Euthanasia and Assisted Suicide of Patients With Psychiatric Disorders in the Netherlands 2011 to 2014

Importance  Euthanasia or assisted suicide (EAS) of psychiatric patients is increasing in some jurisdictions such as Belgium and the Netherlands. However, little is known about the practice, and it remains controversial.

Objectives  To describe the characteristics of patients receiving EAS for psychiatric conditions and how the practice is regulated in the Netherlands.

Design, Setting, and Participants  This investigation reviewed psychiatric EAS case summaries made available online by the Dutch regional euthanasia review committees as of June 1, 2015. Two senior psychiatrists used directed content analysis to review and code the reports. In total, 66 cases from 2011 to 2014 were reviewed.

Main Outcomes and Measures  Clinical and social characteristics of patients, physician review process of the patients’ requests, and the euthanasia review committees’ assessments of the physicians’ actions.

Results  Of the 66 cases reviewed, 70% (n = 46) were women. In total, 32% (n = 21) were 70 years or older, 44% (n = 29) were 50 to 70 years old, and 24% (n = 16) were 30 to 50 years old. Most had chronic, severe conditions, with histories of attempted suicides and psychiatric hospitalizations. Most had personality disorders and were described as socially isolated or lonely. Depressive disorders were the primary psychiatric issue in 55% (n = 36) of cases. Other conditions represented were psychotic, posttraumatic stress or anxiety, somatoform, neurocognitive, and eating disorders, as well as prolonged grief and autism. Comorbidities with functional impairments were common. Forty-one percent (n = 27) of physicians performing EAS were psychiatrists. Twenty-seven percent (n = 18) of patients received the procedure from physicians new to them, 14 of whom were physicians from the End-of-Life Clinic, a mobile euthanasia clinic. Consultation with other physicians was extensive, but 11% (n = 7) of cases had no independent psychiatric input, and 24% (n = 16) of cases involved disagreement among consultants. The euthanasia review committees found that one case failed to meet legal due care criteria.

Conclusions and Relevance  Persons receiving EAS for psychiatric disorders in the Netherlands are mostly women and of diverse ages, with complex and chronic psychiatric, medical, and psychosocial histories. The granting of their EAS requests appears to involve considerable physician judgment, usually involving multiple physicians who do not always agree (sometimes without independent psychiatric input), but the euthanasia review committees generally defer to the judgments of the physicians performing the EAS.

Some form of assisted death, such as euthanasia or assisted suicide (EAS), receives legal protection in Belgium, the Netherlands, Switzerland, Luxembourg,¹ and Canada,² as well as in several American states.³ Although the origins of legalization of EAS centered on patients with terminal illness, many do not believe that the principles of autonomy and beneficence (relief of suffering) limit EAS to terminal conditions and argue that EAS should be extended to psychiatric conditions.^4,5 Euthanasia or assisted suicide for such persons in Belgium and the Netherlands^6-8 has received increasing attention.⁹ The recent Supreme Court of Canada ruling permitting physician-assisted death may not limit it to individuals with terminal illness,² and no such limitation exists in Switzerland.¹⁰ Although the numbers remain small, psychiatric EAS is becoming more frequent. In the Netherlands, a 1997 study¹¹ estimated that the annual number was between 2 and 5, and in 2013 there were 42 reported cases.⁷

Although the debate over psychiatric EAS typically focuses on persons with treatment-resistant depression,^4,5,12 little is known about individuals receiving EAS for psychiatric conditions. Aside from a 1997 study¹¹ describing 11 cases in the Netherlands, there is one review of 100 psychiatric EAS requesters evaluated by a Belgian psychiatrist.¹³ Furthermore, requests for EAS to relieve suffering from psychiatric conditions require special scrutiny.⁷ Psychiatric disorders contribute to suicides (a major public health problem¹⁴), can sometimes impair decision making,¹⁵ and are stigmatized.¹⁶ Thus, the regulation of psychiatric EAS is of great interest, as courts cite evidence from countries with established practices.² In the United States, the trend of legalizing physician-assisted death is already accompanied by discussions about broadening the practice beyond individuals with terminal illness.¹⁷

Because of the Dutch system’s commitment to transparency, summaries of most cases of psychiatric EAS are available online⁷ (Box 1). Our study sought to address 2 questions. First, what are the clinical, personal, and social characteristics of persons who receive EAS for psychiatric conditions? Second, how are the rules that regulate such EAS cases (Box 2) applied by physicians and by the Dutch regional euthanasia review committees?

