Legalization of assisted suicide in Oregon ushered in a new approach to evaluating suicidal patients with serious medical illnesses. Two competing paradigms—the traditional model (1–4) and the assisted suicide competency model (5)—now exist. This report compares the traditional approach to evaluating and treating suicidal symptoms with the assisted suicide competency model delineated in an assisted suicide guidebook used in Oregon (5), the only state where such a practice is legal.
In the chapter "Suicide, Assisted Suicide, and Euthanasia" from the Harvard Medical School Guide to Suicide Assessment and Intervention(1), Herbert Hendin pointed out that medically ill suicidal individuals are no different from other suicidal individuals. Although physical illness may be a precipitating cause of despair, these patients usually suffer from treatable depression and are always ambivalent about their desire for death. This conclusion is consistent with epidemiological evidence showing that poor health is not an independent risk factor for death by suicide but is correlated with depression or other mental illness as a key intervening variable (4, 6). A noted large-scale study (7) also demonstrated that terminally ill individuals expressing an interest in assisted suicide all suffered from symptoms of depression or irrational hopelessness. Kissane et al. (8), who studied assisted suicide cases in Australia during that country’s brief experiment with this practice, termed this latter factor "demoralization syndrome."
The traditional clinical approach to dealing with assisted suicide requests is the same as that of treating any other suicidal symptom. It begins with assessment. After an open-ended portion of the interview aimed at empathically understanding the patient, the psychiatrist typically inquires about the onset and recurrence of psychiatric symptoms, previous similar episodes and treatments, recent stresses, social and economic difficulties, and religious or spiritual concerns. Symptoms of depression and substance abuse are noted. In this population, the psychiatrist must pay particular attention to medications that can cause or exacerbate psychiatric disturbance, cancers, or other illnesses known to cause depression or anxiety; the adequacy of pain control; and whether or not the patient has been reassured about the effectiveness of aggressive pain management and other palliative care interventions.
Thoughtful clinicians consider it equally important to explore sources of hope, self-esteem, and strength. At some point, the clinician directly asks about the seriousness and urgency of suicidal intent and the availability of means, including access to firearms and potentially lethal medications. The psychiatrist must also explore the patient’s ambivalence about dying, which is virtually always present (1).
When it comes to treatment, the approach in this population emphasizes an effort to "understand and relieve the desperation that underlies the request for assisted suicide" (1, p. 553). To do so, the clinician must resist assuming the role of "gatekeeper," someone who would focus on issues of competence alone (1, 8, 9). Such patients often suffer from feelings of worthlessness, demoralization, or guilt and may be making a plea for reassurance; supportive psychotherapy should address these issues.
As Varghese and Kelly (10) so clearly demonstrated in their chapter "Countertransference and Assisted Suicide," physicians typically must deal with their own feelings of helplessness in the face of death (11). Hamilton and Hamilton (3) discussed the process of containing (3, 12, 13) the feelings of troubled patients requesting assisted suicide.
Antidepressant, antianxiety, or psychostimulant medication can play a crucial role in alleviating underlying depression or fear leading to desperation. Most psychiatrists have heard a patient convincingly describe seemingly hopeless circumstances, only to find the patient’s perception of the same circumstances entirely changed after a successful course of antidepressant medication. Although many patients with suicidal symptoms can be treated as outpatients, hospitalization may also be needed in those requesting assisted suicide.
The traditional approach, then, assumes that suicidal symptoms in the seriously ill should be evaluated and treated as they are in all other patients (1) and that such evaluation and treatment can be extremely helpful, often lifesaving.
In sharp contrast to the traditional clinical approach, Oregon’s assisted suicide guidebook emphasizes that mental health consultation, when required at all, should be "a form of a competency evaluation, specifically focused on capacity" (5, p. 30) to make a decision. Ganzini and Farrenkopf, who authored the mental health section, stated, "The evaluation should focus on assessing the patient’s competency and factors that limit competency such as mental disorders, knowledge deficits, and coercion" (5, p. 30). These authors insist that the presence of a mental disorder does not disqualify a patient from assisted suicide. They emphasized, "The presence of depression does not necessarily mean that the patient is incompetent" (p. 31). This opinion is at variance with the majority of 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 (14, p. 595).
In the assisted suicide competency model, as used in Oregon, there is no obligation to treat depression or any other mental illness, even when one is found. The guidebook concludes, "If the mental health professional finds the patient competent, refusal of mental health treatment by the patient does not constitute a legal barrier to receiving a prescription for a lethal dose of medication" (5, p. 31).
