Trends in Fundamental Attitudes

In the 2011 Living Well at the End of Life poll of a national sample of U.S. adults, 86 percent of respondents said they thought end-of-life care should be “a top priority for the health care system in this country” (Regence Foundation and National Journal, 2011a, p. 2). More than 90 percent said hospice care should be a top priority, and once the term “palliative care” was explained, 96 percent said it should be a top priority as well.

A Pew Research Center (2013) survey of a nationally representative sample of nearly 2,000 adults provides recent information about Americans' attitudes regarding a number of aspects of end-of-life care, updating information from similar surveys conducted in 1990 and 2005. Since 1990, the proportion of adults who believe that “medical staff should do everything possible to save the life of a patient in all circumstances” has doubled. In part, this increase is due to a much smaller percentage who replied “don't know” to the same question in 2013. However, the percentage who believe “there are at least some circumstances where a patient should be allowed to die” also fell over the period, from 73 to 66 percent.

Nevertheless, many Americans say they would tell their own doctors to stop treatment so they could die:

• if they had an incurable disease and were suffering a great deal of pain (57 percent),
• if they had an incurable disease and were totally dependent on another for care (52 percent), or
• if they had an incurable disease and it were difficult to function in day-to-day life (46 percent).

These percentages have remained about the same since 1990, although at the same time, a growing number of people—35 percent, compared with 28 percent in 1990—say they would tell their doctor to do everything possible to keep them alive even if they were suffering a great deal of pain and there were no possibility of recovery. Again, much of this shift is due to people previously unsure.

On the other hand, 62 percent of those surveyed believe individuals have a moral right to end their own lives if they are suffering great pain with no hope of improvement—a 7 percent increase from 1990. Substantial numbers also believe people have the right to commit suicide if they have an incurable disease (56 percent), when they are ready to die and living has become a burden (38 percent), or when their care is an extremely heavy burden on the family (32 percent). In this survey, the public was equally divided on the question of whether physicians should be allowed to prescribe a lethal drug to assist a person seeking suicide.

Although the majority of black Protestants (61 percent) and Hispanic Catholics (57 percent) would ask their doctors to “do everything possible” to keep them alive even if they were suffering great pain and had no hope of improvement, large numbers of both groups (39 percent and 43 percent, respectively) feel differently or have not decided. This finding underscores the point made in Chapter 3 that clinicians cannot make assumptions about patients' beliefs and preferences based on race, ethnicity, religion, or culture. They must ask.

Understanding of Care Choices

Expression of Individual Preferences

The 2013 Pew survey referenced above found that 37 percent of Americans say they have given a great deal of thought and 35 percent say they have given some thought to their preferences regarding medical treatment near the end of life; among those aged 75 and older, one-quarter have given the issue little or no thought. Among people who say they have given the issue considerable thought, most (88 percent) have captured their preferences either in writing or in conversation with others. Americans who are older, white, and have more education or higher incomes are more likely to have put their wishes in writing. (Chapter 3 reviews this literature in more detail.)

Nevertheless, among Americans aged 65 and older, more report talking with their children about what to do with family belongings (76 percent) than about how to handle their medical care if they can no longer make their own decisions (63 percent) or can no longer live independently (55 percent). Even fewer adult children of older parents say the discussion about medical care decision making has occurred (57 percent) (Pew Research Center, 2009).

If people find the topic of care preferences difficult to discuss with family members, discussion with their health care providers appears to be no easier. Research conducted in the U.S. population in general (not specifically among those with serious advanced illnesses) has found that most (8 in 10) want their clinician to listen to them and to have the full truth about their diagnosis; yet only 6 in 10 say that their provider listens to them, and fewer than half say their provider asks about their goals and concerns for their health and health care (Alston et al., 2012). And although the cost of health care is of concern to patients with serious illnesses, few physicians (16 percent) report having any education or training regarding financial issues and how to discuss them (Regence Foundation and National Journal, 2011b).

