[Skip to Content]
Sign In
Individual Sign In
Create an Account
Institutional Sign In
OpenAthens Shibboleth
Purchase Options:
[Skip to Content Landing]
January 1, 2003

Advance Directives and Patient-Physician Communication

JAMA. 2003;289(1):96. doi:10.1001/jama.289.1.96

Advance directives provide a legal means for patients to state treatment preferences in advance, should they later become unable to participate in decisions about their care. In 1990, Congress adopted the Patient Self Determination Act, which sought to increase the use of advance directives. However, a 1999 study of members of a health maintenance organization found that only one-third of patients aged 65 years or older reported having filed an advance directive.1 Although such decisions may later be made by others acting as proxies, their judgments of patients' treatment preferences may not be accurate.2 Through communication with their patients about advance directives, physicians can play a key role in making advance directives a more effective part of patient care.

In the previously cited study,1 only 15% of patients reported having been asked about their advance care preferences by a physician or nurse, although patients were 3 times more likely to establish an advance directive if they have been asked about their wishes. Many physicians fail to initiate such discussions because they may worry about spending too much time on them,3 but in fact such discussions average less than 3 minutes in length.4 Physicians may also fear that patients are uncomfortable discussing issues surrounding their own mortality.5 In contrast to this assumption, one study found that 93% of outpatients and 89% of the general public were interested in discussing advance directives, with young and healthy individuals expressing as much interest as older, less healthy people.5 A retrospective study of decedents in a Midwestern community found that 85% had had advance directives, and suggested that the unusually high rate of use was the result of an extensive educational program on end-of-life planning.6 This implies that advance directives can be both widely used and effective, provided the resources exist to increase patient awareness.

While increasing advance directives is an important goal, a greater challenge lies in modifying them to better reflect patient preferences. Advance directive forms that are standard in US health care facilities ask patients general questions, which may or may not be applicable to a patient's individual circumstances. In addition, patients often misunderstand their options to refuse or withdraw treatment or to choose palliative care.7 It is possible that even with an advanced directive in place, patients may not receive the care they would want.

Patients must be educated about treatment options in order to make well-informed decisions about their future care. A simple way to accomplish this goal is to focus on the acceptability of potential treatment outcomes to the patient. A recent qualitative study of elderly individuals found they were more concerned with functional outcome than with the medical techniques used to achieve that outcome.8 One method of assessing the acceptability of treatment is to use scenario-based decision aids describing different treatment options and their potential outcomes.9 Assessing patients' understanding and willingness to accept potential treatment outcomes increases the likelihood of informed consent and enhances the utility of advance directives.

In some cases, advance care decisions are made by someone other than the patient. The accuracy of proxy judgments can be enhanced when the patient and his or her proxy have discussed advance care issues.2 This is in accord with research demonstrating that patients consider planning advance care to be a social interaction between loved ones, rather than simply a matter of signing forms.10 Promoting collaboration between family members and loved ones may minimize the ambiguity often found in advance directives, and may better prepare surrogates to make treatment decisions. Physicians can take an active part in this process by encouraging their patients to discuss advance care issues with their chosen surrogates.

Although systemic changes such as computerized reminders can encourage physicians to integrate advanced planning directives in their routine patient care,11 such discussions will still require meaningful communication with patients. Through such discussions, physicians have the opportunity to play a central role in increasing both the effectiveness and use of advance directives.

Gordon NP, Shade SB. Advance directives are more likely among seniors asked about end-of-life care preferences.  Arch Intern Med.1999;159:701-704.
Sulmasy DP, Terry PB, Weisman CS.  et al.  The accuracy of substituted judgments in patients with terminal diagnoses.  Ann Intern Med.1998;128:621-629.
Emanuel LL, von Gunten CF, Ferris FD. Advance care planning.  Arch Fam Med.2000;9:1181-1187.
Goldblatt D. A messy necessary end.  Neurology.2001;56:148-152.
Emanuel LL, Barry MJ, Stoeckle JD, Ettelson LM, Emanuel EJ. Advance directives for medical care—a case for greater use.  N Engl J Med.1991;324:889-895.
Hammes BJ, Rooney BL. Death and end-of-life planning in one midwestern community.  Arch Intern Med.1998;158:383-390.
Silveira MJ, DiPiero A, Gerrity MS, Feudtner C. Patients' knowledge of options at the end of life.  JAMA.2000;284:2483-2488.
Rosenfeld KE, Wenger NS, Kagawa-Singer M. End-of-Life decision making: a qualitative study of elderly individuals.  J Gen Intern Med.2000;15:620-625.
Dales RE, O'Connor A, Hebert P, Sullivan K, McKim D, Llewellyn-Thomas H. Intubation and mechanical ventilation for COPD.  Chest.1999;116:792-800.
Singer PA, Martin DK, Lavery JV.  et al.  Reconceptualizing advance care planning from the patient's perspective.  Arch Intern Med.1998;158:879-884.
Dexter PR, Wolinsky FD, Gramel-Spacher GP.  et al.  Effectiveness of computer-generated reminders for increasing discussions about advance directives and completion of advance directive forms: a randomized controlled trial.  Ann Intern Med.1998;128:102-110.

Acknowledgment: This work was supported by Medical Scientist Training Program Grant GM07739 from the National Institutes of Health.