What do patients express as their preferences in advance directives?

BACKGROUND: Since the passage of the Patient Self-Determination Act in 1991, there has been interest in urging patients to execute advance directives (ADs) for medical care. There are not much data, however, as to what the ADs that patients execute actually specify. I have investigated the percentage of inpatients and outpatients who are admitted to a community hospital who have executed ADs, and I have tabulated what preferences are actually expressed in the ADs that are in hospital records.
METHODS: A questionnaire is filled out by each patient admitted to this hospital, and their response recorded as to whether they have executed an AD. I have tabulated these responses for inpatients and outpatients for the calendar year 1994. I have also examined the ADs in all available hospital records, and tabulated the wishes expressed in these directives.
RESULTS: For inpatient admissions during the calendar year 1994, of 8727 questionnaires completed, 11% of patients indicated that they had executed an AD. For outpatients, the corresponding figures are 22,966 and 15%. A total of 343 hospital records containing ADs were examined. Of these, 15 were nonmedical directives and were excluded. Of the 328 medical ADs, 86 (26%) were living wills, expressing the wish that if the individual had an incurable disease or irreversible injury that he or she not be given any treatment that would only delay death. There were 210 power of attorney for health care forms completed; these were 64% of all the medical ADs. Of these, 7 did not specify any preference that patients wanted their proxy to follow. The remaining 203 forms were divided as follows: 189 individuals requested that they did not want life-sustaining treatment if the burdens of treatment outweighed the expected benefits; 12 wanted their lives to be prolonged unless they were in an irreversible coma; and 2 wanted their lives to be prolonged to the greatest possible extent regardless of the chances for recovery or the cost. There were 32 do not resuscitate forms executed exclusively by residents of nursing homes that specified that they did not want cardiopulmonary resuscitation or artificial feeding.
CONCLUSIONS: The overwhelming desire expressed by the patients in the ADs was not to have their lives prolonged if their medical condition were such that treatment would merely delay death. Only a minuscule number of patients, less than 0.7%, wanted everything done to prolong life regardless of the chance for improvement or the cost. Because such a small percentage of patients have ADs, it is recommended that each hospital appoint a committee on ADs to do everything possible to encourage patients to execute an AD. A second mission of this committee would be to do everything possible to encourage physicians to pay close attention to their patients' wishes for medical care at the end of life.