CONCEPTS AND TERMS IN PHYSICIAN-ASSISTED DEATH
While laws differ slightly across states, all require that patients requesting physician-assisted death satisfy three criteria: (1) terminal illness through a prognosis of having 6 months or less to live, (2) competence and intact judgment, and (3) voluntariness. Speakers and participants highlighted the challenges to and opportunities for improving how these criteria are defined and operationalized in the clinical setting.
Terminal Illness and the 6-Month Prognosis
Director, Center for Elder Care and Advanced Illness
Laws on physician-assisted death, as well as access to hospice, require a patient to be terminally ill, as defined by having 6 months or less to live. Joanne Lynn, director of the Altarum Institute’s Center for Elder Care and Advanced Illness, explained that the 6-month criterion is not based on empirical evidence and arose in a relatively haphazard manner when the U.S. Congress was establishing eligibility for hospice services. The ambi-
guity of the 6-month criterion means it is difficult to apply clinically—it could mean that a person is nearly certain to die within 6 months, is very likely to die, or that 51 percent of people with a similar condition will be dead within 6 months.
Even with a terminal diagnosis for a single, dominant illness (e.g., cancer), death rarely follows a highly predictable course on a set time line, said Lynn. Three-quarters of Americans will die from a long-term, serious debilitating illness that can take 2 to 5 years to result in death, Lynn noted. Lynn further explained that the lead time for knowing someone has 6 months or less to live is usually a few weeks to a month for a person with cancer when his or her health worsens precipitously, or the lead time could be a matter of a few weeks or as little as a few hours for someone with a long-term debilitating illness who dwindles over the course of 1 to 2 years. Lynn stressed that it is unclear who is meant to be included or excluded through the application of the 6-month prognosis criterion for access to physician-assisted death. For individuals with a dwindling course of death due to a potential constellation of illnesses (e.g., dementia, Parkinson’s disease, frailty, strokes, amyotrophic lateral sclerosis [ALS], or organ system failures), it is more difficult to predict when death will occur and, therefore, if and when these patients could gain eligibility for hospice care or physician-assisted death, said Lynn. Eligibility for physician-assisted death for a large proportion of the population will vary remarkably based on how the 6-month prognosis is operationalized (e.g., nearly certain to die within 6 months, very likely to die within 6 months, or more likely than not to die within 6 months), explained Lynn. Without predictive models, which Lynn said no organization is funding the development of, declaring a prognosis of 6 months will continue to be based on intuition rather than science. Lynn challenged the participants to consider what is really meant by the 6-month prognosis and why we, as a society, are unwilling to make this criterion more precise.
Competency, Decision-Making Capacity, and Voluntariness
Senior Investigator, Department of Bioethics
National Institutes of Health
Scott Kim, a senior investigator in the Department of Bioethics at the National Institutes of Health, began his presentation by agreeing with other speakers (see Daniel Sulmasy’s presentation later in this chapter), that there is difficulty in achieving true objectivity and balance in considering physician-assisted death, even for an organization such as the National Academies of Sciences, Engineering, and Medicine.
Regarding competency, or the capacity of an individual to make decisions on her or his own behalf, Kim said that “decision-making capacity” is a term used in most health care laws and also in laws on physician-assisted death. Assessments of decision-making capacity are based on a set of functional criteria—for example, understanding, reasoning, appreciation—that is reflected in most states’ laws (Grisso and Applebaum, 1998; Kim, 2010). Every adult is presumed to be competent unless there is a good reason to justify an assessment, such as knowing that a person has advanced dementia. Assessing competency, whether by a clinician or a judge, must account for the context of the decision and the seriousness of the decision’s consequences, Kim added. He noted that there are few systematic data on how to assess decision-making capacity in the context of physician-assisted death.
One challenge in determining competency is that clinical reality does not always correspond with the clear legal constructs, said Kim. For example, some laws provide a nearly tautological or empty definition of incapacity as lacking the ability to make and communicate health care decisions, or as Kim put it, “incapacity means you are incapable, which does not guide doctors and judges all that much.” Yet, a survey of psychiatrists in Oregon found that only 6 percent felt very confident that in a single evaluation they could adequately assess whether a psychiatric disorder was impairing the judgment of a patient requesting physician-assisted suicide (Ganzini et al., 1996). The same investigators found that psychiatrists’ own ethical views of physician-assisted death may influence the level of scrutiny used in their assessments (Ganzini et al., 2000), and Kim and his colleagues found that two-thirds of consulting psychiatrists found decision-making evaluations to be more challenging than other types of evaluations they perform (Seyfried et al., 2013).
Kim and others have found that impaired decision-making capacity is, in fact, common at the end of life (Silveira et al., 2010). One study found that nearly half of older, terminally ill cancer patients failed a measure of capacity, with the investigators stating that “without thorough and reliable evaluation methods, doctors may fail to recognize decision-making impairments even when the impairments are pronounced” (Sorger et al., 2007). Other studies have shown that cognitive impairment is relatively common in patients with ALS, with the authors of one recent paper finding 40 percent of ALS patients in the study to have cognitive impairment (Rabkin et al., 2016).
In perhaps the most thorough study of hospice patients (both inpatients and outpatients) with no documented or clinically obvious impairments or cognitive disorders, the researchers found that 54 percent of these individuals had a significant cognitive impairment (Burton et al., 2012), said Kim. Moreover, these individuals did significantly worse than
other study participants on the decision-making capacity measure used in the study. A small study in the Netherlands found that 23 percent (5 of 22) of the people who requested physician-assisted death had psychiatric symptoms which decreased their competence in decision making. Of those who were in fact not competent, two out of five had been judged to be competent by their primary physicians (Bannink et al., 2000).
The reliability of assessments of decision-making capacity can be questionable as well, Kim said. Reliability can be high when all of the people conducting the evaluations share the same training background (e.g., psychiatry) and risk–benefit frameworks (Cairns et al., 2005), but reliability falls when the evaluators have different training backgrounds (Armontrout et al., 2016). Reliability is also low for patients in the middle of a distribution of the degree of impairment (Kim et al., 2011) compared with evaluation of patients in the tails of the distribution (Etchells et al., 1999). Reliability is also low in unsettled or novel areas of decisional capacity assessment, such as in research consent or in the context of physician-assisted death. Kim said that there are disagreements among physicians in the Netherlands in assessing the decision-making capacity of psychiatric patients requesting physician-assisted death (Doernberg et al., 2016; Kim et al., 2016). He noted that contrary to expectation, the level of scrutiny and the threshold for declaring incapacity in non-terminally ill psychiatric patients requesting physician-assisted death in the Netherlands is surprisingly low.
According to Kim, it is unknown how strong a presumption of capacity is currently being used in the United States or what thresholds are being applied in assessments of decision-making capacity for physician-assisted death. Are evaluators using a checklist to determine if a patient makes certain statements, or is there an in-depth clinical interview probing the person’s understanding? Is a stand-alone community physician conducting the assessment on his or her own with little peer oversight, or is the assessment part of a larger institution’s more systematic procedures with greater accountability and transparency? Also, there is a natural flow of referrals to low-threshold evaluators (people who tend to say yes rather than no)—is that desirable or undesirable, and how often does it happen? Kim suggested that all of these issues could benefit from additional research. Kim’s conclusion was that a decision-making competency assessment would require more than just a conversation.
Maddie Katz Professor of Palliative Care Research and Education
University of California, Los Angeles
In California, an individual wishing to take advantage of the new law must be referred to a “mental health specialist” if he or she shows
any signs of a “mental disorder.” The California law does not define mental disorder other than allude to the Diagnostic and Statistical Manual of Mental Disorders (DSM), said Thomas Strouse, the Maddie Katz Professor of Palliative Care Research and Education and medical director at the University of California, Los Angeles (UCLA), Resnick Neuropsychiatric Hospital and the UCLA Health System Palliative Medicine Service. The current version of the DSM defines mental disorder as “a syndrome characterized by clinically significant disturbance in an individual’s cognition, emotion regulation, or behavior that reflects a dysfunction in the psychological, biological, or developmental process underlying mental functioning” (American Psychiatric Association, 2014, p. 20).
California law defines a mental health specialist as either a psychiatrist or licensed psychologist. In Strouse’s view, the mental health specialist has two tasks under the law: to review the referring physicians’ determination of whether the patient has the capacity to make a medical decision, act voluntarily, and make an informed decision; and to discern whether the patient is suffering from impaired judgment and whether that impaired judgment results from a mental disorder. “To do that, you have to first decide whether a mental disorder is present, then evaluate for impaired judgment, and then try to causally link those two things,” Strouse said. “Again, no recipe, no guidelines, no standards that I know of to assist us in doing that.” As examples of what he thinks might qualify as impaired judgment caused by a mental disorder, Strouse created four fictional cases:
- A severely depressed patient who has delusions of deserving punishment or death.
- A patient with advanced dementia who cannot recognize the impact of his or her behavior on others.
- A chronically hypomanic patient with recent impulsivity and an inability to appreciate the consequences of most actions.
- A polysubstance-dependent patient who has been impaired enough in recent years by addictions so as to have been unable to make other important life decisions.
Strouse said that his general sense is that mental health referrals related to the California End of Life Option Act (EOLOA) are made out of an abundance of caution and that there has been a learning curve with regard to referrals. There were more referrals early in the first year after California’s law went into effect than there have been in the past 6 to 8 months. Often, the patients he has seen were referred because they had a history of mental disorder or were taking a maintenance dose of an antidepressant. In most cases, these individuals did not have demonstrable
active or symptomatic mental disorders when he or his colleagues evaluated them, and impaired judgment caused by a mental disorder has not been a disqualifier for these patients.
