Voluntary dehydration as end-of-life option?

Study reports on caregiver experience

Studies of terminally ill patients indicate that a small number of them want the option of physician-assisted suicide (PAS) or other means of controlling the manner in which they die. Yet with PAS legal in only one state, most will not have that option.

Voluntary refusal of food and hydration has been proposed as an alternative to PAS that would allow patients to hasten their deaths if they want, without violating the law. But many clinicians and caregivers fear that such decisions will result in prolonged, severe suffering for patients choosing to do so.

However, a recent study1 performed by researchers in Oregon indicates that a significant number of people do choose to do this and do not suffer a prolonged, difficult death as a result.

The researchers mailed a questionnaire to all nurses employed by hospice programs in Oregon, asking whether they had ever cared for a patient choosing to refuse food and hydration and, if so, to report information about the most recent patient to do so.

They received 307 responses from the 429 questionnaires mailed and 102 of the respondents (33%) reported that in the previous four years they had cared for a patient who deliberately hastened death by voluntary refusal of food and fluids (VRFF).

The nurses reported a number of reasons given by patients for their choice: because they were ready to die, saw continued existence as pointless, and considered their quality of life poor. Eighty-five percent of the patients died within 15 days after stopping food and fluids. On a scale from zero (a very bad death) to nine (a very good death) the median score for the quality of these deaths, according to the nurses’ reports was an eight.

Compared to patients who chose PAS (which is legal in Oregon) during the same time period, the patients choosing to refuse food and fluids were typically older, less likely to want to control the circumstances of their death, and less likely to be evaluated by a mental health professional.

"Some patients do make this choice and health care providers need to be able to advise them about the consequences of the choices and continue to care for them," says Linda Ganzini, MD, professor of psychiatry at Oregon Health and Science University School of Medicine, director of the Palliative Care Fellowship at the Portland Vet- erans Affairs Medical Center, and the lead author of the study report. "Our study demonstrated, somewhat counterintuitively, that patients did not suffer intolerable discomfort with this choice, had good deaths, and died within two weeks."

Ethical issues

During the debate over the passage of Oregon’s Death with Dignity Act, which legalized PAS in that state, many ethicists drew a bright line between withholding and withdrawing treatment from terminally ill patients — which is universally considered acceptable — and physician-assisted suicide, which many ethicists consider unacceptable, Ganzini says.

Voluntary refusal of food and fluids occupies sort of a middle position between withdrawal or withholding of treatment at the patient’s request and PAS, she continues. VRFF involves a deliberate decision by the patient to hasten death and still requires physicians and other caregivers to continue to provide care to the patient.

"VRFF shares some characteristics with each of these other types of choices and forces us to re-examine our thinking about the bright line," she says.

Both the recent study and other studies of PAS highlight the continued need for clinicians to expand their knowledge of how to provide good care at the end of life, Ganzini adds.

"Palliative and hospice care has focused on control of symptoms, spiritual concerns and the family’s needs," she explains. "But our data suggest that there is a group of patients for whom the most important goal is remaining in control and not being dependent on other people. We need to do a better job of recognizing these people and helping them plan for the end of life."

Although their research found a number of reports of patients choosing VRFF, not much is known nationwide about how many patients choose to do this.

In Oregon, the state, health systems, and public health officials have conducted several different educational initiatives focusing on patient needs and rights at the end of life. So terminally ill patients there may be more proactive in deciding they want to hasten their death, and they may be more inclined to be open about such decisions, Ganzini says.

Because there are no other studies of the experiences of patients choosing VRFF, the data obtained from the Oregon study cannot reliably be generalized nationwide.

"Patients can choose VRFF without ever discussing it with their family or care providers," she says. "In addition, loss of appetite is common enough in patients with terminal illness that patients can hide their motives."

It may even be difficult to determine which patients are deliberately stopping food and hydration with the goal of hastening death, and which are simply not eating or drinking because of loss of appetite or difficulty swallowing, both of which are common at the end of life, says J. Andrew Billings, MD, head of the Palliative Care Service at Massachusetts General Hospital in Boston.

"Despite how this survey was described, I suspect that there may be some overlap in the study between simply choosing to stop eating and drinking, which is quite rare in my practice, vs. stopping because of loss of appetite, thirst, poor energy, difficulty or pain with swallowing, mouth sores, etc.," he explains. "At the same time, these patients may choose not to start or to discontinue nutritional or fluid support."

However, the important information provided by the Oregon study is that the caregivers reported that patients choosing VRFF did not seem to suffer, he says.

