Articles from The Journal of Clinical Ethics are copyrighted, and may not be reproduced, sold, or exploited for any commercial purpose without the express written consent of The Journal of Clinical Ethics. 212
The Journal of Clinical Ethics
Fall 2015
Craig M. Klugman and Nicole M. Tolwin, “Assessing Advance Care Planning: Examinig Autonomous Selections in an Advance Directive,” The Journal of Clinical Ethics 26, no. 3 (Fall 2015): 212-8.
Assessing Advance Care Planning: Examining Autonomous Selections in an Advance Directive Craig M. Klugman and Nicole M. Tolwin
ABSTRACT We examined the management of completed advance directives including why participants completed a document, what procedures and values they chose, with whom they held end-of-life conversations, and where they stored their document. Participants elected to complete a SurveyMonkey survey that was made available to individuals who wrote an advance directive through TexasLivingWill.org; 491 individuals elected to fill out the survey, aged 19 to 94 years. The survey asked multiple questions about why participants completed an advance directive, where they would store it, and with whom they had conversations about their end-oflife wishes. A list of procedures and values allowed participants to indicate what they refused or requested in their advance directive. Advance directives are most often completed to allow patients to prepare and control the healthcare they wish to receive when dying. One-half to two-thirds of individuals refuse common end-of-life medical procedures, and the rest request the procedures. We found a correlation between the choice to refuse or request a procedure and the age of the participant. Participants reported that their end-of-life conversations most often occurred with their spouse. Respondents often reported that their advance directive was stored with their physician and power Craig M. Klugman, PhD, is Professor in and Chair of the Department of Health Sciences at DePaul University in Chicago. Cklugman @depaul.edu. Nicole M. Tolwin, BS, is an MSN Candidate at the University of Illinois at Chicago College of Nursing. ©2015 by The Journal of Clinical Ethics. All rights reserved.
of attorney for healthcare, conversations with those individuals rarely happened. Advance directives document patients’ requests for and refusals of end-of-life care. Physicians and surrogates need to be better educated so that the documents are part of a meaningful conversation with the patient. Because patients’ choices change over their lifespan, these documents need to be revisited regularly and not completed as a onetime event. INTRODUCTION The advent of modern advance directives can be traced to the 1977 passage of the California Natural Death Act. In the succeeding decades, most states enacted similar laws. In 1991, the United States Congress passed the Patient Self-Determination Act, which provided for information and inquiry about advance directives upon admission into a hospital. Advance directives allow an individual to explain his or her wishes for medical care at the end of life and to appoint a surrogate decision maker. The use of an advance directive has been associated with patients choosing less aggressive care at the end of life and incurring lower costs for their care.1 However, beginning with the “study to understand prognoses and preferences for outcomes and risks of treatments” (that is, SUPPORT),2 some researchers found fault with advance directives. Scholars criticized advance directives as ineffective and confusing, as patients may not understand the documents, medi-
Articles from The Journal of Clinical Ethics are copyrighted, and may not be reproduced, sold, or exploited for any commercial purpose without the express written consent of The Journal of Clinical Ethics. Volume 26, Number 3 cal procedures, and patients’ experiences involved; directives require patients to imagine unrealistic future competency and focus more on procedures than on health outcomes or quality of life; as standardized forms, directives may ignore conversations, emotional preparations, and nuances in decision making.3 Many responses to these criticisms and shortcomings revolved around building a better form. For example, ethicists employed computer-based programs,4 online registries,5 physician orders for life sustaining treatment (POLST), or dispensing with documents in favor of directed conversation.6 Many of those proposals were not evidence-based, however, or lacked an understanding of how patients actually used the documents. The rate of completion of advance directives by adults ranges from 9.8 percent to 72.0 percent, depending on the study. In 1996 Hanson and Rodgman reported on 1986 data, which found that 9.8 percent of respondents to a national survey had completed an advance directive; completion was correlated with race, sex, insurance status, level of education, and presence of a chronic disease.7 In 2001, Braun and colleagues reported that 29.0 percent of Hawaiian residents said they had a formal advance directive, and 22 percent said that they had a written power of attorney for healthcare document.8 A 2003 Canadian study reported that 46.0 percent of subjects said they had a written directive.9 In 2010, Silveira and colleagues reported that the rate of completing