California law does provide a definition of an informed decision as it pertains to physician aid-in-dying, Strouse said, but voluntariness is an area that needs research in order for the concept to be better defined as it pertains to physician-assisted death. Most people would agree, he said, that voluntariness requires two conditions: intentionality and freedom from controlling influences. Coercion, he explained, is generally construed to mean undue influence by another person or entity. Some scholars, though, propose that the illness itself is coercive (Garrison, 2007). Strouse suggested that there might be an important difference between voluntariness to consent to a proposed procedure, such as an emergency appendectomy, and voluntariness to request a desired procedure (Nelson et al., 2011).
While there is reason to be concerned that a single 1- or 2-hour evaluation might not meet a comfortable standard for assessing capacity, Strouse said he is reassured by the fact that these patients have been followed closely by a large network of caring, attentive people who have done extensive individual and family psychosocial assessments long before they are referred to him. “In many settings, and certainly in ours, it is a much richer fabric from which we draw information that leads to decisions,” he said.
Regarding capacity versus competence, Strouse said that health professionals would generally assert that they are conducting clinical evaluations of capacity, with competency being the term used when the courts become involved, particularly to assign a proxy decision maker where lack of competency is adjudicated. So far, he said, California’s courts have not yet become involved in capacity evaluations for end-of-life decisions, but he expects that might happen eventually.
The components of a capacity assessment, he said, include an assessment of functional abilities, an assessment for the presence or absence of psychopathology, an evaluation of the complexity of the task demand at hand, and an assessment of the understanding of the consequences of the decision. Reassessments are also conducted in order to confirm original evaluations. When it comes to assessing the abilities needed for capable decision making, mental health professionals rely on a substantial body of research that has identified four essential components: (1) understanding treatment information, (2) appreciating the significance of that treatment information for one’s own situation, (3) displaying the ability to reason with relevant information, and (4) demonstrating a logical weighing of options and expressing or communicating a durable choice (Berg et al., 1996; Eckstein and Kim, 2017). Several instruments exist to perform capacity assessments, and the relative performance of these instruments
has been evaluated (Sessums et al., 2011). However, Strouse said, these tools were organized or conceived around affirmatively offered proposed treatments, and a patient-initiated request for physician aid-in-dying is a different circumstance.
In Strouse’s opinion, there are several reasons why physician aid-in-dying might need to abide by different standards for evaluating capacity than are applied to other medical decisions. In physician aid-in-dying, the patient’s goal is death, while with other medical decisions the patient’s goal is health. By law, patients must initiate the discussion about physician aid-in-dying, while the clinician usually initiates the discussion about most other medical procedures. In addition, for physician aid-in-dying, the law specifically outlines the process for determining a patient’s capacity, mandates reporting and sign-off by a consulting physician, and requires a consulting mental health professional in the case of a mental disorder—none of which are required for other medical procedures.
In response to a question from a workshop participant about whether he believes that the number of referrals for mental health evaluations in California has been too low, Strouse said that the referral rate is currently about 5 to 6 percent, approximately the same as in Oregon. Given that the prevalence of mental disorders in an older population is approximately 20 to 25 percent, he acknowledged that the referral rate might be too low and suggested that additional research is needed to understand the decisions that physicians are making in terms of patients’ mental health and competence. Pointing to another evidentiary gap, David Orentlicher, the Cobeaga Law Firm Professor and co-director of the Health Law Program at the University of Nevada, Las Vegas, added that the same type of research should be conducted on withdrawal of treatment, given that there may be the same issues regarding competence and potential coercion.
Comparative Analysis of Legal Rules: Withdrawal of Treatment Versus Physician-Assisted Death
Cobeaga Law Firm Professor
University of Nevada, Las Vegas
Eight U.S. jurisdictions (California, Colorado, Hawaii, Montana,1 Oregon, Washington, Washington, DC, and Vermont), as well as Swit-
1 A 2009 Montana Supreme Court decision ruled that state law protects Montana physicians from prosecution for helping terminally ill patients die. See Baxter v. Montana, 224 P.3d 1211 (2009). This information was added after prepublication release.
zerland, have authorized physician-assisted death, in which a physician prescribes a lethal dose of medication that the patient self-administers, said David Orentlicher, the Cobeaga Law Firm Professor and co-director of the Health Law Program at the University of Nevada, Las Vegas. Belgium, Canada, Luxembourg, and the Netherlands also allow this practice, he said, as well as allowing physicians to administer the lethal dose of medication, a practice also known as euthanasia. Every U.S. state, as well as many if not most countries, allows the withdrawal of treatment, such as discontinuing ventilator-assisted breathing, dialysis, or other health care necessary to sustain a patient’s life.
Orentlicher explained that there are two types of legal rules regarding end-of-life practices: Who is eligible for the death-hastening practice? And what process is required to qualify for the death-hastening practice? At first glance, he said, it seems that the law views the withdrawal of treatment very differently than physician aid-in-dying. Closer examination, though, reveals that there are more similarities than differences between these two practices.
Regarding who is eligible for these practices, there is little restriction when it comes to the withdrawal of treatment. A competent patient has an open-ended right to refuse any treatment regardless of the patient’s prognosis or type of care. Withdrawal of treatment is generally permitted for incompetent patients, but living will laws may require a “terminal condition”2 and no pregnancy. By comparison, Orentlicher said, there are significant limits for who is eligible for physician-assisted death, including decision-making capacity, the ability of the patient to perform the life-shortening act, a life expectancy of 6 months or less, and the requirement that the patient be a resident of the state where the practice is legal.
The process of qualifying for life-ending actions also differs between the two practices. For the withdrawal of treatment in the case of a competent patient, some courts require confirmation by two independent physicians of the patient’s prognosis and that the patient has decision-making capacity. For incompetent patients, Orentlicher said, every state allows withdrawal of treatment when clear and convincing evidence exists, such as from a living will or the patient’s discussions with family members, that the patient would have wanted treatment to be withdrawn under the circumstances. When clear and convincing evidence is absent, the law in some states allows the family to decide, while other states vary their rules
2 “Terminal condition” for purposes of withdrawal of treatment for incompetent patients is a rather broad definition that requires an incurable and irreversible condition that will result in death in a short period of time, with or without the administration of life-sustaining treatment. For instance, a patient with insulin-dependent diabetes would die within a short period of time without insulin.
depending on the patient’s prognosis, which, Orentlicher commented, starts to look more like aid-in-dying. For example, when a patient has a terminal illness, laws generally allow families to authorize the withdrawal of treatment. However, when the patient is not terminally ill (e.g., neurologic injury), the states are more likely to err on the side of life and require treatment to continue. In contrast, physician-assisted death requires an independent physician to confirm the diagnosis, prognosis, capacity, and genuine consent and may also require a psychological examination. The laws in every aid-in-dying state also require multiple disclosures to the patient, as well as three requests, two oral and one written, by the patient over a 15-day period.
Why the difference in rules between the withdrawal of treatment and physician-assisted death? “Is it because there are meaningful moral differences between ‘passive’ and ‘active’ practices that hasten death,” Orentlicher asked, “or do the different sets of rules reflect concerns about how to ‘operationalize’ the relevant moral principles?” In his opinion, he said, the differences are about operationalizing more than about meaningful differences between the two practices. “I think what drives end-of-life law is the desire to allow relief of suffering from serious and irreversible disease,” he said, an opinion that he said was based on reading the normative arguments in court decisions, articles in the medical and legal literature, and statements by religious organizations about why it is okay to refuse life-sustaining treatment. In Orentlicher’s view, end-of-life law is designed so that patients can choose a quicker death when they are suffering greatly from serious and irreversible illness.
Orentlicher noted that one approach to turning that moral principle into law, which Belgium and the Netherlands have taken, is to simply have the law follow the moral principle—if a patient is suffering greatly from serious and irreversible illness, the patient would be allowed to die. What would matter is why the individual wants to die (e.g., unbearable suffering), not how the death happens (e.g., withholding therapy, taking a lethal dose of medication, or having a physician administer the lethal dose), said Orentlicher.
For context as to why the approach of the Netherlands and Belgium has not been adopted in the United States, Orentlicher reminded the workshop participants that in the 1960s and 1970s, people disagreed about the right to have life-sustaining treatment withdrawn. In the Quinlan case, the court ruled in 1976 that treatment could be withdrawn when a person had a “dim prognosis.” This became the key moral principle—that the patient be suffering from serious and irreversible disease. To Orentlicher, the right to refuse treatment in 1976 looked much like the right to physician-assisted death today. Physician-assisted death currently is permitted only for terminally ill patients; withdrawal of treatment in
1976 was permitted only for patients with a serious and irreversible illness. Treatment withdrawal law followed directly from this moral principle, explained Orentlicher.
The problem with that way of operationalizing the moral principle of allowing relief of suffering from serious and irreversible disease, said Orentlicher, is that the government is in the position of deciding who must live and who may die based on judgments about the patient’s quality of life, which is not the kind of power government should exercise. As a result, the courts abandoned the Quinlan standard of a “dim prognosis,” and the law now allows any patient, regardless of diagnosis or prognosis, to refuse treatment. It is up to the patient to weigh considerations about quality of life and length of life, Orentlicher said. He explained that this legal standard is acceptable in terms of the moral principle that you can hasten death when seriously and irreversibly ill because the typical withdrawal of treatment scenario involves a patient who is suffering from a serious and irreversible illness.3 In other words, he said, the legal rule does a good job of operationalizing the underlying moral principle.
In Orentlicher’s view, the law’s distinction between withdrawing treatment and physician-assisted death represents an important moral difference. The distinction provides a legal proxy to sort the morally justified death from the morally unjustified death. To explain the concept of legal proxy, he offered the example of a legal proxy in the form of speed limits. In the case of driving, the goal is to have people drive safely. The law could allow drivers to drive at any speed, as long as it is a safe speed. But drivers and police officers may have different views about safe and unsafe speeds, and different police officers would also vary in their views of the appropriate speed. Thus, to avoid problems with enforcing speed limits, Orentlicher continued, a specific limit is set, while recognizing that it is not a perfect reflection of a moral principle but rather a good approximation.
