Too Soon to Give Up? Re-examining the Value of Advance Directives

Abstract

In the face of mounting criticism against advance directives, we describe how a novel, computer-based decision aid addresses some of these important concerns. This decision aid, Making Your Wishes Known: Planning Your Medical Future, translates an individual’s values and goals into a meaningful advance directive that explicitly reflects their healthcare wishes and outlines a plan for how they wish to be treated. It does this by 1) educating users about advance care planning; 2) helping individuals identify, clarify, and prioritize factors that influence their decision-making about future medical conditions; 3) explaining common end-of-life medical conditions and life-sustaining treatment; 4) helping users articulate a coherent set of wishes with regard to advance care planning—in the form of an advance directive readily interpretable by physicians; and 5) helping individuals both choose a spokesperson, and prepare to engage family, friends, and healthcare providers in discussions about advance care planning.

When Nina Klinger, 78, went to see her doctor, she learned that she was in the early stages of Alzheimer’s Disease—the same disease that had killed her husband the previous Spring. Her husband’s illness had been very difficult, not only due to his progressive dementia, but also because of the multiple medical complications he endured, including pneumonia, a flare-up of his diabetes, and surgery to place a feeding tube when he would no longer eat. So, too, there had been some challenging interactions with doctors involved in his medical care. To their credit, the doctors tried to fix every problem that arose. But because they didn’t know what her husband would have wanted, they also didn’t know when enough was enough. And because she had never really talked with her husband about his specific wishes for end-of-life medical care, Mrs. Klinger didn’t feel she could tell the doctors when to stop. As a result, her husband was kept alive for months, first on a mechanical ventilator in the intensive care unit, and then with insulin drips, feeding tubes, and other invasive treatments—all the while appearing agitated and distressed. Now as Mrs. Klinger is thinking about her own relationship with the medical care system, she wonders what can she do to avoid unwanted medical treatments if and when the time comes that she cannot speak for herself.

In the complex world of modern medicine, it is not uncommon for patients to receive prolonged and questionable medical treatment at the end of life. (SUPPORT 1995; Dickenson 2000; Breen, Abernethy et al. 2001) For the past 20 years, physicians, ethicists, and lawyers have typically advised patients like Mrs. Klinger to document their wishes in advance, using a state-approved living will form or some other type of advance directive. But in recent years, advance directives have been the subject of a growing number of critiques, (Schneiderman, Kronick et al. 1992; Teno, Lynn et al. 1997; Teno, Licks et al. 1997; Ott 1999; Shalowitz, Garrett-Mayer et al. 2006) some of which have argued that key aspects of this enterprise are flawed. (Schkade and Kahneman 1998; Gilbert and Wilson 2000; Fagerlin, Ditto et al. 2001; Fagerlin, Ditto et al. 2002; Fagerlin and Schneider 2004; Ubel 2006) In this paper, we will describe an interactive, computer-based decision-aid that was designed to overcome some of the major limitations of standard advance directives, (Green and Levi 2009). In doing so, we will defend the broader concept of advance care planning as a worthwhile and achievable aspiration while acknowledging that limitations remain, including various deficits of advance directives.

To start, it’s important to distinguish advance directives from advance care planning. Advance directives are documents that formally convey an individual’s wishes about medical decisions to be made in the event that he or she loses decision-making capacity. In contrast, advance care planning refers to a process that involves preparing for future medical decisions in the hypothetical event that individuals are no longer able to speak for themselves when those decisions need to be made. (Doukas and McCullough 1991; Miles, Koepp et al. 1996; Prendergast 2001; Schwartz, Wheeler et al. 2002; Lo and Steinbrook 2004) As such, advance care planning includes thinking about, and communicating many things, and often culminates in the creation of an advance directive that addresses:

• what is most important to the individuals and why

• what kinds of medical conditions they would (or would not) want treated

• what kinds of medical treatments they would (or would not) be willing to endure

• what kinds of outcomes they would (or would not) wish to live with—where this includes specific disabilities as well as more general quality-of-life issues

• who they would want to serve as spokesperson if and when they could not make their own medical decisions

