• When assisted suicide (MAiD) was legalized in Canada, courts and legislators expected that safeguards could be put in place to protect vulnerable groups, such as persons with physical or mental disabilities, from heightened risk or inordinate impact.
• This report examines whether these expectations have been borne out in reality, using the available national and provincial data and peer-reviewed medical studies.
• This report concludes that Canada’s legalization of assisted suicide has led to an intensified risk of premature death for vulnerable groups, and that the expected safeguards have failed to materialize.
• From 2019 to 2023, at least 42 percent of all MAiD deaths were of persons who required disability services, including over 1,017 persons who required but did not receive these services. MAiD providers in Canada have euthanized disabled persons who needed disability supports and were unable to access them.
• A study of MAiD deaths between 2016 and 2019 at a tertiary care centre in Toronto found high rates of psychiatric comorbidity among patients requesting MAiD; most of these requests were granted. Of 155 patients requesting, sixty (39 percent) had a documented psychiatric comorbidity, most commonly depression.
• According to MAiD providers in 2023, the suffering of almost half of patients to whom they provided an assisted suicide included the perception of being a burden to others, 10 percent more than in the previous year.
• In 2023, isolation and loneliness were reported as a source of suffering by 22 percent of MAiD recipients, up 5 percent from the previous year.
• Degenerative neurological illnesses pose significant risks for patient coercion toward MAiD. In 2023, almost 15 percent of persons who died by MAiD had a neurological condition as a qualifying illness.
• In 2023, 241 persons with dementia received a MAiD death; in 106 of these cases, dementia was the sole underlying condition.

In 2012, Justice Lynn Smith of the British Columbia Supreme Court concluded in Carter v Canada (Attorney General) that an appropriately safeguarded physician-assisted dying program could be adopted in Canada without creating a “heightened risk” or an “inordinate” impact on vulnerable groups, such as persons with disabilities. 1 1 Carter v Canada (Attorney General), 2012 BCSC 886 at paras 9 and 626. On appeal, in 2015, the Supreme Court of Canada affirmed Justice Smith’s reasoning, based on the trial judge’s review of evidence from international jurisdictions with assisted dying programs. 2 2 Carter v Canada (Attorney General), 2015 SCC 5 at para 3. The Supreme Court upheld the trial judge’s ruling, which struck down the prohibitions against assisted suicide and euthanasia in the Criminal Code, 3 3 Criminal Code, RSC 1985, c C-46, sections 14, 241(b), https://laws-lois.justice.gc.ca/PDF/C-46.pdf. resulting in the decriminalization of assisted dying in Canada. 4 4 Carter v Canada (Attorney General), 2015.

Thus the decriminalization and subsequent legalization of MAiD in Canada was based on some specific expectations established by the Carter decision: 5 5 Carter v Canada (Attorney General), 2015 at paras 47, 105-107, 114-121; Carter v Canada (Attorney General), 2012 at paras 418, 436, 667, 852.

• There would not be a disproportionate impact on the right to life for vulnerable groups.
• The people who would die from physician-assisted suicide would primarily be those imminently terminally ill.
• Physicians would be careful to scrutinize the complexities of requests for assisted suicide for persons living with disabilities or depression.
• Those who felt like a burden, were socially isolated, or suffered from neurological illnesses and disabilities would continue to be protected through a scrupulous review process that protected their right to personal autonomy along with their right to life.

This report examines whether these expectations about the impact of MAiD on vulnerable groups—specifically, persons living with disabilities—have been borne out in reality. It argues, using the available national and provincial data on MAiD deaths, that the removal of the complete ban on euthanasia and assisted suicide has led to an intensified risk of premature death to persons with disabilities in Canada and that the expected safeguards have failed to materialize. The MAiD regime has a highly disproportionate impact on persons with disabilities.

The data used in this report are taken from interim and annual Health Canada reports (2016 to 2023), annual reports from the independent monitoring authority of Quebec for end-of-life (2016 to 2023 fiscal years), peer-reviewed medical studies, the Office of the Chief Coroner of Ontario’s internal summary of the 2023 Ontario data, and the resulting public reports of the MAiD Death Review Committee established by the Coroner’s office. 6 6 Health Canada interim reports for 2016–2019 and annual reports for 2019–2023, https://www.canada.ca/en/health-canada/services/health-services-benefits/medical-assistance-dying/monitoring-reporting.html; Government of Quebec, Commission sur les soins de fin de vie, Rapports annuels d’activités for 2015–2024, https://csfv.gouv.qc.ca/publications; Ministry of the Solicitor General, Office of the Chief Coroner of Ontario, MAiD Death Review Committee Reports, Medical Assistance in Dying (MAiD): Marginalization Data Perspectives, MAiD Death Review Committee (MDRC) Report 2024-2, and MAiD Death Review Committee Report 2024-3 (2024), https://arpacanada.ca/ontario-maid-death-review-committee-reports-2024/; MAiD Death Review Committee Report 2024–1: Waivers of Final Consent (2024), https://macdonaldlaurier.ca/wp-content/uploads/2025/02/MDRC-Report-2024.1_Waivers-of-Final-Consent_Final.pdf; MAiD Death Review Committee Report 2024–4: Complex Same Day / Next Day Provisions (2024), https://macdonaldlaurier.ca/wp-content/uploads/2025/02/MDRC-Report-2024.4_Same-Day-Next-Day-Provisions_Final.pdf. A slide deck created within the Chief Coroner’s office and presented at a 2024 conference was made available to the author but is not available publicly. D. Huyer, “Lessons Learned from the Coroner,” presented at the Fifth Annual National Conference of the Canadian Association for MAiD Assessors and Providers: “MAiD 2024: The Changing Landscape.”

The result of this comparison between expectations and reality is stark. While all jurisdictions that have legalized assisted dying have seen dramatic growth in the number of assisted deaths, the exponential increase in the incidence of MAiD among those with disability status is a uniquely Canadian phenomenon. The same population-level data that show the unexpected and significantly higher number of assisted deaths internationally and especially in Canada, as discussed in a previous Cardus report, “From Exceptional to Routine,” 7 7 A. Raikin, “From Exceptional to Routine: The Rise of Euthanasia in Canada,” Cardus, 2024, https://www.cardus.ca/research/from-exceptional-to-routine/. also show the systematic—and lethal—weakening of safeguards for the most vulnerable groups in Canada.

Since the enactment of Canada’s first Criminal Code in 1892, Canada prohibited assisting a suicide or committing nonculpable homicide, on the basis of the state’s interest in protecting the vulnerable and the value of human life. 9 9 Rodriguez v British Columbia (Attorney General), 1993 SCC 3, p 521 at h and i, https://decisions.scc-csc.ca/scc-csc/scc-csc/fr/1054/1/document.do. The Supreme Court of Canada, in Rodriguez v British Columbia, upheld this complete ban in a 1993 split decision, with the majority finding that, “given the concerns about abuse and the great difficulty in creating appropriate safeguards, the blanket prohibition on assisted suicide is not arbitrary or unfair.” 10 10 Rodriguez v British Columbia (Attorney General), 1993, p 522 at d.