Box Section Ref ID Box Section Ref ID

We reviewed all online EAS summaries identified by the Dutch regional euthanasia review committees (RTE) as psychiatric cases that were available as of June 1, 2015. At that time, there were 85 reported cases of psychiatric EAS mentioned on the RTE website (https://www.euthanasiecommissie.nl/oordelen/) for the years 2011 to 2014: 13 cases in 2011, 14 in 2012, 42 in 2013, and 16 in 2014 (the final number for that year was not available at that time), with 66 of those cases published online. After completion of our study, the total number of psychiatric EAS cases for 2014 was reported on October 7, 2015, as 41 patients,¹⁸ bringing the total for 2011 to 2014 to 110. The RTE has changed their publication practice (Box 1), resulting in only one more case from 2014 being published. Translations were obtained through the National Institutes of Health Library’s translation services, which uses companies to provide certified medical translations. Subsequent questions about specific passages were addressed by a Dutch-speaking member of the research team (R.G.D.V.), who further conferred with native Dutch-speaking academics.

The case reports were analyzed using directed content analysis.²² The coding scheme was developed iteratively by a bioethicist-psychiatrist (S.Y.H.K.) and a consultation psychiatrist (J.R.P.) as they independently read the reports, repeatedly comparing variables of interest in light of the main research questions of the project. The former read and coded all the reports, and the latter confirmed the coding by reading through the reports again. Discrepancies were resolved by discussion. The data were analyzed using statistical software (SPSS, version 21.0; IBM Corp). Analysis consisted of frequencies and ad hoc cross-tabulations, without hypothesis testing given the descriptive goals of the study.

Seventy percent (n = 46) of patients were women (Table 1). Thirty-two percent (n = 21) were 70 years or older, 44% (n = 29) were 50 to 70 years old, and 24% (n = 16) were 30 to 50 years old. Fifty-two percent (n = 34) had made suicide attempts, and 80% (n = 53) had been psychiatrically hospitalized in the past. Many had multiple suicide attempts or admissions.

Most patients had more than 1 condition, with 36 having at least 2 conditions, 11 having at least 3 conditions, and 4 having at least 4 conditions (Table 2). Depressive disorders were the primary psychiatric issue in 36 cases (55%). Eight cases with depression had psychotic features. Therefore, 17 of 66 patients (26%) had some form of psychosis. Posttraumatic stress disorder–related and other anxiety disorders were prominent, occurring in 28 of 66 patients (42%). Cognitive impairment was present in 4 patients, one of whom (case number 2014-83 from RTE case summaries) had a legal guardian but was judged competent by 2 independent consultants, including a psychiatrist. Four women had a long-term eating disorder, in addition to borderline personality disorder.

The nature of symptoms and suffering varied. Some patients with chronic, severe, difficult-to-treat depressions had received repeated electroconvulsive therapy (ECT) treatments. One patient (case 2012-26) underwent experimental deep brain stimulation (DBS), and a patient (case 2013-04) with obsessive compulsive disorder also received DBS. On the other hand, a woman in her 70s without health problems (case 2011-120044) and her husband had decided some years before that they would not live without each other. She experienced life without her husband, who had died 1 year earlier, as a “living hell” and “meaningless.” A consultant reported that this woman “did not feel depressed at all. She ate, drank, and slept well. She followed the news and undertook activities.”

The patients’ psychiatric conditions were chronic. In 10 patients (15%), the duration of their illness was described qualitatively (“years,” “decades,” or “longstanding”). In the remaining cases, only approximations were possible. The psychiatric history was approximately 5 years or less in 5 patients (8%), approximately 6 to 10 years in 6 patients (9%), approximately 11 to 30 years in 27 patients (41%), and longer than 30 years in 18 patients (27%).

Fifty-two percent (34 of 66) of patients had personality-related problems, sometimes without a formal diagnosis but indicating significant effect on the EAS evaluation (eg, “damaged development,” resulting in “low tolerance for frustration” and “reduced ability to…cope” [case 2014-77]). Personality disorders were more common in individuals 60 years or younger (66% [44 of 66] vs 41% [27 of 66], P = .05 by Fisher exact test).

