Physician-Assisted Suicide in Oregon: A Medical Perspective

Herbert Hendin & Kathleen Foley. Issues in Law & Medicine. Volume 24, Issue 2. Fall 2008.

In 1997, five months after the U.S. Supreme Court ruled that there was no right to assisted suicide in the Constitution but implied that states have the right to decide for themselves whether to permit or prohibit physician-assisted suicide, the Oregon Death with Dignity Act, having survived its own legal challenges, took effect. It was thought that Oregon would serve as a “laboratory of the states,” showing us how physician-assisted suicide (“PAS”) would work. This has not occurred, in large part because the Oregon Public Health Division (“OPHD”), charged with monitoring the law, has interpreted its mandate in an extremely restrictive manner.

OPHD limits its yearly reports to general epidemiological data and collects limited information from physicians who have prescribed lethal medication. Physicians who declined to prescribe the lethal medication, as well as nurses and social workers who cared for the patients, pharmacists who filled the prescriptions, and family members, are not interviewed. Not all the information collected is made public, and after a year “all source documentation is destroyed.”

Since the passage of Oregon’s Death with Dignity Act, however, various sources—patients, families, healthcare professionals, physicians, nurses, social workers, chaplains, and advocacy groups—have supplied more detailed information that suggests that the implementation of the law has had unintended, harmful consequences for patients.

The Oregon law seems to require reasonable safeguards regarding the care of patients near the end of life, which include presenting patients with the option for palliative care; ensuring that patients are competent to make end-of-life decisions for themselves; limiting the procedure to patients who are terminally ill; ensuring the voluntariness of the request; obtaining a second opinion on the case; requiring the request to be persistent, i.e., made a second time after a two week interval; encouraging the involvement of the next of kin; and requiring physicians to inform OPHD of all cases in which they have written a prescription for the purpose of assisted suicide.

The evidence strongly suggests that these safeguards are circumvented in ways that are harmful to patients. Addressing and correcting the situation would require more information than OPHD has been willing to obtain. Instead, based on the inadequate information it collects, OPHD has been issuing annual reports declaring that terminally ill Oregon patients are receiving adequate care. The available evidence, which we will present in this article, suggests otherwise.

Nothing in the Oregon law prevents OPHD from collecting needed information.

During the second year of the law, OPHD did undertake a survey of the family members of patients who had been assisted in suicide. Apart from not permitting independent investigators to examine the data, the Oregon law gives OPHD great flexibility. OPHD has not taken advantage of this opportunity.

This article draws on six cases previously published, three of them by us. In four of them there was independent information from more than one source. In two of the cases the information is provided by one source only—in one case by a proponent, and in the other by an opponent, of assisted suicide. This article differs, however, from our earlier treatments of the subject since it focuses on the implementation of the Oregon law and not on the law itself.

Part I of this article examines OPHD’s failure to ensure that palliative care alternatives to PAS are made available to patients. Parts II and III discuss the adequacy of safeguards to ensure a patient’s psychiatric health and the voluntariness of the decision. Part IV discusses the emphasis on protecting physicians, rather than patients. Part V examines the role of advocacy groups for assisted suicide. Part VI describes how Oregon’s current approach to patient requests for PAS differs from the accepted medical approach both to suicide and to requests for assisted suicide. Part VII concludes with an analysis of the main concerns raised by OPHD’s monitoring of the Oregon law and suggests how these concerns could be addressed.

I. Providing Alternatives

In Oregon, intolerable suffering that cannot be relieved is not a basic requirement of assisted suicide as it is in the Netherlands, the only country to give legal sanction to assisted suicide and euthanasia. A diagnosis of terminal illness with a prognosis of less than six months to live is considered a sufficient criterion.

The unintended consequence of this provision is that it enables physicians to assist in suicide without inquiring into the source of the medical, psychological, social, and existential concerns that usually underlie requests for assisted suicide, even though this type of inquiry produces the kind of discussion that often leads to relief for patients and makes assisted suicide seem unnecessary. When a terminally ill Oregon patient makes a request for assisted suicide, physicians are required to indicate that palliative care and hospice care are feasible alternatives. They are not required, however, to be knowledgeable about how to relieve physical or emotional suffering in terminally ill patients. Without such knowledge, which most physicians do not have, they cannot present or make feasible alternatives available. Nor in the absence of such knowledge are they required to refer the patient to a physician with expertise in palliative care.

In the absence of adequate monitoring, the focus shifts away from relieving the distress of dying patients considering a hastened death to meeting the statutory requirements for assisted suicide. Physicians can merely go through the motions of presenting the possibility of palliative care for their patients. How this happens is suggested by a case which was publicized by Compassion in Dying (now Compassion and Choices), an advocacy group which promotes legalized PAS, as the first case of assisted suicide under the Oregon law.

A. The First Case: Helen

In earlier works, we gave an account of this case based on a news conference given by Compassion in Dying, our own correspondence with the doctor who assisted in the suicide, and other sources of information to which we will refer. The conference described how a patient in her mid-eighties, who had been diagnosed with metastatic breast cancer and who was then living in a hospice, came to choose assisted suicide.

Helen’s own physician had refused to assist in her suicide for unspecified reasons. A second physician refused on the grounds that Helen was depressed. Helen’s husband then called Compassion in Dying and was referred to a physician who would assist her.

The medical director of Compassion in Dying said that he had spoken by phone with Helen at the time of the referral and also spoke by phone to her son and daughter. He described Helen as “‘rational, determined and steadfast'” and questioned the opinion of the physician (with whom the medical director also spoke by phone) who described her as having a depression that was affecting her desire to die. He said Helen was “‘frustrated and crying because she was feeling powerless.'” He said she had been doing aerobic exercises up until two weeks before she contacted him but told him she could not do them anymore. She was also unable to continue to garden, which had been one of her favorite activities. The medical director said she was not bedridden, was not in great pain, and was still able to look after her own house. He said the “‘quality of her life was just disappearing,'” and he thought it prudent to act quickly before Helen lost the capacity to make decisions for herself. He said she was “‘going downhill rapidly…. She could have had a stroke tomorrow and lost her opportunity to die in the way that she wanted.'”