Less than two decades later, in Carter v Canada, several plaintiffs challenged again Canada’s complete ban on assisted dying. The plaintiffs asserted that the prohibition on physician-assisted dying was discriminatory. 11 11 The plaintiffs were the family of Kay Carter, a woman suffering from degenerative spinal stenosis who chose to die in a Swiss assisted suicide clinic; Gloria Taylor, a woman suffering from amyotrophic lateral sclerosis; a Canadian physician willing to administer physician-assisted dying in Canada if it were legal; and the British Columbia Civil Liberties Association. Supreme Court of Canada, Lee Carter, et al. v Attorney General of Canada, et al. (British Columbia) (Civil) (By Leave), Hearing (webcast), October 15, 2014, https://www.scc-csc.ca/cases-dossiers/search-recherche/35591/. This time the courts agreed. On June 15, 2012, Justice Lynn Smith of the British Columbia Supreme Court, a lower court judge, overturned the precedent in Rodriguez by striking down the Criminal Code prohibition of assisted suicide on the basis that it violated section 7 (“life, liberty, and security of person”) and section 15 (“equality rights”) of the Canadian Charter of Rights and Freedoms. 12 12 Carter v Canada (Attorney General), 2012. While Justice Smith’s decision was first overturned on appeal by the British Columbia Court of Appeal, 13 13 The role of a trial judge is to weigh and reach conclusions based on the evidentiary record before them. Appellate courts may only overturn these factual findings where they are deemed to amount to a “palpable and overriding error,” which means no new evidence is usually admitted on an appellate level, according to H.L. v Canada (Attorney General), 2005 1 SCC 401 at para 4, https://decisions.scc-csc.ca/scc-csc/scc-csc/en/item/2226/index.do. Although some limited affidavit evidence was admitted on appeal in Carter by the Supreme Court, it was largely discounted by the Court as irrelevant. See Carter v Canada (Attorney General), 2015 at para 112. the Supreme Court of Canada in Carter v Canada ultimately agreed in a unanimous decision to uphold Justice Smith’s reasoning on section 7 grounds, as risks could be managed through “strict limits that are scrupulously monitored and enforced.” 14 14 Carter v Canada (Attorney General), 2015 at paras 27 and 93. Part of the argument was that legalizing assisted dying would lead to fewer suicides and therefore protect a constitutional right to life, because the criminal prohibition “has the effect of forcing some individuals to take their own lives prematurely.” For example, a person with a degenerative condition who wanted to commit suicide would be able to exercise this option only while they were physically able to, rather than after that time as well. 15 15 Carter v Canada (Attorney General), 2015 at p 335.

The trial judge’s justification for rejecting the Rodriguez decision included her analysis of the new evidence since 1993 from the legal assisted dying regimes in other countries. 16 16 Carter v Canada (Attorney General), 2012 at para 944. The Supreme Court agreed with the reasoning of the trial judge, referring to her “exhaustive review of the evidence.” 17 17 Carter v Canada (Attorney General), 2015 at para 120. The Supreme Court likewise stressed, using Justice Smith’s wording, that a “stringently limited, carefully monitored system of exceptions” was needed. 18 18 The Supreme Court, in Carter v Canada (Attorney General), 2015 at para 29, quoted Justice Smith: “Permission for physician-assisted death for grievously ill and irremediably suffering people who are competent, fully informed, non-ambivalent, and free from coercion or duress, with stringent and well-enforced safeguards, could achieve that objective in a real and substantial way [para 1243 in Carter v Canada (Attorney General), 2012].”

The subsequent legislation (C-14) sponsored by the Minister of Justice and passed by Parliament in 2016 likewise intended to install adequate safeguards to avoid disparate impacts of MAiD on vulnerable communities, including by limiting MAiD to only those with a terminal illness. 19 19 Bill C-14, An Act to Amend the Criminal Code and to Make Related Amendments to Other Acts (Medical Assistance in Dying), 42nd Parl, 1st Sess, 2016, https://www.parl.ca/LegisInfo/en/bill/42-1/C-14. As described by the Justice Department, the introduction of assisted dying intended to balance “the autonomy of those individuals seeking access to medical assistance in dying and the interests of vulnerable persons and of society.” 20 20 Government of Canada, Department of Justice, Legislative Background: Medical Assistance in Dying (Bill C-14) (2016), Part 1, https://www.justice.gc.ca/eng/rp-pr/other-autre/ad-am/p2.html. Central to achieving this balance was limiting the eligibility of MAiD only to persons with an impending natural death.

Just three years later in Truchon, Justice Baudoin, writing for the Superior Court of Quebec, struck down restrictions on MAiD eligibility for non-terminal illnesses, relying on and extending the trial judge’s original argument:

As the Supreme Court stated in 2015, there is no indication that a permissive regime in Canada with properly designed and administered safeguards cannot protect vulnerable people from abuse and error. That remains the case today. . . . Neither the national data in Canada or Quebec nor the foreign data indicate any abuse, slippery slope or even heightened risks for vulnerable people when imminent end of life is not an eligibility criterion for medical assistance in dying. 21 21 Truchon c Procureur général du Canada, 2019 QCCS 3792 at paras 464, 466 (translation from the original French), https://canlii.ca/t/j4f8t.

In 2021, in response to Truchon—and without appealing the lower court decision, as requested by virtually every national disability organization—the federal government, in Bill C-7, liberalized existing safeguards and expanded assisted dying to include those with non-terminal illnesses. 22 22 Bill C-7, An Act to Amend the Criminal Code (Medical Assistance in Dying), 43rd Parl, 2nd Sess, 2021, https://www.parl.ca/LegisInfo/en/bill/43-2/c-7. The bill came into force in the same year and, among other expansions, currently includes the provision that MAiD on the basis of mental illness as the sole qualifying condition will come into effect in 2027.

The trial judge considered evidence from the assisted dying programs in Oregon, Washington, Belgium, Luxembourg, the Netherlands, and Switzerland. Nonetheless, the trial judge’s findings on risks to “socially vulnerable populations” and the efficacy of safeguards was narrower, focusing on the evidence she accepted on this topic from several experts, primarily: 23 23 Carter v Canada (Attorney General), 2012 at paras 621 and 652.

• Margaret Battin and her co-authored study (2007) that summarized the publicly available data for the Netherlands for the years 1995–2005 and for Oregon for 1998–2006. 24 24 Carter v Canada (Attorney General), 2012 at para 667; M.P. Battin, A. van der Heide, L. Ganzini, and G. van der Wal, “Legal Physician‐Assisted Dying in Oregon and the Netherlands: Evidence Concerning the Impact on Patients in ‘Vulnerable’ Groups,” Journal of Medical Ethics 33, no. 10 (2007): 591–97, https://doi.org/10.1136/jme.2007.022335. Critical flaws in the survey cast doubt on the Battin et al. findings. For instance, there are apparent mistakes in the number of natural deaths in Oregon, and the methodology that assumes that patients with serious chronic or terminal illnesses are by default not “vulnerable” (unless they have HIV/AIDS) is questionable. Justice Smith noted some critiques but ultimately dismissed them, finding that “the study withstands that critique” and accepting “that the conclusions stated in the Battin et al. study are soundly based on the data.” See Carter v Canada (Attorney General), 2012 at paras 665 and 667. Justice Smith was not presented with other apparent mistakes in the study.
• Linda Ganzini and her studies of Oregon physicians (2000) and of depression rates among those requesting assisted suicide in Oregon (2008). (Ganzini was also a co-author of the above Battin 2007 study.) 25 25 L. Ganzini et al., “Physicians’ Experiences with the Oregon Death with Dignity Act,” New England Journal of Medicine 342, no. 8 (2000): 557–63, https://doi.org/10.1056/NEJM200002243420806; L. Ganzini, E.R. Goy, and S.K. Dobscha, “Prevalence of Depression and Anxiety in Patients Requesting Physicians’ Aid in Dying: Cross Sectional Survey,” BMJ: British Medical Journal 337 (2008): a1682, https://www.bmj.com/content/337/bmj.a1682; Carter v Canada (Attorney General), 2012 at paras 416–26.