Thirty-eight patients (58%) had at least 1 comorbid medical condition, 22 patients (33%) had at least 2 comorbidities, and 12 patients (18%) had at least 3 comorbidities. The comorbid conditions included the following: cancer, suspected malignancy, chronic obstructive pulmonary disease, cardiac disease, diabetes mellitus, stroke, prior brain tumor surgery, arthritis, orthopedic problems, chronic fatigue, fibromyalgia, migraines, neurological disorders (stroke, Meniere disease, pain syndrome, Parkinson disease, diaphragm paralysis, or gait disturbance), pancreatitis, medical complications of severe weight loss, vision loss, hearing loss, incontinence, and decubitus or other ulcers.

The case reports contained little social history. They often mentioned family members in passing, but we could not reconstruct anyone’s immediate family structure. Marital status, occupation, education level, race/ethnicity, and nationality were rarely mentioned. In 37 patients (56%), the reports mentioned the patients’ social isolation or loneliness, some with striking descriptions such as the following: “The patient indicated that she had had a life without love and therefore had no right to exist” (case 2012-46), and “The patient was an utterly lonely man whose life had been a failure” (case 2013-21).

Twenty-six patients (39%) received ECT at some point. In 7 cases (11%), monoamine oxidase inhibitor treatment was mentioned (or implied when the report explicitly said all medications in the Dutch Psychiatric Association Guidelines²¹ had been tried). Although most patients had extensive treatment histories, 37 (56%) also refused at least some treatment, because of no motivation in 18 cases, concern about adverse effects or risk of harm in 12 cases, and doubts about efficacy in 10 cases (some gave multiple reasons).

The circumstances of refusal varied. In 2 patients who had clearly undergone very extensive treatments, one (case 2012-20) rejected nonstandard treatment (DBS), and the other (case 2012-26) decided to stop it after 1 year. It was common for a personality disorder to have a role in refusals. Patients refused a variety of treatments, including ECT, medications, and various psychotherapies.

Twenty-one patients (32%) had been refused EAS at some point. In 3 patients, the physicians changed their minds and later performed EAS. In the remaining 18 patients, the physician performing the EAS was new to the patient. In 14 cases, the new physician was affiliated with the End-of-Life Clinic, a mobile euthanasia practice. There was one additional case involving the End-of-Life Clinic, for a total of 15 End-of-Life Clinic cases. The time from the first meeting with the clinic’s physician to death was 3 weeks in one case (the case not meeting legal due care criteria), less than 3 months in 7 cases, and 5 to 12 months in 7 cases. The End-of-Life Clinic cases increased, representing 1 of 12 cases in 2012, 6 of 32 cases in 2013, and 8 of 16 cases in 2014.

In 27 cases (41%), the physician performing EAS was a psychiatrist (Table 3), and the rest were usually general practitioners. In half of the cases, more than 1 official EAS consultant was involved, and all official consultants except one were Support and Consultation on Euthanasia in the Netherlands (SCEN) physicians (Box 1). Psychiatrists served as one of the official independent EAS consultants in 39 cases (59%). Consultation with an independent psychiatrist as the EAS consultant or as a second opinion occurred in 59 cases (89%). In 7 cases (11%), no independent psychiatric expert was involved, and in 5 of these cases, the EAS physician was not a psychiatrist. In 4 of these 5 cases, psychiatric input came from clinicians already involved in the patient’s care.

There were disagreements among the physicians in 16 cases (24%). There was one disagreement about the unbearable suffering criterion. The remaining disagreements were about competence (8 of 16) and futility (13 of 16) (cases could have more than 1 reason). In a few cases, disagreement was provisional (the first consultant, a general practitioner, did not believe that the due care criteria were met and recommended a second, specialist consultation), but EAS proceeded with the disagreements unresolved for most cases. In 8 cases, a psychiatrist consultant believed that the due care criteria were not met, while a primary care consultant believed that the criteria were met. In 7 of these 8 cases, the EAS physician was a psychiatrist.

