Abstract
Background: Establishing irremediability of suffering is a central challenge in determining the appropriateness of medical assistance in dying (MAiD) for patients with a psychiatric disorder. We sought to evaluate how experienced psychiatrists define irremediable psychiatric suffering in the context of MAiD and what challenges they face while establishing irremediable psychiatric suffering.
Methods: We conducted a qualitative study of psychiatrists in the Netherlands with experience assessing irremediable psychiatric suffering in the context of MAiD. We collected data from in-depth, semistructured interviews focused on the definition of irremediable psychiatric suffering and on the challenges in establishing irremediability. We analyzed themes using a modified grounded theory approach.
Results: The study included 11 psychiatrists. Although irremediable psychiatric suffering is a prospective concept, most participants relied on retrospective dimensions to define it, such as a history of failed treatments, and expressed that uncertainty was inevitable in this process. When establishing irremediable psychiatric suffering, participants identified challenges related to diagnosis and treatment. The main diagnostic challenge identified was the frequent co-occurrence of more than 1 psychiatric diagnosis. Important challenges related to treatment included assessing the quality of past treatments, establishing when limits of treatment had been reached and managing “treatment fatigue.”
Interpretation: Challenges regarding the definition, diagnosis and treatment of irremediable psychiatric suffering complicate the process of establishing it in the context of MAiD. Development of consensus clinical criteria for irremediable psychiatric suffering in this context and further research to understand “treatment fatigue” among patients with psychiatric disorders may help address these challenges. Registration: This study was preregistered under osf.io/2jrnd.
Medical assistance in dying (MAiD), also known as physician-assisted death, has been legalized in an increasing number of jurisdictions around the world.1 In 2023, Canada will join a small group of countries that allow MAiD for people with mental illness, more commonly referred to as persons with a psychiatric disorder. 2 In the Netherlands, MAiD for irremediable psychiatric suffering has been approved by jurisprudence since the 1990s and it has been regulated by law since 2002. The last decade has seen a marked increase in MAiD for irremediable psychiatric suffering; in 2020, MAiD was performed 88 times for psychiatric suffering (1.3% of all MAiD cases), compared with just 2 instances in 2010.3 The number of requests by patients with a psychiatric disorder is much higher, but 90% of requests do not end in MAiD. Sometimes, they are retracted by patients, but most are denied by psychiatrists.4
The main legal requirements for MAiD in the Netherlands are that the patient must be able to make a competent request, that the patient’s suffering must be unbearable and irremediable, and that the patient and physician agree that there are no other reasonable treatment options. The process for MAiD requires an assessment by an independent physician and, in the case of psychiatric suffering, a third assessment by an independent psychiatrist, preferably one with specific expertise regarding the patient’s disorder.5 The Canadian legal requirements under the amended Bill C-7, which will come into effect in 2023, will be largely similar to the Dutch requirements.6 However, Canadian legislation, which does not yet permit MAiD for mental illness, has 1 important difference: patients with decision-making capacity can qualify for MAiD if they refuse treatments that they do not find acceptable. It is as yet unknown whether this approach to refusal of treatment will also be applied to MAiD where mental illness is the sole underlying condition.
Although there are concerns about decision-making capacity, the central dilemma of MAiD for patients with a psychiatric disorder appears to revolve around applying the concept of irremediability to psychiatric disorders. The 2018 guideline by the Dutch Psychiatry Association defines irremediable psychiatric suffering in the context of MAiD as follows: “irremediability means that there is no longer any prospect of alleviating, mitigating, enduring or removing suffering. There is no longer a reasonable treatment perspective.”7 Elsewhere, the guideline states that reasonable treatment perspective means that “there is a prospect of improvement with adequate treatment, within a foreseeable period, and with a reasonable ratio between the expected results and the burden of the treatment for the patient.”7
A recent scoping review identified a multitude of conceptual articles addressing irremediability in the context of psychiatric MAiD but few empirical studies.1 This suggests that the assessment of irremediable psychiatric suffering is particularly difficult relative to suffering arising from other types of conditions.
Surveys estimate that 46% of psychiatrists in the Netherlands have received an explicit MAiD request at least once in their career, and 4% actually assisted in the death of a patient with a psychiatric disorder.8 The experiences of psychiatrists who have handled MAiD requests can be seen as an important source of knowledge about the challenges of establishing irremediable psychiatric suffering in practice. The aim of this study was to learn how experienced psychiatrists define psychiatric suffering as irremediable in the context of a MAiD request and what challenges they face while establishing irremediable psychiatric suffering.
