|Home | About | Journals | Submit | Contact Us | Français|
KB: Acquisition of data, analysis and interpretation of data, final approval of manuscript.
LI: Acquisition of subjects and data, interpretation of data, final approval of manuscript.
JRO: Analysis and interpretation of data, final approval of manuscript
To explore whether models of health behavior change can help to inform interventions for ACP.
Qualitative cross-sectional study.
Sixty-three community dwelling persons age ≥ 65 years and 30 caregivers with experience as surrogate decision-makers.
In focus groups conducted separately with older persons and with caregivers, participants were asked to discuss ways they had planned for future declines in health and why they had or had not engaged in such planning. Transcripts were analyzed using grounded theory.
Four themes illustrated the potential of applying models of health behavior change to improve ACP. 1) Participants demonstrated variable readiness to engage in ACP and could be in different stages of readiness for different components of ACP, including consideration of treatment goals, completion of advance directives, and communication with families and physicians. 2) Participants identified a wide range of benefits of and barriers to ACP. 3) Participants used a variety of processes of change to progress through stages of readiness, and ACP was only one of a broader set of behaviors that participants engaged in to prepare for declines in their health or for death. 4) Experience with healthcare decision-making for loved ones was a strong influence on perceptions of susceptibility and engagement in ACP.
The variability in participants' readiness, barriers and benefits, perceptions of susceptibility, and use of processes to increase readiness_for participating in each component of ACP suggests the utility of tailored, stage-specific interventions based on individualized assessments to improve ACP.
Although advance directives (ADs) have some important limitations,1, 2 they are the cornerstone of advance care planning (ACP), the process by which individuals can ensure that their preferences guide the care they receive during times of decisional incapacity.3 ADs include instructional directives, health care proxy designations, and living wills. There is growing recognition that the completion of documents alone is insufficient to ensure that patients receive care consistent with their values and preferences.4-7 Additional components critical to ACP are the consideration and articulation of goals of care and the facilitation of communication between individuals and their loved ones, in order to promote a shared understanding of these goals.
Both ADs specifically and advance care planning more generally are underutilized.3 A recent review of studies designed to improve the rates of completion of advance directives in the ambulatory setting found that these interventions had only modest effects.8 The authors of this review suggest that interventions consider patients' readiness to engage in advance directive discussions according to stages of behavior change, and it has been suggested that health behavior models hold the promise of providing a useful framework in which to study ACP.9 These models have been used as the theoretical grounding for successful strategies based on individualized patient assessments and tailored interventions to increase the rates of a range of health behaviors.10-13
Although there are many different models of health behaviors, including the Health Belief Model,14 Theory of Reasoned Action/Planned Behavior,15 Social Cognitive Theory,16 and The Transtheoretical Model (TTM),17 they all share the goal of explaining why people engage in a given behavior, and they share an overlapping set of constructs.18 Several of the key shared constructs that are particularly relevant for ACP include perceived susceptibility, the belief that one is vulnerable to developing the condition the behavior will protect against; self-efficacy; and the barriers to and benefits of changing one's behavior.18 The TTM includes a temporal component, represented by stages of change, which include precontemplation (no intention to change behavior in the near future), contemplation (thinking about changing behavior in the near future), preparation (commitment to changing behavior soon), action (a recent change in behavior); and maintenance (ongoing behavior change). The TTM also includes the concept of processes of change, strategies used to increase readiness for participation.19
There is a growing effort to apply health behavior models to ACP. Conceptual work includes a tool to stage persons regarding their readiness for participation20 and suggestions for what kinds of interventions are most suitable for persons at different stages.21 Empirical work includes an observational study proposing a stages-of-change model for ACP22 and an intervention study incorporating the concept of stages of change that successfully increased advance directive completion.23 However, there has been little study of ACP directly examining individuals' readiness to change or examining behavior change constructs other than readiness to change and barriers to participation. The purpose of this study was to explore ACP as a health behavior, exploring a full range of health behavior constructs, by asking participants to describe their experiences with ACP. The goal was to determine_whether the constructs of these models are meaningful to the process of ACP and, if so, to provide new insights into how older persons can be helped to complete a full process of clarifying their goals, communicate these goals to loved ones, and complete advance directives.