The physician who agreed to prescribe the medication had met Helen two and a half weeks before she died and described her as having more physical discomfort than Compassion in Dying had indicated. He said that after twenty years the cancer had spread to her lungs, causing some pain and shortness of breath. He followed a protocol that included an anti-nausea medication that Helen had taken before he arrived to be with her and her family when she died. She then took a mixture of barbiturates (nine grams) and syrup followed by a glass of brandy. She is said to have died within thirty minutes.

B. The Medical Decision

Helen’s case was presented by Compassion in Dying as an example of how well the Oregon law is being implemented. The organization did not seem aware that, contrary to their expectations, their presentation would raise troubling questions.

The physicians who evaluated Helen offered two contradictory sets of opinions about the appropriateness of her decision. As the decision-making process progressed, there was no mechanism in place for resolving the disagreement based on medical expertise. An ethics committee that would hear the facts of the case before going forward could have resolved this dispute. Instead, the opinions of the two doctors who did not support the patient’s decision—one who had known her for some time and another who considered that she was depressed—were essentially ignored. As Barbara Coombs Lee, then the executive director of Compassion in Dying, expressed it, “‘If I get rebuffed by one doctor, I can go to another….'”

Patients, of course, have the right to obtain other opinions and to seek out physicians who will provide the therapy that the patients choose. We wondered at the time if the prescribing physician consulted either Helen’s physician or the physician who diagnosed her as depressed. In reply to a journal article we wrote that asked this question, we received a response from Dr. Peter Reagan, who had been anonymous but who now identified himself as the prescribing physician. He wrote:

Before my patient died I didn’t personally discuss the case with her regular physician and had only a very cursory contact with her second. I regret this. I don’t think either of the previous MDs disagreed with her qualification, but at the time I would have clarified it. Had I felt there was a disagreement among the physicians about my patient’s eligibility, I would not have written the prescription.

It is noteworthy that Reagan used words like “qualification” and “eligibility” to justify his actions rather than discussing the appropriateness of the decision.

C. Making Options Available

No information indicates Reagan was trying to find any feasible alternatives to suicide. In the taped interview with Helen, he told her that it is important she understand that there are other choices she could make that he will list for her, and in three sentences covering hospice support, chemotherapy, and hormonal therapy, he did:

[Reagan]: There is, of course all sorts of hospice support that is available to you.

There is, of course, chemotherapy that is available that may or may not have any effect, not in curing your cancer, but perhaps in lengthening your life to some extent. And there’s also available a hormone which you were offered before by the oncologist—tamoxifen—which is not really chemotherapy, but would also have some possibility of slowing or stopping the course of the disease for some period of time.

[Helen]: Yes, I didn’t want to take that.

[Reagan]: All right, OK, that’s pretty much what you need to understand.

During the taped remarks, Helen expressed concern about being artificially fed, a concern that may have contributed to her request for assisted suicide, and suggests greater anxiety and uncertainty about her course of action than the physician perceived. One would expect him to have assured her that this need not happen in any case. Instead he ignored the remark and changed the subject by asking a question about her desire to die.

Reagan was impressed by Helen’s determination to die. In an interview with Oregon Public Broadcasting, he described talking to her as “‘like talking to a locomotive'” in her desire for death even though she was not in great immediate distress. Although Reagan was troubled by her haste, and with good reason—such stubborn urgency is often a sign of irrational motives—he was unable to resist it. As striking as Helen’s determination is, Reagan’s haste in presenting and moving past the alternatives to assisted suicide is even more striking. Reagan subsequently wrote an article for the British journal Lancet, in which he stated that he liked Helen immediately. He continued:

The thought of Helen dying so soon was almost too much to bear… On the other hand, I found even worse the thought of disappointing this family. If I backed out, they’d feel about me the way they had about their previous doctor, that I had strong them along, and in a way, insulted them.

Neither reluctance to disappoint her family nor embarrassment at backing out should have been such a significant factor in the decision to end her life.

D. The Palliative Care Alternative

The difference it makes if a cancer patient is seen by someone who has experience in providing palliative care is suggested by the following excerpt from a letter written by a practicing oncologist in response to the law:

As a practicing gynecologic oncologist in Portland, Oregon, where physician-assisted suicide is legal, I informed patients of my views by having a clear statement, based on the Hippocratic Oath posted in my waiting room. This reassured most patients, however, I had two patients who objected. The first was afraid that I would prolong her life beyond her wishes. This conversation helped me to meet her needs and she had a peaceful, comfortable death at home with her family. The second patient wanted me to prescribe lethal medications in case her cancer pain became unbearable. Prior to this conversation, she had been minimizing her pain. This conversation allowed us to work together to better control her pain, after which her desire for assisted suicide disappeared. She died comfortably and naturally two months later.

The OPHD’s yearly progress reports contend that patients who requested assisted suicide were receiving adequate end-of-life care, citing the frequency with which patients were in hospice care as evidence. However, a referral to hospice care cannot be regarded as a substantive palliative care intervention without knowing what care the hospice provided.

Moreover, available data contradict the OPHD’s contention of adequate care. A study at the Oregon Health & Science University indicated that there has been a greater percentage of cases of inadequately treated pain in terminally ill patients since the Oregon law went into effect. However, among patients who requested PAS but availed themselves of a substantive intervention by a physician, forty-six percent changed their minds about having PAS.

Surviving family members surveyed by the Oregon Board of Medical Examiners (“BME”) found a trend of increasing rates of moderate to severe pain reported among patients dying in acute-care hospitals throughout Oregon. This trend led the BME to conclude that inadequate palliative care was a problem in the state.