Recently, critics have assailed advance directives, and in doing so, questioned some of the premises on which advance care planning is based. One such criticism is that people cannot accurately predict in advance what medical decisions they would want in the future. In some instances, this is because people do not have sufficient knowledge and experience to know what they are making decisions about—what Forrow has called The Green Eggs and Ham Phenomena. (Forow 1994) This can be due to lack of information, inaccurate preconceptions, or misunderstandings about the nature of the decisions at issue. For example, many people have strong views about being “hooked up to a breathing machine.” Yet in reality, whether one would be willing to endure mechanical ventilation likely will depend on the circumstances—short- versus long-term treatment, good versus bad prognosis for recovery, short versus long life-expectancy, etc. Unfortunately, the time and patience needed for physicians (or other healthcare providers) to put such treatment decisions into proper context (much less to explain their many nuances) are often in short supply. Furthermore, physicians are sometimes reluctant to initiate the discussions that are needed to help prepare their patients for future medical exigencies—either due to lack of skill (Tulsky, Fischer et al. 1998; Cherlin, Fried et al. 2005) or concern over adverse consequences. (Kahana, Dan et al. 2004) Hence preconceptions and misunderstandings about many medical conditions and their treatment persist.

Another reason why individuals might inaccurately predict future medical decisions is their overestimation of the impact that specific debilities will have on their lives (Loewenstein and Schkade 1999; Loewenstein 2005; Fried, Bradley et al. 2006; Ubel 2006) —either because they discount the many aspects of their lives that are not altered by the debility, or they miscalculate the actual impact of various debilities. Relatedly, individuals may underestimate their ability to constructively adapt to even the most severe debilities—think Christopher Reeve. (Schkade and Kahneman 1998; Loewenstein and Schkade 1999; Gilbert and Wilson 2000; Coppola, Ditto et al. 2001; Ubel 2006) So, too, preferences may simply change, due to a change of mind, change in one’s underlying values, change in context or perhaps in response to realizing the impending consequences of one’s decision. (Fagerlin, Ditto et al. 2002; Ditto, Jacobson et al. 2006; Fried, Byers et al. 2006) In these circumstances, previously communicated wishes may not reflect this change of heart. Taken together with the issue of inadequate knowledge and experience, these concerns over the accuracy of advance care planning can be referred to as problems with future forecasting, and they are not insignificant. (Fried, O'Leary et al. 2007)

A second set of problems raised by critics of advance directives involves communication and logistics. Many advance directive documents are narrowly constructed, and leave little room (conceptually and/or physically) for individuals to express the full scope of their wishes. (Teno, Licks et al. 1997; Holley, Hines et al. 1999) Even when documents are broadly conceived and designed, summarizing and accurately expressing one’s thoughts can be difficult. Moreover, documentation that may appear crystal clear to the person creating an advance directive can seem confusing, if not opaque, to those who must interpret and use the advance directive as the basis for making medical decisions. (Ditto, Danks et al. 2001; Fagerlin, Ditto et al. 2002; Upadya, Muralidharan et al. 2002) The involvement of surrogate decision-makers can further compound these difficulties insofar as their decisions on behalf of a patient may be complicated by conflicts of interest and/or surrogates’ own treatment preferences. (Pruchno, Lemay et al. 2005; Pruchno, Lemay et al. 2006; Wrigley 2007) Equally challenging is the mundane task of keeping track of the necessary information so that a patient’s advance directive is available when needed. Documents are not always where they are supposed to be, files go missing, and memories fail to recall a location. (Lynn and Goldstein 2003; Fagerlin and Schneider 2004)

A third set of problems has to do with translating advance directive documents into appropriate medical decision making at the bedside. This has several levels: 1) the challenge of matching a general (or specific) wish to a viable treatment option; 2) accurately interpolating when no direct correlation exists; 3) reconciling conflicting and/or illogical wishes so that treatment decisions make sense and are consistent with the genuine values and preferences of the individual; and 4) knowing whether an individual would have wanted to delegate decision-making to their spokesperson’s judgment. (Teno, Stevens et al. 1998; Upadya, Muralidharan et al. 2002; Fried, O'Leary et al. 2007; Fried, Van Ness et al. 2007)

