COLOFON The art of letting go Printed by: Gildeprint | Cover and lay-out by: Ilse Modder | Copyright © 2022, Sisco van Veen All rights reserved. No part of this thesis may be reproduced, stored or transmitted in any form or by any means without the prior permission of the author, or when applicable, of the publishers of the scientific papers.

VRIJE UNIVERSITEIT THE ART OF LETTING GO A study on irremediable psychiatric suffering in the context of physician assisted death ACADEMISCH PROEFSCHRIFT ter verkrijging van de graad Doctor of Philosophy aan de Vrije Universiteit Amsterdam, op gezag van de rector magnificus prof.dr. J.J.G. Geurts, in het openbaar te verdedigen ten overstaan van de promotiecommissie van de Faculteit der Geneeskunde op donderdag 9 juni 2022 om 13.45 uur in een bijeenkomst van de universiteit, De Boelelaan 1105 door Sisco Marinus Pieter van Veen geboren te Baarn

promotoren: prof.dr. G.A.M. Widdershoven prof.dr. A.T.F. Beekman copromotoren: dr. A.M. Ruissen dr. N.C. Evans promotiecommissie: prof.dr. A.J.L.M. van Balkom prof.dr. B.D. Onwuteaka-Philipsen prof.dr. J. Legemaate dr. D.P. Touwen prof.dr. J.J.M. van Delden

Voor mijn moeder José van der Sman 1953 - 2016

TABLE OF CONTENTS Chapter 1 General introduction PART 1 ESTABLISHING THE RELEVANCE OF IRREMEDIABILITY Chapter 2 Last-Minute Recovery of a Psychiatric Patient Requesting Physician-Assisted Death Chapter 3 Euthanasia of Dutch Patients with Psychiatric Disorders between 2015 and 2017 PART 2 IDENTIFYING THE CHALLENGES OF ESTABLISHING IRREMEDIABILITY Chapter 4 Irremediable psychiatric suffering in the context of physician assisted death: a systematic review of arguments Chapter 5 Irremediable psychiatric suffering in the context of medical assistance in dying: a qualitative study PART 3 ESTABLISHING CRITERIA OF IRREMEDIABILITY Chapter 6 Irremediable psychiatric suffering in the context of medical assistance in dying: a Delphi-study. Chapter 7 General discussion Appendices Summary Samenvatting Dankwoord List of publications 11 27 29 37 51 53 77 95 97 117 136 140 146 150

CHAPTER GENERAL INTRODUCTION Parts of this chapter are translated from an earlier Dutch publication: van Veen, S.M.P. & Widdershoven, G.A.M. Euthanasie in de psychiatrie. Nederlands Tijdschrift Voor Geneeskunde (2020). Also, a modified version of this introduction is currently undergoing peer review as: van Veen, S.M.P., Widdershoven, G.A.M., Beekman, A.T.F. & Evans, N. Physician assisted death for psychiatric suffering: experiences in the Netherlands. 1|

INTRODUCTION Physician assisted death (PAD) for patients with a psychiatric disorder (PPD) is a controversial topic of increasing relevance. In the Netherlands, PAD is possible for patients who are competent in their death wish and who suffer fromamedical condition that is unbearable and irremediable. Although PAD has been accessible for PPD in the Netherlands since the 1990s, until 2010 only a few cases were reported annual. Since 2011 however, there has been a remarkable rise of cases. (Dutch Regional Euthanasia Review Committees – Annual Report 2020) This increase was accompanied by a resurgence of the debate on psychiatric PAD. From the growing clinical experience and accompanying debate it is becoming clear that it is especially challenging to apply the concept of irremediability to psychiatric suffering. (Nederlandse Vereniging voor Psychiatrie, 2018; Onwuteaka-Philipsen et al., 2017) This dissertation aims to address this challenge. In this chapter I will first give a background about irremediability in the context of PAD for PPD. After which I will describe the aims, research questions, and methods. Finally, I will give an outline of this dissertation. BACKGROUND History In some ways the story of assisted death for psychiatric suffering in the Netherlands starts in 1851. This was the year that Jan Slotboom killed Johanna Pluckel, a woman suffering from depression, at her explicit request. This case and the following trial grew into an international scandal. Slotboom was convicted to be ‘hanged in a public place’ but eventually sentenced to imprisonment in a labor camp for 20 years, where he died. This and another high-profile case from the time formed the basis of the criminalization of assisting in someone’s suicide, which still applies to this day. The current debate about PAD however originates in the 1970s. General practitioners started to emphasize the tensions between the duty to alleviate suffering and the duty to not actively terminate life. Some physicians admitted that they sporadically helped people to die at their own request. Anecdotal evidence suggests that these cases mainly concerned terminal cancer patients with a limited life expectancy, people who wanted to die because of psychiatric suffering were not yet discussed. A growing societal movement advocated for more transparency and regulation of assisted death, culminating in the establishment of the ‘Voluntary Euthanasia Foundation’, in 1973. A member of this foundation, Mrs. Wertheim, helped an older lady to die at her request and was charged with assisted suicide in 1983 and was sentenced to six months of suspended imprisonment. In the verdict however the judge gave the first outlines for 12 | CHAPTER 1 1