Because legal proxies are approximations of a moral principle, they may need refinement over time, Orentlicher said. He further explained that people have not changed their thinking as to when it is acceptable to hasten death—society still holds that it is appropriate only when a patient is suffering from serious and irreversible illness. What has changed is the way society turns that moral principle into legal rules, he said. The legal distinction between treatment withdrawal and aid-in-dying, he explained, has provided a useful proxy to sort the morally justified death from the morally unjustified death and avoid the need for case-by-case judgments. The typical withdrawal of treatment involves a patient suffering from serious and irreversible illness, while many suicides involve people suf-
3 Other refusals typically reflect religious belief and these refusals are not always respected.
fering from a depression that should be treated, Orentlicher said. But, there are some patients suffering from serious and irreversible illness who are not dependent on life-sustaining treatment. According to Orentlicher, death with dignity laws reflect the view that the distinction between treatment withdrawal and aid-in-dying does not do a good enough job of sorting between the morally justified and morally unjustified death. By allowing aid-in-dying only for terminally ill persons, said Orentlicher, the legal rules serve as better proxies for the principle that death-hastening actions can be chosen by people who are seriously and irreversibly ill. For Orentlicher, the requirement for terminal illness places a useful and effective legal limit on aid-in-dying, one that he believes is critically important, which is why every state with aid-in-dying laws uses the criteria. By restricting aid-in-dying to terminal illness, he said, society directly limits its access to people suffering from a serious and irreversible illness.
According to Orentlicher, the restriction to terminally ill patients is reasonable even if one believes that a right to choose aid-in-dying should rest simply on patient autonomy. The requirement of terminal illness limits the risk of “false positive” cases in which a person is seriously depressed but allowed to go forward with aid-in-dying because of the wrong conclusion that they are making a genuine expression of autonomy. Orentlicher said that he expects that more states will legalize physician-assisted death if the empirical evidence continues to be reassuring, which could also lead the Supreme Court to recognize a constitutional right to aid-in-dying.
Legal and Regulatory Landscape
Rose Kennedy Professor, Kennedy Institute of Ethics
In most jurisdictions, said John Keown, the Rose Kennedy Professor in the Kennedy Institute of Ethics at Georgetown University, criminal law prohibits a doctor from intentionally administering a lethal drug to terminate a patient’s life, even to end suffering (which he defined as “voluntary euthanasia” if the patient requested it and “non-voluntary euthanasia” if the patient was incapable of requesting it). It is also illegal in most jurisdictions for a physician to intentionally assist a patient to end his or her life by prescribing or providing a lethal drug, which he called “physician-assisted suicide.” He regarded the phrase “physician-assisted death” as both euphemistic and ambiguous: “We are not talking about assisting dying,” he said. “We are talking about either intentionally ending somebody’s life or intentionally helping them to end their own life.” Keown noted the ambiguity in the term “physician-assisted death”: some
use it to mean euthanasia and physician-assisted suicide; others used it to mean only physician-assisted suicide; and still others used it to include withholding or withdrawing life-prolonging treatment.
Keown noted that the 1997 Supreme Court rulings in Glucksberg4 and Quill5 drew a legal distinction between withdrawing treatment at a patient’s request and physician-assisted suicide. Chief Justice Rehnquist explained that whereas everyone was entitled to refuse unwanted treatment, no one was permitted to assist suicide and that a doctor who withheld or withdrew treatment need not intend to hasten death. The chief justice also cited several state interests justifying laws against physician-assisted suicide including the preservation of life; the prevention of suicide; avoiding arbitrary, unfair, or undue influence; and avoiding any future movement toward euthanasia and other abuses. The chief justice added that state interests went beyond protecting the vulnerable from coercion and extended to protecting disabled and terminally ill people from prejudice, negative and inaccurate stereotypes, and societal indifference.
Keown also noted that a number of legal scholars, including Yale Kamisar and Neil Gorsuch (now a Supreme Court Justice) have argued that it is important to keep the concepts of refusal of care and physician-assisted suicide separate (Gorsuch, 2006; Kamisar, 1958). Gorsuch, for example, has argued that the refusal of care is not logically equivalent to a right to hasten death and that to equate the two is to conflate two very different things, both morally and legally (Gorsuch, 2006).
Senior Investigator, Department of Bioethics
National Institutes of Health
Rose Kennedy Professor, Kennedy Institute of Ethics
In his presentation, Scott Kim, a senior investigator in the Department of Bioethics at the National Institutes of Health, also discussed the concept of the slippery slope, as had been requested by the workshop organizers. He discussed two types of slippery slope: first, the expansion of physician-assisted death within an accepted category of practice
4Washington v. Glucksberg, 521 U.S. 702 (1997).
5Vacco v. Quill, 521 U.S. 793 (1997).
(e.g., “terminally ill”); second, expansion in the categories of persons who can receive physician-assisted death (e.g., children, non-terminally ill, advance requests for physician-assisted death). Kim argued that data are needed to assess these two types of expansion, and he questioned whether the jurisdictions that permit physician-assisted death collect the types of data needed to evaluate how decisions are made regarding, for example, how strong a presumption of capacity is used, how terminal illness is determined, and who serves as the second opinion on those determinations. Such data are critical, said Kim, because there is no natural feedback loop to assess how the current laws are working, given that the procedure is final and the dead cannot provide feedback the way a person who has had a wrong limb amputated can.
Kim said that because physician-assisted death laws function primarily as protections for physicians (i.e., as protected exceptions to criminal prohibitions against homicides), no country or state has a monitoring system that can assess whether the two expansions in practice are actually occurring since all jurisdictions rely on the self-reporting of those performing the procedure. He added that retrospective reviews of physician self-reports by regional euthanasia review committees in the Netherlands have not proven to provide rigorous oversight (Miller and Kim, 2017) and that the large national studies in the Netherlands and Belgium will be informative but are mostly epidemiological and do not provide insight into decision making by doctors. In the United States, he noted, it is difficult to get funding to study physician-assisted death since most clinical researchers from academic medical centers are funded by the National Institutes of Health, which is organized by, and primarily interested in, specific disease areas.
Expansion of Practice
Kim said that the U.S. practice of physician-assisted death places the autonomy of the patient at the forefront—the physician writes a prescription, and it is the patient’s responsibility to fill the prescription and use it as he or she chooses, or if he or she chooses. Although the term “physician-assisted suicide” is contested by some, the term “suicide” (taking it only in the descriptive sense of self-caused death) brings out one particular fact about the U.S. system, Kim said—that it requires that the patient exhibit a high level of self-determination. Kim suggested that the term “physician-assisted death” obscures this fact because the term can cover both euthanasia and physician-assisted suicide, which leaves the role of the patient unspecified.
In Belgium, Canada, Luxembourg, and the Netherlands, Kim said, the distinction between physician-assisted suicide and euthanasia is blurred
and the important concept is “physician-assisted death by appointment” (although it is not commonly referred to by this term). In those four countries, the distinction between a physician-assisted death that is completed by injection and that is carried out by a patient ingesting a medication is mainly symbolic since all physician-assisted death is done by appointment. Kim explained that in the Netherlands, even if a patient makes an appointment for physician-assisted death with a stated preference for ingestion, the doctor brings a drink and the necessary materials for a lethal injection in case the ingestion does not work (or does not work fast enough). Self-ingestion no longer has the strong implication regarding self-determination that it does in the United States, Kim said, because it is the doctor who is providing the procedure in either scenario. Kim further suggested that this is why some Canadian institutions have the policy of providing only euthanasia (Li et al., 2017).
Expansion of Eligible Categories: Unbearable Suffering
The categories of those eligible for physician-assisted death have expanded, Kim said. Although virtually all discussions of physician-assisted death began as a debate over how to ease the process of dying (i.e., “how to die” rather than “whether to die”), the practice has tended to expand in terms of the categories of people who are considered as candidates, he said. This has occurred by (1) removing the terminal illness requirement and substituting unbearable suffering, (2) allowing children to be eligible (as in Belgium, Luxembourg, and the Netherlands), or (3) allowing physician-assisted death for those who are incompetent to make decisions through the mechanism of advance requests.
Focusing on the first possible expansion of categories, Kim said that expanding the availability of physician-assisted death to those who are suffering unbearably could in theory require intrusive quality-of-life judgments by physicians concerning whether a person is suffering enough. “There is something not quite right about having doctors have that kind of authority given by the state,” Kim said. In the United States, this question has been avoided by limiting physician-assisted death to terminal illness; no physician-assisted death law in the United States mentions a quality-of-life judgment or suffering requirements, he said. In jurisdictions that do use the suffering requirement, the problem of a state-mandated evaluation of suffering is evaded by using a subjective definition of unbearable suffering: unbearable suffering is based solely on the patient’s account, said Kim. Thus, in practice the unbearable suffering criterion has been reduced to an autonomy-based, strongly libertarian justification in which the individual’s preferences rule, Kim said.
Kim also described the implications of an expansion of physician-
assisted death by using the unbearable suffering criterion rather than the terminal illness criterion in terms of “red flags becoming green flags” phenomenon. For example, when the terminal illness category is in use, mental illness, despair, and hopelessness are a “red flag” that urges caution in allowing physician-assisted death. However, if a standard of unbearable suffering is used, mental illness, despair, and hopelessness become markers of unbearable suffering and thus a “green flag,” or part of the justification for physician-assisted death.
Kim cautioned that the type of physician-assisted death practiced in Oregon—with terminal illness required for access and a strong focus on patient autonomy—represents a small minority (approximately 1 in 12 by rough estimate)—of physician-assisted deaths worldwide. Kim noted that even in the United States, there are signs that the advocacy for a terminal illness-based physician-assisted death is only a strategic one in which the eventual goal is a more expansive suffering-based system.