an advance directive by seniors in their study increased in the previous decade from 47.0 percent to 72.0 percent.10 Rao and colleagues report that 26.3 percent of U.S. adults had an advance directive.11 Similar to Hanson and Rodgman, Rao and colleagues associated completion of an advance directive with education, race, and presence of disease; they also found correlations with advancing age and higher income. Scholars have argued that the continued low level of completion is a sign that people do not want an advance directive.12 Others suggest that the low rates of completion are rooted in fear, lack of education, and mistrust of the healthcare system.13 This prompts the question, “Why do people choose to complete or not to complete an advance directive?” Reasons can be understood by loci of control: external and internal. The external is comprised of race/ethnicity, education, income, health status, religious affiliation, martial status, experience with death, conversations about death with others, relationships with offspring, and physicians’ communications.14 The internal includes personal beliefs and psychological continuity.15
The Journal of Clinical Ethics
213
Most studies of reasons are edic in their approach, that is, they examine data external to lived experience. Perkins and colleagues used an emic approach: they surveyed 58 individuals from three ethnic groups and found that most believed that a patient should have a say in choices of treatment, and that advance directives “improve the chances that a patient’s wishes will be followed.”16 But these findings are from a small sample size and may not be generalizable. Other studies examined whom people chose as their surrogate decision maker. Using health survey data, Hopp found that 87.2 percent (n=520) of individuals older than 70 years chose to rely on spouses or adult children and grandchildren to make endof-life decisions when they could not.17 Some studies report that patients said they had trouble discussing end-of-life issues with their adult children.18 A study conducted in Wisconsin of 1,883 older adults found that those who were happily married were more likely to appoint their spouse as their designated power of attorney for healthcare, and that those who received emotional support and had a close relationship with their children were more likely to appoint an adult child.19 Note that a surrogate decision maker was appointed does not necessarily mean a patient discussed end-of-life preferences with the surrogate. An Ohio study found that fewer than half of 231 older adult study participants, aged 65 to 99, discussed their end-of-life wishes with another person.20 Of 513 patients in the Hospitalized Elderly Longitudinal Project (HELP), 70.8 percent stated that they would prefer to have a family member make a decision about resuscitation, rather than have their pre-stated wishes be followed.21 In addition, a study of 646 SUPPORT patients found that 78.0 percent said they would prefer to have their family and physician decide their medical wishes rather than having their advance directive followed.22 There is little to no literature, however, on what choices people make in their advance directive. Nor has much been written about where people keep their document, despite concerns about the availability of patients’ advance directives when they are needed.23 METHODS Procedure TexasLivingWill.org is an advance care planning website that provides end-of-life education, an interactive interview on what choices one might make when facing a terminal illness or irreversible condi-
Articles from The Journal of Clinical Ethics are copyrighted, and may not be reproduced, sold, or exploited for any commercial purpose without the express written consent of The Journal of Clinical Ethics. 214
The Journal of Clinical Ethics
tion, and an opportunity to complete and store an advance directive. Those who visited the website were presented with a link to a voluntary and anonymous program evaluation on SurveyMonkey where they were asked 20 multiple choice questions, including how they learned of the website, why they used the website, and their experience using it.24 Respondents also were asked why they completed an advance directive, what choices they made in the document, where they would store it, and with whom they had conversations about their end-oflife wishes. Respondents were asked to provide their age at their last birthday. The multiple-choice selections were developed from the published literature as well as from educational sessions held with community members and health professionals prior to the launch of the website. Questions on procedures (resuscitation, artificial nutrition and hydration, antibiotics, dialysis, blood and blood products, respiratory support, and surgery) asked respondents to choose procedures that they would “request” or “refuse”; other questions asked respondents to select all that applied from a list of possible responses. For the latter inquiries, when a respondent skipped the question, the response was coded as though the respondent did not select a choice. This retrospective analysis was declared exempt by the Institutional Review Board at DePaul University on 28 April 2014. Frequency analysis and one-way ANOVAs were conducted using SPSS version 22.0.