due diligence requirements that could be grounds for acquittal, one of them was that a physician should be involved. The judge also remarked that psychiatric suffering can cause unbearable suffering and that the patient does not have to be close to death in order to be eligible for PAD. The first psychiatrist to publish about PAD for PPD was Dr. F. van Ree. In 1982, he wrote a commentary in the Dutch Journal of Psychiatry describing three cases about patients with persistent suicidal ideations. Two of the described patients were eventually ‘allowed’ to commit suicide in a humane manner and one patient eventually recovered after long and involuntary clinical treatment. This dissertation will focus on the topic of irremediability of psychiatric suffering, and although van Ree does not explicitly identify this as an important challenge in the context of PAD, he does however implicitly raise questions that are related to irremediability. For instance, he mentions that psychiatric suffering, just as somatic suffering, can be ‘therapy-resistant’. He also describes that for him performing PAD is only possible if it is preceded by long and intensive treatment. The described cases show that in his view PAD should be regarded as a last resort and his moral justification of allowing patients to die relies implicitly on the irremediability of suffering. In a 1983 article, van Ree addresses irremediability in more detail. He writes that accurate diagnosis and prognosis are challenging in psychiatry and argues that due expertise by psychiatrists is essential during a PADprocedure. Or as he puts it: ‘anyone who has never seen how people can recover from a very deep, vital depression and a seemingly hopeless state, cannot make a sound judgment about assisted suicide.’ He also states that intercollegial consultation should always be sought before PAD is performed. (van Ree, 1983) At the start of the 1990s, the Dutch Psychiatry Association and the Royal Dutch Medical Association issued statements that PAD should be allowed for psychiatric suffering. (van Pinxten, 2012) This viewpoint was confirmed in 1994 by the ruling of the supreme court in the Chabot arrest. This case concerned a 50-year-old woman with a depressed mood who wanted to end her life after both her sons died. Over a period of two months, psychiatrist Chabot met with her to discuss her death wish, he also let her stay in his private guesthouse. She refused psychotherapeutic or psychopharmaceutical treatment. After establishing that she was mentally competent and that there were no treatment options left he assisted in her death by prescribing medication. Although he later stated that he consulted seven colleagues, none of these other psychiatrists examined the woman themselves. Chabot was found guilty of assisted suicide, but no penalty was imposed. In the ruling the judges made clear that, although extreme caution is advised, psychiatric suffering can be a justified ground for PAD. Furthermore, this ruling inspired the practice that in the case of psychiatric PAD, an independent consultation from another expert is mandatory. (Berghmans, 1998; Schoevers, Asmus & Van Tilburg, 1998) INTRODUCTION | 13 1

Four years later, in 1998, the first guideline regarding PAD for PPD was published by the Dutch Psychiatry Association. It already contained an elaboration of many of the due diligence requirements that are still important today, including irremediability. In the guideline, suffering was seen as irremediable when there is no prospect of relieving, alleviating or removing it. It also stated that ‘absolute irremediability’ is virtually non-existent in psychiatry, unless the psychiatric disorder is based, for example, on demonstrably irreversible brain damage. For this reason, the guideline committee chose to speak of irremediability if “no realistic treatment perspective” was present. Three requirements were set for a realistic treatment perspective: (a) according to current medical insight, with adequate treatment, there is prospect of improvement, (b) within a foreseeable period, (c) and with a reasonable ratio between the expected results and the burden on the patient. The guideline further stated that a psychiatrist may conclude that there is no prospect of improvement when all relevant biological, psychotherapeutic and social intervention options have been exhausted. Finally, the guideline stated that, in principle, suffering cannot be irremediable if a realistic alternative to alleviate suffering is rejected by the person seeking help. In 2001, the Dutch parliament voted in favor of the ‘Termination of Life on Request and Assisted Suicide Act’ that firmly established and clarified the due diligence procedures for PAD that had developed over the previous years. These legal requirements will be discussed in greater detail below. The Act became law in 2002 and at the time, the Dutch Minister of Health, Els Borst, emphasized that psychiatric suffering could also be grounds for PAD. This historic legislative change was followed by a mostly quiet decade in the field of PAD for PPD. The number of reported cases remained very low and the debate appeared dormant. The guideline underwent minor revisions by the Dutch Psychiatry Association in both 2004 and 2009, mainly regarding procedural demands. This relatively uneventful period ended when, in 2011, the number of reported cases started rising. This rise coincided with the foundation of The End-of-Life Clinic; a special facility specifically aimed at patients who wanted PAD but did not find help from their own physician. The End-of-Life Clinic has quickly become the center of psychiatric PAD in the Netherlands. In the years following its establishment, an increasing portion of all reported PAD’s for PPD have been performed by physicians and psychiatrists working in The End-of-Life Clinic (figure 1). The End-of-Life clinic saw this as a problem and, in 2019, renamed themselves as the Expertise Centre Euthanasia (ECE) and started focusing more on supporting regular psychiatrists in performing PAD themselves. With the rising number of cases, the societal attention and the academic interest returned. This led to a major revision of the PAD-guideline in 2018, which we will further discuss below. This renewed interest and the new guideline in particular provided the motivation for this dissertation. 14 | CHAPTER 1 1