A fourth kind of problem that is at least as damning as the preceding critiques is the empirical evidence that the presence of advance directives have little effect on the actual decisions that are made on behalf of patients. Most famously, the SUPPORT study —a 4-year study conducted in the 1990s at 5 hospitals that examined the relationship between the presence of advance directives and the end-of-life care received by over 9,100 patients— found that advance directives had no discernible effect on the nature or extent of medical care that patients received at the end of their lives. (Teno, Lynn et al. 1994; SUPPORT 1995; Goodman, Tarnoff et al. 1998) More recent studies have likewise demonstrated that without intensive, community-wide efforts, few people complete advance directives and they have minimal impact on end-of-life care. (Hammes and Rooney 1998; Marchand, Fowler et al. 2006)

As we describe the decision-aid we created, we will explain how we attempted to address these various concerns. But first, there is one additional criticism that sometimes underlies the more specific concerns: that the emphasis on respect for patient autonomy is misguided, and that when individuals are at their most vulnerable, our primary responsibility should be to protect them from harm, not engage in a quixotic quest to discern their exact, authentic wishes regarding healthcare. (Wrigley 2005; Wrigley 2007)

In certain respects, responding to this criticism is fundamental to any defense of advance directives, or advance care planning more generally. Although a detailed defense is beyond the scope of this paper, ample responses have been made by ourselves and others. (Mill 1947; Abram 1982; VanDeVeer 1986; Levi 1999) Our starting point is that respect for autonomy is so central to health care ethics that those who contend that temporary incapacity should diminish its importance need to demonstrate why this might be so. For, despite minority views that individuals lose the capacity to make autonomous decisions simply by being ill, (Komrad 1983) most would not argue that, as a rule, the self-regarding wishes of autonomous patients should be disregarded for paternalistic reasons. But if that is so, then why give less weight to a patient’s wishes simply because they lose the ability to communicate those wishes for a period of time? It would be as if we claimed to respect the wishes of a dinner companion to choose their entrée, but thought it OK to disregard their wishes (and choose a different meal for them) if the waiter came while they were in the restroom and so couldn’t give the order themselves. The accepted standard is that an autonomous individual orders for him- or herself. If this is not possible (or they wish to delegate the responsibility), they tell us what we should order for them. In the absence of these options, we try to figure out what they would have wanted us to order for them. Only if we have no reliable information about their wishes do we order what we think (all things considered) is best for them to have.

There are, of course, limitations to this analogy. Ordering a meal at a restaurant is not the same as choosing whether or not to undergo a life-saving medical intervention. Further, while it may be reasonable to decide what food to order in times of severe hunger, it might not be a good time to make portentous decisions about potentially end-of-life medical care when a person is severely ill. As such, people sometimes talk about advance care planning in terms of “pre-need planning”—which brings us back to the question whether there is a method of advance care planning and documentation that can overcome the challenges mentioned above.

Description of Our Decision-Aid

With this background about the limitations of advance directives in mind, we developed a computer-based decision-aid, Making Your Wishes Known: Planning Your Medical Future, to help individuals with the process of advance care planning. Making Your Wishes Known is an interactive, self-directed computer program that uses a question-answer format involving audio, text, graphics, patient vignettes and videotapes of “professional experts.” This program takes an educational approach, simulating the kind of idealized discussion one might have with an experienced, reflective health-care professional who is well-informed and has ample time. As such, its purposes are to:

1. educate users about the various components of advance directives (including their uses and limitations);

2. help individuals identify, clarify, and prioritize factors that influence their decision-making about future medical conditions;

3. explain several common medical conditions (stroke, coma, dementia, terminal illness) that often require life-sustaining medical treatment;

4. describe and demonstrate several common forms of life-sustaining medical treatment (dialysis, mechanical ventilation, feeding tube, cardio-pulmonary resuscitation) including indications, side effects, and complications;

5. help individuals reflect on the acceptability (to them) of various symptoms (e.g., nausea, pain, shortness of breath) as well as specific disabilities (e.g., inability to ambulate, communicate, live independently, or think clearly);

6. help individuals consider and articulate the relation between a condition’s prognosis and one’s wishes for life-sustaining medical treatment;

7. help individuals choose an appropriate spokesperson and substitute spokesperson(s);

8. help individuals articulate a coherent set of medical wishes in the event they could not communicate in the future;

9. create a tailored advance directive that accurately represents the individual’s views and wishes (with regard to future medical decisions), using a format that is readily and accurately interpretable by physicians;

10. prepare individuals to engage in discussions about their values and wishes with family, friends, and healthcare providers.

