3

Experiences with and Reflections on Physician-Assisted Death in the United States

THE OREGON EXPERIENCE

Oregon’s Death with Dignity Act first passed by popular vote—51 to 49 percent—in 1994 but was not enacted until 1997 after a series of legal challenges and a second vote. For 14 years Oregon remained the only U.S. state to have legalized physician aid-in-dying, and Linda Ganzini, a professor of psychiatry and medicine at Oregon Health & Science University and associate director of the Health Services Research and Development Center of Innovation in the Veterans Affairs (VA) Portland Health Care System,¹ said she had thought the law would always be “one of these quirky Oregon things.” That changed, however, when Washington State passed a nearly identical law in 2008, followed by Vermont in 2013, and

___________________

¹ The views expressed in these workshop proceedings are those of Dr. Ganzini and do not necessarily reflect the position or policy of the Department of Veterans Affairs or the U.S. government.

California, Colorado, and Washington, DC, more recently.² By Ganzini’s calculation, almost 60 million people now live in U.S. jurisdictions in which physician aid-in-dying is legal. In addition, as a result of a judicial decision, Montana does not prosecute physician aid-in-dying for competent, terminally ill patients.

Statistics and Research Results on PhysicianAssisted Death in Oregon

Linda Ganzini
Professor of Psychiatry and Medicine
Oregon Health & Science University

Katrina Hedberg
Health Officer and State Epidemiologist
Oregon Health Authority

The Oregon law, as well as others modeled on it, allows a competent, terminally ill patient to receive a lethal prescription from a physician for self-administration. A second physician must confirm that the patient is both competent and terminally ill, and the patient must be informed of all feasible alternatives, such as hospice care. If there is concern that a mental illness such as depression is affecting the decision, the patient must be evaluated by a mental health professional. Physicians and other health care providers are not required to participate. The law does not allow for lethal injection or for aid-in-dying to occur as a result of a request in an advance directive. Physicians who prescribe the medication are required to notify the state and provide documentation that the legal requirements have been met.

Over the past 20 years, physicians have written 1,967 prescriptions, and 1,275 people—65 percent—have taken those medications to hasten their deaths, said Katrina Hedberg, a health officer and state epidemiologist with the Oregon Health Authority’s Public Health Division. On an annual basis, approximately 0.2 percent of deaths in Oregon result from taking prescribed lethal medications. The median age of those taking the medication is 72, with the large majority over age 55 (Oregon Health Authority, 2018).

More than three-quarters (78 percent) of those who die by lethal prescription have cancer as the underlying cause of death, Hedberg said.

___________________

² After the dates of the workshop, Hawaii passed a law in April 2018 legalizing physician-assisted death. Our Care, Our Choice Act. H.B. 2739, 29th Legislature, State of Hawaii (April 2018).

Amyotrophic lateral sclerosis (ALS) patients have the highest rate of participation, at nearly 440 per 10,000 people (Hedberg and New, 2017). Physicians report that approximately 90 percent of the individuals who died from lethal medication said their end-of-life concerns were losing autonomy and being unable to participate in activities that make life enjoyable. More than 75 percent said loss of dignity was a major concern, and almost half feared losing bodily control. More than 40 percent worried they would be a burden to their family, 26 percent were concerned about inadequate pain control, and under 4 percent had financial concerns (Oregon Health Authority, 2018).

Among individuals with the same underlying disease, individuals with college or graduate degrees are far more likely to have a physician-assisted death (Hedberg and New, 2017), Hedberg said. Between 2000, the first year that data were collected, and 2017, 374 physicians, or 0.6 percent of all licensed physicians in Oregon, wrote at least one prescription (with 85 being the most). Twenty-two physicians have been reported to the Oregon Medical Board for incorrect documentation, incomplete written consent, lack of two witnesses, or not complying with the mandated waiting period; in each of these instances, the medical board found that the physicians had acted in good-faith compliance with the law.

The referral rate for psychological evaluation has been approximately 5 percent, but it has been declining (Hedberg and New, 2017). Hedberg said that the percentage of deaths at which the health care provider was present has also declined since 2000 (Hedberg and New, 2017). Nearly 90 percent of patients were in hospice at the time they took the lethal medication. The patient’s home, followed by assisted living facilities, were the two most common locations of death (Oregon Health Authority, 2018). The median time from ingestion to death was 25 minutes, with a range of 1 minute to 104 hours, Hedberg said. Seven people regained consciousness, which Hedberg said reflects the fact that the medications do not work uniformly (Oregon Health Authority, 2018).

