Patients' understanding and use of advance directives

The Patient Self-Determination Act was implemented in December 1991. Before and after its implementation, we used a structured interview of 302 randomly selected patients to determine their awareness, understanding, and use of advance directives. Implementation of the Act did not have a major effect on these. Although more than 90% of patients were aware of the living will, only about a third selected the correct definition or the correct circumstances in which it applied, and less than 20% of patients had completed one. About a third of patients were aware of a Durable Power of Attorney for Health Care and chose the correct definition, and about half identified the correct circumstances in which it applies; less than 10% had completed such a document. Surprisingly, patients who said they had completed advance directives did not demonstrate better understanding of these documents. Our results indicate that many patients, including some who have completed advance directives, do not fully understand them. It may be unwise to regard these documents as carefully considered, compelling statements of patients' preferences. Appropriate responses to our findings include increased public education, revising state statutes to bring them into congruence with public perception, and expanding the dialogue between physicians and patients.

232 Articles Patients' Understanding and Use of Advance Directives JAY A. JACOBSON, MD; BARBARA E. WHITE; MARGARET P. BATIIN, PhD; LESLIE P. FRANCIS, PhD, JD; DAVID J. GREEN, MD; and EVELYN S. KASWORM, Salt Lake City, Utah The Patient Self-Determination Act was implemented in December 1991. Before and after its imple­mentation, we used a structured interview of 302 randomly selected patients to determine their awareness, understanding, and use of advance directives. Implementation of the Act did not have a major effect on these. Although more than 90% of patients were aware of the living will, only about a third selected the correct definition or the correct circumstances in which it applied, and less than 20% of patients had completed one. About a third of patients were aware of a Durable Power of Attorney for Health Care and chose the correct definition, and about half identified the correct cir­cumstances in which it applies; less than 100/0 had·completed such a document. Surprisingly, patients who said they had completed advance directives did not demonstrate better understanding of these documents. Our results indicate that many patients, including some who have completed advance directives, do not fully understand them. It may be unwise to regard these documents as carefully considered, compelling statements of patients' preferences. Appropriate responses to our findings include in­creased public education, revising state statutes to bring them into congruence with public percep­tion, and expanding the dialogue between physicians and patients. Oacobson JA, White BE, Battin MP, Francis LP, Green DJ, Kasworm ES: Patients' understanding and use of advance direc­tives. West J Med 1994; 160:232-236) The Patient Self-Detennination Act of 1990 is a pub­lic policy intended to infonn patients about some of the steps they can take to exercise control over their medical care if they become incompetent or unable to express their wishes.l The Act, implemented December 1991, requires that health care institutions infonn all adult patients about the relevant state laws and hospital policies that apply to advance directives. Advance direc­tives such as the Living Will and Durable Power of At­torney for Health Care putatively enable patients who wish to exercise control over their future health care to do so. The Patient Self-Detennination Act also requires that hospitals inquire whether the patients have com­pleted advance directives. The Utah Personal Choice and Living Will statute specified, as do the statutes of 44 states, that a living will applies only when the person who completed it is in­competent and tenninally ill. The Utah statute required that artificial nutrition and hydration be administered to patients who have a living will, unless they specifically noted that they refused this treatment. The statute allows a person to appoint someone to make medical decisions for him or her if the granter of this durable or special power of attorney for health care becomes incompetent.l In the relatively short period of time since advance directives have been available, several studies have inves­tigated patients' attitudes about them.3,4 Patients appear to be receptive to questions about their preferences for med­ical care, and they seem attracted to the concept of a doc­ument that would ensure that their preferences would be honored. On the other hand, few patients-less than 15% in most studies-have actually completed a living will.4.5 Few studies have investigated how many patients have designated a durable power of attorney for health care, but the highest estimates do not exceed 15%.5 No studies have deeply probed patients' knowledge about advance directives. We used extended structured in­terviews that included clinical vignettes to investigate how well patients admitted to a hospital understood ad­vance