Referring to Lynn’s comments on the imprecision of the 6-month prognosis criterion, Kim defended the criterion as still being useful as a “natural backstop” for the difficult discussion of medical futility. He said that in Canada the attempt to limit physician-assisted death in the general realm of end-of-life practices by using the criterion of “reasonably foreseeable death” has led to a slippery slope. Kim described how some doctors in Canada are now openly using death within 10 years as a definition of “reasonably foreseeable death.” Kim further suggested that if the standard becomes unbearable suffering, medical futility becomes tied to social and health policy priorities in terms of what we are willing to devote to the treatment of various disorders. Unlike a terminal illness about which modern medicine can do nothing to stop eventual death, whether someone finds suffering from a non-terminal illness “intolerable and irremediable” will depend greatly on the quality and quantity of medical and social programs available to that person. To permit physician-assisted death for such persons is less resource intensive and less costly than making high-quality care universal, Kim said. In addition, he said, determining whether such a person qualifies for physician-assisted death will always be an implicit judgment by society about the worth of some lives over others. Kim suggested that considering these moral questions is important for society for various reasons, not least of all is to inform the kinds of data collected on the practice of physician-assisted death.
Practical and Logical Slippery Slopes
Keown argued that it is not feasible to have effective legal control of either voluntary euthanasia or physician-assisted suicide. He invoked two “slippery slope” arguments: the “empirical” and the “logical.” The former
suggests that it is not feasible either to draft or to enforce effective safeguards; the latter holds that safeguards (such as the requirement of a voluntary request or a “terminal illness”) are vulnerable to logical extension.
Keown referred to Capron’s conclusion that the safeguards in Oregon’s Death with Dignity Act are “largely illusory” (Capron, 1996). Capron pointed out that any doctor could act as one of the two practitioners who must approve the request; neither physician needs to have any prior knowledge of the patient or expertise in psychological evaluation, and any physician inclined to set up a practice specializing in physician-assisted suicide could become a magnet for terminally ill people from around the world. Moreover, the act does not require the second doctor to be independent of the first, so both could be partners in the same physician-assisted suicide practice. Furthermore, Keown said, the Oregon law relies on self-reporting after the fact by the physician involved, and the Oregon Health Authority has acknowledged that it cannot detect or collect data on issues of noncompliance with any accuracy. As Gorsuch observed, the Oregon law made reliable and relevant data and case descriptions difficult to obtain (Gorsuch, 2006).
Gorsuch had suggested that answers to the following questions were essential to understanding the effects of the Oregon law and necessary to provide a thoughtful assessment of the law’s worthiness for emulation elsewhere, but that there was little chance that these questions would be answered anytime soon given the many limitations that the law placed on the Oregon oversight agency (Gorsuch, 2006):
- What role was depression, as opposed to terminal illness, playing in patients’ decisions to die?
- Were alternative options, including treatment for depression, being fully presented, or presented at all?
- Were the doctors who prescribed death even knowledgeable about the alternatives?
- To what extent were family members unduly influencing patient choices and physician evaluations?
- Do physicians and psychologists have a duty to perform more than a cursory examination?
- Should prescribing physicians consult the patient’s primary care providers and other doctors who may have declined to provide a lethal prescription?
- Do health maintenance organizations have a conflict of interest, given that assisting suicide is undeniably cheaper than continuing care?
- How many cases were not reported, and how accurate were the reports that were filed?
Keown also discussed whether the principles underlying the Oregon law made it vulnerable to extension. He cited Capron’s questions, including: Why should physician-assisted suicide be available only to the “terminally ill” but not the chronically ill? Why physician-assisted suicide but not voluntary euthanasia? Did denying an injection to those unable to end their lives even with assistance not constitute unjust discrimination on the basis of disability? Keown noted that Chief Justice Rehnquist had observed in Washington v. Glucksberg that what might seem a limited right to physician-assisted suicide was actually a much broader license that could prove extremely difficult to police and that the expansion of this practice seemed all but inevitable.
Keown described another difficult moral and legal question that emerges if laws like Oregon’s are changed to allow voluntary euthanasia. Once laws allow for voluntary euthanasia, why not extend them further to allow non-voluntary euthanasia? Why deny incompetent patients a merciful death? Keown noted that the Dutch law allowing voluntary euthanasia has not prevented non-voluntary euthanasia. He observed: “Six comprehensive national surveys in the Netherlands have disclosed that since it was declared lawful in 1984, thousands of patients have been given lethal injections without their explicit request and thousands of cases have not been reported by physicians.” He added, “It is difficult to conclude, therefore, that the Dutch system, which has now twice been criticized by the UN [United Nations] Human Rights Committee, has been a model of effective control.”
Keown reported that one Dutch ethics scholar, Theo Boer, who served on a euthanasia review committee from 2005 to 2014, has turned from a supporter to a critic of the Dutch law and declared that the “explosive” increase in euthanasia cases showed that “some slopes truly are slippery” (Center for Bieothics & Human Dignity, n.d.; Doughty, 2014). Moreover, in 2016 Boer observed that in approximately 45,000 cases of euthanasia and assisted suicide reported in the Netherlands since 2002, only 75 had been referred by the review committees to the public prosecutor and not a single case had resulted in prosecution.6 Boer explained that public opinion in the Netherlands has begun to interpret “assisted dying” as a right, not an exception, and that legalization has led to a normalization and a general expansion of the practice. Boer concluded that supply creates demand and that many who would never have considered euthanasia 10 years ago now say, “Why not”?7
Keown also noted that in 2016 the Dutch government announced its proposal to extend the law to allow assisted suicide for older individu-
6 Personal communication between John Keown and Theo Boer, August 16, 2016.
7 Personal communication between John Keown and Theo Boer, August 16, 2016.
als who may be healthy but feel their life is “completed” and that Dutch courts have, logically, used the same justification they used to justify voluntary euthanasia—namely the doctor’s duty to alleviate unbearable suffering—to permit non-voluntary euthanasia in the case of disabled infants (Keown, 2018).
Keown concluded that the wording, interpretation, and application of laws—and the principles underlying those laws—are at least as important to an understanding of the end-of-life landscape as statistical data, not least because whatever data are available may largely reflect changes in the laws or their interpretation.
REFLECTIONS ON THE ETHICS OF PHYSICIAN-ASSISTED DEATH
Empirical Research and Controversial Medical Practices
André Hellegers Professor, Kennedy Institute of Ethics
Addressing the subject of empirical research about controversial medical practices, Daniel Sulmasy, the André Hellegers Professor of Biomedical Ethics at Georgetown University’s Kennedy Institute of Ethics, said it is important to be aware of both the potential contributions and the methodological limitations of “descriptive” ethics research (Sugarman and Sulmasy, 2010). For example, one must take note of how one’s personal and political views affect that research and may even influence the types of hypotheses one generates. In the case of physician-assisted suicide, said Sulmasy, opponents will look for evidence of abuse, whereas proponents will look for signs of reassurance. The scientific ideal is a dispassionate, disinterested desire to know the true state of affairs, noted Sulmasy. For example, a 2017 paper from Canadian researchers presented a straightforward, formal cost–benefit analysis of medical assistance in dying in Canada (Trachtenberg and Manns, 2017). However, Sulmasy said, bringing up the linkage between cost containment and medical assistance in dying in the United States is largely off-limits, and we should be considering why that is the case.
Sulmasy questioned why the National Academies found this topic worthy of holding a workshop. Physician-assisted death is not a public health crisis, said Sulmasy, given how few people in the United States are taking advantage of these laws. Thus, Sulmasy said, an argument can be made that this topic is not critical and even distracts from more pressing issues such as improving health care for the nation’s aging population.
Sulmasy cautioned that taking a “health services research” approach to physician-assisted suicide, such as in this workshop, presumes that the goal of physician-assisted suicide is “good” and that, alternatively, the “bad” becomes assessed in terms of efficacy, safety, cost, and access. This takes the central ethical question of whether the service ought to be delivered in the first place off the table, suppressing critical debate, Sulmasy said.
A potential result, intended or not, of having a workshop such as this is the normalization of physician-assisted suicide, Sulmasy said. If the subject is worthy of examination by the National Academies, he said, it becomes medicalized as a subject for empirical study. He wondered if the workshop was to be a prelude to a National Academies consensus study that would serve to promote physician-assisted death.
Language is important because it shapes attitudes and approaches to research, he said. There is no scientific basis for deciding on terminology, other than perhaps marketing science, he said, adding that he chooses to use the term “physician-assisted suicide.” From his perspective as an ethicist and social sciences researcher, he said, he finds that there are important distinctions between physician-assisted suicide, euthanasia, vigorous symptom control, and forgoing life-sustaining treatment, all of which, Sulmasy said, could be included by a reasonable person under the umbrella term “assisted death.” In that regard, he urged caution when using terms that are more political than ethical or scientific in the way they are constructed.
Community engagement is a new standard for research involving human participants and serves to inform the agenda and approach for research, Sulmasy said. In determining who gets to represent “the community,” he said, it is desirable to avoid using advocacy groups for this purpose. Or, if advocacy groups are used, it is important to ensure a balanced representation of proponents and opponents of a particular issue. Sulmasy suggested that opening this workshop with a presentation from a leading advocate of physician-assisted suicide was not a balanced, fair representation of the issues intended to inform a serious engagement of the questions surrounding physician-assisted suicide (see Box 3-1 for the presentation by Dan Diaz, Compassion & Choices).
A study’s design must also be carefully considered, Sulmasy said. He criticized proponents of physician-assisted suicide for relying on unscientific online polls that have found physicians to be in favor of legalizing physician-assisted suicide when there are other, equally unscientific, polls reaching the opposite conclusion. In fact, the only national, weighted, probability-based random sampling study—from 2008—found that 69 percent of physicians opposed legalization (Curlin et al., 2008). Sulmasy criticized as well the use of “ambiguous, complex, and leading questions” in surveys used by proponents of physician-assisted suicide.