FIGURE 1. Number of physician-assisted deaths performed inside and outside of the Euthanasia Expertise-Centre. Dutch laws, regulations and guidelines In the Netherlands, assisting in the death of another person is currently still punishable by law. However, if physicians follow the due care criteria as described in the ‘Termination of Life on Request and Assisted Suicide Review Act’ they are exempt from prosecution. From this point on we will refer to this act as ‘the euthanasia act.’ Both assisted suicide and euthanasia fall under this act, and a practical guideline prescribes how the respective procedures should be performed. In the case of assisted suicide, the patient ingests a fluid with a heavy sedative that will suppress the respiratory stimulus and stop the heart in a matter of minutes to hours. In euthanasia, the physician intravenously administers a sedative that induces a coma, and consequently a muscle relaxant is given after which death follows almost immediately. (KNMG, 2012) The six due care criteria in the euthanasia act are the following: the physician must be convinced that the patient is making a voluntary and well-considered request [A] and that the suffering is irremediable and unbearable [B]. The patient must be educated about the situation they are in and their prospects [C]. The physician and the patient must be convinced that there is no reasonable alternative to the situation [D]. At least one other doctor must see the patient and give a written opinion on the above due diligence criteria [E]. The assisted death or euthanasia must be performed in a medically prudent manner [F]. (Regional Euthanasia Review Committess RTE: Code of Practice, 2015) When it concerns psychiatric suffering, an additional due care INTRODUCTION | 15 1

requirement applies. Based on jurisprudence and guidelines, another second opinion must be performed by an appropriate expert. This will usually be a psychiatrist working in an academic setting who specializes in the disorder the patient is suffering from. (Regional Euthanasia Review Committess RTE: Code of Practice, 2015) To support the physician throughout this challenging and long process, a guideline is drawn up by the Dutch Psychiatric Association; the first version was published in 1998, the current version is from 2018. (NVvP, 2018) The guideline distinguishes four phases: the request phase, the assessment phase, the consultation phase, and the implementation phase. The request phase begins when a patient expresses a wish for euthanasia. Important goals at that time are: to create an open and safe atmosphere in which to discuss the death wish, to carry out an assessment of possible acute suicidality, to check whether the relatives of the patient are aware of the request, and to provide information about the extensive euthanasia procedure that may follow. In the assessment phase, the physician assesses all due care criteria and requests the mandatory second opinion from an appropriate expert. In the consultation phase, a second physician is consulted. This is normally a physician, often a general practitioner, that received additional training in assessing PAD-requests (called a SCEN-physician). In the implementation phase, the assisted death takes place. (1) In this guideline, irremediability is conceptualized as follows: “irremediability means that there is no longer any prospect of alleviating, enduring or removing suffering. There is no longer a reasonable treatment perspective. In patients with a psychiatric disorder, irremediability is strongly personal and individually determined.” (page 78). The Euthanasia Act also specifies how the PAD-procedure should be evaluated after the patient’s death. Because it concerns an unnatural death, the body should be examined by a coroner directly. The physician who performed the PAD must report to the regional euthanasia review committee. In practice this means that the physician fills out a standardized form describing how the due diligence demands were followed, accompanied by relevant medical correspondence, the reports of the independent physicians and the coroner’s report. The regional euthanasia review committee exists of three members: a physician, an ethicist and a lawyer. If the review committee is satisfied that all due diligence demands were adequately followed, the physician is discharged from further prosecution. If there are doubts, the physician is sometimes asked to appear before the committee to give further explanation. If there are serious doubts about the legality of the PAD, the review committees can transfer the case to the public prosecutor for further investigation and possibly prosecution. (RTE, Code of Practice, 2015) 16 | CHAPTER 1 1

International laws, regulations and guidelines PAD remains highly controversial around the world. A clear example of this is that the World Medical Association recently renewed their viewpoint that PAD is not acceptable under any circumstances. (WMA, 2019) However, a growing number of countries are legalizing a form of PAD. Different countries and states have set up divergent procedural and legal frameworks, often leading to an implicit or explicit exclusion of psychiatric suffering as justifiable grounds for PAD. Countries that allow PAD only for terminal somatic suffering include: Colombia, Japan, New Zealand, South Africa and several states in Australia and the United States of America. (Shaffer, Cook, & Connolly, 2016) In all these jurisdictions, psychiatric disorders are implicitly excluded as legitimate grounds for PAD because the suffering must be terminal or death must be reasonably foreseeable. The countries that do allow PAD for psychiatric suffering are the Netherlands, Belgium, Luxemburg, Switzerland, Spain and Canada. (Rada, 2021) The Netherlands, Belgium and Luxembourg have clear and comprehensive PAD laws that explicitly allow PAD for psychiatric suffering. The Dutch situation is described in detail above and the laws and procedures in Belgium and Luxembourg are very similar, meaning that in all three countries irremediability is a central demand. The Flemish association for psychiatry published a guideline in 2017 in which they further elaborate on the legal requirements, this largely follows earlier Dutch guidelines and states that the bar for irremediability of psychiatric suffering should be set high. All reasonable treatment perspectives have to be exhausted according to the diagnosis specific guidelines. According to the Flemish guideline, reasonable treatment perspectives meet three main requirements: improvement with adequate treatment must be possible, within a manageable time period and with a reasonable relationship between the expected results and burdens of the treatment. Finally, the guideline states that attention must also be given to social interventions and recovery-based approaches. (VVP, 2017) The Netherlands, Belgium and Luxemburg are regularly seen as the first countries to allow PAD for psychiatric suffering. However, strictly speaking, PAD for psychiatric suffering has been possible for a longer time in Switzerland. Article 115 of the Swiss penal code, which came into effect in 1942, states that anyone can assist in the death of another competent person. The only two legal demands are that the person assisting should not have selfish motives and that the person who wants to die ultimately ends their lives themselves, for instance by opening a prepared infusion pump with a lethal drug. In spite of these sparing legal demands, in practice more criteria have to be met. This is mainly because the preferred lethal drug is only available on doctor’s prescription, and the Swiss Academy for Medical Sciences has drawn up guidelines for when such a prescription may be issued. In these guidelines irremediability plays a central part: INTRODUCTION | 17 1