directives. We examined the effect of the implemen­tation of the Patient Self-Detennination Act on patients' understanding and use of advance directives. We found that patients' understanding of these documents, espe­cially the living will, was poor and not improved by the implementation of the Act. We propose several steps that could ensure a better match between end-of-life care and patients' preferences. Patients and Methods At the LOS Hospital, Salt Lake City, Utah, the Patient From the Division of Medical Ethics, Departments of Internal Medicine, LDS Hospital and University of Utah School of Medicine, Salt Lake City. Reprint requests to Jay A. Jacobson, MD, Div of Medical Ethics, LOS Hospital, 8th Ave and C St, Salt Lake City, UT 84143. WJM, March 1994-Vol 160, No.3 Self-Determination Act was implemented by providing all patients a one-page summary of state statutes and hos­pital policies regarding advance directives. Patients' pri­mary care nurses ask patients if they have completed an advance directive. For this study we planned to complete interviews with at least 200 persons before the Patient Self-Determination Act was implemented and 100 afterwards. We conducted bedside interviews with randomly selected patients ad­mitted to LDS Hospital, a 500-bed acute-care hospital. The interviews were obtained from August 12 to Novem­ber 22, 1991, before implementation of the Act, and from February 24 to May 15, 1992, after the Act was imple­mented. The interviews were conducted by trained per­sonnel of the Division of Medical Ethics. The questionnaire was designed to assess patients' awareness and understanding of a living will and of a durable power of attorney for health care. We asked pa­tients to choose the definition for each document from a list of seven choices (Figure 1), to identify the clinical cir­cumstances in which each document applies, such as in­competence, terminal illness, or both, and to determine the effect of each document in a series of clinical vi­gnettes. We also inquired whether the patient had com­pleted either or both of these documents. The interview required about half an hour to conduct. The questionnaire was pretested and revised to ensure the greatest possible clarity. To be eligible for this study, patients had to be admit­ted within the previous 48 hours and to be 18 years of age or older. Starting at the top of a computer-generated ran­dom list of eligible patients, interviewers investigated if What do you understand a Living Will is? [Same question and choices for Durable Power of Attorney for Health Care] • A document that says how your money and property will be distributed after you die • A document that says how your body should be handled if you die • A document that specifies what life-sustaining treatment you don't want if you have a terminal ill­ness and are unable to communicate • A document that authorizes someone else to act for you in financial and other personal matters • A document that specifies who should make medical decisions for you if you are unable to do so • A document that says you are willing to donate your organs if you die • A document that specifies a way you can give money to someone or an institution and receive a tax benefit • Don't know • Other Figure 1.-Shown enclosed in the box is the question patients were asked to indicate their understanding of a living will or a durable power of attorney for health care. Advance Directives-jacobson et al 233 TABLE 1.-Percentage of Patients Who Are Aware of and Who Understand Advance Directives Before and After Im~ementation of the Patient Self-Determination Act (PSDA) (1=302) Advance Directives LM~ Will Durable Power of Attorney Awareness and Before PSDA AfterPSDA Before PSDA AfterPSDA Understanding (11=202), 96 (n=100),96 (n--202), 96 (n=100),96 Awareness ............. 91 94 32 39 Chose single correct definition ............ 31 41 29 37 Chose correct drcumstance(s) •...... 28 32 46 54 the patient was available to be interviewed and if the pa­tient, as judged by his or her nurse, was mentally and medically able to participate in the study. Patients were eligible to be interviewed regardless of the acuity of their illness or unit to which they were assigned. Patients who did not speak English were not interviewed. Available pa­tients deemed capable were approached and asked if they would consent to be interviewed. Midway through the in­terview, patients were asked if they wished to continue. Factors associated with answers to questions and the completion of advance directives were statistically ana­lyzed using X2 and Fisher's exact test. Results Before our hospital implemented the Patient Self-De­termination Act, we completed interviews with 202 of 361 (56%) patients that we approached. Afterwards, 100 of 195 patients (51 %, P = not significant [NS]) completed the interview. Our study population, 41 % male, with a mean age of 48 years, did not differ significantly from the inpatient population during the interview periods. Like the patients at our hospital generally, our subjects were predominantly white, and 48% said they belonged to the Latter Day Saints religion. Most of our interviewed patients (91 %) had graduated from high school; 68% had some addi­tional education, and 29% had