A second issue concerning study design involves the use of the metaphor “scanning the landscape” in describing the study of this subject, said Sulmasy. A landscape appears differently to people who are differently situated. Whether you are sitting in front of a desk in a think tank or sitting in a wheelchair matters in terms of evaluating the data, he noted. Biases also exist in recruiting participants for a study, particularly when using advocacy groups to help with the recruitment process. Sulmasy also noted that cognitive dissonance can lead to biases in retrospective surveys of surviving family members.
Relying on results from legal reporting in state-sponsored databanks has various problems as well, Sulmasy said. The data are collected in a manner that maximally protects patient privacy. Given these limitations, the data are “thin,” he said, and give the bureaucratic appearance of monitoring, but the reports are not well validated, and they do not capture unreported cases. The European experience, he added, has been that there are a significant number of unreported cases.
Data interpretation can also be subject to bias, the direction of which depends on whether one is an opponent or proponent of physician-assisted suicide, said Sulmasy. Abuse is inevitable for any law, Sulmasy said, but the question is, How much abuse is too much? “That is not going to be decided by the data themselves,” he said. “It is going to be dependent upon people’s attitudes and their ethical sensibilities.” For example, studies have appeared to show that there is no slippery slope toward undesirable expansions of physician-assisted suicide because the disabled and minorities are not overrepresented among those who seek assisted suicide. However, that does not answer the question from the point of view of those who are disabled. For many of these individuals, Sulmasy said, the fact that dependence on others has become a socially sanctioned reason to be made dead is itself a threat to their dignity even if they are not themselves seeking assisted suicide. Regarding what the pitch of the slippery slope actually means, he said, this will not be determined by research, but rather by serious discussion from an ethical and policy perspective.
Though many are reassured by the fact that only small numbers of people are taking advantage of these laws, Sulmasy questioned why it would be bad if the numbers were large if this practice is thought to represent good care. He also noted that the year-to-year increases seen in Oregon show that, to some extent, there is an increasing social comfort in prescribing lethal medications and using them. A topic deserving of study is the psychological slippery slope, or how practitioners and patients begin to see this as part of normal practice and whether there is pressure to participate.
Sulmasy stressed the importance, particularly for people attending a workshop such as this, of realizing that scientific facts are not ethical arguments (Sugarman and Sulmasy, 2010). “We have to pay attention to what is called the naturalistic fallacy or the fact/value distinction, the fact that you cannot derive what we ought to do from a series of facts,” he said. “Good policy is based on both facts and ethics.” He said that one must be aware that invalid causal inferences can lead to invalid conclusions, citing as an example the inference that the legalization of physician-assisted suicide improved palliative care in Oregon because palliative care improved after Oregon legalized the practice. It is not valid to claim that palliative care improved as a result of legalized physician-assisted suicide in Oregon, Sulmasy said. Correlation is not causation, he concluded.
He argued, too, that ethical issues should be decided based on ethical arguments, not polls, and that the attitude that “the practice is legal, so just deal with it” can lead to a political effect that promotes a controversial practice. Laws that are ethically wrong can be changed, he said. Quoting Aristotle, he said, “Ethics is about what to do when what to do is up to us.” He cautioned as well about drawing the conclusion that science and medicine are moving toward accepting assisted suicide based on what appears in the literature. Journal editors have a bias toward what is new. “That means defense of the status quo is not new and does not get published,” he said.
Based on data collected thus far, Sulmasy contended that what is known about physician-assisted suicide in the United States is:
- The number of reported cases that follow the law is small, but increasing;
- Those who make use of the law tend to be white, wealthy, and educated and to have a “dismissive” personality style focused on control;
- A small but growing number of physicians write lethal prescriptions;
- A substantial proportion of patients seeking physician-assisted suicide are depressed;
- Very few requesting patients are sent for psychiatric referrals;
- Individuals are not requesting physician-assisted suicide because of unbearable symptoms related to their disease, but rather due to loss of autonomy, independence, and control; and
- Approximately one-third of patients who fill a lethal prescription die without taking the drugs, either because they died too soon or decided not to use the drugs. Sulmasy cautioned that, based on this observation, it cannot be assumed that these patients only wanted the security of having “a way out.”
Going forward, Sulmasy said, he would like to see serious deliberative polling, not just snapshots with loaded questions. He also called for research on the validity of the legal reporting mechanisms in order to gain a better understanding of whether those data are true or not and of the incidence of abuse. Research should also look at the psychiatric effect of assisted suicide on families, given the data from Switzerland reporting posttraumatic stress disorder in families that have witnessed assisted suicide (Wagner et al., 2012), and on incomplete suicides and whether, how, and where they are completed. More data are needed on the effects of physician-assisted suicide on individuals with psychiatric disorders, given the reports that publicity about the practice triggers an increase in suicides (Marzuk et al., 1993, 1994) and studies showing increases in suicide in the general population in states that have legalized physician-assisted suicide (Boer, 2017; Jones and Paton, 2015). Concluding his comments, Sulmasy proposed a thought experiment. “If this is just normal medicine, why are we not doing randomized controlled trials to find out how well, efficiently, and cost-effectively we can implement physician-assisted suicide? If that is disturbing, then maybe it would be a good study to find out why.”
Disability and Physician-Assisted Death
San Francisco State University
New England Regional Director
Not Dead Yet
Anita Silvers, a professor in and the associate chair of the Philosophy Department at San Francisco State University, addressed the particular ethical concerns for individuals with severe physical disabilities. The history of repression and suffering that those with disabilities have experienced from programs of institutionalization and eugenics-driven euthanasia, all driven by health care professionals, recommends vigilance regarding physician-assisted death, Silvers said. However, raising administrative barriers that must be hurdled by individuals seeking equitable access to medical services just because they are identified as having a disability is mainly about health care institutions protecting themselves from criticism, she said, arguing that such exclusion is an improper response to ableism.
Silvers noted that for disabled individuals, the requirement that the lethal prescription be self-administered can pose a problem. The self-
administration eligibility requirement, said Silvers, who herself is a post-polio limited quadriplegic who said she doubted that she could open 100 capsules without spilling the medication, is intended to guarantee that the final act is voluntary. She pointed out, though, that being able to control the movements of one’s body does not guarantee that what one’s body does is voluntary, nor that such physical control is necessary, given that U.S. courts have ruled that individuals who lack manual control can use equipment to exercise their right to refuse treatment (Applebome, 1989).
Silvers noted that a model being used for California hospital policies explicitly requires that all patients with disabilities must undergo a thorough assessment for consenting ability beyond what is required by the California aid-in-dying act. She hopes that in the same spirit of protectiveness, the added cost to the disabled patient in time and money will be weighed when considering policies such as this one. “Suffering this kind of legally endorsed exclusion, having to bear such socially imposed loss of commonplace options due to disability nearly every day, that is what my life is like, and it remains one of the hardships of living with a disability,” she said. The current legalization approach to self-administration eligibility is counterproductive, Silvers concluded, because it coerces terminally individuals with a progressive deterioration of bodily control into shortening their own lives prematurely in order to maintain their options.
John Kelly, the New England regional director for Not Dead Yet, an organization that has opposed assisted suicide for 20 years, said that people with terminal conditions and those with disabilities share much in common in that both need assistance from others and are both identified as terminal. At the same time, he said, while non-disabled people are often considered terminal, people with terminal illnesses are almost never described as disabled, which he said helps explain why proponents sometimes say that no disabled person has been affected by these programs. He noted, too, that two-thirds of the people who Jack Kevorkian helped die in the late 1990s were non-terminal disabled individuals even though his victims were sometimes reported in the media as being terminally ill.
Kelly said that popular culture carries the message and educates the public that disabled lives are not worth living. In one film, Me Before You, the main character—a disabled man—insisted on being euthanized in Switzerland because he did not want to inconvenience his girlfriend by having her accompany him to multiple hospital visits.
Kelly said that advocates of physician-assisted suicide are pushing for non-dying, severely disabled people such as himself as well as those with long-term depression to be considered eligible under the laws. He
also said that data from Oregon show that non-terminal people do receive lethal prescriptions, with one person living 603 days after being declared eligible and another person living 1,009 days. “For me, the fact that so many people who are non-terminal are getting these drugs should be reason to stop it,” Kelly said. “A public health program with such an error rate would never be tolerated.” In his view, he said, assisted suicide is beginning to be considered a right, which means it would no longer be a medical practice.
Kelly reported that the leading suicidal motives reported by doctors in Oregon revolve around distress about disability and the loss of autonomy through dependence on others, the loss of abilities, a loss of dignity, suicidal despair over incontinence, and the devastation of feeling like an emotional or financial burden to others. Kelly said that these reasons reflect disability and that there are many people living with such disabilities who are fine with their lives. In his view, he said, the name of the Oregon law, Death with Dignity, means dying to avoid the so-called indignity of being disabled and dependent on others. Assisted suicide, he added, promotes the belief that people would rather be dead than be disabled like he is.
The economics of assisted suicide, Kelly said, make for a deadly combination with the nation’s broken, profit-driven health care system. Some have predicted that economics, and not the quest for broadened individual liberties or increased autonomy, will drive assisted suicide to become acceptable practice (Humphry and Clement, 2000). Kelly mentioned several individuals who were denied treatment because of cost but offered assisted suicide, including one woman who found her copay for assisted suicide would be $1.20. He also said that if the nation were to provide fully funded home care for everyone, people would not have to weigh their own legacy versus their care and perhaps end up choosing assisted suicide.
Provisions in the Oregon and Washington laws also create the possibility of abuse, particularly for older individuals and the disabled, Kelly said. In those states, he said, a friend, relative, or heir can encourage an older person to make the request for assisted suicide, sign the forms as a witness, pick up the prescription, and even administer the drug with or without consent because no objective witness is required at death. Another shortcoming of the laws, he said, is that neither doctors nor witnesses need to know the patients more than superficially. Witnesses can simply check the person’s identification, and doctors who decline for medical reasons are not interviewed, which means that people can doctor shop until they find someone willing to prescribe the lethal medication. In Oregon, he added, the Oregon Health Authority lacks the ability to investigate violations of the law.