The guidebook mentions the importance of determining the presence or absence of coercion as a part of the competency determination. As these guidelines are applied, however, coercion is narrowly defined. Such was the case in the widely discussed assisted suicide of Kate Cheney (15–17), an 85-year-old cancer patient with growing dementia whose psychiatrist believed she was being pressured by her family. Nevertheless, she was approved for an overdose by a psychologist and was given assisted suicide in Oregon.
Oregon law requires that a patient who makes an initial assisted suicide request be judged to have less than 6 months to live. A second physician must confirm the prognosis. An assisted suicide doctor typically chooses this consultant. There is no requirement for a psychiatric evaluation. Only if the doctor intending to write the prescription for the overdose or the consultant believes that the patient has seriously impaired judgment because of a mental disorder is there any requirement for referral to a psychiatrist. In actual practice, the percentage sent for mental health consultation prior to assisted suicide in Oregon has steadily dropped over 5 years to 5% (18).
No more dramatic illustration of the inconsistencies in these two distinct paradigms for dealing with suicidal ideation in the seriously ill can be found than in the case we present here. To our knowledge, this is the first reported case of a patient legally prescribed assisted suicide drugs for which medical records have been made available.
Mr. A, a 63-year-old lung cancer patient, made a telephone call to Physicians for Compassionate Care, a medical group dedicated to improving the care of the seriously ill and opposed to the practice of assisted suicide. He told the volunteer who answered the phone that he had just arrived home from his doctor’s office after receiving a diagnosis of terminal cancer, and he wanted to find out how he could get the assisted suicide drugs. Although he did not say so, apparently he thought he had dialed the Compassion in Dying Federation, a politically active assisted suicide group that shepherds the vast majority of assisted suicides in Oregon. One of the authors (C.A.H.) was the volunteer who answered the phone. Believing that this man needed to talk, she started asking questions: What kind of cancer? Did his doctor discuss treatment options? What about chemotherapy?
Mr. A sounded distraught. He explained that he saw no purpose in undergoing chemotherapy. Although he had just received his prognosis, he was already making funeral arrangements. He said, "I might as well just end it." When C.A.H. empathized with how upset he must be having just received such a dire prognosis, he became tearful. She asked him if he had a family support network. He replied that he had a daughter but did not want to tell her about his cancer because she was moving out of the state to attend graduate school, and he did not want to interfere with her education. He said he lived alone.
C.A.H. explained her views on assisted suicide and assured him that good palliative care was available for any symptoms he might develop. He wasn’t experiencing pain at the time and was working as a computer technician. As she would have with any other suicidal individual, she told him she did not want him to kill himself. She offered to help him find treatment for his depression and promised to advocate for him if he needed assistance finding a doctor who could treat pain or any other symptoms that he felt were not being addressed.
In subsequent conversations, Mr. A revealed that he had been haunted by suicidal feelings ever since his mother died from a self-inflicted gunshot wound when he was 21. Shortly after her death, he 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. Later, he developed alcoholism but joined Alcoholics Anonymous and had been sober for more than 20 years.
C.A.H. kept in monthly contact with Mr. A over the telephone during the next year. With encouragement, he started chemotherapy and radiation treatment, which alleviated his shortness of breath significantly, he reported. He also started taking an antidepressant that his primary care doctor prescribed for his depression. With urging, he told his daughter about his cancer. She gladly postponed her graduate studies and returned to Oregon.
Although he continued to be suspicious, kept a variety of loaded assault weapons, and maintained elaborate computerized surveillance cameras around the perimeter of his property, Mr. A began, for the first time in years, to allow a few people, including his daughter and C.A.H., into his home. He provided permission to communicate with his primary care doctor, if it became necessary, and C.A.H. initiated monthly supportive home visits while continuing the telephone calls. Additionally, she did such things as call his doctor’s office concerning pain care and talk with family members or friends about his care. His daughter began to reside with him to help with transportation and meals and walk his beloved 100-lb yellow laborador.
Near the anniversary of receiving his terminal prognosis, Mr. A announced, "I have the pills." He said he had obtained the overdose from Dr. B, a known assisted suicide activist who had previously published an article that described giving an overdose to a female patient, one who also had been diagnosed with depression (3, 19). Reports indicated that the woman had been referred for a competency evaluation (3, 16, 19), which had cleared her for assisted suicide approximately 2 weeks after she had met Dr. B. In contrast to that case, Mr. A was not referred for a mental health evaluation. According to Mr. A’s daughter, who accompanied him to the appointment, Dr. B said that he did not think a psychiatric consultation would be "necessary." Years later, in a telephone conversation between one of the authors (N.G.H.) and Dr. B about this case, it was revealed that Dr. B had not known about Mr. A’s history of prior suicide attempts and would have gotten a lot more evaluation if they had been known.