One of the first surveys that Ganzini and her colleagues conducted—in 1999—went to more than 4,000 Oregon physicians, of which 2,649 responded (Ganzini et al., 2000b). By then, 143 physicians had received a request for a lethal dose of medication, of which one in six resulted in the physician writing the prescription. At the time of the request, almost one-third of the patients were already in hospice, three-quarters had a life expectancy of less than 6 months, and 59 percent were confined to a bed or chair for over half of their waking hours. In almost half of the cases, she said, the physician responded to the request by implementing a substantive palliative intervention such as treating pain, referring to hospice, or referring the patient to see a mental health professional. Nearly half of the Oregonians with substantive interventions changed their mind about

physician-assisted death, compared with 15 percent of those individuals who did not receive those interventions (Ganzini et al., 2000a). The intervention most likely to result in a patient withdrawing the request for aid-in-dying was referral to hospice. Small-town physicians were less likely to write a prescription for lethal medication, as were physicians whose patients viewed themselves as a burden or had symptoms of depression. Physicians were more likely to prescribe if the patient was enrolled in hospice and wanted control over his or her death. Ganzini and her colleagues found that 34 percent of physicians were willing to prescribe lethal medication (Ganzini et al., 2001).

In all of her studies, Ganzini measures religiousness on a scale of 1 to 10, where 10 means religion is very important in the respondent’s life and 1 means religion is not important in the respondent’s life. In Oregon, hospice nurses and caregivers of ALS patients and cancer patients had an average score of 6 to 7, but patients requesting physician aid-in-dying had an average score of 2.3, the lowest score of any group she and her team have measured. This finding, she said, may indicate that religiousness protects one from pursuing physician aid-in-dying and is likely one reason that Oregon was the first state to pass a physician-assisted death law, given that it is among the least religious of all states. Ganzini added that the strongest predictor of who among the terminally ill will request physician aid-in-dying was a low score on a measure of spirituality that combines meaning, hope, and purpose in one’s life (Smith et al., 2015). Another correlate was the patient having a dismissive attachment style, which is characterized by lifelong values of self-sufficiency and independence (Oldham et al., 2011).

One concern about physician aid-in-dying is that it is a form of suicide, Ganzini said. Given that mental disorders, and particularly depression, are the most important risk factors for suicide, a question to answer is how much depression compels patients to pursue physician aid-in-dying, said Ganzini. Ganzini said that the results of her research are mixed. In a study of Oregon hospice practitioners, Ganzini and colleagues asked social workers to rate their clients’ depression on a scale of 1 to 5, where 5 denotes a very important reason for requesting physician aid-in-dying. The practitioners interviewed had a great deal of clinical experience in assessing depression in hospice patients and rated depression as an average of 1: they did not feel that patients were asking for physician aid-in-dying because of depression (Ganzini et al., 2002). However, a study of requesting patients found that one in four met the criteria for significant clinical depression, although about half of those patients did not feel that depression was influencing their request (Ganzini et al., 2009a). Of the 18 patients in the study who received lethal prescriptions, 3 had clinical depression but had not had a mental health evaluation, although

2 of those said they did not believe depression was influencing their decision. The third patient was successfully treated for depression and still chose physician aid-in-dying. Ganzini said that the results, in total, leave unanswered questions regarding the role of depression in requests for lethal prescriptions.

Ganzini’s research has shown that the desire to maintain control is among the strongest reasons that patients request physician aid-in-dying. Also rating highly, she said, were the loss of independence, future physical symptoms and inability to care for one’s self, not wanting to be cared for by others, and wanting to die at home (Ganzini et al., 2009a). Studies that included qualitative interviews with physicians and hospice nurses about their patients have supported this finding (Ganzini et al., 2000a, 2003; Miller et al., 2004).