According to Alston and colleagues (2012, p. 7), “Unsurprisingly, 30 percent of people said they ‘very often’ get health information from a source other than their health care provider. The most common sources were their spouse or partner (15 percent), the Internet (9 percent), and a friend or family member who works in health care (6 percent).” Likewise, with respect to palliative and end-of-life care in particular, people said they received most of their information from family members and friends (49 percent) (Regence Foundation and National Journal, 2011a). In this study, only about one-third of respondents said they received such information most often from their doctor or health care provider or the news media, and 45 percent said they received very little or no such information from their physician. When asked how much they trusted the information they received, respondents gave high ratings to information that came from doctors or other health care providers (76 percent) and from family and friends (69 percent). The information that came from the news media received high trust ratings from only 17 percent of respondents.

Surveys indicate that most people (70-80 percent) want a patient experience that includes deep engagement in shared decision making; however, a substantial gap exists between what people want and what they receive. Those who do become more engaged report a better experience (Alston et al., 2012). Written public testimony gathered for this study through an online questionnaire (see Appendix C) indicates the often poor communication patients and families have with clinicians. People feel that explanations are rushed, issues are not explained, choices are not understood, and clinicians do not listen. Good communication, by contrast, is greatly appreciated. The way to establish good communication, one caregiver said, is to “ask patients and families what they want,” which is a message of Chapter 3.

Experts in health care communication believe a combination of three elements—clinician expertise, patient and family goals and concerns, and medical evidence²—is necessary for truly informed health care decisions, and in general, Americans strongly value all three (Alston et al., 2012). What is needed is to mobilize that general support in the specific context of advanced illness. Death and dying can be a difficult, emotional issue for both public engagement and private discussion. For most people, until a family member is actually facing a serious illness, interest is just too low, the psychological barriers are too high, and preoccupation with the demands of daily life is a ready excuse not to engage. At least in the short term, that reluctance must be acknowledged and societal means found to help people understand that, in most instances, good information is available when they want it.

Public education and engagement efforts should aim to normalize difficult conversations and help people from diverse communities acquire the information and skills needed to participate meaningfully in those conversations. As a result, more people might obtain the care they want and need as they near the end of life.

Sponsorship

Especially with topics as sensitive as advance care planning and end-of-life care, the choice of a credible and trustworthy entity to sponsor a public education and engagement effort is critical. In many communities, coalitions of organizations have come together to sponsor a project, bringing in more people and providing assurance that the effort is broad based.

As reviewed in Chapter 1, a great many stakeholder groups have a professional or civic interest in end-of-life issues. They include health and social services professionals, clergy, volunteers, and others who provide direct care and counseling; those who manage health care institutions and programs, run public and private insurance programs, and advocate for better care; state and federal policy makers; and business executives and union leaders. Health care systems and voluntary health organizations may be able to deliver credible messages, but sponsorship and participation in coalitions are often broader than that. Other organizations—such as labor unions, religious organizations, public health agencies, and insurers—interested in the health and welfare of their members or the community also may become interested. A sponsor or coalition partner can be as large as a national mental health organization or as small as one of its local chapters, a national religious body or an individual congregation.

Excellus BlueCross BlueShield, serving Upstate New York, has supported a Community-Wide End-of-Life/Palliative Care Initiative since 2001 that engages a broad array of professionals, consumers, and other collaborators from diverse backgrounds. This collaboration has achieved heightened awareness of the value of advance care planning and greatly increased the percentage of people who have completed health care proxies (through its Community Conversations on Compassionate Care Program); has implemented Medical Orders for Life-Sustaining Treatment (MOLST) (New York's Physician Orders for Life-Sustaining Treatment [POLST] paradigm initiative) and eMOLST programs, described in Chapter 3; has led regional and statewide discussions and encouraged improved policies related to high-quality palliative and end-of-life care, including pain management and use of feeding tubes; has engaged in additional community-wide education efforts; and has conducted a number of surveys to establish baseline data and determine progress toward measurable goals (CCCC, 2014).

Audiences

Different end-of-life messages are relevant to different audiences, and there are various ways to segment audiences for the purpose of crafting messages with maximum appeal. The general rule of thumb in audience segmentation is to have relevant sameness within groups and relevant differences between groups (Andreason, 1995; Noar, 2006). Policy makers and health care leaders might respond to one message, whereas members of racial and ethnic minority groups might be interested in another; likewise, older and younger adults would likely respond to different messages.