Fall 2015 their family and 56.4 percent (n=251) said they wanted to avoid a non-sentient state. A lesser number indicated they were motivated by a desire to “protect myself from others who might make decisions I would not choose” (41.74 percent, n=192), “I’m getting older” (40.87 percent, n=188), and from the personal experience of observing another person’s death (39.78 percent, n=183). Several respondents said that they completed the document because of their own illness (14.35 percent, n=66) or a hospital stay (9.57 percent, n=44). Table 1 lists the prospective choices participants made in regard to the end-of-life procedures that they desired or refused, as well as their values regarding quality of life. In most cases, two-thirds to threequarters of participants chose to refuse interventions. One option allowed participants to share whether they indicated a willingness to allow a healthcare surrogate to consent for participation in clinical trials at the end of life; 44.2 percent included this selection. Participants described keeping their advance directive document in a number of places as listed in table 2. No one place was selected by the majority. The most common locations were with a spouse/ partner or with their will of inheritance and other important papers. The evaluation also asked participants to indicate with whom they had conversations about ad-
TABLE 1. Self-reported end-of-life values and procedures chosen Requested
Refused
RESULTS Participants Of the 543 individuals who visited the TexasLivingWill.org website and clicked on the evaluation, 491 completed some portion of the survey between April 2009 and April 2014. Since the survey was anonymous there was no way to know why 52 people chose not to complete, although many may have clicked the survey link by accident. The respondents ranged in age from 19 to 94 years, with a mean of 55.53 years and a median of 57.00 years. With three exceptions, all of the respondents were residents of Texas. Varied reasons were given for why participants completed advance directive documents (n=460); however, 80.70 percent (371) of respondents indicated they wanted to “be prepared” and 65.65 percent (302) said “to have control of my own decisions.” Of the 460, a total of 61.8 percent (n= 275) stated that they completed an advance directive for
Choice
n
%
n
%
n
Resuscitation Artificial nutrition and hydration Antibiotics Dialysis Blood and blood products Respiratory support Surgery To be free from pain* To have a period of support before withdrawal of medical interventions* Quality over quantity of life* Quantity over quality of life*
382
26.4
101
73.6
281
376 374 372
30.1 36.6 25.8
113 137 96
69.9 63.4 74.2
263 237 276
371 374 366 441
33.4 124 26.5 99 30.3 111 95.06 367
66.6 73.5 69.7 --
247 275 255 --
440 396 396
18.9 84.1 4.3
83 371 19
----
----
* These choices could be “requested” but not refused. Not choosing indicates that the participant did not agree with the statement.
Articles from The Journal of Clinical Ethics are copyrighted, and may not be reproduced, sold, or exploited for any commercial purpose without the express written consent of The Journal of Clinical Ethics. Volume 26, Number 3
The Journal of Clinical Ethics
TABLE 2. Self-report of advanced care planning conversations and the location of AD documents Location of AD
Discussed AD with
Category
%
n
%
n
Spouse/partner Child Sibling Power of attorney for healthcare (surrogate decision maker) Parents Primary care physician Attorney Clergy With my will of inheritance In a safe deposit box Online registry With my local hospital Motor vehicle glove box On my smartphone In my purse In my wallet Friend Coworker Other healthcare provider
44.4 29.5 16.6
195 130 73
63.9 42.4 31.3
282 187 138
38.5 9.8 31.3 4.5 2.3 39.9 19.0 14.3 13.4 12.2 4.5 4.1 4.1 36.5 9.3 3.2
170 43 138 20 10 176 84 63 59 54 20 18 18 161 41 14
27.4 20.9 18.8 4.3 2.5 ------------
121 92 18 19 11 ------------
215
vance care planning; spouse/partner was the most common answer, followed by friends and children (see table 2). The lack of a response indicates that the participant had no conversations or that the participant did not have someone in his or her life in this category. Those older than age 50 were significantly more likely to indicate that they would refuse resuscitation, antibiotics, blood and blood products, respiratory support, surgery, artificial nutrition and hydration, and dialysis than were those younger than age 50 (see table 3). The differential choices made by each age group are listed in table 4. DISCUSSION Participants used advance directives to request as well as to refuse treatment: 26.4 percent of the participants requested resuscitation, 30.1 percent requested artificial nutrition and hydration, 36.6 percent requested antibiotics, 25.8 percent requested dialysis, 33.8 percent requested blood and blood products, 26.5 percent requested respiratory support, and 30.3 percent requested surgery. This contradicts the more common belief that advance directives are used solely to refuse medical treatment.25 Subjects older than 50 were most likely to refuse aggressive care, and those younger than 50 were more likely to request it. In general, older people were more likely
TABLE 3. One-way ANOVA results correlating age (>50 and 50
%