Suicide in a family can be associated with a sense of shame, rejection, and stigma on the part of family members, but family members of physician-assisted death patients scored no differently on depression, prolonged grief, the peace they felt at the end of the patient’s life, or other measures of mental health and grief than family members of people who died of natural causes. In fact, Ganzini said, the families of patients who requested physician aid-in-dying felt much more prepared for the death of their loved ones and felt that the patient’s end-of-life preferences were honored. Ganzini said that these outcomes appear different than in the case of suicide (Ganzini et al., 2009b).

In summary, Ganzini said, the primary reason Oregonians request physician aid-in-dying is to increase their sense of control and avoid dependence on others. Qualitative interviews show that this is a lifelong value and not something that occurs just at the time of a terminal illness. Patients are also more motivated to request physician aid-in-dying because of worries about future symptoms than because of worries about the symptoms they are experiencing at the time of their request. While one in four of requesting Oregonians studied have had clinical depression, Ganzini said it is unclear how this mental disorder influences the request. Finally, pursuing physician aid-in-dying is not associated with an increased risk of adverse mental health outcomes in family members (Ganzini et al., 2009b).

Data Collection

Oregon’s law, Hedberg said, requires the state to monitor compliance and issue an annual report containing data collected up to the point a prescription is written for a patient requesting physician-assisted death. The law does not address what happens once the prescribing physician files notification with the state that a prescription has been

written and the law does not mandate collecting data on conversations between physicians and their patients leading up to the writing of a prescription. The law does not include any requirements for who needs to be present at the time of ingestion, for the prescribing physician to reevaluate the patient for mental competency prior to ingestion, or to monitor the patient until death. Data on when a patient eventually dies are determined by matching the patient who received a prescription with the death certificates, which also provide some information on demographics, education, and underlying illness, Hedberg said. Although death certificates are not the best source for demographic information, she said, they are the nationwide standard source of such data. Hedberg said that in Oregon, prescribing physicians are required to complete a patient follow-up form, on which they indicate whether the patient ingested the medications, and provide information about the circumstances surrounding the death.

During the discussion period a workshop participant noted that physician-assisted death laws specify that it is illegal for physicians to note in the death certificate that the cause of death is physician-assisted death and questioned the intellectual integrity of this requirement. Neil Wenger of the University of California, Los Angeles, said that some physicians in California have expressed a similar concern that they must write untruthful things on death certificates in cases of physician-assisted death. Timothy Quill of the University of Rochester School of Medicine said that the underlying disease (e.g., cancer, heart disease) is reported on the death certificate. In domains similar to physician-assisted death—withdrawal of treatment or palliative sedation—the same is true in that the underlying disease is most often reported as the cause of death, Quill said.

Hedberg spoke of the importance of balancing the right to privacy of a patient who is dying, and the associated confidentiality issues, with the need for society to understand discussions at the end of life and decisions about participating in assisted death. While the physicians report that patients have met the qualifications outlined in Oregon’s law, they do not report specific details of their evaluation, said Hedberg. For example, data elements not included under the requirements of the law include how the diagnosis and prognosis were made and how the patient’s mental capacity was ascertained. Hedberg said that data are also not collected on patients who were denied their request and the reasons for that denial.

During the discussion, John Kelly of Not Dead Yet shared reports of disabled and terminally ill individuals believed to be coerced or involuntarily euthanized in Oregon and asked if the state could include instances of violations of the law in their annual reporting. In response, Hedberg

said she recognized that many would like to see more information available through the state’s oversight mechanism but that as currently written, the statute is silent on any data collection efforts in the period after a prescription is written. She clarified that the information about potential abuses of the law reported in the blogosphere is not seen by the Oregon Health Authority and that therefore the members of the authority do not have a role in potential criminal charges related to the act. Hedberg said that a certain degree of tension exists in that many advocates of physician-assisted death want less reporting and data collection, whereas opponents want more.

Dan Diaz of Compassion & Choices said that any requirement to collect data on physician aid-in-dying should balance the privacy of patients and families. Additional data collection efforts should not burden patients or families, or intrude unnecessarily in their lives, Diaz said, and in the end, data collection efforts should benefit the patient.

Surveys of Patients, Families, and Health Care Practitioners

To partially remedy the shortcomings of Oregon’s official reporting system, Ganzini has conducted a series of research surveys that collect information directly from patients and families. The greater detail in the resulting data is useful for hypothesis testing, she said, although she acknowledged that not everyone invited to participate returns the surveys, so there could be a selection bias which might limit generalizability. She noted, too, that these surveys were administered 10 to 20 years ago and that it is possible the data have changed since then.