Clinician audiences are a vital complement to public audiences. Opportunities for engagement exist within the clinical community. For example, clinicians who believe health professions schools should do a better job of teaching about palliative care might work through a professional group—and possibly even attempt to engage segments of the public—to advocate for increased palliative care training. Likewise, senior centers, employer groups, and other entities might want to tie messages for caregivers to programs aimed at healthy living, retiree benefits, and so on.

Audiences defined as the “general public” or “all Americans 65 and older” are usually too diffuse to be maximally useful in campaign planning. The more carefully a market can be segmented into different groups, with different messages and appropriate media for each, the more likely the members of those groups are to respond as desired (Andreasen, 1995; Grier and Bryant, 2005; Walsh et al., 1993). Given the usual limited funding for educational and engagement efforts, audience segments can be prioritized by importance or likelihood of response. Thus, “segmentation can help managers achieve both efficiency and effectiveness” (Andreasen, 1995, p. 177).

Audiences can be segmented by socioeconomic strata, by role (e.g., caregivers, clergy), by their involvement with the issue, and in many other ways. For example, research has shown that people who have had a recent hospitalization or who have been a caregiver for someone who recently died are particularly receptive to engaging in advance care planning (Carr and Khodyakov, 2007). They would be a natural audience for messages on that topic.

Timing may be important in identifying a target audience. The point at which a serious illness is diagnosed may be the most critical period of attention and focus for individuals who are ill and their families, and the time when they are most in need of useful information. People in this audience may be seeking relatively in-depth information about what to expect as their illness progresses and how to respond to increasing care requirements.

An additional consideration is whether and how to target by behavior. People who actively seek out health information—for example, on the Internet—might constitute a discrete audience segment. Or people may reach a stage of behavior change that prompts them to seek information (the “stages of change” decision-making model is described in Chapter 3). Even within audience groups, there may be significant subaudiences. Online information seekers may be inclined to view their doctors as collaborators to whom they can bring relevant health care information for review and follow-up, or they may make use of the information independently. Making good use of online information sources requires that people “have skills to effectively seek out the desired information, evaluate it, and then apply the information they find toward solving their health problems” (IOM, 2009, p. 10). Almost 60 percent of the very large number of Internet users who seek out online health information say they have used it in making health decisions, and almost 40 percent say they have used it to change the way they manage a chronic condition or pain (Fox, 2006). Nevertheless, information provided in Chapter 3 about the low level of health literacy raises some question as to whether people can use the information they find in a way most useful to them. By contrast, people who have difficulty finding answers to their health questions or understanding complex issues may prefer to rely on their physicians to provide them with information. Although both groups are active information seekers, they differ in where they look for information, the way they relate to physicians, and their demographic characteristics.⁶

Messages

Not until the target audiences have been chosen and some research on their current views on relevant themes has been conducted can the work of crafting specific messages begin. A number of past public engagement campaigns around end-of-life issues have focused on the concept of consumer/patient “control of your destiny” and the ability to “make your own decisions on your own terms” (for example, the public engagement campaign that accompanied Bill Moyers' public television series On Our Own Terms in 2000). Research on the efficacy and outcomes of this messaging has been insufficient to allow assessment of audience response to this specific approach. Message development often proceeds in stages, beginning with in-depth personal interviews or focus groups, to gain a deeper understanding of the views held on a subject by members of a target audience. Those insights can be part of the basis for survey questionnaires administered to larger groups to learn what aspects of an issue are most meaningful. Survey responses can be either programmed (forced choice) or open-ended. Open-ended comments are another rich source of opinion data.

Several recent end-of-life campaigns—aided by the ability to make resources available via the Web—are providing much more in the way of supporting materials to back up their basic messages (see Box 6-1 and Table 6-1). An example is The Conversation Project, discussed in Chapter 3, whose basic theme is that people need to talk about the important but difficult topic of end-of-life care. Its message is the provocative “Have you had the conversation?” It supports that message by providing a Conversation Starter Kit and personal stories that can help normalize the discussion of the topic (Bisognano and Goodman, 2013). Lately, The Conversation Project has extended its message to emphasize the importance of conversations with clinicians as well.