alternative solutions have to be discussed and all options acceptable for the person requesting PAD have to be tried. In 2018, a new guideline was introduced removing the requirement that death must be reasonably foreseeable, which enables PAD for psychiatric suffering. The new guideline does prescribe a mandatory consultation by a psychiatrist when the death wish stems from psychiatric suffering, this consultation is used, in part, to assess irremediability. (van de Wier, 2021) At the time of writing, no official numbers are available, but PAD for psychiatric suffering appears to remain rare in Switzerland. In July 2021 Spain passed a law which is similar to that of the Netherlands and Belgium, it allows PAD for people who ‘suffer from a serious and incurable disease or suffering from a serious, chronic and disabling disease’, leaving room for PAD for psychiatric suffering. In March 2021, the Canadian parliament voted in favor of removing the demand that death has to be reasonably foreseeable from their PAD law. This means that PAD for psychiatric suffering will become possible in Canada. A two-year sunset clause is installed, so that a due diligence procedure can be drafted. The debate is ongoing, and it is unclear what this procedure will look like exactly at the time of writing. It is possible that irremediability will be interpreted more leniently in Canada as compared to the Netherlands, Belgium and Luxembourg. Canadian law states that only the treatments that are acceptable to the patient have to be tried, meaning that in theory a patient with a psychiatric disorder could refuse all treatment and still be eligible for PAD. (Gaind, 2020) The Dutch practice in numbers Although the Supreme Court ruled in 1994 that psychiatric suffering can also be a reason for PAD, for almost two decades, it remained rare. Since 2011, however, there was a remarkable increase in the number of patients that receive PAD due to psychiatric suffering (Figure 1). (Dutch Regional Euthanasia Review Committees - Annual Report 2020, n.d.) The number of requests based on psychiatric suffering is many times higher than the number that is actually performed. It is estimated that 56% of all psychiatrists have had a request for euthanasia during their career, and that about 95% of all requests were rejected. (Kammeraat & Kölling, 2020; Onwuteaka-Philipsen et al., 2017) Detailed quantitative empirical research into PAD for PPD, until recently, remained scarce. For a few years, the only available source for research were the case summaries that were published on the website of the Dutch Regional Euthanasia Review Committees. Kim et al. studied 67 summaries of patients with a psychiatric disorder that received PAD between 2011 and 2014. (Kim, de Vries, & Peteet, 2016) Chapter three of this dissertation describes a study that followed up on this research using the summaries that were published between 2015 and 2017. Both studies found that 18 | CHAPTER 1 1

most patients were diagnosed with multiple psychiatric disorders (71-79%). Common diagnoses were depression (46%-74%), personality disorders and personality problems (52%-54%), anxiety disorders (11-23%), and PTSD (20%-23%). Both studies were limited by the relatively low number of publicly available cases and the fact that they only concerned patients receiving PAD, not patients requesting PAD. This limitation does not apply to a report by the ECE that was presented at the beginning of 2020. (Kammeraat & Kölling, 2020) For this report, 1308 files of patients who requested PAD for psychiatric suffering were analyzed. The report, that has not been peer reviewed, shows that patients requesting PAD often had severe and long-standing psychiatric complaints, 60% had a treatment history of more than 10 years. The mean age of applicants was 50 years. 60% of the applicants were women, 70% were single, 76% had a low or secondary education level and 88% were receiving benefits. 70% of the applicants had more than one psychiatric diagnosis. The most common main diagnosis was depression (35%). When comorbidity is taken into account, common diagnoses were depression (50%), cluster B personality disorder (22%) and trauma- and stressorrelated disorders (20%). 90% of the PAD requests due to psychiatric suffering did not end in PAD, 20% withdrew the request, 68% were rejected. Patients with a cluster B personality disorder as the main diagnosis were relatively often rejected. The study also mentions 8 patients who died by suicide after their request for PAD was rejected. Most patients whose PAD request was granted were between the ages of 50 and 60, 28% had a diagnosis of major depressive disorder and 13% a trauma- or stressorrelated disorder. The challenge of irremediability As mentioned above, the irremediability of suffering serves as an important justification for PAD. This justification is also known as the conflict of duties; a physician has both the duty to relief suffering and to protect life. Usually these duties coincide, but there are situations where the only way to relief the suffering of patients is to help them to die. In these cases, when the suffering is irremediable and there are no subsidiary means of relieving it, the duty to relieve suffering may transcend the duty to protect life and therefore PAD may be justified. However, when we apply this reasoning to psychiatric suffering, several challenges concerning the concept of irremediability arise. To further illustrate this we will first discuss the ongoing conceptual debate and secondly we will discuss the scarce empirical evidence on the topic. Over the past four decades, conceptual papers on the topic of PAD for PPD continuously identify irremediable psychiatric suffering (IPS) as an important ethical challenge. (Berghmans, Widdershoven, & Widdershoven-Heerding, 2013; Blikshavn, Husum, & Magelssen, 2017; Schoevers et al., 1998; van Ree, 1983) Several aspects of psychiatric suffering appear to contribute to the complexity of this topic. First, INTRODUCTION | 19 1