In addition to research surveys, Ganzini has conducted qualitative research interviews which produced vivid descriptions of participants’ experiences. She said that these interviews can be used to generate hypotheses and can support quantitative data, although they lack generalizability and have a risk of bias (Dobscha et al., 2004; Ganzini et al., 2003; Harvath et al., 2006). Ganzini studied all physicians in Oregon eligible to prescribe under the law and found that 5 percent of the state’s physicians had received a request for physician aid-in-dying by 1999 (Ganzini et al., 2000b). Another early study of hospice nurses and social workers found that almost one-third of respondents had cared for a client who had received a lethal prescription (Ganzini et al., 2002). An advantage of studying that particular population, Ganzini explained, is that hospice nurses and social workers see patients more frequently than physicians and have more discussions with them in the last few weeks of life. All of these surveys, Ganzini said, had response rates of more than 65 percent, making it more likely that the results are generalizable. Other studies she

has conducted included measuring depression, hopelessness, social support, burden to others, religiousness, and reasons for request in 56 Oregonians who had explicitly requested physician-assisted death (Ganzini et al., 2009a) and interviewing 95 family members of 83 Oregonians who had requested physician-assisted death; interviews took place an average of 14 months after their loved one’s death (Ganzini et al., 2009b).

Surveys of Oregon health care providers, some of which were conducted more than 20 years ago, found that physicians and hospice nurses were divided in their support for the law, with about half supporting it, one-third opposing it, and the rest holding a neutral opinion. Hospice social workers tended to be much more supportive of the option, which Ganzini said might reflect the focus on client autonomy in their education (Miller et al., 2004). Hospice chaplains were surprisingly divided, she said, with a similar number supporting and opposing the law, but 54 percent had worked with a requesting patient (Carlson et al., 2005). Psychiatrists have a core professional role in preventing suicide, she noted, but over half supported this law, and psychologists overwhelmingly supported the law (Fenn and Ganzini, 1999b; Ganzini et al., 1996). Table 3-1 describes the attitudes of various Oregon health care practitioners toward Oregon’s Death with Dignity Act or physician-assisted death as collected through different surveys.

Her team’s study of the attitudes of other health care practitioners found that only 3 percent of hospice nurses (Miller et al., 2004) and 14 percent of hospice chaplains would actively oppose a client’s choice for physician aid-in-dying (Carlson et al., 2005). None of the chaplains surveyed would transfer a patient who received a lethal prescription to another chaplain, while 12 percent of hospice nurses said they would transfer such a patient to another hospice nurse (Carlson et al., 2005; Miller et al., 2004). Ganzini said she believes that these numbers reflect the increased focus on non-abandonment among hospice providers. The most recent figures show that 90 percent of patients who die by physician-assisted death in Oregon are enrolled in hospice (Oregon Health Authority, 2018). Ganzini said that this is important because there is general agreement that physician aid-in-dying should be the option of last resort after other palliative interventions have failed.

Dan Diaz, Latino Leadership Council, Compassion & Choices, shared the experience of his wife, Brittany Maynard, in seeking access to medical aid-in-dying (see Box 3-1).

TABLE 3-1 Oregon Health Care Practitioners’ Attitudes Toward the Oregon Death with Dignity Act or Physician-Assisted Death

SOURCES: Ganzini presentation, February 12, 2018. Data from Carlson et al., 2005; Fenn and Ganzini, 1999a; Ganzini et al., 1996, 2001; and Miller et al., 2004.

OTHER U.S. EXPERIENCES

Washington

Helene Starks
Associate Professor of Bioethics and Humanities
University of Washington

More than 20 years ago, Helene Starks, an associate professor of bioethics and humanities and an adjunct associate professor of health services, family medicine, and pediatrics at the University of Washington, conducted a study in which she and her research team asked patients and families in Washington, where physician-assisted death was still illegal, and in Oregon, where physician-assisted death had recently been approved, about their experiences in helping family members with terminal illnesses hasten their deaths (Back et al., 1996, 2002; Starks et al., 2005, 2007). She and her colleagues interviewed 35 families, and in the 12 cases where they could recruit the patient while he or she was still alive, they followed the patient until death. In all cases, they followed the families for another year to gain some insight on their bereavement experiences.