A study, Means to a Better End, by the Last Acts Program of the Robert Wood Johnson Foundation, evaluated end-of-life care in all fifty states and gave Oregon a mediocre grade. The Foundation and the Last Acts Program have no position on assisted suicide, but they have a strong commitment to improving end-of-life care. Oregon received good marks for its use of advance directives, for not overusing intensive care units in ways that only prolong the dying process, and in training registered nurses in palliative care. Oregon did poorly on five other measures utilized in the evaluations, including the large number of its nursing home residents in persistent pain, the small number of its hospitals providing hospice or palliative care, and the lack of state policies encouraging pain control and palliative care.

Supporting these findings regarding the quality of palliative care in Oregon is an anonymous survey of Oregon physicians regarding their experience in dealing with patients’ requests for assisted suicide. Physicians recommended a palliative care consultation in only thirteen percent of the first 142 requests for assisted suicide after the Oregon law went into effect; we do not know how many of these recommendations were actually implemented.

II. Psychiatric Concerns

Because Oregon was the first state to legalize suicide as a treatment for medical illness, it would seem to have a special responsibility to protect the significant number of patients who become suicidally depressed in response to serious or terminal illness. Although pain and other factors, such as lack of family support, contribute to the wish for death, researchers have found hopelessness, which is strongly correlated with depression, to be the factor that most significantly predicts the wish for death.

Patients requesting suicide need psychiatric evaluation to determine whether they are seriously depressed, mentally incompetent, or for whatever reason do not meet the criteria for assisted suicide. Oregon law, however, does not require it for patients requesting assisted suicide.

A. The Second Case: Anonymous

The next case is an example of what can happen when psychiatric consultation is not provided:

[A] woman in her mid-fifties with severe heart disease … requested assisted suicide from her cardiologist, despite having little discomfort and good mobility She was referred to another doctor, who in turn referred her to a physician willing to provide assisted suicide. That doctor determined that the woman had more than six months to live, according to his best estimate. Therefore, she was eventually dismissed as ineligible. Rather than inquire further into possible causes of [her] suicidal despair [or refer her for psychiatric treatment], the physician apparently considered … his responsibility ended…. [H]e told her to go back and make yet another appointment with her original physician and dismissed her. She killed herself the next day.

Under the Oregon law, only if the physician believes a patient requesting assisted suicide is suffering from a psychiatric or psychological disorder or from a depression causing impaired judgment must the physician refer the patient to a licensed psychiatrist or psychologist. The caveat about impaired judgment is strange, since impaired cognitive function is one of the characteristics of a depressive disorder; a rigid tendency to see only one possible solution (such as suicide) to their problems is also characteristic. In any case, a number of studies have shown that physicians are not reliably able to diagnose depression, let alone to determine whether depression is impairing judgment.

More than pain, depression, or current distress is often involved in patients’ requests for assisted suicide. Many patients who request assisted suicide are doing so out of fear of what will happen to them. Such fears often derive from the patient’s past experience with the death of someone close to him or her, so a history of these experiences should be part of any physician’s evaluation of a request for assisted suicide. That evaluation must reflect an awareness of risk factors for suicide, such as alcoholism, a past history of depression, and, of course, any prior suicide attempts.

Patients attempting assisted suicide are usually ambivalent about dying; so too are patients requesting assisted suicide. Physicians inexperienced in dealing with suicidal patients tend not to hear this ambivalence. Therefore, they are likely to take such requests to die literally and concretely and to act on them.

A guidebook for health care professionals written by the Oregon University Center for Ethics advises physicians to refer all cases requesting assisted suicide for psychiatric evaluation, even though physicians are not legally required to do so. Oregon physicians are not following that advice. The percentage of cases referred for psychiatric evaluation dropped from thirteen percent in the eight years between 1998 and 2005 to four percent in 2006.

B. The Third Case: Joan Lucas

The psychiatric evaluation when employed in Oregon, however, like the palliative care recommendations, has the tendency to be utilized to protect clinicians rather than patients, as the following case illustrates.

Joan Lucas, a patient with amyotrophic lateral sclerosis, attempted suicide. Paramedics were called to Joan’s house, but her children sent them away, explaining, “‘We couldn’t let her go to the ambulance. They would have resuscitated her.'” Joan survived her attempt and was assisted in suicide eighteen days later by a physician who gave interviews about the case to an Oregon newspaper on condition of anonymity He stated that after talking with attorneys from the Oregon Medical Association and agreeing to help aid Joan in death, he asked Joan to undergo a psychological examination. The doctor reported that “‘[i]t was an option for us to get a psychological or psychiatric evaluation. I elected to get a psychological evaluation because I wanted to cover my ass. I didn’t want there to be any problems.'”

The doctor and the family found a cooperative psychologist who asked Joan to take the Minnesota Multiphasic Inventory, a standard psychological test. Because it was difficult for Joan to travel to the psychologist’s office, her children read the true-false questions to her at home. The family found the questions funny, and Joan’s daughter described the family as “‘cracking up'” over them. Based on these test results, the psychologist concluded that whatever depression Joan had was directly related to her terminal illness, which he considered a completely normal response.

When Oregon psychiatrists were surveyed, only six percent felt very confident that, absent a long-term relationship with a patient, they could satisfactorily determine whether a patient was competent to commit suicide. The psychologist’s report in Joan’s case is particularly disturbing because without taking the trouble to see her, and on the basis of a single questionnaire administered by her family, he was willing to give an opinion that would facilitate ending Joan’s life. The physician’s attitude toward the consultation surely played a part in his receiving a report that did not meet professional standards.

