Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
N Engl J Med. Author manuscript; available in PMC 2013 September 3.
Published in final edited form as:
PMCID: PMC3760676

Discussing Overall Prognosis with the Very Elderly

Alexander K. Smith, M.D., M.S., M.P.H., Brie A. Williams, M.D., M.S., and Bernard Lo, M.D.

It’s late in the day in the office of a busy primary care physician, who is relieved to see that his last patient, although 86-years old, is a woman with multiple stable medical problems visiting for her annual exam. The patient is joined at the appointment by her daughter, who helps her mother with several Activities of Daily Living including bathing, dressing, and balancing her checkbook. During the visit, the daughter asks about preventive health measures for her mother: “It’s been years since her last colonoscopy, hasn’t it?” The physician explains that national guidelines recommend against cancer screening for patients over age 85.1 The patient responds, “What do you mean, doctor? From what I see on TV, cancer is a disease of the elderly.” The doctor now faces an important decision. He could skirt the question of prognosis and explain the issue primarily in terms of the risks and benefits associated with cancer screening in the very elderly. Or he could talk with the patient about her overall prognosis.

Paradoxically, despite knowing that life expectancy inexorably decreases with advancing age, we tend to avoid discussing overall prognosis with elderly patients, particularly those who do not have a dominant terminal illness. By avoiding such discussions, however, we may undercut the ability of patients and their caregivers to make informed choices for their future.2 To improve the quality of decision making for the very old, we believe we should radically alter the paradigm of clinician–patient communication: offering to discuss overall prognosis with our very elderly patients should be the norm, not the exception.

We would suggest that clinicians routinely offer to discuss the overall prognosis for elderly patients with a life expectancy of less than 10 years, or at least by the time a patient reaches 85 years of age. By 85 years, the average remaining life expectancy in the United States is 6 years; 85-year-old Americans have a 75% chance of living 3 more years and a 25% chance of living 10 more years, with variation due primarily to variable functional status and the presence or absence of various medical conditions.3 Since the harms of many health interventions are immediate, whereas the benefits of preventive interventions may accrue slowly over time, clinical priorities should and do vary with life expectancy.3 For patients with a life expectancy of more than 10 years, cancer screening, intensive blood-pressure management, and tight control of glycated hemoglobin levels will have high priority, whereas for patients with a shorter life expectancy, priority might be given to reducing the pill burden and engaging in advance care planning. Most very elderly patients place great emphasis on the harms as well as the benefits of medications.4 Avoiding burdensome and potentially risky interventions of limited benefit may improve a patient’s functional abilities and quality of life.3 And for primary care doctors who may be overwhelmed by multiple health issues in the elderly, considering the bigger picture may simplify decision making.

Clinicians may be reluctant to discuss with patients prognostic estimates based on population-level epidemiologic data that lack precision and accuracy. However, clinicians can move beyond population-based averages by using prognostic indices and taking into account individualized clinical factors that are not incorporated in prognostic models, such as poor functional recovery from a debilitating fall. In decisions regarding renal dialysis or treatment for cancer or stroke in the elderly, it is recommended that physicians offer to discuss probabilistic estimates of prognosis. Why not similarly offer to discuss overall prognosis in the very old? If we avoid talking about the possibility of death or functional decline until prognostic tools are greatly improved, we may never do so, even though we know that all elderly people will eventually die.