This study found that patients do not make this decision quickly and that there are three basic concerns that patients take into account when considering physician-assisted dying as a serious option: (1) the enduring diminishments that come with terminal illness; (2) the exhaustion that comes with the process of dying and living with the side effects of treatment, including pain medication; and (3) the more existential threat of a person losing his or her sense of self (Pearlman et al., 2005). In addition to this “triple threat,” patients also have fears about what the trajectory

of their death will entail, based either on past experiences with others’ death or on what they think or know about the course of their own illness.

Given that the procedure was illegal in Washington at the time of the study, the patients who wanted to hasten their own deaths had to accumulate a lethal dose of medication on their own. Starks analyzed the reasons and triggers for the eventual timing of the patients’ deaths. The research team organized the cases based on their own estimates of the patients’ prognosis given the patients’ clinical situation (Starks et al., 2005). Those individuals with days to 1 week to live waited to take the medication until the day they were likely to die anyway; for these patients it was a matter of asserting their own control over the disease process and not feeling like a victim of the disease. For those with a longer estimated prognosis, Starks said, the fear of what the future might hold characterized why they chose to die when they did.

Regarding families, Starks said that current laws protect the patient and physician but do not address the concerns of patients’ families. People are often told that families should leave the room during discussions about physician aid-in-dying to avoid the possibility of coercion, Starks said, but often these conversations include important education about the process. This means that family members may not be as informed as they could be, even though they are often actively involved with planning and logistics from the beginning. Additionally, in places where the practice is still illegal, family members are often charged with implementing a backup plan in case of a medication failure.

Overall, families have felt very isolated and ill-prepared, especially in places where the practice was illegal, Starks said (Starks et al., 2007). She said that her personal stance toward physician aid-in-dying was originally neutral but that this study swayed her in the direction of legalization because it mandates that people have a protocol and are engaged. In this way, she explained, people are able to access the information that previously they were trying to figure out on their own. She pointed out that assisted death is still happening in places where it is not legal and said that if patients want it, they will find a way.

California

Helene Starks
Associate Professor of Bioethics and Humanities
University of Washington

Starks and colleagues at the University of California, San Francisco, are now conducting a study in California similar to the one she conducted in Washington two decades ago. From the patients and families they have

spoken with so far, they have learned that health systems that provide a navigator make a significant difference for patients and families in terms of working through the bureaucratic complexities of the law. There are problems, however, with the interpretation of what counts as a patient’s first request. Some patients consider the first request to be when they are given the news that they are out of options and they say they want to start the death-with-dignity process. However, this request is not always heard by their physicians, and thus, while they believe they have started the process, they find out later that this request was not formally recognized, said Starks.

When physicians stall, said Starks, patients often go outside their health systems to find another physician who will provide the initial approval. When patients are able to access a willing provider inside their system, it is often a consulting physician who is confirming the prognosis, not someone writing the prescription, stated Starks.

Offering lessons from California’s experience, Starks said that the process takes time and that very few patients or health systems report that they are able to complete the request process in 15 days. The exceptions, she said, occur with what she has started calling the “destination doctors,” practitioners who are providing physician-assisted death either as a specialty within a larger practice or as their sole practice. In a sense, these physicians become self-made navigators who know how to work the system. At the same time, Starks said, she has found that physicians’ ambivalence is a factor that slows down the process, particularly with those who have received a request for lethal medication for the first time. Starks has found that what patients want is for physicians to tell them right away if they are ambivalent about aid-in-dying so that the patient has the opportunity to find another provider without delay. Some physicians took weeks to months to decide if they were going to participate, Starks said, which created additional stress for the waiting patients.

Families have told Starks that they see themselves as being project managers, which was not the job they always wanted to have. This is particularly true when it comes to managing the medication. “There is a great deal of distress about what it takes to open 100 Seconal capsules,” Starks said. Some family members have reported that on the day of death they were so fixated on their roles as project managers that they could not be fully present with their loved ones during the final moments of their lives. One person told Starks that she felt cheated out of the last couple of minutes because she was helping her family member with the medicine.