Among all 110 psychiatric EAS cases reported to the RTE, the RTE found that the due care criteria were not met in only one patient (1%), a woman (case 2014-01) in her 80s with chronic depression who sought help from the End-of-Life Clinic. The clinic physician met with her 2 times (the first time was 3 weeks before her death), and the patient was not alone on both occasions, with family members present. The physician was not a psychiatrist, did not consult psychiatrists, was unaware of the Dutch Psychiatric Association Guidelines,²¹ and yet “had not a single doubt” about the patient’s prognosis. The consultant in the case, a SCEN general practitioner, agreed with the physician that all due care criteria were met.

In another case, the RTE was critical yet judged that the physician acted with due care. The patient (case 2013-27) had attempted suicide, which led to a broken thigh. The patient refused all treatments and requested EAS. The RTE was “puzzled” by the fact that this physician “complied with the patient’s [EAS] wish almost at once” and criticized the physician for prematurely opting for the EAS evaluation because the RTE could “not exclude the possibility that the patient might yet have accepted treatment….” However, the RTE ultimately decided that the case met the due care criteria “at the moment” the euthanasia was implemented.

The mean number of words (in Dutch, excluding abstracts) per report declined yearly between 2011 to 2014 (from 1573 words, to 1248 words, to 1154 words, to 1117 words, respectively). The assessment section of the case report—which discusses whether the notifying physician’s actions conform to the due care criteria—used language without any case-specific elements in 43 reports (65%). In the 7 cases without independent psychiatric opinion, the assessment section addressed that issue in 3 cases. In 16 cases with physician disagreements, the RTE specifically addressed the disagreement in their assessment in 2 cases.

The RTE exercised case-specific flexibility. For example, although the RTE’s stated view is that the intervening time from EAS consultation to death should be less than a “few weeks,”⁷ a lag of 3 months without a revisit by a consultant in one case (because of a vacation) was deemed acceptable owing to case-specific reasons (case 2013-09).

A sociodemographic characterization of Dutch psychiatric patients receiving EAS proved difficult because data on education level, occupation, marital and family status, ethnicity and nationality, and race were lacking. However, a striking finding is that the ratio of women to men was 2.3 to 1, which is the reverse of the suicide ratio of women to men in the Netherlands²³ and almost identical to the ratio of women to men attempting suicide.²⁴ It also contrasts with the ratio of 43% of women to 57% of men among Dutch EAS recipients overall.²⁵ It is possible that the availability of EAS renders the desire to die in women psychiatric patients more effective. This interpretation is consistent with the fact that most patients in the present study had previous suicide attempts, and the request for EAS followed a suicide attempt in several instances.

Although the ethical arguments concerning EAS for psychiatric disorders generally focus on otherwise healthy persons with severe treatment-refractory depression,^4,5,12 the reality is more complicated. First, although depressive disorders were indeed the most common problem, there were many other psychiatric conditions, including psychotic disorders, cognitive impairment, eating disorders, and prolonged grief, among others. Second, even among those with depression, the typical person had at least 1 of the following characteristics: age 70 years or older, at least 1 comorbidity, physical dependence or institutionalization, or prominent personality disorder or problem. Among 29 persons whose primary psychiatric issue was nonbipolar depression, 25 had one of the above cofactors. Therefore, the patients we studied were only somewhat younger than Dutch EAS recipients overall,²⁵ and 61% (40 of 66) of those we studied were 60 years or older. The findings appear consistent with a 1997 study.¹¹ Despite their open attitude toward EAS, Dutch physicians may be self-regulating to limit EAS to such complex cases, or it may be that psychiatric patients with those features may disproportionately seek EAS. A recent study¹³ of 100 consecutive persons requesting psychiatric EAS referred to one Belgian psychiatrist showed that most of her patients were women, with high rates of depression (58%) and personality disorders (50%). However, they were much younger than the Dutch patients we studied (only 6% were >70 years, 59% were <51 years, and 11% were <31 years), with a lower rate of comorbidity (23%) and a surprising 19% with autism spectrum disorder. Although any comparisons are tentative—the Belgian report describes requesters referred to a single psychiatrist, rather than recipients of EAS in an entire jurisdiction—it appears that the Belgian psychiatrist attracted younger psychiatric patients with fewer comorbidities.