The trial judge endorsed the Battin et al. 2007 study and the testimony of Battin and of Ganzini, stating, “I accept that the opinions of Dr. Ganzini and Dr. Battin were based on the evidence and that they had analyzed that evidence very carefully.” Justice Smith concluded, “This evidence serves to allay fears of the practical slippery slope. The evidence does not support the conclusion that, since the legalization of physician assisted death, there has been a disproportionate impact, in either Oregon or the Netherlands, on socially vulnerable groups such as the elderly or persons with disabilities.” 26 26 Carter v Canada (Attorney General), 2012 at paras 652 and 1241–42.

In this paper, the findings of the court on this international evidence and the expectations it created is contrasted with the Canadian experience with MAiD since its decriminalization in 2015. 27 27 MAiD was immediately decriminalized after the Supreme Court’s ruling in 2015, and the first MAiD cases occurred in the interim period between MAiD decriminalization and legalization in 2016. The methodology of this paper is similar to the methodology used by Margaret Battin and Linda Ganzini in examining published annual reviews of MAiD deaths in Canada and comparing trendlines in the data.

The trial judge concluded in Carter that “there is no evidence that persons with disabilities are at heightened risk of accessing physician-assisted dying in jurisdictions where it is permitted.” The judge accepted as evidence Battin et al.’s conclusion that in Oregon, no one with “pre-existing non-terminal physical disabilities or chronic non-terminal illnesses” received an assisted suicide on the basis of disability alone, and testimony that in Belgium “researchers saw no cases of euthanasia of disabled persons.” 28 28 Carter v Canada (Attorney General), 2012 at paras 852 and 628.

Although the judge conceded that people with disabilities are at risk of being harmed in the medical system through bias and prejudice, she rejected the idea that they would be harmed through legalized MAiD if safeguards were put in place for “careful and well-informed capacity assessments by qualified physicians who are alert to those risks.” 29 29 Carter v Canada (Attorney General), 2012 at para 853.

The national and provincial MAiD data support the findings from chart reviews of MAiD deaths, which is that those who died from MAiD were more likely to have been living with a disability than those who did not die from MAiD, even though both groups had similar medical conditions and experienced diminished capability. 30 30 D. Selby, B. Chan, and A. Nolen, “Characteristics of Older Adults Accessing Medical Assistance in Dying (MAiD): A Descriptive Study,” Canadian Geriatrics Journal 24, no. 4 (2021): 316, https://doi.org/10.5770/cgj.24.520. In other words, the evidence indicates that MAiD is increasingly driven by disability status, rather than by underlying illness.

Even if no persons in Oregon, Belgium, and the Netherlands qualified for an assisted death on the basis of disability alone, Canada today permits disability as a sole qualifying condition for MAiD, including for those who are not dying (“Track 2” MAiD). This expansion occurred despite significant opposition from national disability groups, as well as affidavits submitted to Justice Smith and the Supreme Court by the only disability experts who testified in Carter. 31 31 For a detailed list of warnings and predictions made by the disability community, see Catherine Frazee’s 2024 edited collection: Canadian Journal of Disability Studies 13, no. 2 (2024): “Medical Assistance in Dying: Resistance in Canada,” https://cjds.uwaterloo.ca/index.php/cjds/issue/view/47. Frazee also was an expert witness in the Carter decision. In December 2023 more than fifty Canadian disability organizations wrote a collective letter to the federal government voicing their concerns: “We know, as do you, that the existing law is not working and has not worked, and that people with disabilities have been dying by MAiD due to their life circumstances and oppression.” G. Alexiou, “Canada’s Permissive Euthanasia Laws Spark Debate on the True Meaning of Disability,” Forbes, January 10, 2023, https://www.forbes.com/sites/gusalexiou/2023/01/10/canadas-permissive-euthanasia-laws-spark-debate-on-the-true-meaning-of-disability/. Parliament was presumably concerned with the optics, with the majority of the Special Joint Committee on Medical Assistance in Dying having recommended “replacing references to ‘disability’” in the Criminal Code exceptions for MAiD to “avoid stigmatizing persons with disabilities”—as long as persons with disabilities would continue to have the same access to MAiD. 32 32 Parliament of Canada, Special Joint Committee on Medical Assistance in Dying, Medical Assistance in Dying in Canada: Choices for Canadians, 44th Parl, 1st Sess (February 2023), 42, https://www.parl.ca/Content/Committee/441/AMAD/Reports/RP12234766/amadrp02/amadrp02-e.pdf. More recently, several bodies of the United Nations—the United Nations Committee on the Rights of Persons with Disabilities, the Committee on the Elimination of Discrimination Against Women, and the Special Rapporteur on the Rights of Persons with Disabilities—have criticized Canada for failing to protect the “right to life” of disabled persons through allowing MAiD on the basis of disabilities. 33 33 United Nations Committee on the Rights of Persons with Disabilities, “Concluding Observations on the Combined Second and Third Periodic Reports of Canada,” April 15, 2025, https://docstore.ohchr.org/SelfServices/FilesHandler.ashx?enc=IK6N7o%2F2Yv0ieE7CZGijzQJwv0zld0LfAp60sJB4mAJmPti7qrrpELhJPQ1wU4VwHkZsnY%2F8MQ5q5iAXPNOuoQ%3D%3D.

Data on the number of Canadians with disabilities who have received MAiD are muddled, as there are inconsistencies between Health Canada’s data on MAiD recipients who required disability support services and the available provincial data from the Chief Coroner of Ontario. Notwithstanding these problems in data collection and reporting, Health Canada’s data demonstrate that people with physical disabilities are overrepresented in MAiD deaths when compared to the expectations established in Carter. From 2019 to 2023, 42 percent of all MAiD deaths involved people who required disability services, including over 1,017 people who required but did not receive these services. 34 34 Calculated by the author. Health Canada, First Annual Report on Medical Assistance in Dying in Canada, 2019 (July 2020), https://www.canada.ca/en/health-canada/services/publications/health-system-services/annual-report-medical-assistance-dying-2019.html; Second Annual Report on Medical Assistance in Dying in Canada 2020 (June 2021), https://www.canada.ca/en/health-canada/services/publications/health-system-services/annual-report-medical-assistance-dying-2020.html; Third Annual Report on Medical Assistance in Dying in Canada 2021 (July 2022), https://www.canada.ca/en/health-canada/services/publications/health-system-services/annual-report-medical-assistance-dying-2021.html; Fifth Annual Report on Medical Assistance in Dying in Canada, 2023 (December 2024, updated February 1, 2025), https://www.canada.ca/en/health-canada/services/publications/health-system-services/annual-report-medical-assistance-dying-2023.html. During this period, the type of person who was most likely to die from MAiD was one who required disability supports or who had an unknown disability status. 35 35 The data on necessary disability supports also show a significant and unexplained variance between provinces. According to Health Canada’s Fifth Annual Report on MAiD, disability supports were required for 54.7 percent of all MAiD deaths in British Columbia but only 13.2 percent in Quebec (see p. 74). The data on disability supports, therefore, should be taken with reservation, yet unfortunately there is no other comparable dataset or any explanation by Health Canada on this systematic variance. Moreover, nationwide in some years, and in Ontario in 2023 (the only data point in Ontario), people with disabilities were the most likely type of person to die from MAiD.