Channels

Healthy People 2020, the federal government's initiative to promote disease prevention and health promotion goals for the nation, acknowledges the important role of communications media in forming public views on health and disease. According to the U.S. Department of Health and Human Services (HHS, 2013), “Health communication and health information technology (IT) are central to health care, public health, and the way our society views health.” Information on health and medical care is widely available through traditional print and broadcast outlets, as well as through the Internet and social media, with more than 70 percent of Americans using the Internet to acquire health information (Fox and Duggan, 2013). Typical mass media campaigns are only one way to reach and influence key target audiences on end-of-life topics. Communicating through intermediaries, such as faith communities, health care providers, and consumer affinity groups, is another approach. The Robert Wood Johnson Foundation's 10-year Last Acts campaign was built on a model of involving trusted organizations as “message carriers,” and by its close, organizations of all sizes were participating and sharing information with their constituents (Patrizi et al., 2011). Social media can be useful channels as well, especially when they engage people who feel connected to the message source. Offering high-quality websites at moments of readiness is another potentially useful strategy.

Entertainment television has not been overlooked as a vehicle for carrying health-related messages (Singhal and Rogers, 2004), including messages about death and dying. Both long story arcs—such as the illness and death of Dr. Mark Greene at the end of E.R.'s eighth season—and specific episodes of a program (e.g., N.Y.P.D. Blue, Gideon's Crossing) have portrayed dilemmas and decisions that arise at the end of life. Some audience members who would never watch a documentary or attend a lecture on the topic are thereby exposed to valuable information (and sometimes misinformation). Social modeling theories suggest that audiences learn from fictional characters with whom they identify, making them more likely to emulate behavior that has positive outcomes and avoid behavior that has negative outcomes (Singhal and Rogers, 2004).

Evaluation

Evaluation is essential for campaigns to assess progress, make course corrections, and achieve meaningful results. Finding campaigns that have been well evaluated is difficult, however, and many of those that have been evaluated tackle topics that are so different from end-of-life care that comparisons may be elusive. The 2013 Pew Research Center survey cited earlier provides baseline data on consumer awareness, attitudes, and behavior regarding a number of issues related to advanced illness and end of life. Such national baseline information is a good starting point and with some refinement could be used to gauge the effectiveness of public awareness and engagement initiatives at the state or local level. Organizations that sponsor end-of-life public education and engagement initiatives might be able to adopt at least some of these tracking survey questions for their projects for both program improvement and accountability purposes, as well as the ability to compare their project's results with national trends. See Annex 6-1 at the end of this chapter for brief descriptions of selected health-related public information and engagement campaigns and their results.

Physician-Assisted Suicide

Supporters refer to the ethical principle of autonomy in advocating state laws and policies to allow physician-assisted suicide. However, many clinicians and others believe the practice violates a different fundamental principle: “Do no harm.”

Opposition to public policy support for physician-assisted suicide goes beyond religious objections. Allowing people “a choice” of whether to end their lives may be fraught with opportunities for coercion and disruption of patients' trust absent vigorous attempts to ensure that all Americans have access to high-quality care that would meet complex needs near the end of life, as well as systemic changes that would lessen burdens (financial and otherwise) on family members. As the American Geriatrics Society (AGS) stated in its 1996 Supreme Court amicus brief in the case of Vacco v. Quill, “The health care available to the terminally ill may be the most important factor influencing the care provided and may result in requests for [physician-assisted suicide] that could be avoided if appropriate care were available… . The AGS has been opposed to legalizing [physician-assisted suicide] or physician involvement in euthanasia, primarily on the grounds that our frail elderly patients are especially vulnerable to social coercion and that the well-being of those who are old and sick is not being carefully considered” (Lynn et al., 1997, pp. 497-499).

In 1997, the U.S. Supreme Court ruled that assisted suicide is not a constitutionally protected right, although it did not bar states from formulating their own statutes to address it, and five now allow it under state law or court authorization.⁷ Legislatures in Oregon, Vermont, and Washington have enacted laws that permit state residents to end their lives voluntarily with a lethal dose of medication prescribed by a physician if they are “terminally ill” (Oregon), have a “terminal condition” (Vermont), or have a life expectancy of less than 6 months (Washington).⁸ During 2013, similar legislation was introduced in at least six states, and proposals to specifically outlaw the practice were introduced in two states. In Montana, legislation was proposed on both sides of the question, and the state Supreme Court ruled that physicians who help a person end his or her own life voluntarily will not be subject to trial for homicide.⁹ In early 2014, a New Mexico judge ruled that terminally ill, mentally competent patients have the right to “aid in dying” under the state constitution.¹⁰ If upheld, that decision may apply statewide. Continuing efforts to revise state laws guarantee that this controversial issue will be recurrent, at least locally if not nationally (Eckholm, 2014).