Two groups participated in the study: a group of older persons (≥ 65 years of age) and a group of caregivers for older persons. We focused on older persons because the majority (77%) have 2 or more chronic conditions24 and therefore have experience with decision-making. The caregivers were persons who responded “yes” to the question of whether they had faced treatment decisions, but who were unrelated to the older persons included in the study. We selected this group to explore the concept of susceptibility among persons who had directly experienced the need to participate in healthcare decision making on behalf of a loved one. We excluded persons who were non-English speaking, were severely hard of hearing, or who had cognitive impairment, defined as the inability to remember ≥ 2 items on a 3-item test of short-term recall.25 Older persons were recruited from three senior centers and the outpatient clinics of a VA hospital, selected to promote purposeful sampling by providing access to participants of diverse ethnic/racial and socioeconomic status. Caregivers were recruited from these sites as well as bereavement groups at a community hospital and funeral home. Participants were recruited by one of the investigators providing a brief introduction to the study at a group being conducted at the recruitment site (e.g. meals or recreational activities at senior centers, education or support groups at clinics, hospital, funeral home) and then approaching group members for participation.
Focus groups were conducted separately for older persons and for caregivers by trained moderators at the site of recruitment. The use of focus groups allowed participants to reflect upon and respond to the ideas offered by other members of the group. In addition, the focus group methodology encouraged participation from those who may have been reluctant to be interviewed on their own,26, 27 which was particularly important given that the discussion included potentially difficult topics. The groups were conducted with the use of a discussion guide, consisting of open-ended questions. The guide first asked participants how they had planned or thought about planning for their future health and future care. Subsequent questions asked participants their understanding of advance directives, whether and why they had or had not completed them, and the real or perceived consequences of completing them. The next set of questions asked participants to participate in an exercise in which they considered their broader goals of care by asking their thoughts about trade-offs between treatment burdens and outcomes and between desirable and undesirable outcomes in the context of uncertainty.28 They were asked about their willingness to undergo treatment given different likelihoods of a return to current health versus dying or experiencing severe physical or cognitive disability. The final set of questions asked participants about their communication regarding advance directives and care goals with family members and physicians. Caregivers were asked to consider these questions from the perspective of having been a surrogate decision-maker, and then asked how these experiences had influenced their own thoughts and behaviors regarding ACP for themselves. The questions included in the discussion guide for older persons are provided in the Appendix. Prior to starting the focus group, participants completed a questionnaire regarding sociodemographic and health status.
Focus groups were audiotaped and transcribed by an experienced medical transcriptionist. Using the constant comparative method, preliminary analysis took place simultaneously with data collection. This allowed the discussion guide to be modified in order to explore concepts arising in earlier groups in greater depth.29 Participants were enrolled until the point of theoretical saturation was reached. The transcripts were analyzed according the methods of Strauss and Corbin.30 Utilizing multiple close readings, small segments of the transcripts were coded into discrete concepts by two of the investigators (TRF and KB) independently. A coding book was developed and revised until the investigators reached >80% agreement on the application of codes. Initial codes emerged directly from the data. These codes were then reviewed from the perspective of the health behavior constructs to determine whether additional codes were necessary to address the question of the applicability of these models to ACP. The investigators then compared codes within and across interviews in order to elucidate the larger themes that emerged. Although the original intent was to analyze the older person and caregiver groups separately, the process of analysis revealed that many of the older persons spoke from the perspective of having been a caregiver themselves. As a consequence, the themes arising from each of these two groups overlapped sufficiently to permit a combined analysis.
The results of the focus groups revealed several themes related to how and why participants engage or do not engage in the process of ACP. These themes included: 1) variable readiness to participate in ACP, 2) the wide range of benefits of and barriers to participating in ACP, 3) the variety of processes used by participants to increase their readiness for ACP participation and the broader spectrum of behaviors associated with planning for future declines in health, and 4) the influence of experiences with loved ones on perceptions of susceptibility and engagement in ACP.