The Health Canada report of 2023 indicates that, of those persons with disabilities who did not receive disability supports before their MAiD deaths, in five cases care was not accessible, in 158 cases care was accessible, and in 259 cases it was unknown if care was accessible. 36 36 Health Canada, Fifth Annual Report, 49. These data confirm that MAiD providers in Canada have indeed euthanized disabled patients who needed disability supports and were unable to access them. The data also show that a large number of MAiD deaths occurred even when the provider did not know if disability supports were available, despite the legal requirement to inform MAiD applicants of available disability support services. 37 37 Criminal Code, RSC 1985, section 241.23.1.(g). The adequacy of these disability supports is not assessed, however, in any of these metrics, and therefore these data should be understood as highly limited and partial in regards to gauging access to disability supports.

The trial judge in Carter assumed that “to be accurate and reliable, clinicians who perform [MAiD] assessments would have to be aware of the risks of coercion and undue influence, of the possibility of subtle influence, and of the risks of unconscious biases regarding the quality of the lives of persons with disabilities or persons of advanced age.” 38 38 Carter v Canada (Attorney General), 2012 at para 815. Health Canada’s annual reports enable us to calculate that for more than one in five MAiD deaths (21.2 percent) between 2019 and 2023, the MAiD provider was unable to determine whether the patient even required disability support services. Combined with the 7.3 percent of additional MAiD deaths with an unknown length of disability support provision, the national data show that MAiD providers were frequently unable to ascertain the disability status of their patients, even after their deaths by MAiD. On average, MAiD providers identified only 28.1 percent of those dying by MAiD as persons needing disability supports and for whom they could indicate the approximate length of disability supports. This is virtually the same rate (28.6 percent) as the number of MAiD deaths for which MAiD providers responded “unknown” to questions regarding whether supports were required and the length of time these supports were provided.

These concerns also seem to contradict a study of MAiD providers, which found that “unmet needs” by MAiD applicants were “rare,” although the researchers also noted:

There were some cases in which the provider was worried that unmet needs were driving the request for MAiD. These situations included poor quality or inappropriate housing, inadequate home care in someone who refuses to go into residential care, long waitlists for publicly funded multidisciplinary chronic pain clinics and no local care available requiring unacceptable travel. 39 39 E.R. Wiebe et al., “Are Unmet Needs Driving Requests for Medical Assistance in Dying (MAiD)? A Qualitative Study of Canadian MAiD Providers,” Death Studies 47, no. 2 (2023):207, https://doi.org/10.1080/07481187.2022.2042754.

The length of time that disability supports were provided is likely exaggerated, due to the reporting mechanism. The reporting mechanism measures only whether any disability services are provided, and not whether service provided is at an adequate level. 40 40 Health Canada, Guidance Document: Reporting Requirements Under the Regulations Amending the Regulations for the Monitoring of Medical Assistance in Dying (December 2022), 20. Moreover, the mechanism can lead to double counting, since the length of time that each disability support was provided is cumulatively added. For instance, a patient that received three different types of disability supports over a three-month period would be recorded as having received nine months of support.

Notwithstanding this generous way of counting, 34.3 percent of persons who required disability support services received six months or more of supports, according to the federal data for 2022; 35.6 percent received less than six months of support, 23.0 percent had an unknown length of disability support, and 7.0 percent received no disability supports. 41 41 Calculations by author. Health Canada, Fourth Annual Report, 33. In 2023, the reference period was changed to one year and only 10.9 percent of MAiD recipients were provided more than one year of disability supports. 42 42 Health Canada, Fifth Annual Report, 49. At the same time, the percentage of those deemed not to require disability services prior to their MAiD deaths increased in 2023, yet remained a minority of all MAiD deaths.

The trial judge noted inherent risks for persons with mental illnesses, specifically depression, which can be missed by assessors and may impair decision-making. However, she cited the conclusions of Battin et al., noting that in Oregon, while 20 percent of assisted suicide requests came from depressed patients, not one such patient died from assisted suicide. In the Netherlands, while 3 percent of all requests for euthanasia came from patients with a predominately psychiatric illness, not one of them died from euthanasia. 43 43 Carter v Canada (Attorney General), 2012 at paras 785, 630, and 633.

The judge also referred to a 2008 study by Battin, Ganzini, et al., which included a survey of fifty-eight persons in Oregon who requested assisted suicide. It found that while “one in four had clinical depression” (thus, about fifteen patients), only three patients were approved for and subsequently died by assisted suicide. 44 44 Carter v Canada (Attorney General), 2012 at para 430, citing Ganzini, Goy, and Dobscha, “Prevalence of Depression and Anxiety in Patients Requesting Physicians’ Aid in Dying,” 3. Ultimately though, the trial judge expressed confidence in safeguards that could mitigate the risk.

MAiD deaths due solely to mental disabilities and to chronic non-terminal illnesses were happening in Canada even before Bill C-7 expanded MAiD to include non-terminal illnesses and disabilities. Although not included in federal reporting, we know of at least two cases.

In 2016, the Alberta Court of Appeal ruled in Canada (Attorney General) v E.F. that conversion disorder, a psychosomatic condition, qualified patient E.F. to die from MAiD. This case was not appealed to the Supreme Court. 45 45 Health Canada, Final Report of the Expert Panel on MAiD and Mental Illness (May 2022), https://www.canada.ca/en/health-canada/corporate/about-health-canada/public-engagement/external-advisory-bodies/expert-panel-maid-mental-illness/final-report-expert-panel-maid-mental-illness.html. The same psychiatrist who testified in the case—who characterized E.F.’s condition as “irremediable” despite never having spoken with E.F. or assessed her either in person or remotely—also claimed to have assessed a patient with anorexia nervosa who would go on to die from MAiD. 46 46 Canada (Attorney General) v E.F., 2016 ABCA at para 64; Parliament of Canada, The Standing Senate Committee on Legal and Constitutional Affairs, “Evidence,” 43rd Parl, 2nd Sess (February 2, 2021), 103, 118-120. https://sencanada.ca/en/Content/Sen/Committee/432/LCJC/11ev-55129-e. Both of these cases occurred before the 2021 expansion, let alone the expansion of MAiD to solely mental illnesses in 2027. As Parliament was deliberating on whether Canada was ready to expand MAiD to mental illnesses, MAiD on this basis was already happening, despite the eligibility criteria prohibiting it.