most psychiatric disorders, with the possible exception of severe eating disorders, are not fatal. This means that patients with untreatable mental illnesses potentially have decades to live in which they can recover spontaneously or new treatment options can be developed that might lead to recovery. (Kirby, 2017) Second, little is known about the etiology of psychiatric suffering, complaints can increase and decrease without a clear explanation and practitioners regularly arrive at different diagnoses. As a result, prognosis is very difficult to establish, even in patients who have had many treatments. (Blikshavn et al., 2017) Another point of discussion is whether PPD who refuse certain treatments suffer irremediably. Some authors argue that a patient must try all treatments before suffering can be seen as irremediable. (Appelbaum, 2018) Others state that only treatments that are acceptable to the patient should be tried. (Dembo, Schuklenk, & Reggler, 2018) The ongoing debate indicates that, although the topic has been the subject of substantial moral deliberation, consensus has not been reached among commentators. This lack of consensus is summed up appropriately in 2016 by the ‘Special Joint Committee on physician assisted dying’ of the Canadian Psychiatric Association (CPA): “there is no established standard of care in Canada, or as far as CPA is aware of in the world, for defining the threshold when typical psychiatric conditions should be considered irremediable.” (Gaind, 2020) Apart from the conceptual debate, a few empirical studies suggest that the topic of IPS also poses a challenge in clinical practice in jurisdictions where PAD is possible for psychiatric suffering. The earlier mentioned study by Kim et al. from 2016, analyzing 66 case summaries of Dutch PPD who received PAD between 2011 and 2014, showed that in 20% of these cases psychiatrists disagreed about irremediability of suffering. Furthermore, 56% of patients had refused some sort of treatment, raising questions whether suffering can be seen as irremediable when there are still treatments available. (Kim et al., 2016) Also in 2016, a survey among 248 Dutch psychiatrists, showed that 56% of psychiatrists thought it possible to establish irremediability. This implies that 44% of psychiatrists doubted the possibility or thought establishing irremediability impossible, which indicates a notable dissensus among experts. (Onwuteaka-Philipsen et al., 2017) Finally, a 2016 qualitative interview study among 10 Dutch psychiatrists, identifies irremediability as a central challenge in the context of PAD for PPD. (Pronk, Evenblij, Willems, & van de Vathorst, 2019) OBJECTIVE The aim of this dissertation is to reach a better understanding of the relevance and challenges of establishing irremediability in the context of PAD for PPD. To do this we will take three steps. First, we will investigate whether irremediability is indeed an 20 | CHAPTER 1 1

important clinical challenge in the context of PAD for psychiatric suffering. Second, we will identify the main challenges that psychiatrists face when establishing IPS in the context of PAD. Finally, we will develop criteria on how to address these challenges. This leads to the following research questions: I. Is establishing irremediability a morally relevant issue in the context of physician assisted death for patients with a psychiatric disorder? II. What are the main challenges when establishing irremediable psychiatric suffering in the context of physician assisted death? III. What are suitable criteria for establishing irremediable psychiatric suffering in the context of physician assisted death? METHODOLOGY In order to answer these research questions and to come to a better understanding of the concept of IPS in the context of PAD, we used several research methodologies, inspired by empirical ethics. Empirical ethical research, and hermeneutic approaches in particular, use the lived moral experience of stakeholders to guide moral reflection. (Ives, Dunn, & Cribb, 2016) By exploring the different, and sometimes conflicting, perspectives on a morally complex issue, one can try to develop new and informed ways of dealing with real moral problems. (Widdershoven, Abma, & Molewijk, 2009) In the context of PAD for PPD, the views of psychiatrists who have to establish whether suffering is irremediable constitute an important source of knowledge. Therefore, we will use the moral experience of these professionals to enrich the conceptual debate and to come to clinically applicable consensus criteria about IPS in the context of PAD. In part one of this dissertation, we aim to discern whether establishing irremediability is a clinically relevant issue. First, we will use an ethical-case-report describing an individual case of a PPD requesting PAD (chapter 2). This case report should not be seen as representative for the larger group of patients, but it helps to identify key dilemmas and challenges. Second, we will study the characteristics of PPD who died through PAD and the due diligence procedure that preceded their death, using a series of casefile summaries (chapter 3). In part two, we aim to clarify the challenges that arise in this context. To do this we will start by performing a scoping review on this topic (chapter 4). This type of review combines the rigor of a systematic search that is best suited to identify all relevant literature on a topic, while preserving the openness to synthesize divergent article types in a suitable way. (Munn et al., 2018; Tricco et al., 2018) The knowledge gained from INTRODUCTION | 21 1

assessing the literature and the abovementioned studies is used as basis for a qualitative interview study among psychiatrists who have experience with patients requesting PAD (chapter 5). In the context of PAD for PPD their views are highly relevant because they belong to a small group that have practical experience with establishing IPS and the challenges that arise when actually having to do this. Qualitative research often uses interviews or focus-groups to contribute new knowledge and new perspectives in health care. In general, qualitative studies are suitable to explore complex phenomena encountered by clinicians. (Tong, Sainsbury, & Craig, 2007) From an empirical ethics perspective, qualitative research enables us to elucidate the experience of psychiatrists and compare these with theoretical literature. Finally, in part three of this dissertation, we will perform a Delphi-study to provide guidance for policy and practice (chapter 6). The main goal of Delphi-studies is “the formation of consensus or exploration of a field beyond existing knowledge and the current conceptual world”. (Jünger, Payne, Brine, Radbruch, & Brearley, 2017) In health care research specifically this method is suitable for developing consensual guidance on best practice, in this case on what clinical criteria are suitable when establishing IPS in the context of PAD. From an empirical ethics perspective, Delphi studies provide a means to foster consensus between practitioners on morally sensitive issues, by exchanging views and arguments. The analysis of arguments between various rounds can be regarded as a hermeneutic endeavor, in which the researchers contribute to the consensus formation process by identifying crucial issues and proposing options for reflection. OUTLINE In accordance with the aim, this dissertation consists of three parts. First, we will establish irremediability as a clinically relevant concept in the context of PAD for PPD. Secondly, we will identify the most important challenges that arise when establishing IPS in the context of PAD. And finally, we will draft criteria for IPS that can guide clinical decision making. Part 1: Establishing the relevance of irremediability in the context of PAD for PPD Chapter 2 describes an ethical case report of a man requesting PAD on the basis of psychosis, but who recovers after a new diagnosis and treatment. We use this case to critically reflect on the complexity of IPS in the context of PAD and give suggestions for future research. In chapter 3 we describe a study of 35 casefile-summaries of patients with a psychiatric disorder that died through PAD between 2015-2017. We looked at the clinical characteristics and reviewed the due diligence procedure, confirming that establishing irremediability is a challenge in these cases. 22 | CHAPTER 1 1