completed four or more years of higher education. Patients in the two parts of our study did not differ significantly with respect to the above factors. Table 1 shows awareness and two levels of under­standing of living wills and durable powers of attorney for health care by our study subjects. There were no signifi­cant differences after the Patient Self-Determination Act was implemented. When asked to define only a living will or a durable power of attorney, 10% and 2%, respectively, of patients chose to define both documents. Only 6% of patients in both periods selected the correct definition and identified the clinical requirements for both types of ad­vance directives. The 30 patients who correctly answered these questions about the living will were not signifi­cantly different from patients who answered incorrectly with respect to sex, age, religion, or race. They were more likely, however, to have finished high school (100% ver­sus 89%, P < .05). The 41 people who answered these questions accurately about the durable power of attorney 234 WJM, March 1994-Vol 160, No.3 TABLE 2.-Percentage of Patients Who Correctly Indicated the Effect of an Advance Directive in a Paradigmatic Case, or a Case That Involved a Persistent Vegetative State (PVS) Patients Unde~tood Effect of Advance Directives, % Living Will Durable Power Durable Power Patient Self·Determination in Paradigm Livin~ Will of Attorney in of Attorney Act Implemented COle in VS Paradigm Case inPVS Before Act (n = 202) ..... 50 24 75 71 After Act (n = 100) ....•. 58 16 76 78 Significance, P •.•••••••• NS NS NS NS NS = not significant differed from those who answered incorrectly by age (76% versus 47% were younger than 51, P < .001), and they were also more likely to have finished high school (100% versus 88%, P < .05), but they did not differ sig­nificantly by sex, religion, or race. Most patients who had heard of advance directives re­ported that the media was the source. Less than 10% said they heard about them from a physician or nurse. During the second study period when written information about advance directives was given to patients by the hospital, 24% and 16%, respectively, recalled receiving that infor­mation about the living will and durable power of attorney. We also assessed patients' understanding of advance directives with clinical vignettes. In one vignette, a para­digmatic case, "The patient has cancer, cannot communi­cate his or her wishes, and is expected to die within one month. The patient has respiratory failure and will die sooner unless a ventilator is used." In a second vignette, "The patient has been in an acci­dent and is now in an irreversible coma (persistent vege­tative state). Nutrition and hydration (food and water), delivered intravenously or by a tube inserted into the stomach, are necessary for survival." Table 2 shows the proportion of patients who correctly indicated the effect of each advance directive in these cases. Notably, about 80% of patients made an incorrect choice about the effect of a Utah living will in the case of the patient in a persistent vegetative state: 44% indicated that the living will would prevent the provision of nutri­tion and hydration, 11 % thought the living will would re­quire these measures, and 22% said they did not know the effect of the living will in this case. Only 21 % chose the correct answer, namely, that the living will did not apply in this case because the patient was not terminally ill. Before the Patient Self-Determination Act was imple­mented, 15% of our subjects reported completing a living will and 5% a durable power of attorney for health care. Afterwards the proportions were 18% and 7%, not a sig­nificant change. Of the 48 patients who said they had a living will, 16 said they also had a durable power of attor­ney. Two other patients reported that they had only a durable power of attorney. Although Table 2 includes the answers of all the re­spondents to our survey, we separately examined the re­sponses of patients who reported that they had completed either a living will, a durable power of attorney, or both. When we examined the demographics of these patients Advance Directives-Jacobson et al and those in our survey who had not completed such doc­uments, we found that patients with living wills were sig­nificantly more likely to be older than 51 years than those without them (P < .001). These patients were also signifi­cantly more likely to have graduated from high school (98% versus 89%, P = .04) and to have completed addi­tional years of education (P = .03). We noted no signifi­cant differences between those patients who reported completing a durable power of attorney and those who did not, although those with the document were more likely to be older than 51 years. Surprisingly, patients who said they had completed a living will did not understand this document's definition or clinical application better than patients who said they had not completed one. Of the 48 patients who reported having such a document, 16 (33%) identified the single correct definition, and 19 (40%) correctly identified the circumstances in which the Utah Living Will applied. Presented with the paradigmatic vignette, living will holders did not perform better