The participants demonstrated variable readiness to participate in ACP, corresponding to the concept of “stages of change.” One caregiver, reflecting on why she had not completed ADs, recognized that readiness was a necessary precursor to action, commenting, “You have to be ready. I don't know what makes you ready.” Some participants were clearly not ready to participate in the process of ACP, and could therefore be considered to be in the “pre-contemplation” stage:
I'm going to tell you I don't want to talk about it. I don't want to discuss it. I don't want to hear that morbid talk.
Others commented that, based on their participation in the group, they were ready to start doing some planning. Asked whether she might complete AD forms in the future, one participant stated:
After listening to you talk today, I think I will. I got some good points just listening.
One participant spoke of his intentions to talk to his family about his wishes for his end of life care, saying:
I now feel, and I made this decision myself, that it is now time to tell them the truth…I've made up my mind, and I will talk to them about it.
Another participant discussed having considered the specific actions she needed to take, consistent with being in the “preparation” stage:
I have been thinking about this lately, that I need to get [a health care proxy]. I wake up at night thinking about a few things I need to get straightened out before something happens, and it's “he didn't tell me this or that.” I want it in black and white. I am going to take care of that.
Participants who had completed living wills or health care proxies or who had discussed their attitudes toward life-sustaining treatment with their loved ones could be considered as being in the “action” stage. One participant stated, “We review our [ADs] every couple of years,” indicating that he and his wife were in the “maintenance” stage.
Participants could be at different stages of readiness for the different activities involved in ACP. For example, some participants had completed a living will but had not discussed its contents with their loved ones or provided the form to their physicians. Notably, although virtually all groups contained participants who had engaged in completing advance directives and communicating their preferences to their loved ones, participants had indicated their preferences only in circumstances in which it would be clear that the patient was dying, without hope for recovery. Only a single participant had thought about future scenarios of health and decision-making in which the likelihood of recovery was not known with certainty.
Participants discussed a wide range of reasons for engaging or not engaging in ACP. The benefits of doing ACP, along with representative quotes, are provided in Table 3, and the barriers in Table 4. The benefits of doing ACP, tended to be general ones that were applicable to all forms of health care planning. While some barriers were general, others were specific to certain forms of planning, such as barriers to communication with loved ones or to appointing a health care proxy.
Asking participants to consider scenarios involving different likelihoods of returning to their current health or experiencing disability elicited its own set of barriers and benefits. Some participants welcomed the opportunity to specify that they would not want to receive treatment if the likelihood of an acceptable outcome became too low:
Well, I feel like if it's a 50% [chance of recovering], go for it; 99% go for it, but 1%, no; 10% no.
Other participants were made uncomfortable by the discussion of probabilities and/or felt that any probability of survival did not merit consideration of forgoing treatment:
It is too negative…how could you answer it because you don't know what you have and what is the percentage.
I think people of our age have thought some about [these issues] before these questions ever came up, but nobody ever wants to face their own mortality…. I'm the optimist…. Out of 1%, I am the 1%.
Participants reflected that this exercise was much harder than completing documents because of the uncertainty and trade-offs between quantity and quality of life:
OK, but to make a decision that you can't …. Like a living will, that is a black and white situation.
If you had a bad stroke or a heart attack and you are out of it, well then, that is it for me…but those odds … too many issues come up in that.
Participants described a number of different cognitive, emotional, and behavioral strategies they used to engage in the different components of ACP. These processes and illustrative quotes are presented in Table 5. Several of these processes, such as individualizing communication to the needs of a loved one, were also reflections of self-efficacy. They represented strategies developed by participants to overcome the challenges to completing ACP.
When asked in an open-ended fashion what they had done or thought about doing in order to plan for declines in health in the future, participants mentioned ACP, but also talked about a much broader range of activities. These included planning for a move to a site with increased personal assistance, taking out long-term care insurance, taking care of “affairs,” such that loved ones knew where important papers were, and funeral planning. Several participants spoke in great detail about the need to do funeral arrangements in advance:
Well, I already have things paid for. I don't want [my son] to be spending money for an expensive casket…. Well, you know I told him just give me a plain casket…. And I told him I don't want him changing that. I want it to go just the way I planned it.
