Am Fam Physician. 2006 Jun 1;73(11):2062-2067.
A 65-year-old woman is receiving chemotherapy for recurrent breast cancer, which has metastasized to the bone, pleura, and liver. Her physician has delayed several treatment courses and modified the chemotherapy dosage because, despite supportive treatment, she has developed persistent anemia. Bone marrow biopsy and aspiration revealed hypercellularity with excess blast cells (20 percent). The patient has developed “preleukemia” (i.e., refractory anemia with excess blast cells), and her physician has told her that she has six months or less to live.
How did her physician come up with this six-month time frame? I have been told that a patient's first question after receiving a terminal illness diagnosis is commonly, “How much time do I have left?” How accurate can a physician be when predicting death? What factors should be considered when determining a prognosis?
Physicians often overestimate a terminal patient's life expectancy. One study1 showed that physicians overestimated survival by a factor of 5.3. A few clinical guideposts and a basic understanding of disease progression can improve a physician's ability to offer patients and their families more accurate time frames in which to organize their priorities.
Although medical decisions often are based on a patient's prognosis, an accurate diagnosis and definitive treatment have become more important in the modern medical setting. Patients usually receive little prognostic information, and advocacy groups traditionally focus only on survival. This is changing, however, as medicine becomes increasingly evidence-based. Researchers are evaluating treatment outcomes based on factors beyond simple survival (e.g., cost of treatment, quality of life, severity of symptoms, caregiver and societal burdens).2
Specific to the case scenario, the most important prognostic factors to consider when treating a woman with recurrent breast cancer (e.g., locally recurrent in the chest wall, regional lymph nodes or, most commonly, distant metastasis to the bone) are the disease-free interval (i.e., the time between completion of primary treatment and disease recurrence) and the extent and locations of metastases. These factors can help predict the clinical behavior and progression of the disease.
A patient with a long disease-free interval (i.e., one year or more) and limited metastasis will have a better prognosis than a patient who has been disease free for less than one year and has more widespread metastasis. Metastasis to the bone or pleura is not imminently life threatening; however, metastasis to the liver, lung, or brain considerably reduces the patient's life expectancy. A patient with metastasis to the liver and lung has a median life expectancy of less than six months. A patient with widespread metastasis or with metastasis to the lymph nodes has a life expectancy of less than six weeks. A patient with metastasis to the brain has a more variable life expectancy (one to 16 months) depending on the number and location of lesions and the specifics of treatment. Certain cancer complications (e.g., recurrent hypercalcemia and septicemia, malignant pericardial effusion, and abdominal carcinomatosis with ascites and partial bowel obstruction) are associated with less favorable prognoses. Many patients with comparable diseases will respond similarly; however, some will either do much worse or much better than expected. A patient's prognosis also is based on his or her response to treatment.
Other than cancer histopathology, metastasis, and tumor complications, the two most important factors to look at when determining a prognosis are: (1) how well the patient is able to perform activities of daily living (e.g., bathing, dressing, toileting), and (2) how burdensome the symptoms are to the patient.3 Physicians can use the Karnofsky performance scale (KPS)3 to help determine a patient's ability to perform activities of daily living. If a patient is sedentary for 50 percent or more of his or her waking hours and is becoming increasingly dependent on caregivers, or if he or she has a KPS score of less than 50 percent, the life expectancy is two to three months.3 Symptoms such as difficulty swallowing, dry mouth, shortness of breath, lack of appetite, and weight loss may portend a poorer prognosis4 and may accompany anorexia-cachexia syndrome (wasting). Increasing symptoms and diminished functionality cause progressive debilitation. Progressive debilitation (and its subsequent complications) is the underlying mechanism that causes death in patients who are terminal.
Other important prognostic considerations include comorbidity, age, and nutritional status. Psychosocial and spiritual factors also should be considered. Recent evidence5 has called into question the notion that patients with a strong “will to live” often live longer than those who have accepted that they are going to die; however, patients with inadequate social support, and patients who are impoverished or poorly adherent to prescribed regimens, may have shorter life expectancies. Unresolved relationship issues or existential distress may prolong the dying process.
In the final stages of cancer, patients typically deteriorate rapidly.6 Patients can live several months with little food intake, but they cannot live without adequate fluid intake. Therefore, a decrease in fluid intake and urinary output can help determine when a patient should receive palliative care.
Ultimately, the family physician's role in determining a prognosis depends on his or her communication with the patient's oncologist and how involved he or she is in the patient's care. A future Curbside Consultation piece will address issues related to delivering a poor prognosis to a patient.
1. Christakis NA, Lamont EB. Extent and determinants of error in doctors' prognoses in terminally ill patients: prospective cohort study. BMJ. 2000;320:469–72.
2. Reynolds T. Prognostic models abound, but how useful are they? Ann Intern Med. 2001;135:473–6.
3. Mor V, Laliberte L, Morris JN, Wiemann M. The Karnofsky performance status scale. An examination of its reliability and validity in a research setting. Cancer. 1984;53:2002–7.
4. Reuben DB, Mor V, Hiris J. Clinical symptoms and length of survival in patients with terminal cancer. Arch Intern Med. 1988;148:1586–91.
5. Young DC, Hade EM. Holidays, birthdays, and postponement of cancer death. JAMA. 2004;292:3012–6.
6. Lunney JR, Lynn J, Foley DJ, Lipson S, Guralnik JM. Patterns of functional decline at the end of life. JAMA. 2003;289:2387–92.
Case scenarios are written to express typical situations that family physicians may encounter; authors remain anonymous. The commentary in this issue was written by John W. Finn, M.D., Detroit, Michigan.
Please send scenarios to Caroline Wellbery, MD, at firstname.lastname@example.org. Materials are edited to retain confidentiality.
Copyright © 2006 by the American Academy of Family Physicians.
This content is owned by the AAFP. A person viewing it online may make one printout of the material and may use that printout only for his or her personal, non-commercial reference. This material may not otherwise be downloaded, copied, printed, stored, transmitted or reproduced in any medium, whether now known or later invented, except as authorized in writing by the AAFP. Contact email@example.com for copyright questions and/or permission requests.
Want to use this article elsewhere? Get Permissions