Evidence of overrepresentation of depression in MAiD cases comes from a retrospective chart review of all MAiD requests at a single tertiary care centre in Toronto between June 2016 and April 2019. The review found “high rates of psychiatric comorbidity among requesters of medical assistance in dying,” though unlike in the Ganzini study, most of these requests ended in MAiD. 47 47 E. Isenberg-Grzeda et al., “High Rates of Psychiatric Comorbidity Among Requesters of Medical Assistance in Dying: Results of a Canadian Prevalence Study,” General Hospital Psychiatry 69 (2021): 7–11, https://doi.org/10.1016/j.genhosppsych.2020.12.017. Of the 155 patients requesting, sixty (39 percent) had a documented psychiatric comorbidity (most commonly depression); 117 patients in total received MAiD. 48 48 Fourteen patients of the 155 assessed were deemed ineligible. Isenberg-Grzeda et al., “High Rates of Psychiatric Comorbidity Among Requesters of Medical Assistance in Dying.” Moreover, these sixty patients had a statistically indistinguishable rate of eligibility compared to patients without a psychiatric illness (p=0.363). Compared to the Ganzini findings, patients with a psychiatric comorbidity were much more likely to have requested MAiD than those without a psychiatric comorbidity. A study of Canadian Association of MAiD Assessors and Providers members also found high rates of concurrent mental illness among Track 2 requestors. 49 49 E. Wiebe and M. Kelly, “Medical Assistance in Dying When Natural Death Is Not Reasonably Foreseeable: Survey of Providers’ Experiences with Patients Making Track 2 Requests,” Canadian Family Physician 69, no. 12 (2023): 856, https://doi.org/10.46747/cfp.6912853. Case studies also found that rapid responses to treat depression led some patients to withdraw requests for MAiD. 50 50 N. Berens and S.Y. Kim, “Rapid-Response Treatments for Depression and Requests for Physician-Assisted Death: An Ethical Analysis,” American Journal of Geriatric Psychiatry 30, no. 11 (2022): 1255–62, https://doi.org/10.1016/j.jagp.2022.07.003.

After Canada legalized Track 2, many of the cases involving persons with mental illness also involved a lack of adequate medical or social supports. According to a study of fifty-four Track 2 patient assessments, two-thirds had concurrent mental illness, one-fifth had difficulty finding “appropriate” treatment, and over one-third had not been offered “all appropriate and available treatments.” 51 51 Wiebe and Kelly, “Medical Assistance in Dying When Natural Death Is Not Reasonably Foreseeable,” 853. In Ontario in 2023, rates of marginalization and vulnerability were much more prominent with Track 2 deaths than with Track 1 deaths. 52 52 Ministry of the Solicitor General, Office of the Chief Coroner for Ontario, Medical Assistance in Dying (MAiD): Marginalization Data Perspectives and MAiD Death Review Committee Report 2024–3: Navigating Vulnerability in Non-Reasonably Foreseeable Natural Deaths.

The trial judge cited Battin et al.’s evidence that people with disabilities were not at risk from assisted dying because researchers found that in the Netherlands the natural remaining lifetime of those who died from euthanasia was more than six months in only 0.2 percent of cases. Consequently, the court found that “there is no evidence that persons with disabilities are at heightened risk of accessing physician-assisted dying in jurisdictions where it is permitted.” 53 53 Carter v Canada (Attorney General), 2012 at paras 629, 667, and 852.

Scholarly and governmental analysis has assumed that most people receiving MAiD would otherwise die in a matter of hours or days. 54 54 E.J. Emanuel and M.P. Battin, “What Are the Potential Cost Savings from Legalizing Physician-Assisted Suicide?” New England Journal of Medicine 339, no. 3 (1998): 167–72, https://doi.org/10.1056/NEJM199807163390306; A.J. Trachtenberg and B. Manns, “Cost Analysis of Medical Assistance in Dying in Canada,” CMAJ 189, no. 3 (2017): E101–5, https://doi.org/10.1503/cmaj.160650; Office of the Parliamentary Budget Officer, Cost Estimate for Bill C-7 “Medical Assistance in Dying” (October 20, 2020), https://distribution-a617274656661637473.pbo-dpb.ca/241708b353e7782a9e5e713c2e281fc5ed932d3d07e9f5dd212e73604762bbc5. Low natural life expectancy for MAiD patients was an important consideration in the Carter decision, because it provided evidence that the only people with disabilities who would be affected would be people who were imminently dying. They would therefore presumably be seeking to hasten their death due to their underlying illness, not because of their disability.

The original MAiD legislation likewise intended to restrict the scope of MAiD deaths to those imminently dying; thus it included the qualifying criterion that natural death be “reasonably foreseeable.” Even as Canada since 2021 has expanded MAiD eligibility beyond reasonably foreseeable death, this remains a qualifying and expediting criterion of Track 1 MAiD, which is subject to fewer safeguards and possible same-day MAiD assessment and provision. 55 55 Ministry of the Solicitor General, Office of the Chief Coroner for Ontario, MDRC Report 2024–4: Complex Same Day / Next Day Provisions. The expansion reversed initial indications that assisted suicide would be restricted only to those with terminal conditions.

As Canada left behind the initial safeguards that restricted MAiD to those who were terminally ill, the number of MAiD deaths of non-terminally ill persons began moving steeply upward. The 2021 expansion to non-terminally ill and disabled persons led to 223 MAiD deaths for non-terminally ill persons in 2021, 463 deaths in 2022, and 622 deaths in 2023. 56 56 Health Canada, Fourth Annual Report; Health Canada, Fifth Annual Report, 15. In 2027, Canada will expand MAiD to permit access by reason of mental illness alone, portending further increase in the numbers of non-terminally ill persons seeking state-administered death.

Even if we set aside the MAiD deaths of those who otherwise would have lived for decades, Canada is still experiencing significantly higher average life expectancies of MAiD recipients than was anticipated, far exceeding the early predictions. 57 57 Trachtenberg and Manns (“Cost Analysis of Medical Assistance,” 2017) relied on Dutch and Belgian data on lost life expectancy in their cost analysis of MAiD, and the Parliamentary Budget Office report (Cost Estimate for Bill C-7, 2020) adopted the Trachtenberg and Manns methodology. According to Quebec data on the forfeited natural life expectancy of MAiD recipients (Quebec being the only jurisdiction in Canada to annually report such data prior to the 2021 expansion of MAiD), a majority of MAiD recipients had an estimated natural life expectancy of over one month. Unlike the Battin et al. evidence accepted by the court, indicating that only 0.2 percent of those who died from MAiD in the Netherlands had a natural life expectancy of more than six months, in Quebec in 2023–24 at least 20 percent of MAiD recipients had a natural life expectancy of more than six months—a one-hundredfold difference. 58 58 Government of Quebec, Commission sur les soins de fin de vie, Rapport annuel d’activités, du 1er avril 2023 au 31 mars 2024 (2024), 29, https://csfv.gouv.qc.ca/fileadmin/docs/rapports_annuels/csfv_rapport_activites_2023-2024.pdf. Calculation by author.

In short, despite the limited data available on disability supports and natural life expectancy for MAiD recipients, there is significant evidence that Canadians with physical disabilities are at much greater risk of an assisted death than Canadians without physical disabilities. Since patients in most Canadian provinces are usually eligible for palliative care services only during the last six months of natural life expectancy, Canadians with physical disabilities are possibly accessing MAiD before they could even theoretically have availed themselves of end-of-life care. 59 59 Canadian Virtual Hospice, “Asked and Answered: How Is Eligibility for Palliative Care Decided? What Procedure Is Followed?,” n.d., https://www.virtualhospice.ca/en_US/Main+Site+Navigation/Home/Support/Support/Asked+and+Answered/Palliative+Care/How+is+eligibility+for+palliative+care+decided_+What+procedure+is+followed_.aspx.

Central to the Supreme Court’s decision in Carter was its finding that “an absolute prohibition” on assisted dying would be necessary only “if the evidence showed that physicians were unable to reliably assess competence, voluntariness, and non-ambivalence in patients,” which in turn relied “on the assumption that physicians who do so [assess and provide assisted dying] apply a rigorous standard of scrutiny given the gravity of the decision.” 60 60 Carter v Canada (Attorney General), 2015 at paras 104–107; Carter v Canada (Attorney General), 2012 at para 795.