Withholding and Withdrawal of Life Support

Over the years, the issue of withholding and withdrawal (especially) of life support has stimulated contentious public debates and led to numerous changes in law and policy. These debates have frequently focused on women or minorities (Holloway, 2011). Many of their cases have prompted unprecedented public intervention in what would ordinarily be considered private family decisions.

The most widely publicized, and protracted, legal cases centered around three young women who left no advance directives—Karen Ann Quinlan, Nancy Beth Cruzan, and Teresa Marie Schiavo. Their sudden, unexpected, and permanent unconsciousness (from different causes) ignited strident public debates.¹¹ In each instance, the courts eventually decided that treatment could be withdrawn, but years of legal wrangling devastated family members and health care team members. According to Holloway (2011, p. 147), “When the body belongs to a woman or a member of a racial or ethnic minority, the privacy [that legal] protections ordinarily grant is already at risk.” The notoriety of such cases may prompt people to consider their own end-of-life preferences.¹²

As of early 2014, two cases involving clashes over cessation of treatment of individuals declared brain dead again received considerable media attention. One in California involved an African American child whose family wanted her to continue receiving treatment (Debolt, 2014). The other was a Latino woman whose family and husband wanted life support withdrawn, but because she was pregnant, Texas state law was believed to prohibit it (Hellerman et al., 2014). In the latter case, after 2 months of public discussion, a judge ruled the Texas law was being misapplied and ordered an end to treatment. As of June 2014, the child in California had been kept alive on life support for 6 months after being declared brain dead (Debolt, 2014). As technology improves the ability to keep body systems functioning even without mental function, more such cases can be expected.

Distributive Justice and Futile Care

As discussed in previous chapters, when patients and families have a clear understanding of the course of a terminal illness, the consequences and experience of cure-oriented versus palliative care, and the quality of life each produces, most choose the palliative approach. A minority of patients nevertheless do want aggressive care, and at present, they usually receive it. As a result, some question whether patients should be entitled to medical interventions that hold no realistic promise of extending life or improving or maintaining quality of life. Because many critical care services are expensive, is this the best way to spend health care dollars?

The principle of autonomy would suggest that individuals should receive whatever services they want. However, the amount of money society has available to spend on health care is not infinite. There are trade-offs and opportunity costs, as spending on truly futile health care services may deprive other people of important benefits, and services that provide the patient no benefit may be thought of as too expensive regardless of their cost (Meisel, 2008).

Other countries have faced this issue and come to different conclusions, while U.S. policy makers have strictly avoided any measures—such as basing reimbursement policy on comparative effectiveness assessments—that might have the effect of limiting care (Satvat and Leight, 2011). Ironically, economists and policy analysts readily acknowledge that allocation decisions are actually common in the United States, based mainly on ability to pay; they are merely implicit and hidden, rather than explicit, transparent, and potentially more fair (Lauridsen et al., 2007; Swanson, 2009).

Tensions surrounding this issue might be alleviated if research continued to show that receiving more medical interventions near the end of life does not produce better outcomes in terms of longevity or quality of life, as other chapters of this report have documented. The nation can no longer pretend that there is no upper limit on what it can afford, and the debate will continue to be divisive if concern about health care costs results in some version of a cap on spending.

Making Dying Visible

Given Americans' acknowledged reluctance to discuss dying, an unlikely controversy arose in early 2014 about the use of social media to discuss the consequences of a serious illness. Lisa Bonchek Adams is an active user of social media—Facebook, Twitter, and her blog¹³—which she uses to talk about her metastatic breast cancer and coping with illness and grief and to urge her online followers to have cancer screenings. Although her illness is advanced and likely will eventually be the cause of her death, she was not dying at the time the controversy erupted.

