Most jurisdictions that have legalized some form of Medical Assistance in Dying (MAD), the term now in vogue that includes Physician Assisted Suicide (PAS) and Euthanasia, have done so only for those who are terminally ill or more broadly, close to dying a natural death. In the few jurisdictions that provide broader access, including for mental illness, in particular Belgium and the Netherlands, the growth of the ‘psychiatric euthanasia’ practice in the last five years has faced particular controversy. Yet, some bioethicists and health law scholars argue that other jurisdictions, including in the US and Canada, should follow their example. They tend to build their argument around paradigm cases of patients with “treatment-resistant depression” (TRD), for which, so they argue, there is no hope of recovery. Not providing access to people with TRD, they put forward, is discriminatory and forces people to either suffer indefinitely or to commit suicide in horrible circumstances.
In Four Reasons Why Assisted Dying Should Not Be Offered for Depression, Thomas Blikshavn, Tonje Lossius Husum, and Morten Magelssen provide an exemplary interdisciplinary and sophisticated response. Although their paper focuses on TRD, many of their points are relevant for the broader mental health context. The paper stands out for its thoughtful reflection on the real-life clinical context in which this practice will play out. Inviting us to reflect on how policymaking needs to account for the complex nature of mental illness and the unique mental health care setting, the paper reveals the dangers of basing sweeping policy changes on well-constructed theoretical arguments that are disconnected from the complex clinical and social context in which they will operate.
The authors wisely don’t waste time discussing two common claims, namely that mental-health-related suffering is as serious as physical suffering and that objectors to psychiatric euthanasia ignore the idea that people with mental illness can have capacity to consent to medical treatment. Proponents of broad access laws keep bringing these up as points of contention, even if no one argues the opposite. Moreover, the authors’ discussion of the complexity of mental-health-related suffering also indirectly reflects their appreciation of its seriousness. Instead, they focus on the argument that it is reasonable to allow MAD for depression when there is no “realistic hope of recovery.” They develop four points: one broader policy point about the impact on mental health services of creating a system that relies on determining that particular patients are unlikely ever to get better; and three related to the impact of introducing MAD on clinical practice in the uniquely difficult context of mental illness.
First, they criticize how the concept of ‘treatment resistant depression’ gives a false aura of ‘objectivity’ and scientific rigor to the argument for expansion. TRD is a technical term that indicates a lack of symptom response to some courses of psychopharmalogical treatment. It does not mean that there are no other treatment options. Anyone familiar with the weak evidence undergirding some of the most widely promoted anti-depressants should be extremely concerned about the argument that failure to respond to three (or even more) such treatments should constitute a reliable basis for agreeing to end a patient’s life. The authors point to the irony that, in cognitive behavioral therapy, the belief that nothing will help is treated as a symptom, and modification of this belief a “crucial therapeutic goal.” One has to wonder what kind of therapy will allow us to deal with this belief at a broader societal level.
The second point relates to the fundamental misunderstanding of what psychiatric diagnoses mean. Outside of psychiatry, pathophysiological processes usually explain diseases and allow health care providers to give some reasonable prognosis (even if uncertainty remains). By contrast, psychiatric diagnoses are broad generalizations based on some shared behavioral traits; the explanatory power of diagnoses and the value of diagnostic prediction remain very limited. We statistically know that a percentage of patients with depression fail to recover, but we cannot know in advance who will fall into this category. In addition, the authors powerfully illustrate how interpersonal relations between therapists and patients, as well as patients’ emotional involvement with their therapy, strongly impact treatment outcomes. More so than in other treatment contexts, patients’ expectations and therapists’ attitudes interact with therapeutic interventions and make the success of treatment context-dependent. In other words, while philosophers and lawyers may construct TRD as an objective, measurable ‘access criterion’, the clinical reality reveals that this is illusory. Moreover, offering the choice of MAD itself affects the likely outcome of the health care intervention.
The third related point is the therapeutic significance of hope. Hope is increasingly recognized as an important contributor to all health care outcomes, but this is particularly true in mental health care. It is essential that health care providers transmit hope to patients whose illness has made them lose hope and long for death. When therapists confirm the patients’ (perceived) reasons for hopelessness, the therapeutic alliance, an essential contributor to a good treatment outcome, is shattered. Giving therapists the task to evaluate the ‘stability’ of the desire to die thus creates a self-fulfilling prophecy: when psychiatrists conclude that hope of recovery is futile, the most crucial tool to counter the patient’s loss of hope (i.e. continued support and the therapist’s persistent hope) disappears.
Their final point is about the broader impact of the normalization of MAD on mental health care. The concern is that normalizing the practice of MAD for psychiatric diagnoses might undermine the resilience therapists currently develop against their own despair when confronted with their most desperate patients. Asking therapists to oscillate between transmitting hope and giving up hope in some cases will undermine the therapeutic alliance, which is crucial for good mental health care. In addition, the institutionalization of assisted dying will undermine patients’ ability to develop an ability to cope with and accept some level of suffering, which is increasingly seen as a crucial component of promising behavioral therapy approaches to mental illness.
That their concern about the normalization of the practice is realistic is in my opinion highlighted by the surge in the demand for euthanasia by psychiatric patients in the Netherlands, as reported in the Third Evaluation of the Dutch euthanasia law. Even though the practice itself remains limited (83 out of a total of 6535 reported euthanasia cases in 2017), 1100 psychiatric patients asked for it in 2015, up from 300 in 1995. This huge surge in demand reveals precisely the kind of broader impact on the therapeutic relation the authors aptly describe. The surge in demand also coincides with an increase in its practice, with specialized end-of-life doctors feeling increasingly comfortable providing access to MAD for mental illness. As the authors conclude: “A wish to die does not arise in an ideological vacuum”; “the very availability of assisted dying may lead some depressed patients to give up the hope that is so vital for therapeutic progress.” Policymakers ought to heed this sophisticated warning about the pernicious impact of MAD as a new therapeutic option in the context of mental health care and should take the recent changes in Dutch and Belgian euthanasia practices as a serious warning sign. And scholars should read this piece to reflect on the serious real life implications of some of their theoretical commitments.