Participants who had engaged in these other activities had not necessarily done ACP, nor did they always see connections between these activities and ACP. However, some participants had done ACP in the context of these other activities:
I was going to the attorney to have a will drawn up anyway, so we did the Living Will while I was there.
Several of the participants talked about the powerful effects of either engaging or failing to engage in ACP with their loved ones:
He didn't have [a living will], and it caused a lot of problems and more worries about making decisions and what to do and not to do and guilt. Who wants to make that final decision? I saw my nieces - two of them wanted to do this and the other one didn't … and I didn't want to be the deciding factor either, so it was very difficult for him not having that.
At the very end when we had to make decisions about the amount of medication that he received, and the hospital was reluctant to administer what he needed, I really had to pull out that paper and say, “You will do this for him,” and that was pretty powerful. It was an experience that I will never forget.
Participants in many of the focus groups talked about a prior experience with loved ones as a major reason why they had done ACP. When asked why she had completed a living will, one participant responded:
We went through that a couple of years ago with my mother. She passed at 106. I had been through the living will with her and … she had everything the way she wanted it, and I just want to follow suit.
A second participant reflected:
What really made me do this is since my mother died. My mother was living with me, so all that fell on me, and it was really hard. When you are grieving for your mother and having to go through all this, it is hard. You don't make good decisions sometimes ….. So that is why I wanted to get this all cleaned out so my son wouldn't have to worry.
This qualitative study exploring how older persons and their caregivers think about ACP illustrates how participation in ACP can be conceptualized as a process of health behavior change. The discussion of ACP by study participants revealed that older persons are at different stages of readiness to participate in the range of activities involved in ACP, and that they possess highly varied barriers to, as well as facilitators of, ACP. Older persons utilize a variety of strategies to engage in ACP, and these strategies are a reflection of their self-efficacy. ACP is only one of a larger set of activities that older persons participate in to prepare for worsening health and the end of life.. Experiences with the health and end-of-life care of family members can provide a powerful sense of susceptibility, or understanding of the consequences of failing to engage in ACP.
The variable readiness of participants to engage in ACP suggests that the most effective programs to increase the rate and improve the quality of ACP would provide tailored interventions, based upon an individualized assessment of a person's readiness to change, barriers to and facilitators of change, and their use of processes of change.10, 11 The provision of action-oriented interventions, for example, such as the provision of education about completing advance directives forms, is unlikely to change the behavior of an older person who finds it difficult to think about death and dying or who endorses other barriers to ACP. Instead, a more effective intervention for individuals who endorses multiple barriers to engaging in ACP might focus on helping them to understand the benefits to ACP, such as decreasing the burden on loved ones. Acknowledging that attitudes need to change before behavior can change, the intervention could also suggest to them the use of cognitive and emotional processes that have been successfully used by other older persons to move from early to later stages of readiness.
In this study, ACP was considered in terms of: a) the completion of documents; b) consideration of broader goals of care; and c) communication with loved ones and physicians. The older persons in this study were in differing stages of readiness for different aspects of ACP, without evidence for a pattern of progression from one component to another. For example, there were participants who had completed advance directives but who had not considered their broader goals of care. This finding differs from a recently proposed model of ACP according to health behavior change. In that model, each of the components of ACP is considered as analogous to a stage of change in the TTM, so that the steps of ACP occur in an ordered fashion. The current study adds a level of complexity to the design of interventions to improve ACP. It suggests a need for assessments and interventions not only tailored to the individual but also developed separately for each of the different behaviors composing the process of ACP.
Older persons' variable readiness to think about their broader goals of care underscores the challenges inherent in helping them to complete this component of ACP. Echoing the results of an earlier study performed among older homebound persons,31 many of the participants, including those who had participated in other aspects of ACP, were particularly reluctant to think about the care they wanted for themselves under conditions of uncertainty. Other participants did not see the need for this deliberation, because, under conditions other than those in which their physicians knew for certain that they were dying, their goal remained focused on survival. These attitudes stand in contrast to the evidence that physicians only rarely judge that patients have reached a point in their illness when death is certain.32 It is therefore likely that ACP will only have a limited effect on the care delivered at the end of life until older persons understand the uncertainties inherent in medical decision-making and are encouraged to incorporate this understanding into their deliberation about their end-of-life care.