Making Your Wishes Known combines multiple strategies, building on the strengths of values histories, (Lambert, Gibson et al. 1990; Doukas and McCullough 1991; Rich 1991) detailed preference instruction, (Emanuel and Emanuel 1989) and durable power of attorney designation. (Orentlicher 1990) Unlike any other advance care planning tool of which we are aware, our program also includes a decision aid based on multi-attribute utility theory (MAUT) (Torrance, Boyle et al. 1982; Jain and Kahn 1995; Torrance, Furlong et al. 1995; Torrance, Feeny et al. 1996) to help translate values and goals into a meaningful, usable advance directive that explicitly reflects an individual’s healthcare wishes and outlines a plan for how they wish to be treated. (see Appendix 1 for a sample advance directive generated by our decision aid)

Because the program is computer-based and can tailor information to the individual, it can be adjusted to accommodate special needs and preferences for visual and/or auditory delivery, as well as pace of presentation. Though voice-over narration is available for virtually all parts of the program, the content can be viewed as text as well, and is written at a reading level of 8^th grade or less. Additionally, the program provides the user with ongoing feedback throughout, can be accessed at an individual's convenience (and done so repeatedly), and is private. (Kahn 1993; Strecher, Greenwood et al. 1999) Finally, because Making Your Wishes Known accomplishes the time-intensive tasks of education, values clarification, and creation of a personalized advance directive, the program can serve as a springboard for patients to have productive, in-depth, and nuanced discussions about their wishes with their doctors and other health professionals.

While no decision-aid is perfect, we believe that our approach addresses some of the core criticisms of standard approaches to advance care planning, as well as various limitations of standard advance directives. We recognize that Making Your Wishes Known is not for everyone, particularly those who are uncomfortable working with computers or unable to access them. That said, computers have been shown to be widely accepted by people regardless of socioeconomic status, educational background, and age –including the elderly. (Clark 2002) Moreover, we believe that Making Your Wishes Known can significantly improve the value of advance care planning for those who have the ability and desire to seriously engage in this process, and that its use can generate a document that accurately reflects the person’s values, goals, and wishes. This belief forms the basis of our response to the critiques leveled at advance directives more generally.

Responses to Critiques

1. There are several responses to the concerns related to predictions/forecasting. First, we contend that with a suitable intervention, motivated individuals can be educated to better understand conditions they have not yet personally experienced. (Dewey 1980) For example, recent research has shown that once people have a more realistic image of what it means to live with advanced Alzheimer’s disease, their views change as to the medical treatment they would want under those circumstances. (Volandes, Lehmann et al. 2007) Such observations do not resolve all concerns over autonomous preference formation, (Christman 1988) and there is clearly much work to be done to ensure that there is alignment between projections and choices and underlying values. But this problem is not unique to advance care planning, and it is generally well-accepted that education about anticipated events not only increases knowledge about what one is getting into, but often leads to greater agreement with the decisions one actually makes (Legare, O'Connor et al. 2003; O'Connor, Legare et al. 2003) —be it buying a car, choosing whether to enter a given profession, or deciding whether to undergo a particular medical treatment. In fact, it is in part this assumption that posits “informed consent” as a mechanism for respecting personal autonomy.

In Making Your Wishes Known, multimedia educational material is provided for common disease states and treatment options, is discussed in terms of multiple perspectives, and is explored in terms of individuals’ own value schemes and priorities. Users are encouraged (but not required) to explore specific medical conditions and treatments in greater detail by clicking on icons labeled “additional information.” Users can learn more about what it feels like to have a given condition (or treatment), what are its consequences and implications, and what options are typically available to patients who experience this particular condition (or treatment). Additionally, the program contains a glossary and links to various internet resources, which they can access at any time.

Making Your Wishes Known also attempts to overcome problems with future forecasting by encouraging users to engage in focused reflection exercises (Ubel and Loewenstein 1997; Nezu 1998; Ubel 2002; Voogt, van Leeuwen et al. 2005; Block 2006) to help dispel preconceptions and/or weigh the anticipated effects of specific disabilities. Such interventions have been shown to help overcome inaccurate forecasting in other contexts. (Ubel, Jepson et al. 2001; Ubel 2006) In our program we ask individuals to think about what life would be like in a wheelchair, or if they suffered from constant pain, or they were forced to deal with the after-effects of a stroke, before making a choice. Making Your Wishes Known prompts users to make decisions only after they have had the opportunity to reflect on what it would mean to live with these conditions.