OPHD’s monitoring procedures do not make it possible for OPHD to evaluate the care Joan Lucas received. To do so OPHD would have to interview Joan’s primary care physician who had refused to assist in her suicide and to assess the quality of her psychological evaluation. Using psychologists or psychiatrists as gatekeepers only to establish a patient’s capacity to make a decision for assisted suicide contributes to pro forma, meaningless consultations.

In the Lucas case, we have no way of knowing if Joan Lucas was seriously depressed or if the doctor or psychologist was disposed to proceed even if she were. Even more troubling is that OPHD does not seem to want to know about the psychiatric status of patients requesting assisted suicide. Under the current monitoring system, OPHD collects no information from psychiatrists who did not find patients to be competent and has no direct communication with psychiatrists or psychologists who did. Its monitoring reflects a lack of concern with the welfare of depressed patients.

C. Context of Mental Health Evaluation

Although a competent professional psychiatric evaluation is necessary to determine if a patient has impaired judgment that would make him or her not “capable” of an “informed decision,” as required by Oregon law, it is needed for other reasons as well. We know that patients requesting a physician’s assistance in suicide are usually telling us that they desperately need relief from their mental and physical distress and that without such relief they would rather die. When they are treated by a physician who can hear their desperation, understand the ambivalence that most feel about their request, treat their depression, and relieve their suffering, their wish to die usually disappears.

If the patient has seen no one knowledgeable enough to undertake to understand and relieve the desperation, anxiety, and depression that underlie most requests for assisted suicide, then even if the patient is capable, an informed decision is not possible.

III. Voluntariness and the Risk of Coercion

The Oregon law requires both that patients requesting assisted suicide voluntarily give informed consent to the procedure and that they have the mental capacity to do so, but it lacks safeguards to ensure that this takes place. OPHD’s monitoring does nothing to correct the problem.

A. The Fourth Case: Kate Cheney

The case of Kate Cheney, as described by both Cheney and those members of her family who told their story to The Oregonian, highlights the deficiencies in the informed consent procedures. An eighty-five-year-old widow, Kate was diagnosed with terminal stomach cancer. Kate wanted the option of assisted suicide in case she was in pain or if the indignities of losing control of her bodily functions became unbearable. Her daughter Erika, a retired nurse who had come from Arizona to care for her mother, went with Kate when Kate made the request to her physician at Kaiser Permanente. Erika described the physician as “‘dismissive'” and requested and received a referral to a second physician. He arranged for a psychiatric consultation, a standard procedure at Kaiser. The psychiatrist, who visited Kate at her home, found that Kate did not “‘seem to be explicitly pushing for [assisted suicide]'” and lacked the “‘level of capacity … to weigh options about [it].'” Although Kate seemed to accept the assessment, Erika became very angry.

Kaiser then suggested that the family obtain a second assessment from an outside consultant. The psychologist consulted noted that Kate had memory defects and that her “‘choices [might have been] influenced by her family’s wishes and [that] her daughter, Erika, [might have been] somewhat coercive,'” but felt Kate had the ability to make her own decision. A Kaiser administrator saw Kate and decided that she was competent and was making the decision on her own. Kate received the lethal drugs, which were put under Erika’s care.

As time passed, Erika and her husband needed a respite, and they sent Kate to a nursing home for a week. When Erika visited, Kate always asked when she would be going home. On the day she returned from the nursing home, she told Erika and her husband that she had considered going permanently into a nursing home but had decided to use the pills instead and asked for their help. Her son-in-law asked, “‘When would you like to do this?'” Kate replied, “‘Now.'” Within a short time, with her family beside her, Kate took the pills and died.

The eagerness of her daughter and son-in-law are likely to have influenced Kate’s decision. One wonders if the decision would have been different if her family had responded to her request by saying, “We love you and want to keep you at home and care for you as long as possible.” Sending Kate to the nursing home conveyed that she was a burden to her family Her poignant and repeated requests to go home expressed her distress, as did her request to end her life on the day she did so.

One can readily see how in the best of circumstances frail, elderly patients can feel coerced to die. Caregiver burden has been identified as a serious issue, particularly for women like Erika who are asked to shoulder the work and responsibility of providing twenty-four-hour care to a parent. This particular case raises the question of what real meaning or value Oregon’s prohibition of coercion has if it can be circumvented so easily

B. Financial Considerations

Since ongoing care for terminally ill patients is far more expensive than assisted suicide, the role of a single health maintenance organization (“HMO”) administrator in making the final decision in a matter in which the HMO might have a financial conflict of interest, as in Kate’s case, was questionable. Would the HMO have asked for a second opinion if the psychiatrist had deemed the patient competent to request assisted suicide? The Kaiser administrator was indignant at a journalist’s implication that financial considerations might have influenced both his recommendation to Kate’s family to seek an outside consultant and his own final decision. Yet this case makes a compelling argument for the need for openness and transparency—and perhaps even judicial review of competency determinations—because of the competing interests in deciding what was appropriate for a vulnerable elderly patient whose competency was in question and whose family may have been seriously burdened by her care.

C. Consulting with Family

Under the Oregon law, physicians are required to suggest that patients inform their families of their request for assisted suicide, but the patients are not required to do so. The law instructs physicians not to deny the request on the basis of such a refusal. Even if the patient complies, the physician is not required by law to ask to see the patient’s family.

How can any physician be sure there is no coercion unless the physician has met the family and seen the interaction among them and with the patient? On the other hand, not informing the family can prevent a caring family from expressing their affection in ways that might alter the patient’s decision. It also opens the family up to the devastating grief and guilt that we see in survivors of suicide. Much of that guilt comes from feeling there were things they could or should have done to encourage the person who committed suicide to want to live. Feeling cut off from what a loved one was going through before the act is a major contribution to such anguish. Advocates of assisted suicide argue that legalization, by permitting the family to be part of the process, should ameliorate such suffering. Not informing the family makes this impossible.

