Cognitively Impaired Older Persons' and Caregivers' Perspectives on Dementia-Specific Advance Care Planning

Abstract

Background/Objectives:

Advance care planning (ACP) traditionally involves asking individuals about their treatment preferences during a brief period of incapacity near the end of life. Because dementia leads to prolonged incapacity, with many decisions arising prior to a terminal event, it has been suggested that dementia-specific ACP is necessary. We sought to elicit the perspectives of older adults with early cognitive impairment and their caregivers on traditional and dementia-specific ACP.

Design:

Qualitative study with separate focus groups for patients and caregivers.

Setting:

Memory disorder clinics.

Participants:

28 persons age 65+ with mild cognitive impairment or early dementia and 19 caregivers.

Measurements:

Understanding of dementia trajectory and types of planning done; how medical decisions would be made in the future; thoughts about these decisions.

Results:

No participants had engaged in any written form of dementia-specific planning. Barriers to dementia-specific ACP emerged, including lack of knowledge about the expected trajectory of dementia and potential medical decisions, the need to stay focused in the present because of fear of loss of self, disinterest in planning because the patient will not be aware of decisions, and the expectation that involved family members would take care of issues. Some patients had trouble engaging in the discussion. Patients had highly variable views on what the quality of their future life would be and on the leeway their surrogates should have in decision making.

Conclusions:

Even among patients with early cognitive impairment seen in specialty clinics and their caregivers, most were unaware of the decisions they could face, and there were many barriers to planning for these decisions. These issues would likely be magnified in more representative populations, and highlight challenges to the use of dementia-specific advance directive documents.

Advance care planning (ACP), the process by which individuals prepare for future treatment decisions at a time of decisional incapacity, is a key component of high-quality serious illness care.¹ Although ACP has been endorsed for persons with dementia,² there is growing appreciation that existing approaches to ACP are not well suited to the challenges that persons with dementia will face. In traditional ACP, patients consider a disease trajectory involving a brief period of incapacity with an imminently terminal event and the use of life-sustaining treatments during that period. Dementia is fundamentally different because it is characterized by a gradual decline in health and cognition during which decisions need to be made about many potential interventions, not just those at the end of life.³ In addition, dementia can profoundly alter the very self it afflicts, suggesting that ACP in dementia requires explicit consideration of how much leeway a surrogate decision-maker should have.⁴

The unique features clinical features of dementia have led to the recent creation of dementia-specific advance directive documents. This approach has divided experts. Proponents describe dementia-specific directives as providing the surrogate with the information necessary to make clinical decisions⁵ by having patients pre-specify treatments they would not want to receive, such as assisted feeding,⁶ or express choices about their desired intensity of care in different stages of dementia.^7,8 Critics have argued that these documents provide a false sense of security and require patients to make decisions without knowing all of the relevant considerations of the situations they may face. They maintain that written directives cannot substitute for the communication between patients and surrogates necessary to prepare surrogates.^9,10 This debate echoes earlier discussions about the limitations of all written directives and the challenges of having patients make decisions in advance of specific clinical scenarios.¹¹

An understanding of patients’ and caregivers’ views of ACP for dementia is critical to informing this debate. ACP is increasingly being conceptualized as a health behavior,^12-14 with engagement dependent upon individuals’ knowledge, beliefs, and attitudes.^15-18 These constructs have not been systematically explored among persons with early cognitive impairment and their potential surrogates as they relate to dementia-specific planning.

The purpose of the current study was to elucidate views on participation in dementia-specific planning among patients with early-stage cognitive impairment and their caregivers. We sought to elicit participants’ understanding of the need for and readiness to plan, and barriers to and facilitators of planning.

METHODS

Participants

To identify patients with an established diagnosis of mild cognitive impairment or dementia, we initially reached out to local assisted living facilities. We learned from facility directors that many residents with evidence of cognitive impairment either had not received a formal diagnosis or that there was resistance to accepting the diagnosis. We therefore focused recruitment on two sites providing evaluation of memory impairment: an outpatient academic geriatric assessment clinic and the Yale Alzheimer’s Disease Research Unit (ADRU). Inclusion criteria for patients were: a) age ≥55 years; b) diagnosis of minor/major neurocognitive disorder with either a formal Clinical Dementia Rating Scale score¹⁹ of 0.5 or 1 (for patients recruited from the ADRU) or assessment by a case manager of only mild impairment in cognition and/or function (for patients recruited from the assessment clinic); c) a response of “yes” in response to, “Many patients who receive care in [name of clinic] have either mild cognitive impairment or early dementia. Do you have either of these?” This was asked during a phone call that followed an opt-out letter. Exclusion criteria were: a) primary language other than English; b) aphasia out of proportion to other cognitive deficits; c) severe hearing loss; d) acute illness. Patients who agreed to participate were asked if a caregiver might also be interested in participation, and, if so, we requested permission to invite that person. Patients did not need to have a caregiver in order to participate. The protocol was determined to be exempt by the Institutional Review Board of Yale University.