The Dutch practice of EAS is regulated by a set of broad criteria. Applying some of these criteria to persons with terminal illness (cancer accounts for >83% of reported EAS in the Netherlands²⁵) arguably requires less judgment than in psychiatric cases because the eventual prognosis of individuals with terminal illness is not in question. For psychiatric cases, one might expect more variability in judgments given the potential effect of some neuropsychiatric conditions on decision-making capacity^15,26,27 and the more complicated determinations of medical futility that must incorporate patients’ treatment refusals in the context of less-than-certain prognosis even among persons with treatment-resistant depression.^28,29 The variability in physician judgments may be reflected in the present study in that almost one-third (21 of 66) of the patients were refused EAS and almost one-quarter (16 of 66) of the cases engendered disagreements among the physicians involved. In 7 cases, the physicians performing EAS apparently perceived the need to seek 3 official EAS consultations (the law requires one consultation), and there were 3 or more physicians (in various roles, not counting the EAS physician) involved in the evaluation in one-third (22 of 66) of the cases.

Only one of 110 psychiatric EAS cases reported to the RTE during 2011 to 2014 did not meet the due care criteria. Four of all 5306 EAS cases (0.1%) in 2014 were judged as not meeting the due care criteria.¹⁸ Furthermore, although the RTE often cites the Dutch Psychiatric Association Guidelines,²¹ it accepts practices less strict than the guidelines (but consistent with the RTE’s Code of Practice³⁰). There were no official EAS consultants who were psychiatrists in 29 of 66 cases (41%), and there were no independent psychiatrists involved as EAS consultants or second-opinion consultants in 7 of 66 cases (11%). When consultants disagreed, the RTE deferred to the opinion of the treating psychiatrists.

The primary limitation of our study is that because the RTE reports are intentionally written in “plain language,”³¹ there is a limit to what can be inferred clinically. Furthermore, although we focused on variables likely to be reliable and valid, the results rely on coding judgments and approximations of quantities that are sometimes described imprecisely. Because the publication practice of the RTE changed in 2014, the results cannot be generalized to the entire period from 2011 to 2014, although we captured a window during which most cases were published. Nevertheless, unpublished cases may be less controversial. Furthermore, the results may not generalize to other countries that allow euthanasia for mental disorders⁶ because the reporting compliance rate and review procedures,^32,33 as well as the availability of mental health services and health insurance, may be different from those in the Netherlands.

Despite some limitations, an important strength of our study is that we examined reports of actual psychiatric EAS cases across an entire jurisdiction, rather than asking physicians to recollect their experiences or opinions. The results show that the patients receiving EAS are mostly women and of diverse ages, with various chronic psychiatric conditions, accompanied by personality disorders, significant physical problems, and social isolation or loneliness. Refusals of treatment were common, requiring challenging physician judgments of futility. Perhaps reflecting the complexity of such situations, the physicians performing EAS generally sought multiple consultations (but not always), and disagreement among physicians—especially regarding competence and futility—was not unusual. Despite these complexities, a significant number of physicians performing EAS were new to the patients. We conclude that the practice of EAS for psychiatric disorders involves complicated, suffering patients whose requests for EAS often require considerable physician judgment. The retrospective oversight system in the Netherlands generally defers to the judgments of the physicians who perform and report EAS. Whether the system provides sufficient regulatory oversight remains an open question that will require further study.

Correction: This article was corrected on April 6, 2016, to fix an error in the text and Table 2.

Submitted for Publication: September 28, 2015; final revision received November 12, 2015; accepted November 14, 2015.

Corresponding Author: Scott Y. H. Kim, MD, PhD, Department of Bioethics, National Institutes of Health, 10 Center Dr, Room 1C118, Bethesda, MD 20892 (scott.kim@nih.gov).

Published Online: February 10, 2016. doi:10.1001/jamapsychiatry.2015.2887.

Author Contributions: Dr Kim had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: All authors.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: All authors.

Critical revision of the manuscript for important intellectual content: All authors.

Administrative, technical, or material support: All authors.

Conflict of Interest Disclosures: None reported.

Funding/Support: The research costs of this study were supplied by the Intramural Research Program, Department of Bioethics, National Institutes of Health (Dr Kim).

Role of the Funder/Sponsor: The funding source had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Disclaimer: The opinions expressed in this article are the authors’ and do not represent the views or policies of the National Institutes of Health, the Department of Health and Human Services, or the US government.

Additional Contributions: Christine Grady, PhD (National Institutes of Health), Sam Doernberg, BS (National Institutes of Health), and Eric Caine, MD (University of Rochester), provided comments on early drafts of the manuscript. No compensation was provided.