Part 2: Identifying the challenges of establishing IPS in the context of PAD In chapter 4 the literature on IPS in the context of PAD is systematically reviewed and discussed. We found 50 relevant articles and identified three main themes, in the discussion important knowledge gaps are discussed and suggestions for further research are given. Chapter 5 consists of a qualitative study for which we interviewed 11 psychiatrists with experience in assessing IPS in the context of PAD. Part 3: Establishing criteria of IPS to guide practice In chapter 6 we use a two round Delphi approach to form 13 consensus-criteria for IPS in the context of PAD with 53 experienced psychiatrists from the Netherlands and Belgium. INTRODUCTION | 23 1

REFERENCES Appelbaum, P. S. (2018). Physician-assisted death in psychiatry. World Psychiatry : Official Journal of the World Psychiatric Association (WPA), 17(2), 145–146. Berghmans, R. (1998). Commentary on “Suicide, euthanasia, and the psychiatrist”. Philosophy, Psychiatry, & Psychology, 5(2), 131–135. Berghmans, R., Widdershoven, G., & Widdershoven-Heerding, I. (2013). Physician-assisted suicide in psychiatry and loss of hope. International Journal of Law and Psychiatry, 36(5–6), 436–443. Blikshavn, T., Husum, T. L., & Magelssen, M. (2017). Four Reasons Why Assisted Dying Should Not Be Offered for Depression. Journal of Bioethical Inquiry, 14(1), 151–157. Dembo, J., Schuklenk, U., & Reggler, J. (2018). “For Their Own Good”: A Response to Popular Arguments Against Permitting Medical Assistance in Dying (MAID) where Mental Illness Is the Sole Underlying Condition. Canadian Journal of Psychiatry, 63(7), 451–456. Dutch Regional Euthanasia Review Committees - Annual Report 2020. Gaind, K. S. (2020). Canada At a Crossroads : Recommendations on Medical Assistance in Dying and Persons With a Mental Disorder. Ives, J., Dunn, M., & Cribb, A. (2016). Empirical bioethics: theoretical and practical perspectives (Vol. 37). Cambridge University Press. Jünger, S., Payne, S. A., Brine, J., Radbruch, L., & Brearley, S. G. (2017). Guidance on Conducting and REporting DElphi Studies (CREDES) in palliative care: Recommendations based on a methodological systematic review. Palliative Medicine, 31(8), 684–706. Kammeraat, M., & Kölling, P. (2020). Psychiatrische patiënten bij Expertisecentrum Euthanasie. Kim, S. Y. H., de Vries, R., & Peteet, J. R. (2016). Euthanasia and Assisted Suicide of Patients With Psychiatric Disorders in the Netherlands 2011 to 2014. JAMA Psychiatry. Kirby, J. (2017). Medical Assistance in Dying for Suffering Arising from Mental Health Disorders: Could augmented safeguards enhance its ethical acceptability? Journal of Ethics in Mental Health, 10(1). Retrieved from KNMG (2012) Richtlijn: uitvoering euthanasie en hulp bij zelfdoding. Munn, Z., Peters, M. D. J., Stern, C., Tufanaru, C., McArthur, A., & Aromataris, E. (2018). Systematic review or scoping review? Guidance for authors when choosing between a systematic or scoping review approach. BMC Medical Research Methodology, 18(1), 143. Nederlandse Vereniging voor Psychiatrie. (2018). Richtlijn: levensbeëindiging op verzoek bij patiënten met een psychische stoornis. Onwuteaka-Philipsen, B. D., Legemaate, J., van der Heide, A., van Delden, H., Evenblij, K., El Hammoud, I., … Willems, D. (2017). Derde evaluatie Wet toetsing levensbeëindiging op verzoek en hulp bij zelfdoding. Pronk, R., Evenblij, K., Willems, D. L., & van de Vathorst, S. (2019). Considerations by Dutch psychiatrists regarding euthanasia and physician-assisted suicide in psychiatry: a qualitative study. The Journal of Clinical Psychiatry, 80(6), 0. Rada, A. G. (2021). Spain will become the sixth country worldwide to allow euthanasia and assisted 24 | CHAPTER 1 1

