Surrogates may be struggling to let go of a loved one, may not understand clinical realities, or may have religious or cultural beliefs about withdrawal or withholding of life-sustaining medical interventions. Regardless of the reason, families sometimes want to “do everything” when the clinical team feels it’s time to stop.

“This scenario is not infrequent, and may increase with the ever-expanding interventional options to keep people alive at the end of life,” says James Kirkpatrick, MD, adjunct assistant professor in the department of bioethics and humanities at University of Washington Medical Center in Seattle.

Advance directives don’t always prevent such conflicts. People can always disagree as to how they should be interpreted. In some cases, surrogates even directly contradict patients’ stated wishes because they’re hoping for a miracle.

“Many clinicians, particularly those from a secular medical world view that differs markedly from the belief systems of many patients and families, find these scenarios particularly vexing,” says Kirkpatrick.

The clinical team likely lacks training in addressing such conflicts. Many struggle to identify the underlying ethical principles. “Many ethics consultants, on the other hand, have had some form of mediation training that ideally positions them to address these conflicts,” says Kirkpatrick.

The earlier the intervention occurs, the better. “Involvement of ethics consultants at the initial points of conflict is preferable to late involvement,” says Kirkpatrick.

Providers seek to provide the care patients and families want, with an emphasis on shared decision-making over paternalism. “In rare cases, though, we just can’t say ‘yes’ to the request of a surrogate decision-maker,” says Tim Lahey, MD MMSc, chair of the clinical ethics committee at Dartmouth-Hitchcock Medical Center in Lebanon, NH.

If the conflict persists even after extensive efforts at resolution, the clinical team tells the surrogate, “Our institution can no longer provide the requested type of care. But we are willing to assist with transfer of a patient to another institution that will, as long as one can be found with reasonable effort.”

A therapy may be reasonable and requested by a surrogate with good intentions, yet clinicians feel it’s causing excessive suffering. “The process of resolution is important here,” says Lahey. “Often, disagreements can be resolved through information-sharing and consensus-building.”

The team recently cared for a patient with sepsis and multiple organ dysfunction. The unanimous impression from all treating clinicians at the hospital was that death was inevitable. “Yet, his loving partner wanted us to continue intensive care, in part because prior clinicians had, in his impression, ‘given up too early,’” says Lahey. When collaborative discussions didn’t result in agreement in next steps in care, clinicians offered to help the partner find another hospital that would provide the level of care he felt was consistent with the patient’s wishes. He could not. “This helped him come to grips with inevitability in a fashion that also allowed our providers not to feel they were mandated to provide care they felt would lead to undue suffering,” says Lahey.

Right to Decline

Some states, including California and Texas, have laws granting hospitals and healthcare providers right to decline treatments that they regard as medically ineffective or medically inappropriate.

Early communication about prognosis can prevent some of these requests in the first place. “One approach is to identify clinical criteria that are indicative of a poor outcome and communicate it upfront to the family,” says David Magnus, PhD, Thomas A. Raffin professor of medicine and biomedical ethics, professor of pediatrics, and director of the Stanford (CA) Center for Biomedical Ethics.

A patient undergoing a bone marrow transplant who ends up in multisystem organ failure is a common example. “In our institution, we have well over 100 of those cases, and none of those patients have survived. If they go to the ICU, it’s a one-way trip,” says Magnus.

Clinicians explain this upfront to families of patients undergoing this procedure. They state that care will be aggressive up to a certain point, but if multisystem organ failure does occur, the patient won’t go to the ICU because it won’t be successful. “We work that out with them in advance, so they know it’s coming,” says Magnus.

Commonly, patients go to the ICU having had no conversation whatsoever about their goals of care. “I’m constantly amazed at seeing patients show up in the ICU with metastatic cancer, and the oncologist has not talked to them about the good chance they would die,” says Magnus.

Many physicians’ mindset is to focus solely on the possibility the patient will get better. Since the possibility of the patient’s death is never addressed, conversations about patients’ wishes don’t occur.

“We tend to avoid that until the patient is much sicker and the family has misunderstood the possibilities,” says Magnus. The family doesn’t understand why the clinical team seems to suddenly be giving up, fostering mistrust. “There needs to be much more aggressive advance care planning, and it needs to happen much earlier in the process,” he says.

If a patient underwent a Whipple procedure and is undergoing second-line chemotherapy, “people can’t be that surprised when things are not going well — and yet, they are,” says Magnus.

Determining which patients would benefit most from palliative care interventions is critical. These families then recognize that death is a realistic possibility. “If people see this coming, it won’t make all conflicts go away completely. But it will certainly reduce the number of times where it comes out of the blue,” says Magnus.

Scripting sometimes is helpful for clinicians who struggle to explain the clinical situation to families. “What’s interesting is sometimes it turns out to be really easy,” says Magnus. “I would say a huge percentage work out that way.”

The clinical team perceives the family as insisting on prolonging the patient’s life, but this isn’t always as it seems. Some families assume the team keeps asking for their consent to continue because treatments are likely to work. When they learn otherwise, says Magnus, “even if the family really doesn’t want to stop, most will acquiesce and start moving forward with the grieving process. A subset of cases are resistant and very unhappy.”

In a recent case, a patient had clearly stated that he did not want life-sustaining treatment if a certain level of functioning was no longer achievable. “We were way past that point of what the patient would have found an acceptable quality of life. But the spouse just could not accept that,” says Magnus. The spouse reported seeing clinical improvement that none of the neurologists or surgeons could see. In situations like this one, says Magnus, “If we send a patient to the ICU who is not going to benefit and has virtually no chance of getting out of the ICU, just to benefit the family, that is not a good use of an ICU bed.” It means that a patient held in the ED waiting for an ICU bed, or for transfer to a higher-level institution, might die. “The conversation should have taken place on the floor, or even as an outpatient,” says Magnus. “Too often, we allow these things to flow downstream.”


  • • James N. Kirkpatrick, MD, Department of Bioethics & Humanities, University of Washington Medical Center, Seattle. Phone: (206) 598-4641. Fax: (206) 543-8584. Email:
  • • Tim Lahey, MD, MMSc, Chair, Clinical Ethics Committee/Dartmouth-Hitchcock Medical Center. Phone: (603) 650-6063. Fax: (603) 650-6110. Email:
  • • David Magnus, PhD, Director, Stanford Center for Biomedical Ethics, Stanford (CA) University. Phone: (650) 723-5760. Email: