Patients who complete advance directives and what they prefer.

OBJECTIVE

To determine health care preferences expressed by patients in advance directives (ADs) and to identify characteristics of patients who completed them.

PATIENTS AND METHODS

A computer-generated random sample of 500 patients was selected from the 25,865 (margin of error, +/- 4.34%) unique patients who submitted ADs to Mayo Clinic Rochester between January 1, 2004, and July 1, 2005. After excluding 24 (4.8%) patients who had submitted documents related to financial and property matters instead of ADs, we analyzed the contents of the 476 ADs and the demographic features of the patients to whom the documents belonged.

RESULTS

The median (range) age of the patients at the time they signed their respective ADs was 67 (19-97) years. Of the 476 study patients, 409 (91.3%) were high school graduates; 339 (71.2%) had submitted a combined AD, which has features of a living will and a health care power of attorney; 434 (91.2%) had designated a health care agent; and most had granted the agent powers to consent for procedures (340 [78.3%]), to access information (327 [75.3%]), and to withhold and withdraw life-sustaining treatments (337 [77.6%]). Most patients expressed a desire for pain control (308 [64.7%]). For the clinical situations of dying or permanent unconsciousness, most patients explicitly expressed a preference to avoid "general life support" (371 [77.9%]) but did not explicitly address common life-sustaining treatments, including cardiopulmonary resuscitation, mechanical ventilation, hemodialysis, blood transfusion, and artificial nutrition and hydration (if they did, nearly all expressed a preference to avoid the treatments).

CONCLUSION

In ADs submitted to our institution, most patients designated a health care agent and granted the agent broad decision-making powers. Although most expressed a desire to avoid "general life support" if dying or permanently unconscious, few expressed preferences regarding specific life-sustaining treatments. Patients, clinicians, and others who use ADs, and investigators contemplating research involving ADs, might find these results informative.