Mr. A mentioned that another doctor associated with the Compassion in Dying Federation had been calling him regularly to talk with him about the assisted suicide option. He had seen this doctor on a television show, and the program had "convinced me it [assisted suicide] was the way to go." When asked if that doctor knew about his depression and suicide attempts, he replied that they "didn’t get into that. Our conversations have been superficial."
Six months later, after failing to reach him for several days, C.A.H. went to his house. No one was there, and the dog was gone. She called his primary doctor to see if something had happened. The nurse told her that Mr. A had begun having more "mental problems" and had been admitted to the psychiatry ward.
According to the hospital’s psychiatric report, Mr. A’s ex-sister-in-law had begun pursuing guardianship, and he had grown upset about it. His doctor had become worried that he was increasingly suicidal—or even homicidal—and the doctor had arranged for him to be placed on involuntary hold.
An emergency room physician’s report showed that Mr. A had been evaluated for "possible suicidal or homicidal ideation." Reports from the same date by a psychiatric social worker and a mental health technician both described possible suicidal and homicidal ideation. The social worker’s report mentioned that threats may have been made by Mr. A toward a hospice nurse and toward his own daughter. Both reports documented the history of a previous overdose after his mother’s suicide. Mr. A was admitted to the hospital that day. When the inpatient psychiatrist dictated a report the next day, however, possible homicidal comments were featured prominently in the chief complaint, and suicidal intentions were all subsumed under discussion of an interest in assisted suicide. The psychiatrist’s report mentioned no history of suicide attempts and no attempt to reconcile this comment with documentation from emergency room personnel that there was a history of suicide attempts. Other than these two omissions and the failure to account for the paranoia mentioned by the emergency room doctor, the psychiatric record was fairly consistent with the known history.
The inpatient psychiatrist noted that Mr. A had a diminished appetite and had lost 70 lb but that his sleep was adequate with the numerous sedative medications he was taking. Mr. A reported confusion and some memory problems. In addition to sertraline, his medications included diazepam, temazepam, hydromorphone, and oral morphine solution.
The mental status report described Mr. A as thin and tearful. His speech was clear. His affect was discouraged. His thought processes were well organized, and he reported no thoughts of harming himself or others. Paranoia was not mentioned in his psychiatric examination. He was alert and oriented and judged to have above-average intelligence. The results of laboratory studies showed that his hemoglobin was mildly low (12.4 g/dl), but his blood count and chemistry screen results were otherwise normal.
In the discharge summary, the psychiatrist recorded this final note: "DSM-IV diagnoses. Axis I: depressive disorder not otherwise specified. Chronic adjustment disorder with depressed mood. Probably delirium, intermittent. Axis II: narcissistic personality traits. Axis III: lung cancer with guarded prognosis. Axis IV: moderate to severe. Axis V: current GAF [Global Assessment of Functioning score] 45."
A social worker home visit revealed that Mr. A’s home was uninhabitable, with heaps of clutter, rodent feces, ashes extending 2 feet from the fireplace into the living room, and a lack of food and heat. Thirty-two firearms and thousands of rounds of ammunition were removed by the police. The lethal medications, however, were left. Concerning the need for attendant care, a palliative care consultation provided by a hospitalist said the fact that Mr. A had "life-ending medications" may make that problem "a moot point."
The day after discharge, the same psychiatrist who said Mr. A kept the assisted suicide drugs "safely at home" wrote a letter supporting guardianship by saying he "is susceptible to periods of confusion and impaired judgment." He concluded that Mr. A was unable to handle his own affairs and that his cognitive impairments were unlikely to improve. As court records later revealed, his primary care doctor had written a similar letter in which he stated, "I do support guardianship…as I think both his disease process and the medications needed to control his level of pain are impairing his judgment and ability to care for himself." On the basis of this and similar information, a judge declared Mr. A incompetent to make his own medical decisions and assigned him a temporary guardian. Nevertheless, the assisted suicide drugs remained in his home. The guardianship officially lapsed after a month and was not renewed. During the same telephone conversation between N.G.H and Dr. B years after this case, it was revealed that Mr. A’s daughter had informed Dr. B about the finding of incompetence but that the doctor was not sure of having the legal right to retrieve the assisted suicide drugs once Mr. A was determined incompetent. At any rate, Dr. B wasn’t sure that would have been the right thing and commented, "That was a dilemma."