Colorado

Matthew Wynia
Director, Center for Bioethics and Humanities
University of Colorado

In its first year of implementation, 69 Coloradoans received prescriptions for lethal medications from 37 physicians, although little is known about how many individuals picked up or used the medications, said Matthew Wynia, the director of the Center for Bioethics and Humanities at the University of Colorado Anschutz Medical Campus. Like other states with physician aid-in-dying laws, said Wynia, physicians are precluded from noting on the death certificate that the individual died as a result of taking these medications; therefore, it is not possible to truly know how many patients in Colorado died from these drugs. He noted that 96.4 percent of those receiving the prescriptions were white and 87.5 percent lived in the Colorado Springs–Denver–Fort Collins corridor (Colorado Department of Public Health and Education, 2018). While Colorado as a whole is largely white, 30 percent of Denver’s population is Latino or Hispanic, so these initial data suggest that racial and ethnic disparities with respect to access may be significant.

One unusual feature of the Colorado statute, said Wynia, is that it allows individual providers, but not organizations, to opt out. Under the terms of statute, organizations are not allowed to tell their physicians or pharmacists that they cannot participate in the law. However, organizations can prohibit patients from taking the lethal medication on their premises in the same way that the state prohibits people from taking these medications in state parks. The provision for organizations stems from lessons that advocates learned from Oregon and other earlier adopters, which is that while allowing organizations to prohibit employees from participating in the program may be respectful of the notion of “organizational conscience,” it creates an access barrier for patients. Despite the law, several health systems in Colorado have nevertheless said that they are prohibiting their employees from participating in the end-of-life process, which could eventually prompt a legal challenge. Given the writings of Neil Gorsuch, the newest U.S. Supreme Court member, Wynia suggested that this provision might not survive a court challenge if it were to be heard by the U.S. Supreme Court (Gorsuch, 2006).

Vermont

Mara Buchbinder
Associate Professor of Social Medicine
University of North Carolina at Chapel Hill

The Patient Choice and Control at End of Life Act was signed into law by Vermont’s governor in May 2013, and as of the end of June 2017 paperwork had been submitted for 52 patients who made use of the law (Englander, 2018), said Mara Buchbinder, an associate professor of social medicine and adjunct associate professor of anthropology at the University of North Carolina at Chapel Hill. She said that the law’s requirements are nearly the same as those in Oregon.

In her qualitative and ethnographic study conducted between 2015 and 2017, Buchbinder addressed three primary questions: How do ordinary people understand, access, and contest medical aid-in-dying once it is legally authorized? How do health care providers and policy stakeholders accommodate or resist medical aid-in-dying as a new end-of-life practice? What are the ethical challenges for clinical communication and the patient–clinician relationship? To answer those questions, Buchbinder conducted 144 semi-structured interviews of nurses, chaplains, social workers, terminally ill patients, lay caregivers, policy makers and activists, other Vermont residents, and physicians, many of whom had direct experience with Vermont’s law either as a prescriber or as a secondary physician. Buchbinder said that because Vermont is so small, she was able to talk to a majority of the prescribers in the state.

Several preliminary findings have emerged from this work, Buchbinder said, although she is still analyzing the results of her interviews. First, the challenges for physicians are much broader than simply grappling with whether to write a prescription and include, among other things, determining patients’ eligibility. Physicians, as well as nurses and social workers, assist with all aspects of the bureaucratic process, including finding a second physician to certify eligibility, identifying a pharmacy, and completing the paperwork. Many providers, particularly those in private practices and in more rural areas, reported that they felt isolated while navigating this process, Buchbinder said. In fact, many providers had the sense that they were reinventing the wheel each time they started this process because there were no protocols to follow and there were few experienced colleagues to turn to for guidance. Some providers were able to connect with advocacy organizations for support, such as the Compassion & Choices Doc2Doc program,³ which enabled them to speak with

___________________

³ For more information, see https://www.compassionandchoices.org/research/doc2doc-program (accessed March 23, 2018).

experienced physicians in other states. Buchbinder said that there are no formal mechanisms in place in Vermont for sharing lessons learned.

Some providers, she said, involved themselves in planning for the death, such as by making contingency plans for possible adverse events. Some doctors and nurses were present at the death to help deal with emergent adverse events or otherwise provide social support, at the patient’s request. In addition, even physicians who did not prescribe lethal medication were sometimes involved in these other steps because they felt committed to supporting a patient’s wishes and to the ethos of non-abandonment.