Adams' use of social media to discuss her disease was questioned in separate opinion pieces in both The Guardian and The New York Times by writers who found that her public discussion of her illness was unseemly (“dying out loud”) and who appeared to counsel passive acceptance as a more humane (and cheaper) alternative (Keller, 2014). People who follow Adams' posts, and Adams herself, were quick to point out the many inaccuracies and misinterpretations of her intent and her own personal choices (Elliott, 2014; Tufekci, 2014), and a widely publicized backlash ensued that resulted in useful media soul searching, namely around using “one woman's story as an occasion for debate about what might be wrong with broader approaches to dying” (O'Rourke, 2014). Following this public outcry, The Guardian removed the controversial article from its website (Elliott, 2014) The controversy highlights Americans' ambivalence about talking about dying, the sensitivity with which social justice principles need to be applied, and the dangers of making facile judgments about the care and treatment choices made by others. From a media analysis point of view, it shows evidence of the ability of story and of social media to engage people in a meaningful way if trust and respect are built over time, as Adams had done over a period of years.

The 2009 Controversy Over “Death Panels”

A significant setback to more effective advance care planning occurred when Section 1233 of a House bill (HR 3200, 111th Cong.) that led to the Patient Protection and Affordable Care Act of 2010 was withdrawn after false and misleading statements that it would establish “death panels.” The provision would have reimbursed clinicians for the time spent in advance care planning with patients. Such conversations would have included discussion of the documents that can help ensure that patients' wishes regarding care are followed in the event they become unable to express them.¹⁴

Too often, clinicians, patients, and families embark on a treatment journey for a serious disease without important information and understanding of the illness and its likely course, and at times with a conscious effort to protect another from the truth about these matters (Kumar and Temel, 2013; Piemonte and Hermer, 2013). Without understanding the likely course of illness and the risks and benefits of treatment choices, patients (and families) cannot make informed decisions about care (Hajizadeh et al., 2013; Weeks et al., 2012). By contrast, a good advance care planning process gives people “a way to think about death and dying”; for some people, that discussion can allow them to confront dying directly instead of its being a “vague, unmanageable concept” (Martin et al., 1999, p. 88).

At least an hour is usually needed to explore such topics thoroughly (Briggs et al., 2004; Detering et al., 2010). According to the National Center for Health Statistics (undated), in 2010, the average U.S. physician visit lasted only 21 minutes, and more than half of visits lasted 15 minutes or less. Lack of time and lack of payment make clinicians even less likely to have difficult conversations they may be reluctant to have in the first place.

The allegation of “death panels” was first made in 2009. Although no aspect of Section 1233 bore any resemblance to the accusation, it was repeated so often that many people came to regard it as truth. Public education efforts by medical and public health authorities were ineffective in countering this misinformation. The Kaiser Family Foundation's Health Tracking Poll of March 2012—2 years after passage of the Affordable Care Act—revealed that 36 percent of Americans erroneously believed the law actually contains a provision to “allow a government panel to make decisions about end-of-life care for people on Medicare,” and 20 percent responded “don't know” (KFF, 2012). When that polling question was repeated in 2013, 40 percent said they believed the law contains this provision, and 21 percent did not know (KFF, 2013). Thus, 3 years after the act's passage, 60 percent of Americans either believed or were unsure whether “death panels” are law.

Section 1233 of HR 3200 (111th Cong.) was strongly opposed by conservative protestors. House Minority Leader John Boehner and Representative Thaddeus McCotten (R-MI) said the legislation “may start us down a treacherous path toward government-encouraged euthanasia” (Boehner and McCotter, 2009; Pear and Herszenhorn, 2009)—opposition that was “startling because the need for such legislation had been recognized by both political parties for some time” (Altman and Shachtman, 2011, p. 282). In the months before the introduction of HR 3200 with Section 1233, two bills were introduced in the House and one in the Senate relating to orders for life-sustaining treatment and advance care planning, two with bipartisan support.¹⁵

Recent polls conducted with the American public also reveal strong support for advance care planning. In one 2011 poll,

• 97 percent agreed or strongly agreed that “it is important that patients and their families be educated about palliative care and end-of-life care options available to them along with curative treatment”;
• 86 percent thought these discussions should be fully covered by health insurance; and
• 81 percent thought they should be fully covered by Medicare (Regence Foundation and National Journal, 2011a).