In 2017 only 5 of 218 (2.3 percent) receiving lethal prescriptions for physician-assisted suicide in Oregon were referred for psychological or psychiatric evaluations (Oregon Health Authority, 2018). In Massachusetts, the pending legislation that would legalize physician-assisted suicide requires a psychiatric evaluation, but, Kelly said, few psychiatrists believe they can diagnose depression in one visit. Also, Kelly said, people who have experienced depression report that it is not difficult to pass as not depressed in a one-off meeting with someone you have never met before. Thus, Kelly expressed concern that depressed people are threatened by this legislation—especially because depression is constructed as a rational response to terminal illness.
Assisted suicide laws, he concluded, are at their core immunity laws. What he means by that, he explained, is that everyone involved in the process of obtaining the lethal drugs and administering them can receive full immunity simply by saying they acted in good faith. He criticized the emphasis put on preventing suicide in young people while simultaneously steering disabled people and those disabled by their illness toward suicide. “If assisted suicide is legal, some people’s lives will be ended without their consent through mistakes, coercion and abuse,” Kelly concluded. “No safeguards have ever been enacted or even proposed that can prevent this outcome, which can never be undone.”
Advocating for the Option of Physician-Assisted Death
Chief Executive Officer
Compassion & Choices
George Washington University
Kim Callinan, the chief executive officer of Compassion & Choices, an organization that works to pass medical aid-in-dying laws in the United States, said that her main concern regaining medical aid-in-dying is its availability, not its usage. Currently, she said, usage of the medical aid-in-dying laws is low, which raises the question of whether that results from a lack of access or because people are not interested in accessing this option. While large numbers of people may not be taking advantage of the option, she said, they still want the option passed.
In the United States there is strong support for medical aid-in-dying laws, Callinan said. She noted that the Colorado law passed by a 30-point margin, larger than any ballot initiative in Colorado history, and it drew
support in every area of the state and among every demographic. When the Gallup polling organization asked Americans in 2015 whether doctors should be allowed by law to assist a person who has a disease that cannot be cured and is living in severe pain to commit suicide if the patient requests it, 68 percent said they should, and 28 percent said they should not (Dugan, 2015). Similarly, a 2016 survey by LifeWay, which Callinan explained is a Christian organization, found that two-thirds of Americans believe it is morally acceptable for terminally ill patients to ask their doctors for help ending their lives and that majority support was found in a variety of demographic groups (Smietana, 2016). In particular, she said, support is strong among the Latino population as well as among Latino leaders, including Dolores Huerta, Mauricio Ochman, and Jorge Ramos, who see this as a civil liberty issue (Compassion & Choices, 2017; Gonzalez-Portillo, 2017; Huerta, 2017). Three Latino organizations have endorsed medical aid-in-dying—the National Hispanic Council on Aging, Hispanic Health Network, and Latino Commission on Aid (Red Latina, 2017; Reisman, 2018). In terms of African American support, Callinan noted, aid-in-dying legislation passed in Washington, DC, a heavily African American city, and all but one African American city council person voted in favor of the law (Markoe, 2016).
Callinan suggested that physicians are increasingly recognizing that medical aid-in-dying is an option people want. Surveys of state medical societies in Colorado, Maryland, and Massachusetts found that a majority of their members supported medical aid-in-dying, which led to those three organizations dropping their opposition to these laws (Colorado Medical Society, 2016; Maryland State Medical Society, 2016; Massachusetts Medical Society, 2017). Seven more state medical societies have since followed suit (Callinan, 2017; Compassion & Choices, 2018). Callinan said that public support for these laws reflects the desire of many to have control and autonomy at the end of life. A 2016 paper recognized this desire by consumers and called on the medical profession to figure out how to bring this option to the public (Frye and Youngner, 2016). In addition, the New York State Academy of Family Physicians has decided it is a duty of medical professionals to help people with this option in a way that is safe and effective and that declining to do so was a form of patient abandonment (Fandl, 2017).
Callinan suggested that one challenge in implementing U.S. laws on medical aid-in-dying is that the determination of a 6-month prognosis often comes too late for many patients. Data show that a large majority of people are not getting a 6-month prognosis until they have 3 months to live, after which time they need to go through a lengthy process to determine their eligibility and receive a prescription for lethal medica-
tion (Brody, 2007; Christakis and Lamont, 2000). In a supportive system, it is possible to go through the process of requesting and receiving a lethal prescription in 15 days, Callinan said, but most people are not in a supportive system, and for those individuals it can take months to obtain their prescriptions. Callinan suggested that an interesting research question is to find the appropriate balance between legal safeguards and access to medical aid-in-dying.
She suggested that any further study of medical aid-in-dying should include other end-of-life options—withdrawal of treatment, palliative sedation, and hospice—because little is known about any of these options. Callinan argued that it is intrusive and stigmatizing to patients who choose medical aid-in-dying to focus only on research on medical aid-in-dying. Practices such as palliative sedation can be more dangerous than medical aid-in-dying. While there are multiple regulatory protections in place for medical aid-in-dying, Callinan said, there are no regulations on palliative sedation.
Callinan said she would also like to see research on why patient demand for medical aid-in-dying has increased in recent years and how that increase should inform the delivery of patient-directed care. She would also like data to illuminate which legal safeguards and regulatory requirements are necessary and which ones create unnecessary delays and stigma. However, she cautioned that additional data collection efforts or requirements might affect patients’ ability and willingness to access medical aid-in-dying by making the request process too onerous for both doctors and patients.
Callinan suggested that another area of research could examine the impact of not passing medical aid-in-dying laws. For instance, are people committing suicide instead of having a peaceful death because medicine is not meeting their needs? Does a lack of medical aid-in-dying laws create a more dangerous underground practice? Callinan questioned whether, given the small number of people who choose to access medical aid-in-dying and the lack of evidence of any abuse or coercion taking place, research dollars are best spent on studying this or other end-of-life options.
Omega Silva, a professor emeritus of medicine at George Washington University and a patient familiar with the District of Columbia’s recently passed medical aid-in-dying program, said that quality of life is a very personal topic which should be discussed between the patient and doctor. She suggested that the subject of death and dying is a topic about which physicians need to be better educated. Silva suggested that medical school curricula should include an early introduction and continuous referral to end-of-life issues.
Values and Ethics in Conversations About Physician-Assisted Death
Co-Founder and President Emeritus
The Hastings Center
Daniel Callahan, a co-founder and the president emeritus of The Hastings Center and a senior scholar at the Institute of Politics and Policy Studies at Yale University, began by reminding the workshop participants that there was a time when a person with the right connections could easily find a physician who would supply a patient with a lethal dose of medication. Physicians would even sometimes euthanize a patient without the patient’s permission. In the 1980s, overt advocacy for euthanasia in the United States became more common, and there were reports, later confirmed by the Dutch government, that euthanasia without a patient’s permission was occurring in the Netherlands. It was around this same time, in 1989, that Oregon passed its Death with Dignity Act.
Callahan said he came to the debate on assisted death with both wariness and curiosity. Through his work on end-of-life care and the emerging hospice movement, he was well aware that many deaths could be painful, psychologically traumatizing, and messy. But he wondered why hospice was not enough. He also became interested in the observation that medical progress was exhibiting the unpleasant feature of moving society from short lives and quick deaths to longer lives and extended dying.
The autonomy of individuals and their ability to chart their own course of life is at the core of arguments in favor of both assisted death and euthanasia, said Callahan. The argument for autonomy raises a key question, according to Callahan: How far can the liberty to die one’s own chosen death go, and when ought the collective good come into play? The values underlying that question were raised in an interesting way when in the early 20th century, laws outlawing suicide were rescinded across the United States, said Callahan, which had the de facto effect of making suicide socially acceptable. Callahan said that some 43,000 Americans commit suicide annually (O’Brien, 2018). Today, suicide is judged to be a public health problem, and there are anti-suicide efforts at federal and state levels.
Suicide prevention and assisted death programs have nothing to do with one another bureaucratically in Oregon, Callahan said. There are, however, what Callahan called provocative new data emerging that the number of “ordinary” suicides increased in parallel with that of assisted suicide (Dugdale and Callahan, 2017). Callahan suggested this
is not coincidental—his hypothesis is that the Oregon assisted-death law brought considerable media attention to suicide as an acceptable way to deal with the travails of one’s life. During the discussion with workshop participants, this hypothesis was contested by a workshop participant, who cited an article in the Southern Medical Journal (Jones and Paton, 2015) as well as his own unpublished analysis using data from the Centers for Disease Control and Prevention, which found that the increase in non-assisted suicide rates was not statistically significant.
Another argument used by proponents of assisted death is that the option is necessary to alleviate pain and suffering. But among those who want to take advantage of Oregon’s law, only 25 percent cite inadequate pain control, Callahan said. Instead, existential reasons, including the loss of autonomy (89 percent), the inability to engage in activities that make life enjoyable (89 percent), and loss of dignity (77 percent), are the main reasons given by those requesting assisted death in Oregon according to physician reports of patient concerns (Oregon Health Authority, 2018). Callahan questioned whether these existential burdens mirror the troubles of many older individuals.
Callahan was baffled by the word “dignity” in the context of assisted death. Are individuals less human because they may have bouts of incontinence, momentarily forget the names of their own children, or no longer be allowed to drive? Referring to Article 1 of the United Nations Declaration of Human Rights (UN, 1948), which says that all human beings are born free and equal in dignity and human rights, Callahan said he assumes that the UN statement encompasses the end of life as well as the beginning and that it is not undone by existential burdens. “Moreover,” he said, “I trust that it does not mean that indignities in any sense destroy our basic dignity.”