Buchbinder said she has been interested in the way that different providers conceptualize this work. Some see it as falling on the spectrum of palliative care and as no different from other clinical services they provide. Other physicians see it as emotionally and morally burdensome, although they were often still grateful for the opportunity to participate. One physician, who had been one of the heavier prescribers, retired early, in part because of the burden of practicing under Vermont’s law. Another physician told Buchbinder about a patient who experienced a brutal death during the 15-day waiting period between requests, which left the physician feeling terrible that they had not started the process sooner. This type of timing issue came up repeatedly in her interviews with both providers and family members, Buchbinder said.

In July 2016 a lawsuit was filed by the Vermont Alliance for Ethical Healthcare and the Christian Medical and Dental Association against the Vermont Board of Medical Practice and several other state agencies alleging that Vermont’s law entailed an affirmative duty to inform terminally ill patients that the law was an option in their state and therefore that the law violated the constitutional rights of these physicians. The lawsuit was not successful, but it did raise an important ethical question of whether physicians should inform patients about assisted dying laws (Buchbinder, 2017).

Buchbinder said that the physicians and nurses in her study expressed a range of perspectives on who should initiate discussion of aid-in-dying and that some providers reported initiating such a discussion under certain circumstances. Buchbinder noted that we need to be thinking about “circumstances in which it might be ethically permissible for physicians and nurses to inform qualifying patients about aid-in-dying.” Buchbinder explained that she did not want to minimize the difficult ethical questions at stake—around the nature of professional responsibility, potential harms to the doctor–patient relationship, and the potential for undue influence—when it comes to talking about physician aid-in-dying. However, she said, it is worth acknowledging that some providers are indeed bringing up this option with their patients and that medical ethics needs to grapple with this.

One reason Buchbinder said she believes it is worth reconsidering the conventional view that communication should always be patient-initiated is that patients who already know about aid-in-dying laws are likely to be more educated. In that case, she said, waiting for patients to initiate a request might create access gaps for patients from less-advantaged populations. On the one hand, waiting for patients to initiate the conversation might appease concerns about the coercion of vulnerable groups to use aid-in-dying. On the other hand, public policies that privilege access for relatively advantaged populations raise justice concerns (Buchbinder, 2017, 2018). Data showing a lower use of aid-in-dying among low socioeconomic status groups might reflect unequal access rather than less of a preference for aid-in-dying among these groups, Buchbinder said (Battin et al., 2007).

REFERENCES

Back, A. L., J. I. Wallace, H. E. Starks, and R. A. Pearlman. 1996. Physician-assisted suicide and euthanasia in Washington State. Patient requests and physician responses. JAMA 275(12):919–925.

Back, A. L., H. Starks, C. Hsu, J. R. Gordon, A. Bharucha, and R. A. Pearlman. 2002. Clinician–patient interactions about requests for physician-assisted suicide: A patient and family view. Archives of Internal Medicine 162(11):1257–1265.

Battin, M. P., A. van der Heide, L. Ganzini, G. van der Wal, and B. D. Onwuteaka-Philipsen. 2007. Legal physician-assisted dying in Oregon and the Netherlands: Evidence concerning the impact on patients in “vulnerable” groups. Journal of Medical Ethics 33(10):591–597.

Buchbinder, M. 2017. Aid-in-dying laws and the physician’s duty to inform. Journal of Medical Ethics 43(10):666–669.

Buchbinder, M. 2018. Access to aid-in-dying in the United States: Shifting the debate from rights to justice. American Journal of Public Health 108(6):e1–e6.

Carlson, B., N. Simopolous, E. R. Goy, A. Jackson, and L. Ganzini. 2005. Oregon hospice chaplains’ experiences with patients requesting physician-assisted suicide. Journal of Palliative Medicine 8(6):1160–1166.

Colorado Department of Public Health and Education. 2018. Colorado End-of-Life Options Act, year one: 2017 data summary. Denver, CO: Department of Public Health and Education Center for Health and Environmental Data.

Dobscha, S. K., R. T. Heintz, N. Press, and L. Ganzini. 2004. Oregon physicians’ responses to requests for assisted suicide: A qualitative study. Journal of Palliative Medicine 7(3):451–461.