This problem, to which OPHD seems oblivious, is cited by social workers in Oregon hospices as providing a serious challenge to hospice care professionals who feel that in protecting patients’ confidentiality they have failed to help the patients’ families, and they feel split in their allegiance between the patients and their families. The provision of the Oregon law stating that a patient who declines to inform his or her family “shall not have his or her request denied for that reason” is too sweeping in scope, and monitoring is necessary to learn its consequences.

IV. Protection for Physicians Instead of Patients

A concern with physician protection, rather than patient protection, pervades the Oregon experience. The statute’s liability shield, its incomplete reporting system, and the excessive secrecy with which the law is implemented protect doctors who assist patients with suicide, but leave patients vulnerable.

A. Lower Standard of Liability

Oregon physicians assisting in a suicide are exempt from the ordinary standards of care, skill, and diligence required of Oregon physicians in other circumstances (e.g., a physician’s conduct when withdrawing life support). Instead, the physician is immunized from civil and criminal liability for actions taken in “good faith” in assisting a suicide irrespective of community standards in other matters and even when the physician acts negligently.

Good faith is a troublesome, subjective standard. Homicide law provides an example of an area where a good faith test might be appropriate. In some jurisdictions, when a person actually—but unreasonably—believes he must kill in self-defense, the person may nonetheless avail himself of the right of self-defense (which means he is guilty of manslaughter rather than murder). Applying a similar good faith standard to physicians seems curious. As Professor Dan Dobbs has noted:

An instruction [in a medical malpractice case] that tells the jury that the physician is not liable for honest error or good faith mistake injects subjective … issues into [what is usually an] objective negligence test and may lead the jury to think that bad faith, not a departure from professional standards, is the test of liability This kind of instruction is now widely condemned by appellate courts.

In professional practices a negligence standard based on objective, established medical guidelines is customary. If the intent of the assisted suicide law is to protect physicians from accountability for violating the statute’s provision, the good faith standard is ideal. But if the intent of the law is to provide protection for patients, a negligence standard would be more appropriate.

B. Reporting System Lacks Teeth

The fact that it may not be possible to punish physicians even if they have behaved irresponsibly is not a reason for not monitoring their behavior. Ensuring adequate care for patients is the aim of monitoring, and without knowledge of the quality of care being given, that is not possible.

However, there is no enforcement mechanism in the Oregon law should physicians not comply with guidelines set up by OPHD for reporting all cases in which medication for the purpose of assisted suicide has been prescribed. The Dutch experience suggests that even if the Oregon law had noncompliance penalties, nonreporting would still be a serious problem. By continually focusing on this problem, the Dutch have slowly been able to improve reporting. But since OPHD has not addressed the question of non-reporting, it is in the position of drawing conclusions from limited data.

C. Excessive Secrecy

OPHD has focused more on patient-doctor confidentiality than on monitoring compliance or abuse. The agency has developed confidentiality measures unique to physician-assisted suicide which appear to be unnecessarily secretive and limit the potential for thorough research into the dimensions and context of this practice as it unfolds. For example, internal memoranda from OPHD to its county vital records offices instructed all employees that they should “neither confirm nor deny if a [physician-assisted] death has occurred in your county.” To underscore “how seriously this matter is being taken” by OPHD, the memo warned that “[a]ny staff within the Center for Health Statistics that reveals any information they are not authorized to release, will immediately be terminated.”

The Oregon law specifically states that although OPHD will issue a report each year based on a selected sample of cases, “the information collected shall not be a public record and may not be made available for inspection by the public.” There is no provision for an independent researcher or evaluator to study whatever data are available.

Medical standards require openness about facts, research data, and records to assess the appropriateness of treatment. The anonymity and secrecy about physician practice of assisted suicide makes such an assessment impossible. If physician-assisted suicide is to be part of the medical treatment for terminal illness, why are existing patient-doctor confidentiality rules not sufficient to protect physicians in this setting? Restricting access to information about the indications for assisted suicide, patient data, radiologic documentation, and specific drug therapy limits the opportunity to establish an objective standard of care, provides excessive protection to the physician and, in the name of confidentiality, leaves the patient vulnerable.

The law sets a low legal standard for physicians’ conduct, and OPHD does not provide a mechanism for ascertaining whether physicians are meeting even this reduced standard, thereby precluding accountability.

V. The Role of Compassion in Dying

Compassion in Dying executives have indicated that the organization has been involved in seventy-five percent of all cases of PAS since the implementation of the Oregon law. In a study of the role of non-governmental organizations in physician-assisted suicide, however, Stephen J. Ziegler and Georg Bosshard observed that advocacy organizations have unresolved problems in their relationship with doctors, as the following case illustrates.

A. The Fifth Case: A Desperate Wife

Information about this case comes from a talk given by George Eighmey, the executive director of Compassion in Dying of Oregon, to state regulators about Oregon’s experience with physician-assisted suicide. Eighmey described a case in which “a woman who was desperate” called Eighmey and said, “‘I can’t take it any more. My husband is begging me to kill him, I cannot stand his continued suffering any more. I love him too much.'” Eighmey describes the call and its aftermath as follows:

I begged her to wait and she said, “Unless you’re at the door with the pills, don’t come.” I said, “I cannot be there with the pills, I don’t do that.” But wait—I arrived at her door, she opened the door, and as with a lot of people who are in emotional states, she saw me and started laughing and crying simultaneously and I hugged her and I walked in and we sat for three hours, talking to her husband and to her at length about the process. Fortunately, her … ah … his physician had already noted in the file that [the patient] had asked ]another physician] for Oregon’s Death With Dignity fifteen days prior, so the time had elapsed. So we said, you have to ask for it a second time and you have to put it in writing. And then forty-eight hours after the writing you may obtain the prescription.

