MDRC), we review cases selected by the Office of the Chief Coroner to identify systemic risks and improve public safety. The Chief Coroner publicly releases reports based on this review. The reports highlight practical challenges in ensuring valid consent and capacity. It is important to note these concerns are only one part of a much wider pattern of safeguard failures – issues the authors have addressed in other writings.

Here, we share our personal views regarding some of the case narratives in the MDRC reports, and the issues they highlight related to capacity and consent. We maintain our commitment to confidentiality, making reference exclusively to publicly available documents.

Case studies show some assessments lacked rigour

In the MDRC Same-Day or Next-Day MAiD Report, Mr. C, a man with metastatic cancer, initially expressed interest in MAiD but experienced cognitive decline and delirium before being assessed. He was sedated for pain management. A MAiD practitioner, who had been contacted following the patient’s initial expression of interest, was told by the palliative care team that they deemed him incapable of making health care decisions. The MAiD practitioner “vigorously rouse[d]” the patient and asked him if he wanted MAiD and took the “mouthing of the word ‘yes’” as a positive response. He subsequently withheld sedation and found him “more alert” after 45 minutes. A second, virtual assessment followed, which the first MAiD practitioner coordinated. Consent was confirmed via mouthing yes and squeezing of the practitioner’s hand. MAiD was administered that same day.

Another report focused on dementia and described Mr. 6D, who had Alzheimer’s and delirium, receiving MAiD during an acute illness while facing imminent long-term care placement due to the loss of his caregiver. In a separate case, Mrs. 6F, with advanced dementia and aphasia, was referred twice for MAiD by her family, allegedly because she expressed a desire to die. A single eligibility assessment was conducted by the MAiD provider with limited documentation of her cognitive state.

These individuals were not afforded genuine opportunities to confirm their wishes. Moreover, it appears MAiD providers did not conduct rigorous capacity assessments before ending lives, raising serious questions about whether patients fully understood and appreciated the MAiD procedure, and whether their choices were truly voluntary.

As early as 2020, Ontario’s Chief Coroner identified a category of concern in MAiD practice related to issues of capacity and consent. There were cases where MAiD was provided despite “incompatible or contradictory conclusions of capacity by MAiD assessors in comparison to other documented clinical assessments in medical records.” In cases where patients had a “known history of dementia or cognitive impairment,” there was a “paucity of formal capacity assessments or further specialist consultation.” And there was “variability in quality of assessments in cases of wavering capacity or evidence of impaired cognition.” Dr. Leonie Herx – a former president of the Canadian Society of Palliative Medicine who was practising in the province of Ontario at that time – also testified in Canada’s Parliament that MAiD was often carried out in the absence of capacity.

Compared to the rigorous standards used for financial capacity assessments, where designated experts conduct detailed interviews and review collateral information, the minimal scrutiny applied to MAiD capacity assessments is alarming, particularly because the consequence is more serious than losing control of financial decisions – it is literally death.

Capacity to consent: what is the current standard?

Under section 4(1) of Ontario’s Health Care Consent Act, a person is capable of making decisions with respect to treatment if they can understand the information relevant to a treatment decision and appreciate the reasonably foreseeable consequences of making that decision. Although capacity is presumed, it must be assessed when reasonable doubt arises.

The Supreme Court of Canada emphasizes that capacity depends on ability. It is not established by the person in question agreeing with a clinician’s opinion, nor does it stem from having a certain diagnostic label. A person must be assessed functionally: understanding, appreciation, reasoning, and expression of choice.

Ontario’s Consent and Capacity Board applies the same standard. It relies on evidence that a person lacks the functional abilities defined in section 4(1). Clinicians must remain vigilant for signs of impaired capacity, such as cognitive or psychiatric symptoms, fluctuating cognition, or unusual decision-making patterns. These triggers should prompt a formal capacity assessment, especially in high-stakes decisions, such as MAiD. The College of Physicians and Surgeons of Ontario explicitly notes that capacity can vary between treatments and across time.

Clinicians structure evaluations around four functional abilities: the capacity to express a choice, to understand, to appreciate, and to reason. For consent to be valid, it must relate to the treatment, be informed and voluntary, and not obtained through misrepresentation or fraud. “Informed” means the patient must receive and understand information about the nature of the treatment, its expected benefits, material risks, side effects, reasonable alternatives, and the likely consequences of not proceeding. The practitioner must answer all questions.