The Supreme Court likewise upheld the trial judge’s conclusion “that vulnerability can be assessed on an individual basis, using the procedures that physicians apply in their assessment of informed consent and decisional capacity in the context of medical decision-making more generally.” 61 61 Carter v Canada (Attorney General), 2015 at para 115.

The trial judge referred to evidence of the low percentage of assisted dying requests that resulted in an assisted death, in other jurisdictions in which the practice was already legal. For instance, the judge accepted the evidence from Battin et al. that “two-thirds of explicit requests for assistance in dying are not granted” in the Netherlands. The judge twice referred to Ganzini’s evidence that “physicians granted approximately one in six requests for a prescription for lethal medication; one in 10 requests actually resulted in suicide.” 62 62 Carter v Canada (Attorney General), 2012 at paras 631, 418, and 436.

The fact that physicians provided assisted dying to only a small percentage of those requesting it in those jurisdictions implied a rigorous review process by physicians through which vulnerable patients would be protected, the judge believed, and she therefore accepted the opinion that safeguards would be sufficient. 63 63 Carter v Canada (Attorney General), 2012 at paras 631, 836–43.

Despite estimates that MAiD completion rates would remain stable, 64 64 T.E. Quill, “Physician Assisted Death in Vulnerable Populations,” BMJ: British Medical Journal 335 (2007): 625–26, https://doi.org/10.1136/bmj.39336.629271.BE. and despite the expectations established in Carter, 65 65 P. Zimonjic and C. Cullen, “Only 1 in 10 Requests for a Medically Assisted Death Granted, Toronto Doctor Says,” CBC News, September 3, 2016, https://www.cbc.ca/news/politics/medical-assisted-death-toronto-rodin-1.3747136; Quill, “Physician Assisted Death in Vulnerable Populations.” a large majority of MAiD requests end in MAiD in Canada (figure 3). The available evidence from Quebec shows that 65 percent of all explicit MAiD requests, including oral requests, ended in MAiD in 2023. In Canada nationally, the rate increased from 59.4 percent in 2019 to 81.4 percent in 2022. 66 66 Canadian data from 2019 to 2022 are from Health Canada’s annual MAiD reports. Data for 2017–18 are from Health Canada’s interim MAiD reports; the fourth interim report only included data from January 1 to October 31, 2018, while the third and second interim reports included all 2017 data. Only the following provinces were included in the interim reports: Alberta, Saskatchewan, Manitoba, Newfoundland, Prince Edward Island, Nova Scotia, New Brunswick.

In 2023, despite Health Canada changing its reporting to include oral in addition to written requests for MAiD—thus expanding the numerator of the equation—over 78 percent of all requests still ended in a death through MAiD. Phrased differently, requests are decreasingly likely to be challenged by clinicians. This is important because patients approved for MAiD may elect not to receive MAiD if they receive adequate levels of medical treatment and support; for both the Supreme Court and Parliament, MAiD eligibility was not seen to be an automatic pathway to dying from MAiD. After all, studies of dying patients show that “in almost all cases [when a patient requests physician-assisted suicide], the patient’s wishes can be met without PAS [physician-assisted suicide].” 67 67 P.B. Bascom and S.W. Tolle, “Responding to Requests for Physician-Assisted Suicide: ‘These Are Uncharted Waters for Both of Us . . . ’,” JAMA 288, no. 1 (2002): 91, https://doi.org/10.1001/jama.288.1.91.

These rates of MAiD completion raise a serious alarm concerning whether any legal safeguards for MAiD in Canada are capable of functioning in practice. It is a stunning reversal of the courts’ expectation of clinicians’ “rigorous scrutiny” of assisted-dying requests. Based on the figures from the Netherlands and Oregon that were accepted as evidence by the courts, for every 100 persons who received an assisted death there should have been between 200 and 900 persons who requested but did not receive it (whether due to a request being denied, or those with approved requests dying naturally for whatever reason). 68 68 Author calculation. I. Haverkate, B. Onwuteaka-Philpsen, A. van der Heide, et al.,  “Refused and Granted Requests for Euthanasia and Assisted Suicide in the Netherlands: Interview Study with Structured Questionnaire,” BMJ: British Medical Journal 321 (2000): 865–66, https://doi.org/10.1136/bmj.321.7265.865; L. Ganzini et al., “Physicians’ Experiences with the Oregon Death with Dignity Act,” New England Journal of Medicine 342, no. 8 (2000): 557–63, https://doi.org/10.1056/NEJM200002243420806. But in Canada, for every 100 MAiD deaths in 2023 there were 28 or fewer requests that did not result in a MAiD death. 69 69 Author calculation. Health Canada, Fifth Annual Report. It should be noted that we do not know how many of these 28 requests in 2023 went on to re-request and ultimately receive MAiD. Thus the true percentage of requests that do not result in a MAiD death is likely lower than this figure.

When most MAiD requests are being approved and administered, it is unlikely that the system is appropriately protecting the vulnerable—especially when the entire MAiD process of request, assessment, and lethal injection is completed within thirteen days on average (in 2023), and occasionally in one day or less. 70 70 Health Canada, Fifth Annual Report; Ministry of the Solicitor General, Office of the Chief Coroner for Ontario, MDRC Report 2024–4: Complex Same Day / Next Day Provisions.

The trial judge assumed that when persons seeking death showed signs of psychological impairment such as depression, physicians would undertake “a serious diagnostic assessment.” 71 71 See Carter v Canada (Attorney General), 2012 at para 499. The evidence presented—that physicians in other jurisdictions provided assisted dying to only a small percentage of total requests—implied that vulnerable patients would be safe from being coerced or pressured into suicide. Accordingly, the trial judge accepted the opinion that “system safeguards and processes . . . substantially decrease the risk of an involuntary, non-decisional or mood-dependent death by lethal ingestion.” 72 72 Carter v Canada (Attorney General), 2012 at para 436, quoting from Dr. Ganzini.

Drawing on witness testimony, the judge noted that psychiatric referrals serve as a safeguard. 73 73 Carter v Canada (Attorney General), 2012 at paras 854–55. Legal scholars have noted that Justice Smith went so far as to suggest that “prohibiting the prescription of lethal medication to anyone who is depressed, without qualification, would be a more cautious approach. This would be even more restrictive than under Oregon law, where such prescriptions are only prohibited if the depression causes impaired judgment.” 74 74 T. Lemmens, H. Kim, and E. Kurz, “Why Canada’s Medical Assistance in Dying Legislation Should Be C(h)arter Compliant and What It May Help to Avoid,” McGill Journal of Law and Health 11, no. 1 (2017): 61, https://canlii.ca/t/slmz. Canadians were led to believe that those whose requests for MAiD were granted would not be pursuing death due to depression.

Since depression is common for those with terminal illnesses and those with chronic diseases and disabilities, 75 75 See Berens and Kim, “Rapid-Response Treatments.” conscientious screening for depression as a motivating factor for requesting an assisted death is crucial. Yet there is no evidence that such assessments are effectively happening in jurisdictions that have legalized assisted dying.