Even so, a significant number of patients will elect to have their spokesperson’s judgment supersede their own written advance directives, (Sehgal, Galbraith et al. 1992; Stocking, Hougham et al. 2006) and as such, Making Your Wishes Known not only encourages users to name a spokesperson, but prompts them to consider 1) circumstances in which they would want treatment decisions left to their spokesperson’s judgment, and 2) whether they wish their advance directive or their spokesperson to prevail in the event of conflict between the two.

Our program also explicitly recognizes the limitations of attempts to anticipate one’s wishes about future circumstances, stressing to users that advance directives can only go so far. Users are encouraged to revisit the decisions outlined in their advance directive. They are also encouraged to regard their advance directive as a starting point for initiating substantive discussions with loved ones and healthcare providers. Without such a document, detailed and explicit discussions often do not occur. Furthermore, the absence of clearly articulated wishes can lead to disputes about an appropriate medical treatment plan—recall the tragic case of Terry Schiavo. (Annas 2005)

Making decisions in advance inevitably raises concern over authenticity and stability. While some people undoubtedly change their minds over time, the alternative to thinking about and communicating one’s wishes about future medical decisions is the relegation of such decisions to the guesswork of others, which research has shown to be haphazard. Multiple studies substantiate that physicians are poor at predicting patients’ wishes, as are close friends and intimate family members. (Suhl, Simons et al. 1994; Coppola, Ditto et al. 2001; Fried, Bradley et al. 2003; Pruchno, Lemay et al. 2005)

2. In response to concerns over the limited nature of advance directives and the problems associated with communication, several replies are in order. First, while advance directives are typically static documents that may inadequately communicate an individuals’ particular wishes or concerns, a broader goal of advance care planning is to facilitate reflective and dynamic discussions about one’s wishes. (Emanuel, von Gunten et al. 2000; Moskop 2004) Making Your Wishes Known is designed to promote the kind of in-depth dialogue that is at the heart of advance care planning, and toward that end, it promotes respect for patient autonomy. (Levi 1999; Briggs 2004) In pilot testing (paper forthcoming) with 85 patients (49 healthy volunteers, 36 with severe chronic disease), users rated our decision aid quite high in terms of providing information, as well as helping them clarify their wishes and prepare to discuss them with others. (see Table 1)

Table 1.

Satisfaction with Components of Computer Intervention

Satisfaction with the way the computer program:	Mean Score (n=85) 1= Very Dissatisfied 5 = Extremely Satisfied
Increased knowledge about advance care planning	4.4
Provided information about medical conditions	4.2
Provided information about medical treatments	4.2
Helped clarify values	4.1
Helped clarify wishes regarding healthcare	4.2
Helped with feeling greater control over future health care	4.3
Helped with choice of spokesperson	4.1
Helped make important end-of-life decisions	4.3
Helped put wishes into words	4.4
Helped to prepare to discuss wishes with doctor	4.1
Helped to prepare to discuss wishes with family	4.3

Second, the mechanism by which Making Your Wishes Known generates an advance directive document provides a number of checks and balances for accurate expression of an individual’s wishes. At several junctures in the program, users are queried whether they wish to rectify inconsistencies in their answers to various related questions. The multi-attribute utility theory (MAUT) model we use employs an algorithm that tracks and logically integrates an individual’s responses so that the document created is conceptually clear and consistent. Third, individuals are given multiple opportunities within the program to amend their choices and/or edit the general statements ascribed to them—which includes being given opportunities to create free-text additions that will be included in the final document.

In addition to documenting an individual’s specific wishes for particular conditions, our program interprets user’s responses to assign one of six General Wishes statements regarding life-sustaining medical treatment, quality of life, and the importance of various burdens to oneself and others. (see Table 2) Users may accept their assigned statement, edit it, or choose another statement option to accept or adapt. In this way, the final advance directive is intended to help individuals articulate not just specific treatment preferences, but their values (more broadly construed) so that surrogate decision-makers better understand the overarching priorities of the individual when unanticipated clinical circumstances arise. Users are encouraged to view their completed advance directive as a tool for prompting and improving communication, and are cautioned about over-reliance on this document for medical decision-making.