than those without the doc­ument (P = NS). Of the 48 patients, 35 (73%) correctly said that it would prevent the use of a ventilator. When presented with the vignette about a patient in a persistent vegetative state, in which the Utah Living Will would not apply, living will holders again did no better than those without the document (P = NS): 31 (65%) said the living will would prevent health care professionals from providing nutrition and hydration, 9 (19%) correctly said the living will does not apply, 6 (12%) said they did not know, and 2 (4%) said the living will would require artificial nutrition and hydration. Like the patients who had completed a living will, most patients who had completed a durable power of at­torney for health care did not have a significantly better understanding of the definition of the document or the cir­cumstances in which it applies than the rest of the sur­veyed population. Of the 18 patients who said that they had appointed a proxy for health care decisions, 4 (22%) chose the single correct definition, and 9 (50%) correctly identified the circumstances in which a durable power of attorney applies. When presented with the paradigmatic vignette, 15 (83%) indicated correctly that the proxy had the legal right to decide about the use of a ventilator. We asked patients whether a living will or durable power of attorney, as the final expression of their wishes, should be the determinant of their care. We asked them un­der what circumstances the preferences expressed in a liv­ing will or by an appointed proxy should be overridden. Their responses are shown in Table 3. These responses did not change significantly between study periods. Remark­ably, a similar proportion of patients who said they had ex­ecuted an advance directive, which is legally binding, said that it should be overridden. Discussion Before drawing conclusions from our study, certain limitations must be acknowledged. Interpreted as a re­sponse rate, the 55% of patients who completed the inter­view would appear to be a rate that could provide results WJM, March 1994-Vol 160, No.3 Advance Directives-Jacobson et al 235 TABLE 3.-Patients' Opinions About Overriding Advance Directives Should Directive Be Overridden? Patients With Durable Patients With Living Wills Powers of Attorney All Respondents (n=302), % (n=48), % (n=18), % Advance Directive Yes Uving Will Physicians should ignore living will instructions if they think to do so is best for the patient.. ..................... 20 Physicians should ignore living will instructions if family members ask them to .. 19 Durable Power of Attorney Physicians should ignore proxy's instructions if they think to do so is best for the patient. ...................... 21 PhYSicians should i~nore proxy's instructions if fami y members ask them to .• 20 representative of a population if there were no systematic and confounding reasons for refusing to participate. We asked patients why they refused and were given a variety of reasons including discomfort, the presence of visitors, anxiety, fatigue, and a lack of interest in participating in such a survey. It is unlikely that patients refused to be in­terviewed because of their extensive knowledge about advance directives. Thus, it seems that our interviewed sample was not likely to underestimate the level of knowledge in our hospital population. Certainly a survey in a single hospital, particularly with a popUlation that is primarily white and well educated, cannot be assumed to be representative of the knowledge and opinions of other segments of the American population. Each of these pos­sible biases in our sample would seem to suggest, how­ever, that our results would overestimate rather than underestimate the level of understanding of advance di­rectives in the rest of the population. Although we inter­viewed patients in only one hospital, our patient group was more inclusive with respect to age than those in other studies of patients' knowledge about advance directives.4 Although a higher proportion of our patients were mem­bers of the Latter Day Saints religion than might be found elsewhere, we did not find religion to be a significant fac­tor with respect to awareness, understanding, and use of advance directives. The correctness of our patients' an­swers was judged with respect to the Utah Personal Choice and Living Will statute as of 1992. Patients in other states may be more or less knowledgeable about their state's laws, or the statutes may be more aligned with public perception (or both). Our study differs from others in that we assessed the knowledge of adult patients of all ages, but who had a common characteristic, namely a need to be in the hospi­tal. Thus, although our sample is obviously different from a random sample of the United States public, a healthy population, nursing home residents, or a group selected based on being older than 65, patients in our study cer­tainly had a reason to be concerned about and affected by decisions regarding their health care. In a survey of pa- No 63 66 53 55 Not Sure Yes No Not Sure Yes No Not Sure 17 17 71 12 15 23 73 4 26 0 78 22 25 17 72 11 tients admitted to a tertiary care hospital such as ours where many admissions are anticipated and elective, we