suicide. BMJ, 372. Regional Euthanasia Review Committess RTE: Code of Practice (2015). The Netherlands. Schoevers, R. A., Asmus, F. P., & Van Tilburg, W. (1998). Physician-assisted suicide in psychiatry: Developments in the Netherlands. Psychiatric Services, 49(11), 1475–1480. Shaffer, C. S., Cook, A. N., & Connolly, D. A. (2016). A conceptual framework for thinking about physician-assisted death for persons with a mental disorder. Psychology, Public Policy, and Law, 22(2), 141–157. Tong, A., Sainsbury, P., & Craig, J. (2007). Consolidated criteria for reporting qualitative research (COREQ): A 32-item checklist for interviews and focus groups. International Journal for Quality in Health Care, 19(6), 349–357. Tricco, A. C., Lillie, E., Zarin, W., O’Brien, K. K., Colquhoun, H., Levac, D., … Straus, S. E. (2018). PRISMA Extension for Scoping Reviews (PRISMA-ScR): Checklist and Explanation. Annals of Internal Medicine, 169(7), 467. van de Wier, M. (2021, January 8). Sterven in Zwitserland: weinig wetgeving, amper controle. Trouw, pp. 4–5. van Pinxten, P. (2012). Menswaardig sterven voor psychiatrische patiënten. Erasmus University. van Ree, F. (1983). Euthanasie en hulp bij zelfdoding in een psychiatrisch ziekenhuis. Medisch Contact, 749–753. Vlaamse Vereniging voor Psychiatrie. (2017). Adviestekst: hoe omgaan met een euthanasieverzoek in psychiatrie binnen het huidig wettelijk kader? Adviestekst van de Vlaamse Vereniging voor Psychiatrie (VVP) over te hanteren zorgvuldigheidsvereisten. Widdershoven, G., Abma, T., & Molewijk, B. (2009). Empirical ethics as dialogical practice. Bioethics, 23(4), 236–248. World Medical Association. (2019). Declaration on euthanasia and physician-assisted suicide. INTRODUCTION | 25 1


CHAPTER 2| S.M.P. van Veen W.F.J. Scheurleer M.L. Ruijsch C.H. Röder G.A.M. Widdershoven A. Batalla Published in Psychiatric Services (2019). LAST-MINUTE RECOVERY OF A PSYCHIATRIC PATIENT REQUESTING PHYSICIAN-ASSISTED DEATH

ABSTRACT Physician-assisted death is becoming legal in an increasing number of jurisdictions, but psychiatric patients are often explicitly excluded. However, in some countries, including the Netherlands, physician-assisted death of psychiatric patients is allowed. This Open Forum describes a patient with schizophrenia and symptoms diagnosed as refractory musical hallucinations. The patient requested assistance in dying only to recover after a mandatory second opinion, where his complaints were recognized as intrusive thoughts and treated accordingly. This case is used to reflect on how to deal with uncertainty about physician-assisted death of psychiatric patients and to argue for implementation of a due-diligence procedure, such as the one proposed in the Dutch Psychiatric Association’s recent guideline concerning this issue. 30 | PART I - CHAPTER 2 2

INTRODUCTION In an increasing number of jurisdictions, physician-assisted death is legal. (Appelbaum, 2017) An important justification for this option is that physicians should be able to relieve the unbearable and irremediable suffering of mentally competent patients, even by assisting their death. (Berghmans, Widdershoven, & Widdershoven-Heerding, 2013) The Netherlands, Belgium, Luxembourg, and Switzerland are the only countries where psychiatric patients are granted legal access to physician assisted death. (Shaffer, Cook, & Connolly, 2016) Since 2012, the prevalence of physician assisted death due to psychiatric suffering in the Netherlands has increased, peaking at 83 deaths in 2017. In 2018, 1.1% (N=67) of all cases of physician-assisted death were due to psychiatric suffering. (Dutch Regional Euthanasia Review Committees - Annual Reports) A majority of these patients were women over age 50 years who suffered from a complex combination of psychiatric disorders and had long treatment histories. About 50% of these patients were diagnosed as having depression or personality disorders. (Kim, de Vries, & Peteet, 2016; van Veen, Weerheim, Mostert, & van Delden, 2018) The increasing number of physician-assisted deaths in psychiatry might be related to the establishment of the Expert Centre on Euthanasia (ECE; formerly called the End-of-Life Clinic) in the Netherlands, an organization of physicians who are specialized in giving advice on and performing complex physician-assisted death requests. In 2018, ECE received 640 requests for physician-assisted death from psychiatric patients and assisted with 56 deaths (9%). What happened to the 91% of psychiatric patients who requested but did not receive physician-assisted death is unclear. (ECE, 2018) This Open Forum introduces the case of a psychiatric patient who asked his physician for assistance in dying but recovered during the assessment process. We will use this case to reflect on how to deal with uncertainty about physician assisted death of psychiatric patients and to argue for implementation of a due-diligence procedure, such as the one proposed in the Dutch Psychiatric Association’s recent guideline concerning this issue. CASE DESCRIPTION A 36-year-old man with a history of attention-deficit hyperactivity disorder, drug abuse, psychological trauma, obsessive-compulsive personality disorder, and therapy resistant schizophrenia had been experiencing psychotic episodes with delusions and imperative acoustic hallucinations (hearing voices) for 10 years. After the second psychotic episode was treated 8 years ago, the patient repeatedly heard songs from his childhood throughout the day. Their effect on the patient and their intensity increased when his mood worsened and when he was in a stressful environment. This new symptom was initially interpreted as an acoustic hallucination, and for several LAST-MINUTE RECOVERY OF A PSYCHIATRIC PATIENT REQUESTING PHYSICIAN-ASSISTED DEATH | 31 2