Dr. B expressed confidence that nothing illegal had been done. In Oregon, the law specifically states that patients must be considered competent to make legal decisions if a doctor is to give them assisted suicide drugs. They cannot be given the drugs if they are deemed incompetent, and drugs can only be given by the assisted suicide doctors or the consultants themselves. No formal psychiatric assessment of mental capacity is required. Neither is there a provision to retrieve the overdose should the patient become incompetent later. Thus, Dr. B’s dilemma. At any rate, assisted suicide doctors, given goodwill, are provided immunity in Oregon from any legal liability, regardless of which way they decide to deal with the dilemma about whether or not to retrieve assisted suicide drugs given to a patient. In this case, Dr. B ultimately decided not to retrieve the drugs.
At a subsequent home visit, Mr. A reminded C.A.H. that he had already far outlived the original prognosis of 6 months to live, and he added that the assisted suicide doctors gave him a new prognosis of 6 months to live so that his assisted suicide would be "legal."
C.A.H. redoubled her efforts to stay in contact with Mr. A and encourage him. Some old friends from Alcoholics Anonymous and others now began to visit him daily. His few friends were clear that they valued him and did not want him to kill himself. Hospice, however, remained "neutral" on the issue, and he grew suspicious of that organization. As he put it, "I’m going to get rid of hospice. I don’t trust them. Then there’s the morphine. I’m not in any pain. I don’t know why they want me to take all this morphine." He dismissed hospice, decreased the pain medication, and both his mood and cognitive clarity improved for several months.
Three weeks before his death, however, his pain became a significant factor again. As Mr. A’s pain increased, he used more controlled-release oxycodone, which contributed to constipation, which in turn became excruciating in itself. Because of abdominal discomfort, he stopped drinking fluids as well as eating and grew confused and more suspicious again. He now wondered what might be in the liquids people gave him and in the pain medications, so he didn’t take them. At a home visit that both authors made together, his medication tray showed that he was taking minimal, if any, pain medication. He said that he was desperate because of the pain and was on the verge of killing himself with the overdose and that Dr. B had recently offered to sit with him while he took it.
He was reassured that he felt so frightened because of his confusion, but that pain medication and fluids, along with 24-hour care, would help him. Fluids might also help alleviate his constipation, which had become so painful. The controlled-release oxycodone tablets from his bedside stand were handed to him, and he took them as prescribed for the first time in days. C.A.H. then insisted that his doctor should prescribe a morphine pump to be delivered the next day so that his confusion and paranoia would not interfere with his receiving the needed pain relief. She also arranged for a 24-hour attendant care, which he could readily afford. With these interventions, his confusion cleared, his pain abated, and he felt much relieved during the remaining weeks of his life, even while his physical condition deteriorated.
During this time, which he had been on the verge of cutting short, he was able to express his gratitude to and say goodbye to the many people who had helped him. Most important, he was able to reconcile with his daughter, from whom he had been alienated since his psychiatric hospitalization. This opportunity was meaningful to her as well as to him.
Mr. A had been hospitalized after he developed depression and was thought to have both suicidal and homicidal ideation. Before discharge from the hospital, the attending psychiatrist noted in his medical record, "The guns are now out of the house, which resolves the major safety issue." Although removal of his guns may have resolved at least one safety issue, it did nothing to address another: a lethal prescription intended for the purpose of suicide remained in the home of this depressed patient. This inconsistency likely did not arise from any oversight on the psychiatrist’s part but from the competing paradigms informing the decisions.
Removing lethal means is central to the clinical treatment of suicidal symptoms (2, 20), but providing lethal means is central to the assisted suicide model.
The volunteer for Physicians for Compassionate Care and members of the Compassion in Dying Federation took philosophically different approaches. C.A.H. considered the patient’s current depression central to his motivation to kill himself. She recommended psychiatric evaluation and treatment, and when he refused referral to such treatment, she kept in contact with him, talking with him and helping arrange for his care. She called his primary care doctor and encouraged him through communication with his nurse to provide antidepressant medication. She interacted with those who could visit with him. When he became desperate because of apparent delirium and poorly treated pain in the last few weeks of his life, she made certain that he received the pain care he needed and arranged for intensified palliative care, including 24-hour attendant care. She consistently reminded him that she did not want him to kill himself.
