Current issues of ACP Journal Club are published in Annals of Internal Medicine

Quality improvement

Mailed materials increased completion of durable power of attorney

ACP J Club. 1994 July-Aug;121:26. doi:10.7326/ACPJC-1994-121-1-026

Source Citation

Rubin SM, Strull WM, Fialkow MF, Weiss SJ, Lo B. Increasing the completion of the durable power of attorney for health care. JAMA. Jan



To evaluate a method for increasing the completion of the durable power of attorney for health care to prevent ethical dilemmas about life-sustaining interventions in patients who have lost decision-making capacity.


5-month randomized controlled trial.


Health maintenance organization.


1101 patients ≥ 65 years of age (mean age, 76 y; 586 men) discharged from the hospital between January and May of 1991. Participants were excluded if their hospital stay was < 3 days or if they were patients of the investigator. Follow-up was 100%.


Participants were stratified by age (65 to 74 y and ≥ 75 y) and length of hospital stay (≤ 2 wk and > 2 wk). They were randomly assigned to an intervention group (n = 552) or a control group (n = 549). Participants in the intervention group were mailed an educational brochure 4 to 8 months after discharge that described the durable power of attorney for health care and a form for durable power of attorney for health care. A covering letter that explained the randomized nature of the study and the fact that receipt of educational materials was not related to participants' prognoses was included with the package. A follow-up mailing was done 1 month later. The participants in the control group received conventional care only.

Main Outcome Measure

Percentage of durable power of attorney forms completed 5 months after mailing of educational materials assessed by review of medical records.

Main Results

18.5% of the participants in the intervention group completed a durable power of attorney form compared with 0.4% of participants in the control group ([numbers calculated from data in article] P < 0.001). The intervention was effective in all subgroups in the stratification (P < 0.01).


Mailing of information about durable power of attorney for health care increased completion rates of durable power of attorney for health care forms compared with conventional care in which information was not mailed.

Sources of funding: Kaiser Foundation Research Institute; University of California, San Francisco, Center of AIDS Prevention Studies; Robert Wood Johnson Foundation.

For article reprint: S.M. Rubin, University of California, Prevention Sciences Group, Box 0886, San Francisco, CA 94143. FAX 415-597-9213.


How much life-sustaining treatment do patients want? We know mostly what we do not know. For example, we know that neither physicians nor spouses are very good at predicting patients' end-of-life treatment wishes. And, although surveys suggest that the public accepts limits on life-prolonging treatments where quality is compromised—and even accepts euthanasia and assisted suicide—people often balk when it comes to actually voting for "humane death" initiatives or filling out advance directives.

The advance directive movement was given bureaucratic support in the United States through the Patient Self-Determination Act (PSDA), which many assumed would encourage patients to execute such documents. This has not proved to be the case. In our own hospital, the effect of the PSDA has been negligible. The proportion of medical records of hospitalized patients that contain a copy of the patient's advance directive continues to be in the range of 1% to 2%.

Investigators have attempted to increase the use of advance directives through various interventions that involve patients, physicians, nonphysician staff, and community workers, and include efforts to deal with cultural and economic obstacles. These efforts have also met with limited success.

Rubin and colleagues used educational material mailed to patients after their discharge from the hospital. Although the effect of this educational effort was highly significant statistically, I believe it is nothing to cheer about.

All of us involved with advance directive research are becoming aware that most patients do not carry through with advance directives, even when presented with the opportunity. Thus, the research agenda for the future must continue to address basic questions about patients' understanding of directives, their fears, and their reticence to act even when they voice general agreement with the concept.

Lawrence J. Schneiderman, MD
University of California La Jolla, California, USA