Even in Oregon, evidence shows that such psychiatric evaluations and referrals described above have significantly decreased. During the first four years of Oregon’s assisted-dying program, 25 percent of its decedents had been referred for psychiatric consultation (twenty-three out of ninety-one). 76 76 Oregon Department of Human Services, Fourth Annual Report on Oregon’s Death with Dignity Act (February 6, 2002), at p 16, https://www.oregon.gov/oha/PH/PROVIDERPARTNERRESOURCES/EVALUATIONRESEARCH/DEATHWITHDIGNITYACT/Documents/year4.pdf. In contrast, only 0.8 percent (three patients) were referred for psychiatric evaluation in 2023, and 1.0 percent (three patients) in 2022. 77 77 The total is seventy-seven from 1998 onwards, or 2.7 percent. Oregon Health Authority, Public Health Division, Center for Health Statistics, Oregon Death with Dignity Act: 2023 Data Summary (March 20, 2024), https://www.oregon.gov/oha/ph/providerpartnerresources/evaluationresearch/deathwithdignityact/Documents/year26.pdf. Such referrals are likewise rare in Canada. In the first years of Ontario’s MAiD program, between June 2016 and October 2018, although virtually all MAiD decedents had suffered from psychological conditions (96.4 percent), psychiatrists were involved in determining MAiD eligibility in very few cases (6.2 percent). 78 78 J. Downar et al., “Early Experience with Medical Assistance in Dying in Ontario, Canada: A Cohort Study,” CMAJ: Canadian Medical Association Journal 192, no. 8 (2020): E175, https://doi.org/10.1503/cmaj.200016.

If depression screenings are not happening through specialist referrals, neither are they happening through the MAiD assessment process. In 2022, the expert panel for MAiD and mental illness recommended the MacArthur Competence Test for Treatment, co-developed by Paul S. Appelbaum. 79 79 Health Canada, Final Report of the Expert Panel on MAiD and Mental Illness. Yet Appelbaum believes that the capacity assessment recommended as part of the MAiD assessment process is grievously inadequate for screening out depression and anxiety, since the questions “are not detailed enough to result in a diagnosis, and even if they did, the impact the answers to these questions are supposed to have on the final decision about authorizing MAiD is unspoken.” 80 80 A. Raikin, “No Other Options: Newly Revealed Documents Depict a Canadian Euthanasia Regime That Efficiently Ushers the Vulnerable to a ‘Beautiful’ Death,” The New Atlantis, December 16, 2022, https://www.thenewatlantis.com/publications/no-other-options.

In Carter (2012), the trial judge defined “situationally vulnerable” people as those who might be motivated to request assisted suicide for reasons such as “personality, emotional distress, untreated symptoms, coercion or the desire not to be a burden.” 81 81 Carter v Canada (Attorney General), 2012 at para 663. The judge distinguished two categories of socially vulnerable people in this regard: those who felt like a burden to themselves and their loved ones, and those who felt socially isolated. In both cases, MAiD physicians would need to carefully protect these patients from being coerced into MAiD.

Drawing on a single survey of Oregon physicians conducted in 1999, the trial judge described how “more than one third of the patients requested assistance with suicide because they perceived themselves as a burden to others, but . . . only three of these patients received prescriptions for lethal medications, suggesting that the physicians were reluctant to accede to requests for assistance under those circumstances.” Nonetheless, she admitted that the real number of those who felt like a burden is unknowable, since “patients may have received assistance in death after experiencing subtle or overt pressure, facing unconscious suggestions by caregivers that their circumstances are hopeless, or sensing that they are a burden on their families. It is impossible to know from statistical evidence whether this has occurred, or how often.” 82 82 Carter v Canada (Attorney General), 2012 at paras 424 and 671.

Regardless, the trial judge found that “the evidence from both Oregon and the Netherlands about actual decision-making practices does not support the conclusion that pressure or coercion is at all wide-spread or readily escapes detection.” 83 83 Carter v Canada (Attorney General), 2012 at para 671.

While the true number of socially vulnerable persons choosing to die through MAiD in Canada is unknowable without a rigorous review process, the current data paint a dismal image (figure 4). The expectation that most patients who feel themselves a burden to others would be prevented from accessing MAiD did not materialize in Canada—and even the initial optimistic data from Oregon (on which Carter was based) have degraded over time. 84 84 The most recent Oregon annual report for 2024 shows that 42 percent of people who had an assisted death felt like a burden. Oregon Health Authority, Public Health Division, Center for Health Statistics, Oregon Death with Dignity Act: 2023 Data Summary.

According to MAiD providers in 2023, the suffering of almost half their MAiD recipients included the perception of being a burden on others, 10 percent more than the previous year. 85 85 Calculated by author. Health Canada, Fifth Annual Report. Because of the nature of the reporting mechanism, we do not know whether this suffering primarily drove the request or was but one contributing factor. Yet federal data reveal that, according to MAiD providers, more than 38 percent of their patients who received MAiD from 2019 to 2023 voiced concerns that they felt like a burden. 86 86 Calculated by author. Health Canada, Fourth Annual Report. Such data imply that Canadian physicians are not reluctant to provide MAiD for suffering that includes social vulnerability.

In terms of social isolation, the trial judge referred to the same Oregon study finding that “lack of social support” was a driving factor for choosing an assisted suicide for only 6 percent of completed cases. Judge Smith concluded that “empirical researchers and practitioners who have experience in [assisted suicide] systems are of the view that they work well in protecting patients from abuse.” 87 87 Carter v Canada (Attorney General), 2012 at paras 419 and 685.

In 2023, however, MAiD providers reported to Health Canada that 22 percent of their patients chose death because of “isolation and loneliness,” up by 5 percent over the previous year. 88 88 Calculated by author. Health Canada, Fifth Annual Report. This is a marked increase, but despite the public nature of these data, it does not appear to elicit concern from Health Canada or other government entities.

These data suggest that not only do MAiD providers know that their patients perceive themselves to be a burden or socially isolated but that these same providers may believe these factors are not an obstacle for MAiD access and are potentially even qualifying reasons for MAiD. Notably, clinicians in charge of Vancouver Coastal Health’s assisted-dying team have told patients with chronic pain conditions that choosing to die from MAiD because of feeling like a burden to loved ones can be considered an “expression of love.” 89 89 D. Coetsee and A. Stewart, “MAiD and Chronic Pain: What You Need to Know About New Legislation and the Assisted Dying Program,” PainBC, June 7, 2021, YouTube webinar, 1 hr., https://youtu.be/AXFQQlRpDa4?si=sXWNtD2biABMtskV. While social isolation is not a medical reason for requesting MAiD, the data indicate that it is an important factor within MAiD requests. 90 90 On the wide-ranging harmful effects of social isolation on health outcomes, including its connection to MAiD, see R. Vachon and M. Allatt, “Social Isolation, Loneliness, and Christian Communities,” Cardus, 2025, https://www.cardus.ca/research/social-isolation-loneliness-and-christian-communities/.

Degenerative neurological illnesses pose significant risks for patient coercion toward assisted dying, due to higher rates of depression and cognitive impairment. 91 91 D. Kitching, “Depression in Dementia,” Australian Prescriber 38, no. 6 (2015): 209–11, https://doi.org/10.18773/austprescr.2015.071; S. Hegde and R. Ellajosyula, “Capacity Issues and Decision-Making in Dementia,” Annals of Indian Academy of Neurology 19, no. suppl 1 (2016): S34–39, https://doi.org/10.4103/0972-2327.192890. The trial judge deemed unlikely the potential for a “slippery slope” that would expand assisted dying to “neurologically impaired patients,” dismissing this key reason for the Canadian Medical Association’s historical opposition to assisted suicide before the Carter decision. Justice Smith pointed to a Belgian study that “does not show elderly patients or patients dying of diseases of the nervous system (including dementia) to be proportionately at greater risks of LAWER [non-voluntary euthanasia] than other patient groups.” 92 92 Carter v Canada (Attorney General), 2012 at paras 274 and 672.