Clinicians should document the patient’s understanding by having them explain, in their own words, the nature and consequences of MAiD and its alternatives. The patient must apply this information to their own situation, reason consistently, and show stability over time. Because capacity can fluctuate, reassessment is required whenever there is a change in mental status, particularly in patients with cognitive disorders or delirium, where decision stability is uncertain. A preliminary request to have a formal MAiD eligibility assessment cannot be considered sufficient to presume capacity.

Mental disorders can significantly interfere with an individual’s capacity for appreciation and reasoning, which are critical for making informed decisions about MAiD. Symptoms such as delusions, depressive nihilism, anosognosia, mania, or impulsivity can distort understanding, judgment, and risk assessment in ways that compromise decision-making. For example, delusional beliefs can prevent recognition of the true consequences of MAiD, leaving the person unable to fully grasp the permanence or implications of the choice. Severe depression, particularly when accompanied by nihilistic or hopeless thought patterns, can erode appreciation of alternative options, creating a sense that no other course of action is viable. Anosognosia, which frequently occurs in psychotic disorders or dementia, can prevent recognition of one’s own illness or impairments, undermining the ability to make informed decisions based on an accurate understanding of one’s condition. Mania or impulsivity distort risk assessment, leading to hasty decisions that do not reflect considered judgment. In all these situations, it is essential that evidence demonstrates, in functional terms, how the mental state concretely impairs the individual’s understanding and appreciation of their choices. The focus must be on the actual impact of the disorder on decision-making capacity in the specific context of MAiD.

The same functional approach applies to voluntariness and social distress. Clinicians must actively explore external pressures and be alert to family or institutional influence, as these factors can influence an individual’s ability to appreciate the consequences of accepting or rejecting a proposed form of treatment. Some impact of contextual factors and family or institutional dynamics is unavoidable, but clinicians must be sensitive to how these may constitute undue influence or coercion. While social factors such as poverty, housing insecurity, or lack of support do not directly alter the legal test for capacity under Ontario law, they deeply affect informed consent and voluntariness. A patient’s decision may be distorted by feelings of being a burden, lack of access to care, or desperation from social deprivation. Annual Health Canada reports document how a significant number of Canadians who receive MAiD consider these factors to be core components of their unbearable suffering. Law and professional guidance, therefore, require exploration of supports and alternatives, especially given the irreversible consequences of an individual decision to accept MAiD.

An example of undue influence or coercion arose in the case of Mrs. B, in her 80s, who had explicitly stated during a first MAiD assessment that she preferred palliative and hospice care because of personal and religious views. Yet, when a request for immediate hospice care was rejected, her husband, who had been mentioned in the clinical file as struggling with caregiver burnout, contacted the MAiD coordination service again. The patient was assessed, approved, and received MAiD the same day that her request for hospice care was rejected.

In our view, the life-ending nature of MAiD makes the rigour of capacity assessments especially critical. There is precedent in Canadian jurisprudence that recognizes the value of a ‘sliding scale’ approach to capacity assessment, where higher-stakes decisions require more thorough evaluation. In AC v. Manitoba (2009 SCC), the Supreme Court emphasized that the more serious and irreversible a decision, the more searching the capacity inquiry must be, in the context of medical decisions made by mature minors. This “sliding-scale” approach provides a compelling parallel for MAiD, where the decision carries an ultimate and irreversible consequence.

Federal MAiD safeguards that impose additional requirements for Track 2 MAiD reflect a recognition that in some circumstances, more rigorous screening of decisional capacity and informed consent is required. Track 1 (where natural death is reasonably foreseeable) already requires discussion of alternatives and careful examination of voluntariness. Under Track 2 (where natural death is not reasonably foreseeable), the law explicitly requires that assessors ensure the person has been informed of counselling, mental health, disability, and community supports, as well as palliative care, and must offer consultations with those providers. A mandatory 90-day assessment period is intended to allow time to explore supports and ensure MAiD remains consistent with the individual’s values only after other reasonable means to alleviate suffering have been considered. It’s important to note this is the minimum time required by legislation, but it often takes much longer to hold consultations and establish care to meaningfully explore and remediate suffering.

