“For Their Own Good”: A Response to Popular Arguments Against Permitting Medical Assistance in Dying (MAID) where Mental Illness Is the Sole Underlying Condition

Abstract

Canada is approaching its federal government’s review of whether patients should be eligible for medical assistance in dying (MAID) where mental illness is the sole underlying medical condition, and when “natural death” is not “reasonably foreseeable”. For those opposed, arguments involve the following themes: capacity, value of life, vulnerability, stigma, irremediability, and the role of physicians. It has also been suggested that those who are able-bodied should have to kill themselves, even though suicide may be painful, lonely, and violent. Opponents of MAID for severe, refractory suffering due to mental illness imply that it is acceptable to remove agency from such patients on paternalistic grounds. After years of efforts to destigmatise mental illness, these kinds of arguments effectively declare all patients with mental illness, regardless of capacity, unable to make considered choices for themselves. The current paper argues that decisions about capacity must be made on an individual-patient basis. Given the rightful importance granted to respect for patient autonomy in liberal democracies, the wholesale removal of agency advocated by opponents of a permissive MAID regime is difficult to reconcile with Canadian constitutional values.

Introduction – A Word on Paternalism

Patients with mental illness are usually portrayed by opponents of permissive MAID (medical assistance in dying) regimes as exceptionally vulnerable. It is argued that because their decisional capacity can be challenging to assess, they should never be considered for MAID unless they are imminently dying. ^1,2 Those in good physical health are therefore expected to either try every possible treatment available, wait for a cure to be developed, continue suffering, or end their own lives. ^3,4

Most of these arguments spring from a single source in medical ethics: paternalism. The justification for paternalism in health care is based on the beneficence principle. Doctors aim to benefit their patients, much like good parents would their children. Beneficence is supported by most mainstream theoretical approaches in bioethics. ⁵

There is a consensus in medical ethics and law that some paternalism is uncontroversial. It is the kind of paternalism where we stop drunk party-goers from driving. We would not only be concerned about their potential to harm others but also to harm themselves. This is where “weak” or “soft” paternalism comes in: we override incompetent patients’ choices to serve their objective best interests.

Proponents of a prohibition on MAID for all non-terminally ill psychiatric patients need to justify “strong” paternalism: the overriding of a competent individual’s choices for what the paternalist considers to be the individual’s own good. Here it would be necessary to show that to override a competent psychiatric patient’s decision to seek MAID is always the right decision. This is a consequence of the weight we rightly ascribe in a liberal democracy to our right to self-determination.

Errors with respect to MAID are possible for any patient, including those with solely a mental illness. ⁶ However, there is good reason to doubt the validity of existing analyses in support of concerns about the slippery slope-type abuses of the vulnerable. ⁷ The same worry has unsuccessfully been deployed against MAID for physical illness. A more appropriate response to such worries would involve the implementation of rigorous safeguards in the process of eligibility and capacity assessments.

Why “Reasonably Foreseeable”?

The Supreme Court of Canada (SCC) ruled that the banning of physician-assisted death was unconstitutional and that it should be legal “for a competent adult person who 1) clearly consents to the termination of life and 2) has a grievous and irremediable medical condition (including an illness, disease, or disability) that causes enduring suffering that is intolerable to the individual in the circumstances of his or her condition.” ⁸ An ethical defence of these criteria was provided in a lengthy report produced by a Royal Society of Canada expert panel years earlier. ⁹ Canada’s government, though, decided not to make MAID available to all patients meeting SCC criteria; instead it limited eligibility to patients whose natural deaths are “reasonably foreseeable.” ¹⁰ The term “reasonably foreseeable” was used instead of providing a specific time frame, such as the 6-mo prognosis requirement in some US jurisdictions. The government wanted to recognize that individual circumstances—including but not limited to degree of suffering and type of trajectory toward death—vary. ¹¹ This eligibility threshold affects 2 groups of patients: people with mental illness as the sole underlying medical condition in the absence of a terminal physical condition, and people with non-terminal physical illness who do not have other factors, such as age or significant co-morbidities, that make death foreseeable.

The Government’s justification for the narrowed eligibility criteria for MAID include 1) the need to “affirm the inherent and equal value of every person’s life and avoiding encouraging negative perceptions of the quality of life of persons who are elderly, ill or disabled”; 2) the concern that “vulnerable persons must be protected from being induced, in moments of weakness, to end their lives”; 3) the recognition that “suicide is a significant public health issue that can have lasting and harmful effects on individuals, families and communities”; and 4) that it “strikes the most appropriate balance between the autonomy of persons who seek medical assistance in dying, on one hand, and the interests of vulnerable persons in need of protection and those of society, on the other.” ¹¹

MAID and Suicide

There is a widely held, but mistaken, notion that permitting MAID in populations where death is not reasonably foreseeable increases suicide rates in general or leads to premature death.