Similarly, when the California HealthCare Foundation polled 1,669 adult Californians (including 393 people who had lost a loved one in the previous year) about their attitudes toward end-of-life topics, respondents were asked, “One idea is to have insurance plans cover a doctor's time to talk with patients about treatment options towards the end of life. Do you think this is a good idea or a bad idea?” Eighty-one percent of respondents thought it was a “very good” or “somewhat good” idea (CHCF, 2012).¹⁶ Although the question was worded neutrally, this level of support was significant—especially because public opinion research indicates that 43 percent of American consumers do not trust their health insurance plan (PRG, 2012).

So why did the important effort represented by Section 1233 fail? Leaving aside the fear that underlies many Americans' unwillingness to contemplate mortality, concerns about paying for advance care planning were exacerbated by opposition to comparative effectiveness research. Opponents claim that such research will prevent Americans from obtaining treatments they want and lead to rationing based on an external view of their likely effectiveness or excessive expense. While other countries do factor cost comparisons into decisions on which services to reimburse with public monies (Satvat and Leight, 2011), the U.S. approach explicitly prohibits any consideration of treatment costs (Altman and Shachtman, 2011).

The objections to facilitating advance care planning may be misplaced when, as discussed in Chapter 3, some evidence suggests that people who do plan for their care most often choose less aggressive care. This choice has been associated with increased survival, better quality of life, and decreased stress and psychological impacts on family members (Mack et al., 2010; Temel et al., 2010; Wright et al., 2008). No association has been found between having an advance directive discussion or document and earlier death (Fischer et al., 2010). Indeed, according to Winter and Parks (2012, p. 741), “Ironically, we found that those who avoid living wills and end-of-life conversations are the least likely to have treatment wishes respected, because their proxies are unlikely to know their wishes.”

Improving the care people receive at the end of life by giving them the care they actually want while saving money by avoiding costly and futile interventions people do not want could have been a strategy with multiple benefits for the health system. But the discussion of these two goals—individual care and collective savings—in the same public conversation may help explain why the proposed provision “became the lightning rod it did” (Kaebnick, 2013, p. 2). Accordingly, some commentators would keep the financial argument completely out of the advance care planning discussion (Fried and Drickamer, 2010). As a Washington Post opinion piece said, because health care reform was promoted in large part on the grounds that it would control rising and unsustainable health care costs (Antos et al., 2009), in that context, “citizens are not delusional to conclude that the goal [of Section 1233] is to reduce end-of-life spending” (Robinson, 2009).

In early January 2011, the Centers for Medicare & Medicaid Services (CMS) withdrew “voluntary advance care planning” as a specified element of the Medicare annual wellness visit (HHS, 2011). This decision was due to the discordant views of stakeholders, “including those who disagreed when the idea of voluntary advance care planning was first proposed under the Patient Protection and Affordable Care Act” (Holley, 2011). The CMS final rule acknowledged as much:

It has since become apparent that we did not have an opportunity to consider prior to the issuance of the final rule the wide range of views on this subject held by a broad range of stakeholders (including members of Congress and those who were involved with this provision during the debate on the Affordable Care Act). Therefore, we are rescinding the provision of the final rule that includes voluntary advance care planning as a specified element of the annual wellness visits providing personalized prevention plan services, and returning to the policy that was proposed, which was limited to the elements specified in the Act. (HHS, 2011)

In August 2013, the bipartisan team of Senators Mark Warner of Virginia and Johnny Isakson of Georgia again pursued the goal of reimbursement for advance care planning discussions by introducing the Care Planning Act of 2013.¹⁷ According to Warner, the bill “is about honoring a patient's choice, not making it for them” (Mundy, 2013). Nevertheless, the staying power of the distortions around “death panels” may have doomed the senators' initiative.

As Piemonte and Hermer (2013, p. 24) advise, “If we are ever to make progress toward creating policy that incentivizes physicians to engage in constructive end of life conversations, we need to do so in a way that appeals to the shared values of those across the political spectrum.” Recent public opinion research suggests there is ample support for advance care planning and insurance that covers it. Targeted public education and engagement efforts may move the needle on these policy initiatives.