Beyond Futility: Good Shared End-of-Life Decision-Making Takes Ongoing Discussions, Realistic Goals

July 1, 2008

Decision-making at the end of life is difficult, and it should be. Rather than face these time consuming and emotionally demanding discussions, doctors too often look to unsuitable conceptual models.

Decision-making at the end of life is difficult, and it should be. Rather than face these time consuming and emotionally demanding discussions, doctors too often look to unsuitable conceptual models. Futility is such a notion. It is an elusive concept, wielded as a “trump card” by doctors looking for an easier solution or administrators hoping to contain health-care costs. Khatcheressian and colleagues recognize the inherent difficulty in defining what is futile, and, properly, look to improving doctor-patient communication for the answer to their dilemma, “What to do when your patient insists on chemotherapy that likely won’t help.”

Though deciding about a particular treatment is ultimately up to the patient, physicians have a duty to ensure that patients are well informed and not unnecessarily harmed. Abiding by patients' values does not mean that doctors should offer treatments or passively accept decisions not supported by medical evidence or where risks exceed the benefits. Good shared decision-making requires the flow of information to go both ways.

Goals of Treatment

Deciding how aggressively to treat cancer is an ethical dilemma that first requires thoroughly understanding the indications, supporting evidence, and possible outcomes of the medical intervention. It is important to be clear about the goals of treatment. In addition to the traditional goals of tumor response, increased survival, and symptom control, patients with advanced cancer have goals related to quality of life. These may include physical and intellectual independence, spending quality time with their families, taking trips, staying out of the hospital, or even economic stability. Doctors truly fulfill their duty when they help patients establish obtainable goals and guide them toward the treatment option most compatible with the evidence and those goals. In addition to anticancer chemotherapy, these options can include palliative care and hospice.

Cancer patients may overestimate the value of treatments, often because they also overestimate their prognosis. The stage is set for conflict when patients and families develop unrealistic expectations because doctors have not adequately discussed whether the goals of treatment are cure or palliation, including life extension or symptom control. Of course, it is also important to communicate the likelihood of achieving these goals and the probable duration of positive effects of treatment, as well as the uncertainty of cancer care and the need for periodically reassessing the treatment plan. The discussion of treatment options should include all of the alternatives, including palliative care and hospice, and should always include assessment of patient understanding.

How Treatment Options Are Presented

We need to be conscious of how we frame these options, and whether we are adequately communicating the progression of disease. Patients will respond differently depending on how chemotherapy or other options like hospice are presented. Throughout the course of care, options should be presented in terms of patients’ values and goals, and all alternatives should be presented clearly. Hospice and other comfort care approaches should be described as treatment. If these alternatives are not presented, patients and their families may see chemotherapy as the only option for palliation.

Rather than responding directly to patients’ requests for chemotherapy that is unlikely to be beneficial, it may be more useful to ask questions to clarify their goals, values, and reasons underlying the request. Considering a differential diagnosis for the request may help to resolve the situation, or at least help staff understand the reason for conflict or differing perspectives.[1] Patients may also request chemotherapy because of fear of abandonment by their physician, lack of understanding, denial, spiritual distress, guilt, or family issues, including family pressure to continue chemotherapy. Multidisciplinary teams that include representatives from the patient’s faith community, can be helpful in addressing family conflict and grief to help resolve decision-making and help families’ bereavement.[2] In many areas, some hospices and insurance plans now allow chemotherapy or other anticancer treatments as part of hospice care. This open-access approach provides palliative care services earlier in the course of the patient’s disease, and can help patients and their families address underlying end-of-life issues.[3]

Defined Endpoints and Timelines

The ability of an intervention to accomplish its goal defines the limits of oncology care. The efficacy of anticancer treatment at the end of life is often limited and usually based on weak evidence. Patients with limited prognoses may be more vulnerable to treatment-related morbidity and mortality. Therefore, it may be preferable to present anticancer treatment options in this context as experimental treatments that are best given as part of a formal research study. This is, however, not always practicable. “N-of-1” trials and time-limited trials are alternative approaches with a clearly defined goal (such as reduction in tumor size or symptom relief) and plans for stopping treatment if that goal is not met, or if significant side effects or burden alter the harm-benefit ratio.

When the goal of treatment is palliative, chemotherapy should never be administered without defined endpoints and timelines. Oncologists must work with patients to balance any potential benefits against burdens, side effects, and risks. These include physical effects and direct costs, but also nonmedical issues such as caregiver time, stress, life disruption, and lost work. When there is evidence for potential benefit but significant potential for harm or burden, discussing patient preferences can help to guide appropriate shared decisions.

Preventive Ethics

End-of-life decision-making doesn’t work well if the possibilities of treatment not working and hospice care are not presented until the end. From the beginning of care, patients and families usually think about the possibility of things not going well, generally want to know prognostic information,[4] and can be reassured to know that there are good comfort care options. Established early in cancer treatment, expectations and alternatives can be built into “next step” scenarios for patients and families.

This preventive ethics approach is well suited to the circumstances of most cancer patients. Cancer patients and their doctors usually know each other quite well. Oncologists have the unique opportunity to understand their patients’ goals and values, and to prospectively develop action plans around obtainable goals. This type of communication opens the door to effective, therapeutic shared decision-making that would be prematurely closed by discussions about “futility.”

Financial Disclosure:Dr. Dy is on the management team of Community Hospices, which has facilities in Washington, DC, Maryland, and Virginia.



1. Goold SD, Williams B, Arnold RM: Conflicts regarding decisions to limit treatment: A differential diagnosis. JAMA 283:909-914, 2000.
2. Lautrette A, Darmon M, Megarbane B, et al: A communication strategy and brochure for relatives of patients dying in the ICU. N Engl J Med 356:469-478, 2007.
3. Jennings B, Rundes T, D’Onofrio C, et al: Access to Hospice Care: Expanding boundaries, overcoming barriers. Hastings Center Rep 33(2):S3-S7, 2003.
4. Mack JW, Wolfe J, Grier HE, et al: Communication about prognosis between parents and physicians of children with cancer: Parent preferences and the impact of prognostic information. J Clin Oncol 24:5265-5270, 2006.