Methods
Study design and participant selection
We conducted a qualitative, in-depth interview study with psychiatrists in the Netherlands who have experience in assessing irremediable psychiatric suffering in the context of a MAiD request. Two of the authors (S.M.P.v.V. and A.M.R.), with extensive networks owing to membership with a foundation of young psychiatrists, made a list of psychiatrists known to them and their colleagues to have performed an independent consultation during a MAiD procedure and psychiatrists who performed MAiD themselves. Proponents, opponents and psychiatrists with moderate views on MAiD for psychiatric suffering were purposively included in the list. We invited psychiatrists from this list to participate in an interview. The participants’ stance on MAiD for psychiatric suffering was confirmed during the selection procedure.
For reporting, we followed the Consolidated Criteria for Reporting Qualitative Research (COREQ).9 The study protocol was preregistered, during data analysis, at the Open Science Framework under osf.io/2jrnd.
Data collection
We assembled a topic list based on our clinical and ethical experience (Appendix 1, Supplement 1, available at www.cmaj.ca/lookup/doi/10.1503/cmaj.210929/tab-related-content). We identified additional items through a thematic analysis from an ongoing systematic review of relevant literature that we undertook simultaneously. 1 The topic list was piloted with 2 senior psychiatry residents, which did not lead to substantial changes. After the fifth interview, we revised the topic list based on our interim analysis and discussion. In the new topic list, the focus shifted from the precise Dutch legal requirements to themes that emerged in the interviews, including hope, powerlessness, doubt and treatment motivation. After the seventh interview, we made additional minor adjustments, including rephrasing some topics and changing the order of topics to better match the participants’ clinical decision-making process.
Participants were interviewed at their place of work or at a place of their choice. We recruited participants between November 2018 and January 2019, and conducted interviews between January and March 2019. During the interviews, only the participant and interviewer (S.M.P.v.V., a male, senior psychiatry resident) were present. The interviewer knew 3 of the participants before the study in a professional capacity. The participants were told that the interviewer held a moderate stance on MAiD for psychiatric suffering, and the aim of the study was disclosed before the interview. We did not conduct any repeat interviews, nor did we return transcripts to the participants for comments. Field notes were made. All interviews were conducted in Dutch, audio-recorded and subsequently transcribed. After coding, an author (S.M.P.v.V ) translated the relevant quotes to English, with the help of a coauthor (N.E.), who is a native English speaker. The original Dutch quotes are in Appendix 2, Supplement 2, available at www.cmaj.ca/lookup/doi/10.1503/cmaj.210929/tab-related-content. The authors of this study have different stances on MAiD for psychiatric suffering, ranging from moderate opposition to conditional support. S.M.P.v.V., A.M.R. and G.A.M.W. are bioethicists. N.E. is a social scientist who specializes in end-of-life research. S.M.P.v.V., A.M.R. and A.T.F.B. are psychiatrists and have been involved in MAiD consultations as experts, but have not performed MAiD themselves.
Data analysis
The method of analysis was a modified grounded theory approach.10 All data were entered into MAXQDA software, version 2018.2. Two authors (S.M.P.v.V., A.M.R.) coded the data independently. They compared codes and resolved disagreements through discussion; attention was given to similar and divergent cases. Coding followed 3 consecutive stages. In the initial coding stage, S.M.P.v.V. and A.M.R. created a set of codes describing elements psychiatrists incorporated in their decision about irremediable psychiatric suffering and the challenges they faced when doing so. During the focused coding stage, the codes were organized into potential themes and overlap between categories was minimized. We reviewed themes and used them for revision of the topic list. In the theoretical coding phase, we explored the relations among, and patterns between, the various themes, after which we refined overarching themes and formulated the final categories. After 11 interviews, we decided that theoretical saturation was reached because no new themes emerged from the last 4 interviews with participants from various backgrounds.
Ethical approval and informed consent
This study was approved by the medical ethics committee of the Amsterdam University Medical Center (2018.661). Participants received an information letter via mail and gave written informed consent.