Ultimately, the trial judge concluded “that it is feasible for properly-qualified and experienced physicians reliably to assess patient competence, including in the context of life-and-death decisions, so long as they apply the very high level of scrutiny appropriate to the decision and proceed with great care.” 93 93 Carter v Canada (Attorney General), 2012 at para 798.

Canada has already expanded MAiD to include patients unable to consent to MAiD at the time of their death, through the 2021 provision for a waiver of final consent. 94 94 Waivers permit patients approved for MAiD who are at risk of imminently losing capacity to consent before their scheduled date of death, thus waiving their final consent. A. Raikin, “How Death Care Pushed Out Health Care,” National Review, September 14, 2023, https://www.nationalreview.com/magazine/2023/10/02/how-death-care-pushed-out-health-care/. In Quebec, the provincial government has now gone much further, having sent an order barring prosecutors from launching criminal investigations into physicians who violate the criminal law by administering euthanasia to patients who have made an advance request for MAiD. 95 95 O. Dyer, “Assisted Dying: Quebec Allows Advance Directives, Defying Federal Ban,” BMJ: British Medical Journal 386 (September 16, 2024): q2029, https://doi.org/10.1136/bmj.q2029. An advance request allows a patient to be euthanized at an unknown future date, under certain conditions the patient has specified. For further explanation, see R. Vachon, “Policy Brief: The Risks of Advance Requests for Medical Assistance in Dying (MAiD),” Cardus, 2024, https://www.cardus.ca/research/health/policy-brief/policy-brief-the-risks-of-advance-requests-for-medical-assistance-in-dying-MAiD/. The “slippery slope” warning that was rejected in Carter has proved prescient.

Health Canada’s annual reports also show that MAiD deaths of persons with dementia have increased dramatically in Canada. The number of MAiD deaths with a neurological condition as a qualifying factor has more than tripled in number from 2019 to 2023, and increased from 10.4 percent to 14.9 percent of all MAiD deaths. 96 96 Calculated by author based on data from Health Canada, First Annual Report, 5 and Fifth Annual Report, 70. In 2022, dementia deaths were 9 percent of neurological MAiD deaths or 150 cases. 97 97 Health Canada, Fourth Annual Report, 25–26. In 2023, the number of MAiD deaths of persons with dementia increased to 241, which included 106 deaths in which dementia was the sole underlying condition. 98 98 Health Canada, Fifth Annual Report.

The data are clear: Since MAiD eligibility has become increasingly broad in Canada, it has increasingly and disproportionately affected Canadians with disabilities.

This report compared MAiD’s impact on people living with disabilities with the findings made by Justice Smith in Carter and upheld by the Supreme Court. Further analysis of the disproportionate impact of MAiD on seniors, the poor, and other vulnerable groups also warrants attention, but is beyond the scope of this report.

This report’s findings are contrary to the assumptions by Canadian courts and the claims frequently made by cabinet ministers and Parliament. 99 99 Truchon c. Procureur général du Canada, 2019; Parliament of Canada, Special Joint Committee on Medical Assistance in Dying, Medical Assistance in Dying in Canada. It corroborates, instead, concerns shared previously and repeatedly by disability activists and groups, including testimony ultimately rejected by the Supreme Court in Carter. 100 100 Frazee, “Medical Assistance in Dying: Resistance in Canada.”

The death of disabled persons is not a rare or incidental effect of Canada’s legalized euthanasia program; instead, disability is a remarkably common characteristic among those who access MAiD. The average natural life expectancy of MAiD patients belies the claim that assisted suicide is restricted to those whose death is “imminent,” and MAiD’s reach is not limited to those who are terminally ill. Those seeking MAiD do not encounter a consistently “rigorous standard of scrutiny” that prevents most requests from ending in death. Specialist screenings for depression do not appear to have materialized, even at the very start of the MAiD program. Persons with neurological conditions are seeking death in high numbers compared to the expectations established in Carter. And the socially isolated, far from being protected, are instead being approved for MAiD at high—and increasing—rates.

The realities presented in this report match the concerns of the Canadian public more closely than the concerns of most academics and government officials. Polling by the Angus Reid Institute, in partnership with Cardus, found that three in five Canadians believe that MAiD may push socially and economically vulnerable Canadians, especially those with disabilities, to opt for death. 101 101 Angus Reid Institute, “Disability & MAID: Three-in-Five Concerned Lack of Adequate Care May Push Vulnerable to Consider Assisted Dying,” November 21, 2024, https://angusreid.org/disability-MAiD-health-care-canada-medical-assisted-death/. Furthermore, two-thirds of Canadians believe that MAiD should be offered only as a last resort, 102 102 W. Choi et al., “When Medical Assistance in Dying Is Not a Last Resort Option: Survey of the Canadian Public,” BMJ Open 14, no. 6 (2024): e087736, https://doi.org/10.1136/bmjopen-2024-087736. a position that major policymakers in Canada also hold, as discussed in our earlier Cardus report. 103 103 Raikin, “From Exceptional to Routine.”

Many cases have appeared in the media, with persons indicating that they are choosing MAiD not due to a condition or illness but due to fatigue with an “inhuman” system. 104 104 M. Walters, “Death by Poverty: Canada’s Assisted Dying Program Exposes Fault Lines in Healthcare,” Left Voice (blog), February 9, 2023, https://www.leftvoice.org/death-by-poverty-canadas-assisted-dying-program-exposes-fault-lines-in-healthcare/; C. Fidelman, “Saying Goodbye to Archie Rolland, Who Chose to Die: ‘It Is Unbearable,’” Montreal Gazette, March 29, 2018, https://montrealgazette.com/news/local-news/saying-goodbye-to-archie-rolland; M. Nicholls, “I Run a Food Bank. We’ve Seen a 60 Per Cent Increase in Demand,” Macleans, November 30, 2022, https://macleans.ca/society/food-bank-pandemic-need/. As early as April 12, 2019, the United Nations Special Rapporteur on the Rights of Persons with Disabilities warned Canadian policymakers,

I have further received worrisome claims about persons with disabilities in institutions being pressured to seek medical assistance in dying, and practitioners not formally reporting cases involving persons with disabilities. I urge the federal government to investigate these complaints and put into place adequate safeguards to ensure that persons with disabilities do not request assistive dying simply because of the absence of community-based alternatives and palliative care. 105 105 C. Devandas-Aguilar, “End of Mission Statement by the United Nations Special Rapporteur on the Rights of Persons with Disabilities, Ms. Catalina Devandas-Aguilar, on Her Visit to Canada,” OHCHR, April 16, 2019, https://www.ohchr.org/en/statements/2019/04/end-mission-statement-united-nations-special-rapporteur-rights-persons.

Given the evidence presented here, policymakers cannot claim that MAiD involves only those who are unencumbered by psychological or social vulnerabilities choosing MAiD. Lacking adequate safeguards, MAiD disproportionately affects people with disabilities. It is important to note that these data are not abstract. They represent thousands of Canadians with disabilities who have died through a government-funded suicide program when their MAiD provider believed they did not have access—or did not know if they had access—to necessary supports and care.