"Regardless of the reason, the terminally ill patients who do stop eating or drinking or reduce their intake below levels consistent with survival — while typically in rather debilitated states — and who do not get fluid replacement, seem to die peacefully and comfortably, as reported by this study and others," Billings says.

For patients who are not eating or drinking, it’s important that clinicians provide an appropriate medical evaluation and instruction to family and other professional caregivers about how to provide supportive care, he notes.

"This involves an explanation of what can be expected, teasing out concerns the family may have, and interpreting bodily changes and providing reassurance that they are doing what the patient truly wanted," Billings says. "The patient can benefit from good mouth care to prevent or minimize thirst and, if necessary, analgesics and sedation for comfort."

Physician support essential

VRFF commonly has been discussed in palliative care circles, although it is believed to be very uncommon, Billings says. Symptoms associated with dehydration or poor nutritional intake are an extremely common topic in the clinical literature and at professional meetings.

"Most palliative care texts have a chapter on this topic and it was first described in the literature as an alternative to physician-assisted suicide or euthanasia, and thus as a way for patients to have their way without involving physicians," Billings explains. "But patient abandonment seems a risk. I think palliative care clinicians should be involved to evaluate the patient and family, assure patient comfort and guideline the family through the process. This sort of procedure should be done with medical supervision in case there are disagreeable symptoms and also to reassure the family, who may misinterpret pre-terminal changes as distress or just be frightened by the dying process. Likewise, in a nursing home or other institutional setting, the staff typically need supervision and support when a patient is dying."

That being said, it’s also essential to determine whether this is truly what the patient wants, he and Ganzini add.

When a patient expresses a desire to hasten death, it’s important that clinicians examine these requests closely and try to determine the cause.

"Most [dying] patients hint at wanting to speed up the process and most frank requests for it, once examined more closely, do not turn out to be persistent requests for assistance in dying more quickly," he says. "The meaning of the request needs to be explored rather than just taken at face value. Common reasons, including poorly treated pain or disagreeable symptoms, depression, delirium, a feeling that they are a burden to others, a loss of dignity, overall frustration, a feeling of meaninglessness, etc.; practically all of these problems can be addressed and satisfactorily alleviated. We noted only two patients in the first 500 or so seen in our hospice program who had a persistent desire to hasten death after receiving an appropriate evaluation and treatment, which included psychosocial and spiritual support."

Ganzini agrees that it is important for patients who express a desire to hasten death — by whatever means — to receive appropriate evaluations of their physical and mental health. The Oregon survey respondents indicated that patients choosing VRFF were much less likely to have been evaluated by a mental health professional. Patients choosing PAS in Oregon must undergo such an evaluation under state law.

Although she believes that some patients do indeed choose VRFF for acceptable reasons, it is important that they be screened for depression.

"All of the patients covered by our study were in hospice and should have access to a hospice social worker who would have skills necessary to assess for depression," she notes. "Whether a depressed person in the final weeks of life should be prevented from deciding to stop food and fluids is more difficult. Some depression treatments take many weeks to be effective."

Research like the Oregon study is helpful because it helps clinicians learn more about how to help patients at the end of life, says Billings. More research is needed to better understand why patients make such choices and how to help them reach the end of their lives without suffering.

"I think we can learn a great deal about the dying experience, partly by attending to the small subset of patients who wish to hasten their deaths, regardless of the methods they consider or actually employ," he says. "Why do so many patients cling to life when they are terribly debilitated and often quite uncomfortable, while others wish to die? What is the nature of the suffering of the latter group, and what allows other to choose to live as long as possible, sometimes in dreadful circumstances, and what should we do to help both groups?"


1. Ganzini LG, Goy ER, Miller LL, et al. Nurses’ experiences with hospice patients who refuse food and fluids to hasten death. N Engl J Med 2003;349:359-365.

Additional Reading

  • Jacobs S. Death by voluntary dehydration — What the caregivers say. N Engl J Med 2003; 349:325-326.
  • Block S, Billings JA. Patient requests to hasten death: Evaluation and management in terminal care. Arch Intern Med 1994; 154:2,039-2,047.


  • J. Andrew Billings, MD, Head, Palliative Care Service, Massachusetts General Hospital, FND 600, 55 Fruit St., Boston, MA 02114.
  • Linda Ganzini, MD, Professor of Psychiatry, Oregon Health and Science University School of Medicine; Director, Palliative Care Fellowship at the Portland Veterans Affairs Medical Center, 3710 S.W. U.S. Veterans Hospital Road, Portland, OR 97201.