Callahan reflected on his understanding of how assisted death in the Netherlands has expanded to undesirable uses of the practice. He suggested that the concept of personal freedom is being applied in the Netherlands to mean that individuals with mental illness and depression as well as general boredom with life are suitable for an assisted death. Despite his opposition to assisted death, Callahan said he believed the option would continue to be legalized in other U.S. states. He warned that increasing access to assisted death is a poor solution, but for many it may be the only solution available because we have not yet learned how to deal with drawn-out death—the downside of medical progress. Callahan predicted that good hospice programs would be able to minimize the harms of the spreading assisted-death movement but that harm will be inevitable. He said that his hope for the future is to increase research and discussion about assisted death and aging societies.
During open discussion periods with workshop participants, speakers and participants addressed a number of aspects of physician-assisted death.
A workshop participant asked speakers to comment on whether they thought the term “facilitated natural death” was a better description than “physician-assisted death.” Keown objected to this terminology and suggested that society should not pretend that physician-assisted suicide is anything other than an unnatural death. Reflecting on this question, Orentlicher said that there is no neutral terminology on this subject and suggested that individuals pick the term that conveys their beliefs or agenda.
Linda Ganzini from the Veterans Affairs Portland Health Care System suggested that whatever terminology is used, there are ways of developing research and evidence to understand how physician-assisted death is similar to or different from suicide in other contexts. This information, she added, would be useful for patient care and for identifying the appropriate intervention to help patients and families. A workshop participant noted that in October the American Association of Suicidology issued a policy statement distinguishing medical aid-in-dying from suicide (Creighton et al., 2017). Kim responded that comparing the features of what the association’s statement calls “suicide in the ordinary, traditional sense” and the features of persons receiving psychiatric euthanasia show that it is in fact very difficult to differentiate between the two, contrary to the claims made in the association’s statement.
Further reflecting on the term “facilitated natural death” suggested by a workshop participant, Orentlicher questioned why a natural death would be preferred over unnatural death, noting that if natural was preferred over unnatural, there would not be heart bypass surgery, chemotherapy, or the medical profession, as each of those takes us away from our “natural” state of being. Drawing on his experience in the UCLA system, Strouse added that one difference between suicide and physician-assisted death is the extraordinary impact of opening a discussion among patients and families about end-of-life options which enables them to prepare and accept their loved one’s impending death.
Workshop Agenda and the Role of Advocacy Organizations
Sulmasy raised an issue about the workshop agenda and questioned the representation on the agenda of the main U.S. advocacy group for
legalizing assisted suicide, Compassion & Choices. He asked what the proper role (if any) should be for an advocacy organization participating in a scientific meeting and whether this might indicate one of the sorts of problems he noted in his initial presentation regarding potential sources of bias in empirical research about ethically controversial health policies.
Physician-Assisted Death and Economically Marginalized or Minority Populations
A workshop participant asked what is known about the impact of physician-assisted death on vulnerable or economically marginalized populations. Sulmasy answered that there is some evidence to suggest that lower socioeconomic status patients are less interested in physician-assisted death, but we know very little about the impact of the practice on this population. Anthony Back of the University of Washington reflected on his experience working in a safety-net hospital in Seattle and suggested that economic factors are not playing a significant role in patients’ decisions to request physician-assisted death. Ganzini said that in an Oregon comparison between physician aid-in-dying deaths and all other Oregon deaths, those who received aid-in-dying were eight times more likely to have had at least a bachelor’s degree. Ganzini said that she has found from her research that financial concerns are rarely a reason why people request physician aid-in-dying; only approximately 2 percent of the people she studied who received aid-in-dying lacked health insurance.
Scott Halpern of the University of Pennsylvania asked if differences in use of physician-assisted death and endorsement of the practice among different racial and ethnic groups suggests that there is a disparity in access or a difference in preferences. Callinan said that in her experience, people support this issue because they have experienced a loved one who had an unpleasant or painful death. In her opinion, she said, the issue comes down to consumer empowerment which comes with knowing that one can get the prescription medication to end one’s life. Callinan suggested that the desire for autonomy and control over the end of one’s life is observed across all groups, although access is an issue within African American and Latino communities. Silva said that people asking for physician-assisted death usually have a relationship with a physician, which is something that many in minority communities do not have. Daniel Callahan said that the majority of people who request assisted death in Oregon are well educated, and he wondered if assisted death is a form of boutique health care for the “1 percent” in the same way that it previously was possible for the wealthy to receive an assisted death from their physician friends. Callinan and Silva suggested that additional research could help to better understand challenges in access to physician-assisted death.
Mental Health Referrals
A workshop participant questioned data from Oregon indicating less than 5 percent of those who received a prescription for lethal medication had also been evaluated by a mental health professional. For instance, could the 5 percent be misleading since it does not count the number of individuals who are disqualified early on from continuing in the request process for physician-assisted death because they exhibit clear mental disorders or lack of capacity? Ganzini confirmed it is true that, in Oregon, it is unknown how many people receive a psychiatric evaluation and are as a result excluded from continuing in the request process. Reflecting on her experience seeing patients, Ganzini said it often happened that a patient in the throes of delirium would mention physician aid-in-dying, but she said that she does not consider that a serious request. Patients must be consistent and convincing in order to receive a prescription for aid-in-dying, Ganzini said. She also argued that there should be better systematic screening in Oregon for who should be referred to a mental health professional. For instance, Patient Health Questionnaire 9 is a tool that could be used to screen individuals for signs of depression and referral to a psychiatrist. Back added that in addition to simple mental health screens, there is more to learn about what people may be experiencing at the end of life—that is, not quite depression, but perhaps demoralization syndrome or a complicated adjustment reaction. Katrina Hedberg, a health officer and state epidemiologist with the Oregon Health Authority, also noted the limitations in the 5 percent mental health referral statistic and said that there is much more to know about patients who are excluded from the request process, but Oregon does not collect data on this (see Chapter 3 for more information about Oregon’s data collection).
Pain and Suffering
Courtney Campbell, the Hundere Professor of Religion and Culture at Oregon State University, said that the prevention of pain and suffering is often cited as a reason why physician-assisted death should be an option but that the empirical research does not indicate that patients are requesting physician-assisted death for this reason. Ganzini clarified that her research indicates that patients request physician-assisted death to avoid anticipated pain and suffering in the future, but not to relieve pain and suffering they are experiencing at the time of the request (Ganzini et al., 2009). She indicated that very sick people usually do not want to participate in research and therefore it could be a limitation of this research that it includes a healthier population with fewer symptoms. In any case, Ganzini reflected that it is a negative view of the dying process and the
fear of what will happen in the future that is the primary motivator for those requesting aid-in-dying.
Dan Diaz of Compassion & Choices described an experience of his wife, Brittany Maynard (see Chapter 3), and said that it was her fear of pain that led her to request the aid-in-dying process. Diaz explained that Brittany’s fear of pain manifested as her disease progressed in the form of more frequent and severe seizures, the inability to sleep for days, and intense nausea and vomiting. Data indicate that loss of autonomy is a more important factor than fear of pain for those requesting aid-in-dying, Diaz said. He said that loss of autonomy is a significant fear and should not be diminished because it does not fit in the category of a current experience of pain.
Back said that it is important to make a closer examination of the reasons why patients do not feel reassured that pain, suffering, and potential loss of autonomy cannot be managed at the end of life. Sulmasy suggested that if the health care system is not delivering the care that patients need in order to feel reassured that they will have their symptoms controlled, then the obligation is to fix the health care system. Patients should have access to quality palliative care to provide that reassurance that these symptoms can, in large part, be addressed, Sulmasy said.
Ganzini suggested that when patients in Oregon make a request for physician aid-in-dying, they are offered hospice if they are not already enrolled in it. She reflected that the unique services hospice can provide go far beyond traditional pain control and offer many ways to maintain control at the end of life.
American Psychiatric Association. 2014. Diagnostic and statistical manual of mental disorders, 5th ed. Delhi, India: CBS Publishers & Distributors.
Applebome, P. 1989. Judge rules quadriplegic can be allowed to end life. The New York Times, September 7, A16.
Armontrout, J., D. Gitlin, and T. Gutheil. 2016. Do consultation psychiatrists, forensic psychiatrists, psychiatry trainees, and health care lawyers differ in opinion on gray area decision-making capacity cases? A vignette-based survey. Psychosomatics 57(5):472–479.
Bannink, M., A. R. Van Gool, A. van der Heide, and P. J. van der Maas. 2000. Psychiatric consultation and quality of decision making in euthanasia. Lancet 356(9247):2067–2068.
Berg, J. W., P. S. Appelbaum, and T. Grisso. 1996. Constructing competence: Formulating standards of legal competence to make medical decisions. Rutgers Law Review 48(2):345–371.
Boer, T. A. 2017. Does euthanasia have a dampening effect on suicide rates? Recent experiences from the Netherlands. Journal of Ethics in Mental Health 9(Special Theme Issue II).
Brody, J. E. 2007. Tough question to answer, tough answer to hear. The New York Times, March 6, F9.
Burton, C. Z., E. W. Twamley, L. C. Lee, B. W. Palmer, D. V. Jeste, L. B. Dunn, and S. A. Irwin. 2012. Undetected cognitive impairment and decision-making capacity in patients receiving hospice care. American Journal of Geriatric Psychiatry 20(4):306–316.
Cairns, R., C. Maddock, A. Buchanan, A. S. David, P. Hayward, G. Richardson, G. Szmukler, and M. Hotopf. 2005. Reliability of mental capacity assessments in psychiatric inpatients. British Journal of Psychiatry 187:372–378.
Callinan, K. 2017. People should have choices as their life is ending (commentary). MassLive, December 26.
Capron, A. M. 1996. Legalizing physician-aided death. Cambridge Quarterly of Healthcare Ethics 5(1):10–23.
Center for Bieothics & Human Dignity. n.d. Theo A. Boer. https://cbhd.org/content/theoboer-phd (accessed March 1, 2018).