Because the SCC criteria neither excluded mental illness nor required a reasonably foreseeable death, government would need to show that including mental illness would increase the suicide rate, but existing evidence does not support this. Data from jurisdictions where MAID is available to these populations do not show an increase in suicide rates. Lowe and Downie analysed these data in 2017, ¹² in challenging a faulty analysis by Jones & Paton, ¹³ whose paper has been cited by opponents of MAID as evidence that MAID increases suicide rates. Lowe and Downie analysed OECD (Organization for Economic Cooperation and Development) ¹⁴ statistics for non-assisted suicide in Switzerland, Belgium, the Netherlands, and Luxembourg. They found that overall, suicide rates either stayed the same or decreased after MAID legislation pertaining to sole mental illness was passed in those jurisdictions. A 2014 Swiss government report shows that the suicide rate per 100,000 individuals has decreased steadily since before the introduction of this legislation. ¹⁵

Capacity

It is well established that all adult patients are presumed capable with respect to medical decision making unless proven otherwise. ¹⁶ One study using the most well-validated standardised capacity assessment tool, the McArthur Competence Assessment Tool for Treatment (McCAT-T), established that 70% to 80% of involuntarily hospitalised patients with mental illness are capable with respect to treatment decisions. ¹⁷ By diagnosis, capacity can vary; another study using the McCAT-T indicated that only 4% of patients with a personality disorder lacked decisional capacity, whereas 80% of patients with acute psychosis due to schizophrenia lacked capacity. ¹⁸ Among those with severe unipolar depression, in this same study, only 31% lacked capacity. Of note, the above studies evaluated acute inpatients; outpatients likely have higher rates of capacity with respect to treatment, though studies in this area are lacking.

Capacity must be assessed on an individual basis. Any regime in which MAID is permitted for refractory mental illness requires careful safeguards, given that mental illness can unduly influence a desire to die. However, to exclude all individuals requesting MAID for psychiatric illness in the absence of a terminal physical condition falsely implies that everyone in that category lacks capacity.

Suffering, Irremediability, and Autonomy

Not only is the term “quality of life” ill-defined, but views vary widely about who should determine whether a life’s quality is acceptable. In Western medical culture, until recently, this difficult decision often lay in physicians’ hands. ¹⁹ However, this is no longer the case. Patient autonomy is now one of the core principles of medical ethics and law. ¹⁹ Patients are encouraged to assert what “quality of life” means to them. The type of illness affecting quality of life is immaterial. Judgments about suffering, irremediability, and quality of life are closely intertwined, and capable patients rightly define these terms as they apply to their own unique situations. Because suffering is a personal experience, only the affected individuals can assess whether it is intolerable. ^20,21

The bias inherent in arguments made by those who might support MAID in non-terminal severe physical illness but not in severe mental illness is that suffering in the former group is more unbearable than in the latter. However, the intensity of suffering in severe mental illness can be equal to that of the most severe physical conditions. For example, in a large study of subjective wellbeing in Berlin, the authors found only end-stage liver disease (a devastating and imminently lethal condition which only liver transplantation can reverse) was subjectively as severe as mental disorders. ²² Mental suffering can frequently cause severe disability and premature death from medical illness and suicide. ²³ We have no reason to think that decisionally capable psychiatric patients’ evaluation of their quality of life is more unreliable than that of other decisionally capable patients. ²⁴ Canadian courts recognise that suffering in mental illness can be unbearable; in the case of Canada vs. EF (plaintiff), ²⁵ the Alberta Court of Appeal granted MAID due to severe pain and disability arising from a conversion disorder. EF received MAID when it was briefly permissible through a court ruling using the Carter criteria; however, she would now not be eligible for MAID.