MAiD clinicians must remain vigilant about their own biases when assessing capacity. Unlike medical decisions aimed at preserving life, MAiD culminates in certain and irreversible death. In this context, biases like ableism and ageism can dangerously distort clinical judgment via confirmation bias. This may lead clinicians to overestimate capacity when a patient’s wish to die aligns with their own assumptions about suffering, disability, or quality of life. When clinicians view certain lives as less worth living, they may unconsciously interpret suicidality or despair as rational rather than symptomatic, overlooking potentially remediable causes of hopelessness such as untreated pain, social isolation, or loss of supports.

Frameworks developed for high-risk medical procedures cannot be applied uncritically to euthanasia, where definitions of harm, benefit, and autonomy are deeply contested. Awareness of these contextual and psychological influences – including how discriminatory beliefs can create a bias toward affirming capacity – is essential to ensure that MAiD assessments are conducted with greater rigour and caution, minimizing harm as much as possible.

Improving capacity safeguards

These findings point to the need for improvements to MAiD capacity and consent standards, alongside robust mechanisms that ensure compliance in practice.

We are not arguing that Ontario’s general capacity and consent framework under its Health Care Consent Act is inadequate. We’re making the case that MAiD requires the most rigorous safeguards that operationalize a higher and more consistent standard of assessment.

In practical terms, this would involve:

Mandatory formal capacity assessments by specialists with formally recognized expertise when cognitive impairment, psychiatric symptoms, or fluctuating capacity are present;
Clear requirements to document the patient’s functional understanding, appreciation, and reasoning in their own words;
Mandatory collateral information, and review of recent clinical assessments when capacity is in doubt;
An explicit prohibition on proceeding with MAiD when treating teams have documented patient incapacity without further specialist evaluation;
Acceptance and utilization of relevant and credible collateral information that points to factors that may indicate compromised capacity to consent to MAiD.

To ensure that safeguards are consistently upheld, systemic measures are required.

These include:

Independent review by an oversight board with expertise in consent and capacity – as it applies to MAiD – prior to proceeding with MAiD;
Strengthened oversight and accountability mechanisms for clinicians who fail to meet professional standards for MAiD.

Implementing these measures would better operationalize the requirements of consent and capacity, rather than leaving enforcement to individual clinical discretion, which has proven problematic.

Conclusion

In the end, law, ethics, and medicine converge on the same imperative: capacity assessment in MAiD must be conducted with the highest possible care. The consequences of error are irreversible. This creates an imperative that safeguards protecting patients’ autonomy and lives must not only exist in law but also be meaningfully upheld in clinical practice. Achieving this requires consistent oversight, transparent reporting, excellence in standard setting, and accountability mechanisms strong enough to ensure that the principles of consent and capacity are not merely theoretical but meaningfully enforced.

Ramona Coelho, MDCM, CCFP, is a senior fellow at the Macdonald-Laurier Institute, an adjunct research professor of family medicine at the University of Western Ontario’s Schulich School of Medicine and Dentistry, and the co-editor of Unravelling MAiD in Canada: Euthanasia and Assisted Suicide as Medical Care.

David Shannon, CM, OC, OOnt, LLM, is a Canadian lawyer practicing health law, and a disability/human rights activist. He has advised governments across Canada and is the recipient of both the Order of Canada and Order of Ontario for his contributions to human rights. He is the co-editor of Medical Assistance in Dying (MAiD) in Canada: Key Multidisciplinary Perspectives.

Trudo Lemmens, LicJur, LLM (Bioethics), DCL, FCAHS, is a professor and Scholl Chair in Health Law and Policy at the University of Toronto’s Faculty of Law and Dalla Lana School of Public Health. He is a leading scholar in health law, bioethics, and public policy, with extensive research on medical regulation, assisted dying, and pharmaceutical governance. He is co-editor of Unravelling MAiD in Canada: Euthanasia and Assisted Suicide as Medical Care.

Disclosure: All authors are members of the Chief Coroner of Ontario’s MAiD Death Review Committee, and publish widely on the topic of MAiD.