The barriers to ACP elicited in the current study fall into 4 broad categories: knowledge, psychological factors, religious beliefs, and social supports. A number of these barriers have been elicited in other studies examining attitudes toward advance directives and ACP.33-39 The current study rounds out the evaluation of ACP as a health behavior by providing new information on the pros, or benefits, of, the processes used to promote, and perceptions of susceptibility and self-efficacy regarding ACP. These concepts provide additional insights into the design of interventions to promote ACP. The theme arising in virtually all the focus groups, that experiences with a family member are a potent motivation for engaging in ACP, is consistent with quantitative research.40 Asking older persons to reflect on their experiences with loved ones is likely to be a useful foundation on which to start the process of ACP, as demonstrated in one model of ACP.5 A second common experience among the participants was that they had engaged in a broad range of activities in order to prepare for declines in future health or for their death. A novel approach to ACP would be to take advantage of the times at which persons are more receptive to thinking about these issues, by making ACP a routine part of other planning processes, such as funeral planning.
Because our population included only Caucasians and African Americans, it is likely that we did not elicit the full range of attitudes toward ACP, many of which have been identified as prevalent in ethnic and cultural groups who were not represented in the current study.41 It is important to note that these attitudes include a focus on the family's responsibility for decision-making,42 which stands in contrast to notions of patient autonomy, which have traditionally formed the basis of ACP. However, the expanded concept of ACP that underlies this study accommodates planning focusing on family-based decision-making. In addition, the results of this study argue for an individualized approach to the assessment of attitudes toward ACP, regardless of cultural or ethnic background. Although the current study provides qualitative support for the application of health behavior models to ACP, models such as the TTM make quantitative predictions regarding behavior change.43 Therefore, quantitative studies are necessary to evaluate ACP more fully according to these models.
ACP can be conceptualized as a set of health behaviors, with older persons having variable readiness, barriers and benefits, and perceptions of susceptibility for participating in each component. They also utilize identifiable processes to progress along the stages of readiness. The application of stage-matched, tailored interventions offers the promise of increasing utilization and improving the quality of ACP.
The authors would like to thank the following people for their time and effort in helping to facilitate and coordinate the participation of study subjects: Filomena Fondella, Janet Klett and the D'Esopo-Pratt Resource Center, Irrita Osborn, Dagmar Ridgway.
Supported by grants R01 AG19769 and K24 AG28443 from the National Institute on Aging and the Claude D. Pepper Older Americans Independence Center at Yale University School of Medicine (#P30AG21342 NIH/NIA).
Filling out a living will does not address all of the situations or treatment decisions a patient may face. I would like to lead the group through a different kind of form. This form will allow you to think about certain situations. These are situations that require difficult kinds of decisions. We will go through this form together. Then we will discuss your reactions to the form.
I'd like for you to listen to the questions and think about how you would react if you were asked to answer them, but I don't want you to answer these questions now.
The questions we just looked at are examples of questions that ask you to weigh the pros and cons of treatment. Unfortunately, sometimes, in severe illness, you might need to decide about whether you would want to have a treatment that puts a lot of burden on the patient, such as being on a machine, when there are not guarantees that it will get you back to how you were before you got sick. So, when you think about these treatments, you need to think about how you feel about having the treatment if there was a chance you wouldn't survive, even if you had it, or if there was a chance you would survive, but be unable to do the things you can do now. The questions we looked at were meant to give you an idea of what it is like to think about these things. Now we'll talk about your reaction to being asked to think about them:
Now, I'd like us to think about how much discussion you want to have with your family about your illness. Some people think that it is important to discuss their feelings and their concerns with their loved ones. Other people think this is too hard. I want to know how you feel about this.
Conflict of Interest: The editor in chief has reviewed the conflict of interest checklist provided by the authors and has determined that the authors have no financial or any other kind of personal conflicts with this paper.
Sponsor's Role: The sponsor had no role in the design, methods subject recruitment, data collection, analysis or preparation of the paper.