Christakis, N. A., and E. B. Lamont. 2000. Extent and determinants of error in doctors’ prognoses in terminally ill patients: Prospective cohort study. BMJ 320(7233):469–472.
Colorado Medical Society. 2016. Physician-assisted death: Polling shows a divided membership. http://www.cms.org/communications/physician-assisted-death-polling-shows-a-divided-membership (accessed April 25, 2018).
Compassion & Choices. 2017. Compassion & Choices launches PSA with Mexican actor Mauricio Ochmann.https://www.compassionandchoices.org/compassion-choices-launchespsa-with-mexican-actor-mauricio-ochmann (accessed April 25, 2018).
Compassion & Choices. 2018. Physician support for medical aid in dying is strong. https://www.compassionandchoices.org/wp-content/uploads/2018/03/FS-Medical-Professional-Associations-that-Recognize-Medical-Aid-in-Dying-3.02.18-1.pdf (accessed April 25, 2018).
Creighton, C., J. Cerel, and M. P. Battin. 2017. Statement of the American Association of Suicidology: “Suicide” is not the same as “physician aid in dying.” http://www.suicidology.org/Portals/14/docs/Press%20Release/AAS%20PAD%20Statement%20Approved%2010.30.17%20ed%2010-30-17.pdf?ver=2017-10-30-230833-020 (accessed March 2, 2018).
Curlin, F. A., C. Nwodim, J. L. Vance, M. H. Chin, and J. D. Lantos. 2008. To die, to sleep: U.S. physicians’ religious and other objections to physician-assisted suicide, terminal sedation, and withdrawal of life support. American Journal of Hospice and Palliative Care 25(2):112–120.
Doernberg, S. N., J. R. Peteet, and S. Y. Kim. 2016. Capacity evaluations of psychiatric patients requesting assisted death in the Netherlands. Psychosomatics 57(6):556–565.
Doughty, S. 2014. Don’t make our mistake: As assisted suicide bill goes to Lords, Dutch watchdog who once backed euthanasia warns UK of “slippery slope” to mass deaths. The Daily Mail, July 9.
Dugan, A. 2015. In U.S., support up for doctor-assisted suicide. Gallup News, May 27.
Dugdale, L. S., and D. Callahan. 2017. Assisted death and the public good. Southern Medical Journal 110(9):559–561.
Eckstein, L., and S. Y. H. Kim. 2017. Criteria for decision-making capacity: Between understanding and evidencing a choice. Journal of Law and Medicine 24(3):678–694.
Etchells, E., P. Darzins, M. Silberfeld, P. A. Singer, J. McKenny, G. Naglie, M. Katz, G. H. Guyatt, D. W. Molloy, and D. Strang. 1999. Assessment of patient capacity to consent to treatment. Journal of General Internal Medicine 14(1):27–34.
Fandl, J. 2017. Physician’s group endorses medical aid-in-dying legislation. The Legistlative Gazette, July 25.
Frye, J., and S. J. Youngner. 2016. A call for a patient-centered response to legalized assisted dying. Annals of Internal Medicine 165(10):733–734.
Ganzini, L., D. S. Fenn, M. A. Lee, R. T. Heintz, and J. D. Bloom. 1996. Attitudes of Oregon psychiatrists toward physician-assisted suicide. American Journal of Psychiatry 153(11):1469–1475.
Ganzini, L., G. B. Leong, D. S. Fenn, J. A. Silva, and R. Weinstock. 2000. Evaluation of competence to consent to assisted suicide: Views of forensic psychiatrists. American Journal of Psychiatry 157(4):595–600.
Ganzini, L., E. R. Goy, and S. K. Dobscha. 2009. Oregonians’ reasons for requesting physician aid in dying. Archives of Internal Medicine 169(5):489–492.
Garrison, M. 2007. The empire of illness: Competence and coercion in health-care decision making. William & Mary Law Review 49(781):781–843.
Gonzalez-Portillo, P. A. 2017. The power of an anchorman known as Jorge. Compassion & Choices Magazine. https://www.compassionandchoices.org/wp-content/uploads/2017/10/2017-Fall-Magazine-online-9.27.17.pdf (accessed April 25, 2018).
Gorsuch, N. M. 2006. The future of assisted suicide and euthanasia. Princeton, NJ: Princeton University Press.
Grisso, T., and P. S. Applebaum. 1998. Assessing competence to consent to treatment: A guide for physicians and other health professionals. New York: Oxford University Press.
Huerta, D. 2017. Voter-approved death with dignity laws threatened by Congress. The Mercury News, July 28.
Humphry, D., and M. Clement. 2000. Freedom to die: People, politics, and the right-to-die movement. New York: St. Martin’s Press.
Jones, D. A., and D. Paton. 2015. How does legalization of physician-assisted suicide affect rates of suicide? Southern Medical Journal 108(10):599–604.
Kamisar, Y. 1958. Some non-religious views against proposed “mercy-killing” legislation. Minnesota Law Review 42(6):969–1042.
Keown, I. J. 2018. Euthanasia, ethics and public policy. 2nd ed. Cambridge, UK: Cambridge University Press.
Kim, S. Y. H. 2010. Evaluation of capacity to consent to treatment and research. New York: Oxford University Press.
Kim, S. Y., P. S. Appelbaum, H. M. Kim, I. F. Wall, J. A. Bourgeois, B. Frankel, K. C. Hails, J. R. Rundell, K. M. Seibel, and J. H. Karlawish. 2011. Variability of judgments of capacity: Experience of capacity evaluators in a study of research consent capacity. Psychosomatics 52(4):346–353.
Kim, S. Y., R. G. De Vries, and J. R. Peteet. 2016. Euthanasia and assisted suicide of patients with psychiatric disorders in the Netherlands, 2011 to 2014. JAMA Psychiatry 73(4):362–368.
Li, M., S. Watt, M. Escaf, M. Gardam, A. Heesters, G. O’Leary, and G. Rodin. 2017. Medical assistance in dying—implementing a hospital-based program in Canada. New England Journal of Medicine 376(21):2082–2088.
Markoe, L. 2016. Washington, DC, approves “Death with Dignity Act.” National Catholic Reporter, November 2.
Maryland State Medical Society. 2016. Results of MedChi survey on assisted suicide/aid in dying—2016. http://www.medchi.org/News-and-Publications/Surveys (accessed April 25, 2018).
Marzuk, P. M., K. Tardiff, C. S. Hirsch, A. C. Leon, M. Stajic, N. Hartwell, and L. Portera. 1993. Increase in suicide by asphyxiation in New York City after the publication of Final Exit. New England Journal of Medicine 329(20):1508–1510.
Marzuk, P. M., K. Tardiff, and A. C. Leon. 1994. Increase in fatal suicidal poisonings and suffocations in the year Final Exit was published: A national study. American Journal of Psychiatry 151(12):1813–1814.
Massachusetts Medical Society. 2017. Massachusetts Medical Society adopts several organizational policies at interim meeting. http://www.massmed.org/News-and-Publications/MMS-News-Releases/Massachusetts-Medical-Society-adopts-several-organizational-policies-at-Interim-Meeting/#.WuDUIC7wZEb (accessed April 25, 2018).
Miller, D. G., and S. Y. H. Kim. 2017. Euthanasia and physician-assisted suicide not meeting due care criteria in the Netherlands: A qualitative review of review committee judgements. BMJ Open 7(10):e017628.
Nelson, R. M., T. Beauchamp, V. A. Miller, W. Reynolds, R. F. Ittenbach, and M. F. Luce. 2011. The concept of voluntary consent. American Journal of Bioethics 11(8):6–16.
O’Brien, A. 2018. Suicide rate is up 1.2 percent according to most recent CDC data (year 2016). Lifesaver Blog. American Foundation for Suicide Prevention. https://afsp.org/suicide-rate-1-8-percent-according-recent-cdc-data-year-2016 (accessed April 23, 2018).
Oregon Health Authority. 2018. Oregon Death with Dignity Act 2017 data summary. Salem: Oregon Health Authority.
Rabkin, J., R. Goetz, J. M. Murphy, P. Factor-Litvak, and H. Mitsumoto. 2016. Cognitive impairment, behavioral impairment, depression, and wish to die in an ALS cohort. Neurology 87(13):1320–1328.
Red Latina. 2017. National Hispanic Council on Aging endorses medical aid-in-dying laws. October 26.
Reisman, N. 2018. Latino group back aid-in-dying bill. NY State of Politics, February 15.
Sessums, L. L., H. Zembrzuska, and J. L. Jackson. 2011. Does this patient have medical decision-making capacity? JAMA 306(4):420–427.
Seyfried, L., K. A. Ryan, and S. Y. Kim. 2013. Assessment of decision-making capacity: Views and experiences of consultation psychiatrists. Psychosomatics 54(2):115–123.
Silveira, M. J., S. Y. Kim, and K. M. Langa. 2010. Advance directives and outcomes of surrogate decision making before death. New England Journal of Medicine 362(13):1211–1218.
Smietana, B. 2016. Most Americans say assisted suicide is morally acceptable. LifeWay, December 6.
Sorger, B. M., B. Rosenfeld, H. Pessin, A. K. Timm, and J. Cimino. 2007. Decision-making capacity in elderly, terminally ill patients with cancer. Behavioral Sciences and the Law 25(3):393–404.
Sugarman, J., and D. P. Sulmasy. 2010. Methods in medical ethics. 2nd ed. Washington, DC: Georgetown University Press.
Trachtenberg, A. J., and B. Manns. 2017. Cost analysis of medical assistance in dying in Canada. Canadian Medical Association Journal 189(3):E101–E105.
UN (United Nations). 1948. The Universal Declaration of Human Rights. General Assembly resolution 217A, adopted December 10, 1948.
Wagner, B., J. Muller, and A. Maercker. 2012. Death by request in Switzerland: Posttraumatic stress disorder and complicated grief after witnessing assisted suicide. European Psychiatry 27(7):542–546.