years, the patient was unsuccessfully treated with antipsychotics, including clozapine. Hearing these songs became increasingly unbearable for the patient and eventually resulted in a request for physician-assisted death. He contacted the ECE. After an assessment period of 1 year, the ECE referred him to an academic hospital for an obligatory second opinion. During admission, the patient’s symptoms were carefully analyzed, and the songs were recognized as intrusive thoughts and not psychotic phenomena. Treatment started with 20 mg of citalopram, which within 3 weeks led to a decrease of the songs the patient heard—a significant clinical improvement. Afterward, cognitive-behavioral therapy was added to the pharmacological treatment, and a few weeks later the patient reached full remission, which continued through the day of this article’s submission, 9 months later. As of this writing, the patient has withdrawn his request for physician-assisted death. DISCUSSION This case highlights the complexity of physician-assisted death for psychiatric patients. The patient’s recovery, of course, was a relief for all parties involved. Yet, an obvious question presents itself: what should clinicians think of the fact that this patient might have died if the diagnosis had not been revisited and no new treatment was started? Two opposing answers can be given. On one hand, this near miss may be used as an argument for banning psychiatric patients from physician-assisted death. On the other hand, it may be concluded that the procedure worked: in the end, the patient was properly diagnosed and adequately treated, after which he withdrew his request. Moreover, if he had not sought the help of the ECE, he may not have been referred for a second opinion, and his suffering might have continued. These arguments provoke several ethical questions, two of which will be discussed here. First, how does one reach a justified conclusion on the irremediability of a patient’s suffering, given the uncertainty about diagnosis and prognosis? Second, what conditions are required for physician-assisted death due to psychiatric suffering? Dealing with uncertainty. Because of the variety of explanatory models and the unclear biological basis of psychiatric disorders, there is always room for uncertainty. Some authors argue that this uncertainty makes physician-assisted death morally inadmissible for psychiatric patients. (Appelbaum, 2018; Schoevers, Asmus, & Van Tilburg, 1998) Others argue that absolute certainty about a prognosis is epistemologically impossible and therefore unreasonable. (Rooney, Schuklenk, & van de Vathorst, 2017) Schuklenk and van de Vathorst mention that uncertainty should be discussed with the patient. (Schuklenk & van de Vathorst, 2015) If he or she understands this uncertainty after adequate 32 | PART I - CHAPTER 2 2

counseling and still requests physician-assisted death, why should the right to self-determination be restricted? The case presented here illustrates the high level of uncertainty common in psychiatric clinical practice. Meanwhile, it shows that discussing the need for a specialized second opinion before a request for physicianassisted death can be met, and actually obtaining this second opinion, can be regarded as a justifiable way of dealing with this uncertainty. The need for a thorough due-diligence procedure If physician assisted death is made possible for psychiatric patients, a thorough due-diligence procedure should be implemented. In 2018, the Dutch Psychiatric Association reformulated the guideline concerning physician-assisted death among patients with a psychiatric disorder, aiming “to provide a contemporary, thorough procedural framework that is both applicable to everyday practice and complies with ethical standards”. (NVvP, 2018) The guideline offers a comprehensive framework for the entire procedure, specifically tailored to psychiatric patients. The procedure is composed of four phases and involves at least three physicians, two of whom must be psychiatrists. The first phase, the request phase, starts when a patient requests physician-assisted death. In this phase, the crucial goals are to create an open and safe environment to discuss the patient’s death wish, to assess whether there is an acute risk of suicide, to ascertain whether those close to the patient have been informed and their views on the matter, and to give clear information about the (extensive) procedure that will follow. In the second phase, the assessment phase, the physician investigates whether all clinical and legal requirements have been met. Above all, the physician ascertains whether the request is voluntary and well considered, whether the patient’s suffering is unbearable and without prospect for improvement, whether the patient is informed about the prognosis, and whether there are other reasonable alternatives. This phase often takes several months and entails at least one obligatory second opinion performed by an independent psychiatrist (preferably one who works in an academic setting) who is specialized in the patient’s disorder. In the third phase, the consultation phase, a physician trained as a physician-assisted death consultant must assess whether all due-diligence requirements have been met. If the primary physician is not a psychiatrist, this consultant must be a psychiatrist. In the fourth phase, the executive phase, the physician assists the patient with dying by either lethal injection or by ingestion of a lethal liquid. After the death of the patient, a public coroner examines the body, and a regional review board consisting of a physician, a lawyer, and an ethicist investigates the case. In our view, the case presented here shows the importance of an obligatory second opinion by a psychiatrist specialized in the patient’s disorder during the second LAST-MINUTE RECOVERY OF A PSYCHIATRIC PATIENT REQUESTING PHYSICIAN-ASSISTED DEATH | 33 2

phase. Psychiatry is a broad specialty, and subtle nuances can be easily missed if a clinician does not have extensive experience with a specific disorder. Utmost care and expert evaluation are required, especially in matters of life or death. Although the Dutch Psychiatric Association’s new guideline further clarifies and sharpens the duediligence process, several challenges remain. For instance, when patients suffer from multiple disorders, it is unclear which specialist should be consulted. Furthermore, the attitudes of Dutch specialized psychiatrists concerning the obligation to seek a second opinion have not yet been investigated. CONCLUSIONS Physician-assisted death among patients with a psychiatric disorder is an emerging and controversial issue around the world that poses substantial ethical challenges. The fact that physician-assisted death is legal for psychiatric patients in some countries provides the opportunity to learn from actual cases. The case presented here of a psychiatric patient who requested physician-assisted death, only to recover after adequate diagnosis and treatment by an independent specialist, can serve as a basis for further debate. On one hand, it raises questions about whether relatively complex psychiatric diagnoses offer a sufficient basis for irreversible decisions such as physician-assisted death. On the other hand, it illustrates that thorough due-diligence procedures can lead to recovery, even at a late stage, which would not have occurred if the patient and the treating physician had continued treatment without a second opinion. This case shows the importance of taking the request for physician assisted death of a patient with a psychiatric disorder seriously and investigating it further by seeking a second opinion from an independent psychiatrist specialized in the patient’s disorder. 34 | PART I - CHAPTER 2 2