Englander, D. C. 2018. Report to the Vermont legislature concerning patient choice at the end of life. Burlington: Vermont Agency of Human Services Department of Health.

Fenn, D. S., and L. Ganzini. 1999a. Attitudes of Oregon psychologists toward physician-assisted suicide and the Oregon Death with Dignity Act. Professional Psychology, Research and Practice 30(3):235–244.

Fenn, D. S., and L. Ganzini. 1999b. How do Oregon psychologists view their role in physician-assisted suicide? Western Journal of Medicine 171(4):242.

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. 2000a. Evaluation of competence to consent to assisted suicide: Views of forensic psychiatrists. American Journal of Psychiatry 157(4):595–600.

Ganzini, L., H. D. Nelson, T. A. Schmidt, D. F. Kraemer, M. A. Delorit, and M. A. Lee. 2000b. Physicians’ experiences with the Oregon Death with Dignity Act. New England Journal of Medicine 342(8):557–563.

Ganzini, L., H. D. Nelson, M. A. Lee, D. F. Kraemer, T. A. Schmidt, and M. A. Delorit. 2001. Oregon physicians’ attitudes about and experiences with end-of-life care since passage of the Oregon Death with Dignity Act. JAMA 285(18):2363–2369.

Ganzini, L., T. A. Harvath, A. Jackson, E. R. Goy, L. L. Miller, and M. A. Delorit. 2002. Experiences of Oregon nurses and social workers with hospice patients who requested assistance with suicide. New England Journal of Medicine 347(8):582–588.

Ganzini, L., S. K. Dobscha, R. T. Heintz, and N. Press. 2003. Oregon physicians’ perceptions of patients who request assisted suicide and their families. Journal of Palliative Medicine 6(3):381–390.

Ganzini, L., E. R. Goy, and S. K. Dobscha. 2009a. Oregonians’ reasons for requesting physician aid in dying. Archives of Internal Medicine 169(5):489–492.

Ganzini, L., E. R. Goy, S. K. Dobscha, and H. Prigerson. 2009b. Mental health outcomes of family members of Oregonians who request physician aid in dying. Journal of Pain and Symptom Management 38(6):807–815.

Gorsuch, N. M. 2006. The future of assisted suicide and euthanasia. Princeton, NJ: Princeton University Press.

Harvath, T. A., L. L. Miller, K. A. Smith, L. D. Clark, A. Jackson, and L. Ganzini. 2006. Dilemmas encountered by hospice workers when patients wish to hasten death. Journal of Hospice and Palliative Nursing 8(4):200–209.

Hedberg, K., and C. New. 2017. Oregon’s Death with Dignity Act: 20 years of experience to inform the debate. Annals of Internal Medicine 167(8):579–583.

Miller, L. L., T. A. Harvath, L. Ganzini, E. R. Goy, M. A. Delorit, and A. Jackson. 2004. Attitudes and experiences of Oregon hospice nurses and social workers regarding assisted suicide. Palliative Medicine 18(8):685–691.

Oldham, R. L., S. K. Dobscha, E. R. Goy, and L. Ganzini. 2011. Attachment styles of Oregonians who request physician-assisted death. Palliat Support Care 9(2):123–128.

Oregon Health Authority. 2018. Oregon Death with Dignity Act 2017 data summary. Salem: Oregon Health Authority.

Pearlman, R. A., C. Hsu, H. Starks, A. L. Back, J. R. Gordon, A. J. Bharucha, B. A. Koenig, and M. P. Battin. 2005. Motivations for physician-assisted suicide: Patient and family voices. Journal of General Internal Medicine 20(3):234–239.

Smith, K. A., T. A. Harvath, E. R. Goy, and L. Ganzini. 2015. Predictors of pursuit of physician-assisted death. Journal of Pain and Symptom Management 49(3):555–561.

Starks, H., R. A. Pearlman, C. Hsu, A. L. Back, J. R. Gordon, and A. J. Bharucha. 2005. Why now? Timing and circumstances of hastened deaths. Journal of Pain and Symptom Management 30(3):215–226.

Starks, H., A. L. Back, R. A. Pearlman, B. A. Koenig, C. Hsu, J. R. Gordon, and A. J. Bharucha. 2007. Family member involvement in hastened death. Death Studies 31(2):105–130.

This page intentionally left blank.