Results
Of 17 psychiatrists invited to participate, 5 did not have the required experience with MAiD, 1 did not respond and 11 participated in the study. The participant characteristics are described in Table 1. The sample included proponents, opponents and those with moderate views on MAiD. The participants lived and worked in different regions in the Netherlands. The mean interview duration was 62 (range 48–77) minutes. The analysis focused on 2 main questions: the definition of irremediable psychiatric suffering and the challenges psychiatrists face when establishing irremediability.
Definition of irremediable psychiatric suffering
Representative quotes from participant interviews about the definition of irremediable psychiatric suffering are in Table 2. In general, irremediability was considered to be a prospective concept because the physician has to make a judgment about the future course of a disease. Establishing irremediability was therefore seen as synonymous to establishing a poor prognosis (Table 2, quote 1).
Various participants argued that making meaningful prognostic claims about psychiatric suffering is challenging or, some feared, impossible. They described how challenges emerge, in part, because the psychiatrist has to make a claim about a patient who potentially has decades to live (Table 2, quote 2). Given the limits of prospectively establishing irremediable psychiatric suffering, the participants emphasized retrospective dimensions. In particular, participants stated that establishing irremediable psychiatric suffering is dependent on the treatment history and underlined that substantial attempts to reduce suffering should have been tried and failed (Table 2, quote 3).
Most participants viewed psychiatric suffering and treatment as complex concepts; therefore, uncertainty was seen as unavoidable. Comparing psychiatric suffering to somatic suffering, participants mentioned that establishing irremediability for the latter is more certain given the clear biological substrate of the disease and because patients will often die from these diseases unassisted (Table 2, quote 4). A consequence of this uncertainty is that it leaves room for substantial differences among psychiatrists (Table 2, quote 5). Another participant, however, argued that uncertainty should be accepted (Table 2, quote 6).
Challenges in establishing irremediable psychiatric suffering
In relation to establishing irremediable psychiatric suffering, participants mentioned challenges regarding diagnosis and treatment (Table 3).
Challenges regarding diagnosis
Participants said that patients who request MAiD often have more than 1 psychiatric diagnosis, which complicates the assessment procedure and the application of evidence-based guidelines. One participant argued that, therefore, a second opinion by an all-round psychiatrist is the best way to determine irremediable psychiatric suffering (Table 3, quote 1). The independent psychiatrist may also come to a different diagnosis in the course of the MAiD procedure, often leading to new treatment options. One psychiatrist described a patient for whom he diagnosed a new disorder, and noted the need to maintain a nuanced view on new diagnostic insights that emerge from a second opinion (Table 3, quote 2).
Challenges regarding treatment
Most challenges in establishing irremediable psychiatric suffering described by participants were related to treatment. For instance, it can be challenging to assess the quality of past treatments, especially when they concern psychotherapy. To evaluate past treatments, participants described reviewing the patient’s file, considering the reputation of the centre where the patient was treated and taking into account the patient’s views on past therapies (Table 3, quote 3). When evaluating past attempts to reduce suffering, participants regarded the relevant treatment guidelines as important resources (Table 3, quote 4). Participants also emphasized the value of interventions that try to diminish suffering by altering the patient’s perception of the complaints. Acceptance and commitment therapy was often mentioned in this context, but participants also referred to recovery-based or rehabilitation-based approaches, or “using the handicap model” (Table 3, quote 5). Based on this view of psychiatric treatment, some participants argued that treatment options are practically endless and, therefore, irremediability cannot be established (Table 3, quote 6). In contrast, other participants stated that endlessly working toward acceptance or recovery was not a reasonable answer to a patient’s request for MAiD (Table 3, quote 7). Various participants noticed “treatment fatigue” when discussing new treatment options with patients. The participants suggest that this fatigue may be owing to the long treatment history that was typical of patients requesting MAiD (Table 3, quote 8).
Within the current legal framework, additional treatment is sometimes necessary to be eligible for MAiD. As a result, patients may try a treatment just to “check the box.” Various participants doubted whether treating unmotivated patients under these conditions was effective, especially when it concerned psychotherapy. Participants mentioned that no research is available about the efficacy of treatment with unmotivated patients (Table 3, quote 9). One participant emphasized that the possibility of MAiD could increase treatment motivation for some patients (Table 3, quote 10). A final challenge regarding treatment was refusal of new treatments by the patient. Many participants described trying to determine whether refusal was reasonable. If not, they would not see suffering as irremediable, and would stop the MAiD review procedure (Table 3, quote 11). Several participants conceptualized reasonability in terms of a balance between burden and possible benefit. The treatment history was also relevant (Table 3, quote 12).