Table 2.

General Wishes Statements

1. I cherish my life regardless of its quality. I would want any and all medical treatments that might prolong my life, even if the result is a quality of life that others regard as very poor. This means that I want all treatments: even if treatment would prolong my life by only hours or days even if their chance of success if very low regardless of the cost of treatment regardless of the burden of treatment on me or others

2. I cherish my life regardless of its quality. I would want all medical treatments that are likely to prolong my life, unless my family and loved ones would consider the burden to them to be unbearable. This means that I want all treatments: even if the result is a quality of life that others regard as very poor even if treatment would prolong my life by only hours or days

3. I cherish my life, so long as my quality of life is acceptable. I want only those medical treatments that are likely to be successful in preserving what I consider a good quality of life. This means that if my quality of life is likely to be poor, I would rather live a shorter period of time than undergo medical treatments that prolong my life. For me, an unacceptably poor quality of life means: (list of conditions/experiences drawn from user’s responses to the program)

4. I cherish my life, so long as my quality of life is acceptable and efforts to prolong it do not impose on my family and loved ones a burden they consider to be unbearable. I want only those medical treatments that would not impose such a burden and are likely to preserve what I consider a good quality of life – even if this means I would live a shorter period of time. For me, an unacceptably poor quality of life means: (list of conditions/experiences drawn from user’s responses to the program)

5. I do not want any medical treatments that would prolong my life, unless the purpose of the treatments is to help other people, such as: to make organ donation possible to use my body for research or education to allow family or friends to say goodbye to me

6. I do not want any medical treatments that would prolong my life, even if the treatments would: return me to my current state of health decrease my discomfort have a high probability of success impose minimal burden on me or on others benefit others (such as organ donation, research or education)

Of course, none of this speaks to the logistical challenges of keeping track of advance directives, accessing them when needed, or separating out the true wishes of the patient, as opposed to surrogates. It is not yet clear to what extent Making Your Wishes Known can ameliorate these difficulties. A study is currently underway that follows individuals with advanced cancer until their deaths, and will compare the end-of-life treatment for those who used our decision-aid with those who received more “standard” advance directive forms.

3. In response to concerns over translating an individual’s advance directive into appropriate medical decision-making, our ongoing research will provide useful information. Our preliminary data about the usefulness of the computer-generated advance directive for guiding physicians’ decisions suggests that Making Your Wishes Known accurately reflects patients’ wishes. In the above-mentioned pilot testing with 85 patients, our decision-aid had a mean score of 8.9 for the question “How accurately did the completed advance directive reflect your wishes for end-of-life care?” (where 1 = Not at all accurate and 10 = Very accurate). These 85 patients were also highly satisfied with the decision-aid’s ability to help them put their wishes into words (8.5, where 1 = Not at all satisfied and 10 = Extremely satisfied).

4. Does this mean that Making Your Wishes Known can succeed where advance directives have failed, by demonstrating an effect on actual clinical outcomes in end-of-life medical care? This is not yet known. But, we believe that to answer this question we must be careful not to define the markers of success too narrowly. If a decision-aid for advance care planning helps individuals come to a better understanding of their goals and wishes; and/or if the process enriches their lives, and/or if it helps them communicate more effectively with their family, friends, and healthcare providers, and/or if it makes their end-of-life experience better than it would have otherwise been, then, advance care planning should be judged a success. The goals of respecting patient autonomy are promoted by helping those involved come to a deeper understanding of what is important to the patient. At the end of the day, an advance directive is just a piece of paper. But an effective program for advance care planning is an opportunity to help people grow, create meaning, and make their lives (and deaths) better.

Conclusions

We invite readers to examine a pilot version of Making Your Wishes Known, and evaluate it for themselves <http://www.makingyourwishesknown.com/>. Users will see that we have tried to strike a balance between in-depth education and practicality. The result is a decision aid that will be too involved for some, and incomplete for others. Our hope, however, is that Making Your Wishes Known will eventually prove to be broadly accessible, clinically useful, and helpful in promoting advance care planning.