BACKGROUND/RATIONALE:
There is considerable debate about whether end-of-life care is delivered in accordance with patients’ and families’ preferences. At the center of this debate is a lack of data about these preferences. Also lacking is data regarding two key prerequisites necessary for an informed preference: namely, discussions about prognosis and about treatment alternatives.

OBJECTIVE(S):
This study of terminally ill older patients, their families, and their physicians examined: 1) the frequency with which information about prognosis and choices about treatment are given to terminally ill patients and their families; and 2) patients’ and family members’ treatment preferences, examined in terms of treatment outcomes, rather than treatment process.

METHODS:
A cohort of 226 patients with potentially terminal diagnoses of cancer, congestive heart failure, and chronic obstructive pulmonary disease (82% participation rate) and their family members received in-home interviews; their clinicians completed a mail survey. These interviews elicited patients’ and caregivers’: 1) knowledge about disease prognosis and treatment alternatives; and 2) preferences about care, based on a newly developed instrument to assess their goals for care, rather than specific treatment interventions, the WALT (Willingness to Accept Life-Sustaining Treatment). Clinician interviews elicited their communication with patients and caregivers regarding prognosis. Briefly, the WALT incorporates different levels of treatment burden, different potential treatment outcomes, and different likelihoods of these outcomes. Participants are asked if they would want to receive a given treatment, first when the outcome is known with certainty, and then with different likelihoods of a given adverse outcome. Participants are told that the outcome with no treatment was dying of the underlying disease. Interviews with the patient's primary physician determines physicians' communication with patients and families regarding prognosis and treatment alternatives.

FINDINGS/RESULTS:
The WALT measures preferences with a high degree of reliability and validity. Treatment burden, treatment outcome, and the likelihood of different outcomes all influence treatment preferences. In the case of low-burden treatment with return to current health, 98.7% of the cohort chose to receive treatment. Of these 223, when the treatment was changed to high-burden, 11.2% no longer chose treatment. When the outcome was changed to severe functional or cognitive impairment, 74.4% and 88.8% no longer chose treatment, respectively. Fewer participants desired treatment with an increasing likelihood of functional impairment versus return to current health than with an increasing likelihood of death versus return to current health. Preferences did not differ according to primary diagnosis. Regarding communication, in 46% of patient/clinician and 34% of patient/caregiver pairs, the clinician reported saying that the patient could die of the underlying disease; the patient or caregiver reported no discussion. In 23% of patient/clinician and 30% of patient/caregiver pairs, the clinician reported giving an approximate life expectancy; the patient or caregiver reported no discussion. Of 205 patients who reported no life expectancy discussion, 40% did not want this discussion, independent of diagnosis. Although clinicians of cancer patients were more likely to report communication, the proportion of patients agreeing that communication had occurred did not differ by diagnosis.

IMPACT:
These results have important implications for advance care planning. Advance care planning that fully captures patient preferences should incorporate patients’ attitudes toward treatment burden, outcomes, and the likelihood of the outcomes. The likelihood of adverse functional and cognitive outcomes of treatment requires particular and explicit consideration. Further, although clinicians believe that they are discussing prognosis, patients and caregivers frequently do not corroborate these reports. In addition, many patients do not want prognostic information. Despite previous reports concluding that patients want full disclosure about their illness, many seriously ill older persons and caregivers may not be ready or able to receive prognostic information.

---

External Links for this Project

---

PUBLICATIONS:

Journal Articles

1. Fried TR, Bradley EH, Towle VR. Assessment of patient preferences: integrating treatments and outcomes. The journals of gerontology. Series B, Psychological sciences and social sciences. 2002 Nov 1; 57(6):S348-54. [view]
2. Fried TR, Bradley EH, O'Leary J. Changes in prognostic awareness among seriously ill older persons and their caregivers. Journal of palliative medicine. 2006 Feb 1; 9(1):61-9. [view]
3. Fried TR, Bradley EH, O'Leary J. Prognosis communication in serious illness: perceptions of older patients, caregivers, and clinicians. Journal of the American Geriatrics Society. 2003 Oct 1; 51(10):1398-403. [view]
4. Fried TR, Bradley EH, Towle VR, Allore H. Understanding the treatment preferences of seriously ill patients. The New England journal of medicine. 2002 Apr 4; 346(14):1061-6. [view]
5. Fried TR, Bradley EH, Towle VR. Valuing the outcomes of treatment: do patients and their caregivers agree? Archives of internal medicine. 2003 Sep 22; 163(17):2073-8. [view]
6. Fried TR, Bradley EH. What matters to seriously ill older persons making end-of-life treatment decisions?: A qualitative study. Journal of palliative medicine. 2003 Apr 1; 6(2):237-44. [view]

DRA: Aging, Older Veterans' Health and Care, Health Systems
DRE: none
Keywords: Caregivers – not professionals, End-of-life, Patient preferences
MeSH Terms: End-Of-Life, Caregivers, Patient Satisfaction