The initial physician was unwilling to provide the patient with a prescription, but the advocates helped the patient to make a second request to another physician. After the patient obtained the prescription, Eighmey and two other members of Compassion in Dying went to the patient’s house. He describes what happened:

[W]e three Compassion in Dying members were present, the wife, the two friends across the street, and we were preparing everything. [The patient] came up and asked, “What do I wear, and where do I go?” We said, “You might do it in bed, or do it wherever you wish, but we recommend that you do it in bed.” [He] crawled into bed, and we left [him] and his wife together for a while. We came in with the medication and we said, “Now you have the choice to change your mind at any time. Please, please do not feel compelled to do this.” And he said, “I want to do it. I have had a beautiful life, I have had a loving wife, and it is my time. I said goodbye to this earth.” We handed it to him; he took it and he turned to his wife and said to his wife, “I love you very much. We had a good life.” In five minutes he was in a deep coma, and died in seventeen minutes. And that is what being open and honest and above-board and regulated by a state statute means in the state of Oregon. We have compassion for people who wish to die with dignity.

The role of an advocacy group and a lay advocate in facilitating assisted suicide with this patient is troublesome. The advocates formed a relationship with the desperate wife and then coached the patient and his wife in how to access PAS by taking advantage of a loophole in the law that does not stipulate that the two requests for assisted suicide must be made to the same physician. The role of the advocates was to help the patient and family get what they wanted, not to assess whether this was an appropriate option for the patient.

Compassion in Dying identifies their role as helping patients find physicians who will provide them with assistance in death. But advocacy can run amok when passionate volunteers interfacing with demanding families see no options but a prescription for lethal medication. We do not think such volunteers are likely to be able to assess competency or complex psychological issues or family dynamics that may influence the patient’s request. Eighmey seems to have an exaggerated idea of his own ability to do so, stating that patients “tell me more in [a] half-an-hour phone call than they sometimes will tell their physician or their spouse. I know more about their life history in that half an hour than a lot of other people close to them.”

The union of Compassion in Dying with the Hemlock Society and the name of the merged organizations, Compassion and Choices, permits them to avoid the word “dying,” and the association of the word “hemlock” with lethality. We need to have a clearer picture of the role of such advocacy groups in coaching patients who seek their help and in helping patients to have real choices.

VI. A Comparison of Two Different Approaches

The implementation of the Oregon law on assisted suicide encouraged physicians to adopt a different approach to patients with serious medical illness. In the medical model, modified by advances in palliative care and practiced in states other than Oregon, patients requesting assistance in suicide are assessed in the same way as other patients intent on suicide. The medical model recognizes that “[a]lthough physical illness may be a precipitating cause of despair, these patients usually suffer from treatable depression and are [almost] always ambivalent about their desire for death.” Study of terminally ill cancer patients has demonstrated that those preoccupied with assisted suicide had symptoms of depression or hopelessness.

To help these patients, we need to understand and relieve the desperation that underlies the request for assisted suicide. Supportive psychotherapy, antidepressant medication, and good palliative care are instrumental in providing relief from distress and making it possible for patients to appreciate and even enjoy whatever time they have left to live.

Oregon’s assisted suicide guidebook takes a totally different approach. It stresses that any mental health consultation should be an evaluation of competency focused on the patient’s capacity to make a decision,  emphasizing that “[t]he presence of depression does not necessarily mean that the patient is incompetent.” Whether or not one agrees with the majority of clinical psychiatrists and forensic psychiatrists, who believe “that the presence of major depressive disorder should result in an automatic finding of incompetence” to make decisions about assisted suicide, reducing the psychiatric consultation to the issue of competency ignores all the other psychological factors that go into the request for assisted suicide.

A. The Sixth Case: Mr. A

A dramatic illustration of the contrast between these two distinct approaches for dealing with suicidal preoccupation in the seriously ill can be found in the case that follows, in which the patient had substantial contact with Physicians for Compassionate Care, an organization that follows the medical model, and with Compassion in Dying, which follows the assisted suicide competency model.

Right after receiving a diagnosis that he had inoperable lung cancer, Mr. A, a sixty-three-year-old computer technician, called Physicians for Compassionate Care requesting information on how he could get drugs for assisted suicide. He said, “‘I might as well just end it.”‘ The volunteer responded by saying Mr. A was understandably upset at this news. In response, Mr. A became tearful.

The volunteer began a series of phone conversations with the patient about his cancer, his treatment options, family support network, and his own personal history. The patient described his state of mind:

[He was] haunted by suicidal feelings ever since his mother died from a self-inflicted gunshot wound when he was 21. Shortly after her death he had attempted suicide himself and was treated for depression in a psychiatric hospital. He made at least two other suicide attempts and remained preoccupied with suicide.

He had a history of alcoholism but had joined Alcoholics Anonymous and had been sober for more than twenty years. In addition, the volunteer learned Mr. A was not currently in pain.

The volunteer assured him that good palliative care was available for any symptoms he might develop. With her support, he began treatment for his cancer, including chemotherapy and radiation, and received medication for his depression.

Prior to contacting Physicians for Compassionate Care, Mr. A had been a suspicious person. He did not allow others into his home, which he protected through extensive surveillance and ownership of assault weapons. After a few months of treatment, however, Mr. A was able to allow people into his home. The volunteer began monthly home visits and regular phone calls; Mr. AS daughter eventually moved in with him to help in his care.

While he had been talking to the volunteer, Mr. A had also sought help from two physicians whom he knew were activists for assisted suicide. The first physician who evaluated him gave him a lethal prescription; the second, who was affiliated with Compassion in Dying, regularly communicated with Mr. A about the assisted suicide option. Neither doctor had thought he needed a psychiatric consultation, but neither had known or asked about Mr. As history of depression and past suicide attempts.