might expect to have encountered a higher proportion of patients who had completed advance directives than was found to be the case. We did find, as did other investiga­tors, that patients are generally aware of advance direc­tives, 4.5 but the level of patient knowledge and under­standing remains poor. There are several other possible limitations to con­sider because the number of persons in our survey who had actually completed a living will and, particularly, a durable power of attorney was relatively small. Therefore, it may be premature to conclude that patients who have completed these documents do not differ significantly in their knowledge about them from patients who have not done so. Our interview subjects did not include patients who were unconscious, unable to communicate, or in­competent. These are the patients for whom advance di­rectives might be useful. We have no information on the frequency with which these persons had previously exe­cuted advance directives or their knowledge about them. The competent patients we did talk with were obviously in an environment that could be stressful, and the accu­racy of their answers may have been adversely affected by that. We also did not insist that patients who reported completing advance directives produce copies of them. Despite these possible limitations, it appears that a significant proportion of the patients admitted to our hos­pital do not understand our state's advance directives very well. Implementing the Patient Self-Determination Act through a relatively passive program of distributing writ­ten information about advance directives has done little to improve patients' understanding. In our hospital, the Act has had little discernible effect on the low rate at which patients complete advance directives. An unexpected result in our study was that even pa­tients with advance directives frequently misunderstand them. Like patients without living wills, patients with them often failed to recognize the requirement for a ter­minal condition. Perhaps more surprising was that a sig- 236 WJM, March 1994-Vol 160, No.3 nificant minority of patients who had executed advance directives thought they should be overridden if physicians or family members thought it best to do so. Completing an advance directive did not seem to in­crease patients' knowledge about the document, but it may have had a benefit for these persons independent of the document's effect on medical decisions. The act of completing the directive may lower a patient's level of anxiety about death and the treatment associated with the end of life. That result has been suggested by the study of Henderson.' Our impression, as we interviewed these pa­tients, was that they were proud and somewhat relieved that they had taken steps that they felt would control their care in circumstances in which they viewed a lack of con­trol as frightening and burdensome. Completing a living will is unlikely to provide the kind of control they wanted in one of the circumstances they feared most. When we recognized the discrepancy between pa­tients' perceptions about the Utah living will and the statute, particularly with respect to the persistent vegeta­tive state, we shared our findings with state legislators. They chose to amend the statute and include persistent vegetative state as an additional criterion that would make the living will operative.7 This certainly will help some patients realize their expectations of end-of-life care. What remains unresolved, however, is the apparent paradox that a substantial minority of patients want their final expres- Advance Directives-Jacobson et al sions of personal preference to be overridden if their physician or family thinks it best to do so. At this point then, despite the Patient Self-Determina­tion Act, health care professionals are still likely to find that most patients who face end-of-life decisions have not completed advance directives. Even if patients have com­pleted advance directives, clinicians should recognize that the directive(s) may be neither complete nor thoroughly understood expressions of patients' preferences. To en­courage patients to complete advance directives without extensive education and counseling, or to strictly adhere to all advance directives without discussion, is unlikely to serve patient autonomy fully. Discussion with all patients before a crisis still seems the best way to ascertain pa­tients' values, preferences, and goals for therapy. REFERENCES I. Patient Self-Detennination Act-Omnibus Budget Reconciliation Act of 1990. Pub L No. 101-508 2. Personal Choice and Living Will Act. Vt Code Ann 75-2-1101 et seq 3. Emanuel LL, Barry MJ, Stoeckle 10, Ettelson LM, Emanuel EJ: Advance directives for medical care-A case for greater use. N Engl J Med 1991; 324:889- 895 4. Gamble ER, McDonald PJ, Lichstein PR: Knowledge, attitudes, and behav­ior of elderly persons regarding living wills. Arch Intem Med 1991; 151:277-280 5. High OM: All in the family: Extended autonomy and expectations in surro­gate health care decision-making. Gerontologist 1988; 28(suppl):46-5I 6. Henderson M: Beyond the living will. Gerontologist 1990; 30:480-485 7. Personal Choice and Living Will Act. VtCode Ann 75-2-1101 et seq (suppl 1993)