With respect to irremediability, although it is impossible to predict response or remission with certainty, ²⁶ capable patients should have the right to make their own judgments based on the best evidence available at the time of decision making. Mental illness can certainly be irremediable. ^27,28,29,30 In the case of major depression, the STAR*D study conducted by the National Institute of Mental Health (NIMH), indicated that 30% or more of patients did not respond to multiple sequential medication trials, and of those who did respond, up to 70% relapsed within 1 y. ³¹ Patients who fail repeated medication trials, hospitalisation, and psychotherapy sometimes go on to try neurostimulatory treatments, such as electroconvulsive therapy (ECT), transcranial magnetic stimulation (TMS), or deep brain stimulation, all of which can be effective to some extent; but none are 100% effective. A recent study examining different types of ECT indicated response rates between 55% to 64%; ³² response to TMS was 29% in a recent meta-analysis; ³³ and deep brain stimulation, according to a Canadian sham-control trial, can achieve a 6-mo response rate of 48% and a 12-mo response rate of 29%. ³⁴ Furthermore, the rates of response in patients who then proceed to ablative surgeries, such as capsulotomy, are only about 50%. ³⁵ Therefore, individuals with refractory mental illness who have undergone many years of unsuccessful therapeutic efforts are not necessarily irrational when they feel recovery is unlikely, or when they refuse to wait for therapeutic advances whilst experiencing intolerable suffering.

The possibility of new and better treatments arising from research and development (R&D) understandably leads to caution amongst regulators and clinicians. If someone with refractory depression has a life expectancy of decades, would it not be reasonable to ask them to live on in hope that therapeutic R&D would succeed during their lifetime? Capable patients with mental illness should certainly consider this possibility in their decision making; however, they must not be forced to wait indefinitely.

The Carter decision was clear that “irremediable” does not only refer to the statistical probability of survival with an illness but also to “enduring physical or psychological suffering that is intolerable to [the individual] and that cannot be relieved under conditions that they consider acceptable.” ³⁶ Patients are permitted to refuse treatments which might otherwise prolong their lives or even reverse the disease process. It is unjust to apply different standards to capable people with true refractory mental illness.

Vulnerability

Much is made by proponents of the current restrictive MAID regime of psychiatric patients’ “vulnerability”. That same label was initially liberally deployed by anti-choice activists opposed to MAID generally. In their interpretation all patients were vulnerable, potential targets of abuse, and in need of protection against that potential abuse, and they should not be given the option to choose MAID. The SCC disagrees that all patients are sufficiently vulnerable as to render them unable to make decisions that are substantially autonomous. Bioethicists are highly critical of the vagueness of the label and its propensity to stereotype groups of quite diverse patients. ³⁷ Indeed, it is doubtful that the label signifies a universally agreed-on concept at all. Rhodes writes, “Instead of trying to respect the autonomy of others by presuming that they are autonomous and trying to see their choices as reasonable from their perspective, classifying people as ‘vulnerable’ denies them respect.” ³⁸ The vagueness of the “vulnerability” trope upon which proponents of a restrictive regime trade makes its use problematic. The vulnerability rhetoric has not been without harmful consequences in the history of medicine. ³⁸

In the context of psychiatric patients—all psychiatric patients no less—vulnerability rhetoric is used to remove agency from a large group of diverse patients, including patients with decisional capacity. All members of this already highly stigmatized group are treated as decisionally incompetent regardless of their individual circumstances. This is achieved by labelling all members of this group “vulnerable.” ⁷ It is incoherent that these same patients are—at the time of writing—eligible for MAID, if they meet the “reasonably foreseeable” standard by virtue of their age or co-morbidities. The Council for International Organisations of Medical Sciences, in the most recent edition of its preeminent international research ethics guidelines, has eliminated reference to “vulnerable” people, as a category, altogether. It notes, “A consensus emerged in recent publications that vulnerability can no longer be applied to entire groups.” ³⁹

Let Them Kill Themselves

Some psychiatrists opposed to MAID for those with sole psychiatric illness have argued that these patients do not need access to MAID because they can terminate their own lives at their own volition. As Maher puts it: “On this view, a doctor is really a sanitized version of a gun. But no matter how you parse it out, people living with mental illness can swallow their own suicide pills.” ⁴ Maher is correct on the latter point but that is not in question. Patients’ eligibility for MAID does not depend on their ability to commit suicide. MAID is available to able-bodied, non-psychiatric, eligible patients. It is difficult to accept at face value expressions of concern about conveniently labelled “vulnerable” psychiatric patients but then to have them asked to resort to gruesome suicide methods in lieu of providing MAID.

Conclusion

Throughout much of history, society and medical professionals have made judgments on behalf of people with severe mental illness, with the benevolent intent of making decisions in their best interests. The MAID debate now highlights where strong paternalistic values persist in psychiatry. When an individual is clearly incapable with respect to medical decision making, medical professionals and substitute decision makers must step in. However, capacity must be determined on an individual basis. It is unjustifiable to exclude psychiatric patients from benefitting from a medical intervention that is designed to preserve autonomy and reduce suffering, without giving due consideration to individual variability.