Interpretation
This study explored how psychiatrists define irremediable psychiatric suffering in the context of MAiD and what challenges they encounter while establishing irremediable psychiatric suffering. The main strength of the study is that it adds in-depth, empirical knowledge to a controversial topic that is gaining global importance and that is the subject of a long-lasting conceptual debate. Although participants saw irremediable psychiatric suffering as a prospective concept, they mostly referred to retrospective dimensions and, more specifically, to the failure of past treatments when defining irremediable psychiatric suffering. This appears logical, for various studies have shown that failed past treatments are a predictive factor for chronicity in psychiatry.11–13 The main opportunity of explicitly adopting a retrospective view on irremediable psychiatric suffering is that it absolves the psychiatrist from the unreasonable task of making highly accurate prognostic claims. It also corresponds with the feeling of “treatment fatigue” that often appears to underly the MAiD request of patients with a psychiatric disorder.14 Furthermore, a retrospective view on irremediability justifies the intuitive uneasiness that various psychiatrists reported with providing MAiD for patients who refuse treatment. Because the patient’s history is central to establishing irremediability, treatment refusal can seriously hamper the psychiatrist’s ability to make a meaningful claim about irremediability. In other words, physicians cannot compel patients to try therapies, but patients cannot compel physicians to conclude that a condition is irremediable if they are not willing to try a sufficient number of therapies. The retrospective view on irremediability therefore also justifies a stricter stance on treatment refusal in psychiatry than in other (somatic) specialties that can use a more prospective view on irremediability, given the biological parameters that lead to a meaningful prognosis. Similar reasoning can be applied to young people requesting MAiD, which happens often in the Netherlands and is seen as a moral challenge.4 If treatment history forms the basis of the decision on irremediability, but relatively little treatment has been possible owing to the fact that the patient has not had symptoms for long, it is reasonable to postpone judgment on irremediability.
This study also confirms earlier findings that psychiatrists struggle with uncertainty as a distinctive element of the definition of irremediable psychiatric suffering. In the conceptual debate about irremediable psychiatric suffering in the context of MAiD, much attention is given to uncertainty, and it is often used as an argument against MAiD for psychiatric suffering.1,15 Both the nature of psychiatric suffering and the lack of a biological substrate are seen as elements that add to the uncertainty surrounding irremediable psychiatric suffering, yet most participants felt that refusing MAiD because of uncertainty alone does not do justice to the individual patient’s request for MAiD. Absolute certainty about the prognosis of any type of suffering is epistemologically impossible. Psychiatrists should therefore aim to find a reasonable balance between the need for certainty and the need to assist individual patients who wish their suffering to be ended.16
Participants mentioned differences in professional opinion when establishing irremediable psychiatric suffering. An earlier case file study of patients with a psychiatric disorder who died through MAiD showed that physicians disagreed about irremediable psychiatric suffering in 11% of cases.17 Although the number of cases in this earlier study was relatively low, differences in professional opinion raise questions about the clarity of the concept of irremediable psychiatric suffering and the potential for a high degree of subjectivity in its use in clinical settings. More clarity and higher levels of consensus, would be of great value to clinical practice in Canada and the rest of the world, and can be achieved by exploring whether psychiatrists with relevant experience can agree upon clinical criteria for irremediable psychiatric suffering in the context of MAiD, such as by using Delphi methodology.18
When establishing irremediable psychiatric suffering, participants faced different challenges regarding diagnosis. The participants mentioned that most patients who request MAiD have more than 1 psychiatric diagnosis. This finding is in line with those of earlier studies reporting that 71%–79% of psychiatric patients who died through MAiD in the Netherlands had more than 1 psychiatric disorder. 4,17,19 This raises questions about whether all distinct treatment protocols for each disorder must be followed completely before irremediable psychiatric suffering can be established. It also raises questions about the current Dutch guideline, which states that an independent psychiatrist with specific expertise about the patient’s disorder should be consulted. In practice, it may prove difficult to identify which expert is best suited; therefore, as 1 participant mentioned, an independent assessment by a psychiatrist with a more generalized view might be more suitable. For Canada, it may be better to adjust the Dutch requirement, because it may sometimes be a challenge to find an independent psychiatrist within a reasonable travel distance, let alone a psychiatrist with specific expertise.