Clinicians may fear talking about or even raising the topic of overall prognosis because it may seem threatening to many patients or family members. Imagine, for example, that the daughter in this case declares, “We don’t want to talk about doom and gloom. We want you to focus on the positive.” Yet, this very elderly patient, and the very old in general, may be less threatened by these discussions than clinicians or family caregivers realize. One study of frail elderly patients with a mean age of 73 years, a life-limiting illness, and a need for assistance in at least one instrumental activity of daily living found that 55% of the 205 patients whose physician had never discussed prognosis in fact wanted to discuss it, while 40% did not want to discuss it and 5% did not know.2 In a study we conducted, 65% of 60 disabled black, Chinese-American, Hispanic, and white elderly people living in the community (mean age 78 years) said they would want their doctor to tell them if they had less than 5 years left to live.5 Thus a majority of elderly patients might want to discuss prognosis, while a substantial minority might not. Clinicians should therefore offer to discuss overall prognosis with very elderly patients but respect those who decline to have such a discussion. The physician in this case could respond to the daughter’s statement by aligning with her, saying, “Yes, I also want to focus on the positive, for example, keeping your mother as active and independent as possible. At the same time, I want to make sure your mother is not worrying about these issues alone.” The physician can them turn to the patient and elicit her concerns and preference regarding a discussion of overall prognosis. If the patient declines, rather than retreating, respectfully exploring her reasons may help build trust. The clinician could say, “So I can help with other important decisions in the future, can you help me understand your reasons for not wanting to discuss this information?”6

Whereas clinicians consider overall prognosis to inform medical decision making, in our experience, many very elderly patients are interested in their overall prognosis because it affects their personal life choices — motivating them, for instance, to arrange finances for long-term care or prioritize spending time with grandchildren and other family members while they are still active. These personal choices require information about prognosis that clinicians can provide.

Opportunities to discuss overall prognosis often flow quite naturally from topics commonly addressed in primary care (see table). In the case described above, the physician could use the decision about screening for colon cancer as an occasion for offering to discuss overall prognosis. If the patient were receptive to such a discussion, the physician could explain how the information would help her make important medical decisions and life choices and explore her concerns about aging. The physician could explain that colonoscopy has important immediate risks, such as perforation of the colon, whereas the benefits might not occur for years. To explain the patient’s overall prognosis the physician might say, “People in their mid-80s with health similar to yours live about 6 more years, on average. Some people live more than 6 years, others less.” In this way, the physician would acknowledge uncertainty. He could then assess how the patient received and processed this information.

Table thumbnail
Common Medical Decisions and Life Choices That Offer Opportunities to Discuss Overall Prognosis with Very Elderly Patients.

If it is difficult to discuss prognosis with patients with a serious progressive illness such as cancer or congestive heart failure, it is likely to be even more difficult to do so with very elderly patients who do not have a clearly terminal condition. Yet, notwithstanding large current gaps in evidence, we believe we should start talking about overall prognosis now, even as we carry out more research on patient preferences and ways of improving such discussions. To make care more patient-centered, we need to start helping our very elderly patients set goals of care that take their overall prognosis into account. We should do so in the ordinary course of clinical practice, letting our patients be our guides.


Disclosure forms provided by the authors are available with the full text of this article at


1. Screening for colorectal cancer: U.S. Preventive Services Task Force recommendation statement. Ann Intern Med. 2008;149:627–37. [PubMed]
2. Fried TR, Bradley EH, O’Leary J. Prognosis communication in serious illness: perceptions of older patients, caregivers, and clinicians. J Am Geriatr Soc. 2003;51:1398–403. [PubMed]
3. Walter LC, Covinsky KE. Cancer screening in elderly patients: a framework for individualized decision making. JAMA. 2001;285:2750–6. [PubMed]
4. Fried TR, Tinetti ME, Towle V, O’Leary JR, Iannone L. Effects of benefits and harms on older persons’ willingness to take medication for primary cardiovascular prevention. Arch Intern Med. 2011;171:923–8. [PMC free article] [PubMed]
5. Ahalt C, Walter LC, Yourman L, Eng C, Perez-Stable EJ, Smith AK. “Knowing is better”: Preferences of Diverse Older Adults for Discussing Prognosis. J Gen Intern Med. 2011 accepted. [PMC free article] [PubMed]
6. Back AL, Arnold RM. Discussing prognosis: “how much do you want to know?” talking to patients who do not want information or who are ambivalent. J Clin Oncol. 2006;24:4214–7. [PubMed]