Eighteen months after initially receiving his diagnosis, Mr. A became increasingly agitated and was admitted to a psychiatric hospital after expressing thoughts that were both suicidal and homicidal. His daughter had to move out because of his combative behavior. Mr. A was given a DSM-W diagnosis of a depressive disorder; when he responded to treatment in the hospital, he was discharged. The volunteer from Physicians for Compassionate Care increased the frequency of her visits. This was a good period for Mr. A because he was able to enjoy regular visits from old friends and reconciled with his daughter.

After a while Mr. A developed excruciating constipation from his pain medication which led him to discontinue the medication, dismiss hospice, and consider using the medication he had been given for assisted suicide. When he was given fluids to relieve his constipation and prescribed a morphine pump and twenty-four-hour attendant care, however, his pain abated and his mood improved. During the weeks he had left, he said goodbye to his friends and expressed his appreciation to the volunteer and others who had helped him.

There are striking differences in the two approaches. The volunteer for Physicians for Compassionate Care understood that the patient’s depression and anxiety were an integral part of his wanting to end his life. By phone and in visits to his home, she maintained a relationship with him and was instrumental in seeing that he received the care he needed. Through communication with his nurse, she encouraged his primary doctor to prescribe antidepressant medication for him. When toward the end of his life he became desperate because of poorly treated pain, she saw to it that he received the care he needed.

The contrast with the two doctors associated with Compassion in Dying is significant. Without inquiring about a past history of depression or suicide attempts, “the doctor who prescribed the assisted suicide drugs … told the patient and his daughter that a psychiatric evaluation would not be ‘necessary.'” He later admitted that he would have obtained such an evaluation had he known of the prior suicide attempts. He did think, however, that “giving Mr. A the assisted suicide drugs may have added to his sense of control and security and may even have prolonged his life.” Yet it seems likely that Mr. A would have used the pills at least a year before his death if it had not been for the caring and knowledgeable intervention of the volunteer from Physicians for Compassionate Care.

B. Predicting When a Patient Will Die

An additional problem in the implementation of the Oregon law is its stipulation that eligibility for assisted suicide depends upon patients having six months or less to live. Predictions regarding terminal illness vary in accuracy depending on the disease involved—somewhat higher accuracy for cancer (although not in Mr. As case) and lower for cardiovascular disease. The majority of Oregon physicians, when surveyed, were not confident they could make such a prediction. The nine-year data suggest that a significant number of patients live beyond their six-month prognosis. OPHD does not indicate the time interval that elapsed until their deaths, thus preventing evaluation of the reliability of this crucial legal criterion and hiding from the public the uncertainty of these predictions.

VII. Ten-Year Perspective

A number of medical, psychological, social, and cultural factors have been influencing attitudes toward physician-assisted suicide in the past decade.

A. Palliative Care

The advance in palliative care in the past ten years that has most diminished the need for assisted suicide and euthanasia is the increasing understanding and acceptance that to relieve suffering, it is often necessary to administer pain medication even in doses that might shorten the patient’s life. The medical profession, the U.S. Supreme Court, and most religious groups have come to this realization. Lack of knowledge by physicians of established guidelines on withholding care and the use of palliative care approaches has led to confusion between foregoing life-sustaining therapy (the legal right of every competent patient) and active euthanasia. Such uncertainty results in inadequate control of distressing symptoms in terminally ill patients. Some clinicians have argued that morphine drips in such cases are a form of “slow euthanasia.” There is a distinction, however, between the intent of palliative care physicians whose goal is to prevent and treat suffering, and those who intend to hasten death. Specialists in palliative care have developed guidelines for the aggressive pharmacological management of intractable symptoms in dying patients, including sedation for those near death.

We now know that that proper use of pain medications in patients with chronic pain, as well as patients at the end of life, does not hasten death. Studies have demonstrated that dying patients who received morphine lived longer than those who did not receive morphine.

Efforts at educating physicians appear to be making a difference in both the United States and the Netherlands. The more physicians know about palliative care, the less they favor assisted suicide; the less they know, the more they favor it. The more critical question is whether it changes the way they practice medicine. In the Netherlands, where there was evidence that interesting doctors in palliative care was made more difficult because of the easier alternatives of assisted suicide and euthanasia, the Dutch undertook a national program to bring palliative care and hospice care to the population.

In 2005, for the first time since the Netherlands legalized assisted suicide and euthanasia, a survey showed a slight drop in assisted suicide and a significant drop in euthanasia. There are now reports by some Dutch doctors who have performed euthanasia that, had they known then what they know now, they would have treated their patients differently. Such a development was hardly conceivable ten years ago.

B. Autonomy and Control

On the other hand, what is most likely to increase the demand for assisted suicide is the impetus to treat the question as one of autonomy and control. Oregon has been in the forefront of this trend. The original impetus for passage of the Oregon law was to help relieve intractable symptoms such as pain, but as the law was written and monitored it has evolved into providing an option for control. Oregon physicians report that the most common reason patients request PAS is not pain or depression but a need for control. This need is usually related to patients’ fears of the future and presents the physician with an opportunity to address their specific concerns and to develop interventions that will relieve the anxiety of most patients. Oregon researchers have described these patients, noting that they were inflexible and “dreaded the thought of being dependent on others.”

The need for control, however, is characteristic of most suicidal patients. They make absolute conditions on life: “I won’t live … without my husband,” “if I lose my looks, power, prestige, or health” or “if I am going to die soon.” Depression, often precipitated by discovering a serious illness, exaggerates their tendency to see life in black and white terms, but for most such people the need for control has been a dominant feature of their lives. They are unable to tolerate dependency on other people. In any case, the good practice of medicine obliges doctors to relieve distress rather than to assume that hastening death is the best or only way of doing so.

C. Oregon: What We Know and Need to Know

What has the Oregon experience with PAS taught us? Given the expectations that Oregon could serve as a laboratory for understanding and assessing physician-assisted suicide, not very much. Sadly, OPHD is wasting the opportunity to study a natural experiment and to provide understanding of the needs of patients and families at the end of their lives.