The participants also described challenges regarding treatment when establishing irremediable psychiatric suffering in the context of MAiD. They mentioned that it can be difficult to evaluate the quality of earlier treatments. They also struggled to find a reasonable limit to demanding more interventions that try to diminish suffering by altering the patient’s perception of the complaints, which is a confirmation of earlier conceptual work that identified this as a specific challenge for psychiatry.20 Participants recognized a form of treatment fatigue among patients with a psychiatric disorder requesting MAiD, owing to long treatment history. The term “treatment fatigue” has been researched in the context of HIV and type 1 diabetes, but it has not yet received attention in psychiatry.21 Clinicians may already be attentive to treatment fatigue in current clinical practice, but more research on this topic could provide new insights and result in better care for patients with treatment-resistant psychiatric disorders. Eventually, this may lead to alternatives to MAiD, such as palliative or recovery-oriented treatments.
The participants also mentioned struggling with treatment refusal while establishing irremediable psychiatric suffering. The finding that treatment refusal is a relevant issue is in line with results from an earlier study showing that 56% of patients with a psychiatric disorder who received MAiD refused some form of treatment.19 This finding also justifies the considerable attention treatment refusal has received in the conceptual literature about irremediable psychiatric suffering in the context of MAiD.1 As previously mentioned, from a retrospective view on irremediability, it is reasonable to assert that when a patient refuses a substantial number of treatments, irremediable psychiatric suffering cannot be established because this refusal hampers the opportunity to properly asses irremediability. In our view, a process of shared decision-making is most suited to decide on the limits of treatment refusal.16 The patient should be aware of the potential benefits and burdens of new treatments, and the psychiatrist should try to understand why the patient refuses certain treatments. Again, intersubjective clinical criteria for irremediable psychiatric suffering in the context of MAiD could help to establish reasonable limits for treatment refusal.
Limitations
The main limitation of this study is the limited number of psychiatrists interviewed. More interviews may have established more confidence in the comprehensiveness and generalizability of our findings. However, given that we had a focused study aim, and substantial repetition of themes started to emerge after 7 interviews, we are confident that we have investigated the topic of irremediability with sufficient thoroughness for this study population. Our study did not capture the perspectives of patients and their families on irremediable psychiatric suffering in the context of MAiD, and future studies should explore this.14,22 Future research should also focus on the views of psychiatrists from other countries and cultures.
Conclusion
Establishing irremediability of suffering is a central challenge in the context of MAiD for psychiatric suffering. Although irremediable psychiatric suffering is essentially a prospective concept, psychiatrists mostly refer to retrospective dimensions when defining irremediable psychiatric suffering. More specifically, they mainly focus on the history of failed treatments. When establishing irremediable psychiatric suffering, psychiatrists face challenges related to both diagnosis and treatment. The main challenge regarding diagnosis is that patients requesting MAiD often have more than 1 psychiatric diagnosis. Important treatment-related challenges are assessing the quality of past treatments, establishing the limits of approaches that try to diminish suffering by altering the patient’s perception and managing treatment refusal. A better understanding of treatment fatigue may result in better care for patients with treatment-resistant psychiatric disorders, may yield a better understanding of treatment refusal and may eventually lead to alternatives to MAiD. Many of the identified challenges could be mitigated by drafting clinical criteria for irremediable psychiatric suffering in the context of MAiD through a consensus process.
Footnotes
Competing interests: None declared.
This article has been peer reviewed.
Contributors: Sisco van Veen, Andrea Ruissen and Guy Widdershoven conceived and designed the study. Sisco van Veen performed the interviews, which were coded by Sisco van Veen and Andrea Ruissen. All authors contributed to interpretation of the data. All of the authors drafted the manuscript, revised it critically for important intellectual content, gave final approval of the version to be published and agreed to be accountable for all aspects of the work.
Data sharing: The data that support the findings of this study are available from the corresponding author upon reasonable request.
- Accepted February 1, 2022.
This is an Open Access article distributed in accordance with the terms of the Creative Commons Attribution (CC BY-NC-ND 4.0) licence, which permits use, distribution and reproduction in any medium, provided that the original publication is properly cited, the use is noncommercial (i.e., research or educational use), and no modifications or adaptations are made. See: https://creativecommons.org/licenses/by-nc-nd/4.0/