In contrast, according to the patient, the doctor with the Compassion in Dying Federation did not seem "very interested" in his psychiatric history and previous suicide attempts. According to the daughter, the doctor who prescribed the assisted suicide drugs, Dr. B, told the patient and his daughter that a psychiatric evaluation would not be "necessary" but later told N.G.H. that such an evaluation would have been obtained had the prior suicide attempts been made known. It is clear that Dr. B agreed with the guidelines that state it is only the competency to make decisions that is required legally. While initially not wanting to comment on this case when invited to do so, Dr. B eventually did consent to discuss the case. Dr. B cited being struck by "how similar our goals were," meaning the patient’s comfort, well-being, and longevity. Dr. B thought that giving Mr. A the assisted suicide drugs may have added to his sense of control and security and may have even prolonged his life. This well-meaning view is somewhat at odds with the traditional psychiatric practice of depriving suicidal individuals of lethal means insofar as possible (20), since doing so makes carrying out suicide more difficult, gives patients time to recover hope, and gives a clear message that those caring for them value their lives. Members of the Oregon assisted suicide movement, however, have frequently commented, undoubtedly in agreement with Dr. B, that they believe they are showing their caring by valuing the patient’s autonomy, and that this act might make completion of suicide less likely in some individual cases. However, they will acknowledge that ready access to lethal means is associated with an increased incidence of death by suicide in the average case (20).
Other clinicians took different approaches. The emergency room doctor, along with the social workers and court investigators at Mr. A’s psychiatric admission, considered his depression and previous suicide attempts most seriously. Even these notes, however, skirted the issue of his possessing a lethal overdose and focused more directly on homicidal than suicidal ideation, although the danger of suicide was clear. The health care professionals seemed placed in a clinical bind when presented with a suicidal mentally ill patient who had been given an overdose by another doctor.
The inpatient psychiatrist seemed to mix both the traditional clinical approach and the assisted suicide competency model with predictably mixed results. The psychiatrist performed an evaluation, diagnosed depression and intermittent delirium, and attempted to treat the depression. One must wonder if the circumstances of the patient having a life-threatening illness and intending to take assisted suicide drugs might have interfered with recognition of the importance of thoroughness in psychiatric diagnosis. Of particular concern was the failure to account for paranoid symptoms. Did the patient have major depression with psychotic symptoms, paranoid schizophrenia with alcoholism and a history of suicide attempts, or a preexisting delusional disorder or paranoid personality disorder with an adjustment disorder with depression developing later? Or would some other diagnostic formulation have been more appropriate? Such a determination, of course, cannot be made at this point.
Nevertheless, the psychiatrist protected the patient from danger to himself or others through inpatient treatment for over a week and recommended antidepressant medication and social support. The psychiatrist took care to make sure guns were removed from the home before discharging the patient. All of this is consistent with the traditional clinical model. In deference to the assisted suicide competency model, however, it was noted that the patient had been given a lethal prescription and that prescription remained at his home. Declaring Mr. A incompetent the day after discharge and apparently doing nothing to retrieve the lethal overdose seems incompatible with both approaches—and is perplexing. The psychiatrist could not be reached by telephone to discuss this case.
Although the Oregon assisted suicide law does not exclude mentally ill patients who are technically competent from receiving overdoses, it clearly states that patients who are not competent to make legal decisions should not be given assisted suicide drugs. What should be done when a patient is considered competent at one point by an assisted suicide doctor and, at another time, is declared incompetent to make legal and medical decisions by two other doctors and a court of law, however, has not been clarified. The assisted suicide doctor, in a case like this, is protected by the Oregon law because it exempts the assisted suicide doctor, acting in goodwill, from legal action—a protection not shared by the other doctors involved in such a case. In particular, a psychiatrist who considers a patient incapable of making medical decisions yet does nothing to retrieve previously prescribed overdose drugs along with household guns may be placed in legal jeopardy, although he or she may have no legal right to remove those drugs.
This case illustrates how difficult it is to combine the two paradigms of response to suicidal ideation. Attempts to mix the traditional clinical model and the assisted suicide competency model in this case resulted in unusual clinical interventions and inconsistent approaches.
Although all physicians would agree that they must respect patients’ rights to hold and express freely their wishes and decisions, there remains considerable controversy over whether physicians should be given the right to respond to those concerns by providing lethal overdoses to patients or should they be held to the traditional standard of care in responding to suicidal patients. The difficulty in determining competency during a phase of life when mental capacity can fluctuate dramatically and of discerning when depression or other mental illness may be affecting decisions about assisted suicide presents psychiatrists wishing to accept the assisted suicide competency model with daunting dilemmas that have not yet been addressed adequately.
We conclude that the attempt to mix the assisted suicide competency model and the traditional clinical model of responding to seriously ill suicidal patients may prove confusing to both clinicians and patients, may contribute to substandard medical care, and may endanger seriously ill patients, particularly those with a history of preexisting mental illness (21).
Presented in part at the 157th annual meeting of the American Psychiatric Association, New York, May 1–6, 2004. Address correspondence and reprint requests to Dr. Hamilton, 2250 N.W. Flanders, Suite 306, Portland, OR 97210; firstname.lastname@example.org (e-mail).