To date, OPHD figures indicate that since the Oregon assisted suicide law was enacted, 292 Oregonians have used PAS to die between 1998 and 2006; 456 received prescriptions to do so. Those who did not use them either died of natural causes or are still alive. Sixteen Oregonians used PAS in 1998, and that number has almost tripled, rising to forty-six in 2006. The ratio of PAS deaths to total deaths in Oregon has increased from 5 in 10,000 in 1998 to 14.7 in 10,000 in 2006.

From the time the Oregon law went into effect, OPHD officials have admitted that they have no way of knowing how many PAS cases are not reported. If OPHD wished to know what is going on, it would need to follow the Dutch example by granting physicians full immunity and then surveying them with questionnaires and interviews.

The OPHD annual reports are marked by overreaching in the conclusions they draw from the limited information they have. As we have previously discussed, most striking and least justified has been OPHDE contention, without substantiating data, that patients who have requested assisted suicide were receiving adequate end-of-life care. Data from Oregon investigators, surveys of and interviews with families who observed the pain or distress of their relatives who received end-of-life care, new surveys of nurses who cared for hospice patients, and new surveys of physicians’ experiences do not support this contention.

It has been possible to learn enough to know the defects that are there, but much more will need to be known if they are to be addressed and corrected. Any effective change in the way physician-assisted suicide is practiced in Oregon will require OPHD to become a more effective monitor.

What should OPHD be doing? As we have noted, apart from the restriction in the Oregon law that prohibits independent researchers from having access to the data, OPHD has been given great flexibility to do its own research.

OPHD could correct its most glaring limitation by expanding its collection of information beyond physicians who have written lethal prescriptions. OPHD should interview doctors who, for whatever reason, declined to prescribe lethal medication; psychiatrists who evaluated these patients (whether or not they found them to be competent); and nurses, social workers, or family members who cared for the patients. Without such information, we have no idea of how many requests for assisted suicide there are each year, why some physicians declined while others agreed to proceed, or what transpired in any particular case.

OPHD justifies obtaining information only from physicians who prescribed medication that patients actually used to end their lives as necessary “[t]o maintain consistency in data collection and to protect the privacy of the patient and the prescribing physician.” Limiting the information collected to one physician when other physicians who saw the patient in connection with an end-of-life decision might have information that would be contradictory runs counter to the basic expectations of research design and undermines the validity of the results. As for privacy, if OPHD collected information from all physicians who received a request for assisted suicide, OPHD would have valuable information about the patients, but the physicians who provided the information and had not written the prescription would not by this process know the outcome, and the privacy of the prescribing physicians and their patients would be protected.

That information could also serve to evaluate the use of a six-month prognosis to define terminal illness; recall that some of the patients receiving prescriptions lived for a year or longer. OPHD should calculate the survival time of patients receiving a prescription and provide summary data on the validity of this criterion.

As the years go by, OPHD is providing less information when it should be providing more. For example, only in OPHD’s reports for 2004 and 2005 were the number of prescriptions written per physician cited. We were told that the maximum number written by any one doctor was seven in 2004 and eight in 2005. We know from other sources, however, that in one hospice that has had twenty-eight PAS cases since 1997, a single doctor was involved in twenty-three of them. Whether the doctor was the prescribing physician or the consultant is not clear, but OPHD could ascertain that information. Another piece of data that was only reported for the first two years but that OPHD continues to collect is the number of physicians a patient must see to obtain a prescription. In 1999, ten of the twenty-seven cases went to a second physician and eight went to a third or fourth physician. This information would clarify whether a small cohort of physicians is involved in a majority of the cases and might suggest study of how physicians’ biases may be influencing patients’ requests.

D. Need for Change

So far OPHD has been collecting limited data and has not tried to address any of the controversial issues the cases we have discussed present. They have rather responded to pressure from advocacy groups not to use the term “assisted suicide.” Since the inception of the Oregon law, OPHD had used the internationally accepted term “physician-assisted suicide” in its annual reports and on its website. In 2006, however, Compassion and Choices, heeding polling data that indicated that the public responds unfavorably to the practice when the term “assisted suicide” is used, lobbied OPHD not to use it. OPHD briefly considered using “physician-assisted death,” but gave that up in response to strong objections from PAS opponents and settled on “death with dignity.”

Patients who take a legally prescribed drug overdose to end their lives are to be referred to as “persons who use the Oregon Death with Dignity Act.” Derek Humphry, the founder of the Hemlock Society, objected strongly to the change, calling the term “death with dignity” “wildly ambiguous.”

As the Oregon assisted suicide law is currently implemented, “Death with Dignity Act” is something of a misnomer. When patients like Kate Cheney die because their relatives no longer want to care for them, they die feeling rejected, not dignified. Mr. A was terrified when he learned he had an illness that would be fatal. He assumed its course would be painful and a hastened death the only solution. He was supported in that notion by Compassion in Dying. He would likely have died feeling distraught, not dignified, had it not been for the fortunate intervention of a truly compassionate volunteer.

As we have previously noted, under the current monitoring system, Oregon physicians appear to have been given great power without being in a position to exercise it responsibly. They are expected to inform patients that alternatives are possible without being required to be knowledgeable about such alternatives or to consult with someone who is. They are expected to evaluate patient decision-making capacity and judgment without a requirement for psychiatric expertise or consultation. They are expected to make decisions about voluntariness without having to see those close to the patient who may exert a variety of pressures, from subtle to coercive. They are expected to do all of this without necessarily knowing the patient for more than fifteen days. Since physicians cannot be held responsible for wrongful deaths if they have acted in good faith, substandard medical practice is permitted, physicians are protected from the consequences, and patients are left unprotected while believing they have acquired a new right.