We sent 68 letters to patients. We were unable to reach 13, 16 refused, two had acute illnesses, and seven answered no to the question about a diagnosis of cognitive impairment. Of the 30 patients who agreed to participate, 21 gave permission for the study team to contact a caregiver, and two of these caregivers refused participation. Because two patients did not show up to their assigned group, the final cohort consisted of 28 patients and 19 caregivers.

Data collection

We utilized qualitative methods to elicit open-ended responses. In order to facilitate a dialogue among participants about their experiences, patients and caregivers participated in separate focus groups. Groups were composed of 3-5 individuals, with the exception of two caregiver groups that were individual interviews and one caregiver group with two participants. The focus groups were led by a trained moderator using a discussion guide. The initial guide was developed based on a review of the literature combined with clinical experience. The guide was modified over time so that salient issues arising in earlier groups could be more fully explored in later groups. The final versions appear in Supplemental File S1.

The guide for both patient and caregiver groups began by asking participants their understanding of what might happen to the patient over time and what planning they had done. If ACP was not mentioned, the moderator made a statement to the group that some people plan for a time when they cannot make their own decisions. Patients were asked how likely it was that someone would have to make decisions on their behalf. Caregivers were asked how likely they were to have to make decisions for the patient. Patients and caregivers were both asked how difficult they thought surrogate decision making would be and about how decisions would be made. Caregivers were asked how they would weigh the patient’s cognitive disorder if faced with making a medical decision. Prior to the start of the group, all participants completed a brief survey asking about sociodemographic and health status.

Focus groups were conducted until thematic saturation was achieved in that no new ideas or concepts arose in the final set of groups. Focus groups were audio recorded and transcribed. The audio file for one patient focus group was unavailable for transcription.

Data analysis

Content analysis was used to code small portions of the text representing unique concepts. Initially, three investigators independently coded two transcripts and met to review coding and develop a coding structure. These investigators used the preliminary structure to code two additional transcripts independently. They met to review application of the structure and to identify additional codes. An additional transcript was coded by the three investigators independently, with only minor differences in application of the coding structure. After this, two investigators independently coded two of the transcripts, and one investigator coded the rest.

Once coding was complete, the coding structure and associated text were examined within and across transcripts to examine relationships among the codes. This process was conducted iteratively through discussion among the investigators to articulate larger themes arising from the analysis of relationships and context. NVivo 12 (QSR International, Melbourne, Australia) was used to facilitate application and review of the coding structure.

RESULTS

Description of participants:

We conducted seven focus groups each with patients and caregivers, with three sets recruited from the geriatric assessment clinic and four from the ADRU. A description of the participants appears in Table 1.

Table 1:

Description of participants

Characteristic	Patients (n=28)	Caregivers (n=19)
Age, mean (SD)	77.4 (6.8)	76.0 (5.9)
Women, %	33	63
White %	100	100
College or higher education, %	75	42
Married or partnered, %	79	100
Self-rated health very good, %	42	--
Spouse/partner of patient, %	--	94

Themes:

Participants’ discussions generated four broad themes: 1) absence of dementia-specific medical planning in current forms of ACP; 2) barriers to dementia-specific ACP; 3) views on future selves and on surrogate decision making as they relate to dementia-specific planning; 4) effect of focus group participation on attitudes toward dementia-specific ACP.

Current forms of planning:

When asked about planning they had done in response to the patient’s diagnosis, all groups discussed non-medical planning, including financial preparations, planned or actual moves to different living situations, and other plans for help with increasing functional needs. In all but three groups, medical planning was discussed only when the group was specifically asked about this type of planning. ACP was generally discussed in terms of the completion of traditional advance directives. Many participants completed these prior to the diagnosis of dementia, often as a part of estate planning, and participants reported that directives included instructions about the use of life-sustaining treatments for severe acute conditions such as strokes or heart attacks or if the individual were in a coma:

No participants had engaged in any written form of dementia-specific planning. The topic only came up without probing in two of the groups. In one group, a patient expressed recognizing a point in his trajectory at which he would want limits on his treatment:

Patient: Yeah, I have talked about it with my kids and they respect my wishes on this, but they hope it doesn’t happen any time soon. I think they agree though that for me to go on clinging to life when—for you it’s you can’t take things apart anymore and put them together. For me it’s reading and writing.

In a second group, a caregiver had received a dementia-specific directive but had not acted on it:

Caregiver: We got something forwarded by our lawyer … I really haven’t reviewed yet, but it was some suggestions for directives for Alzheimer’s.

Barriers to Dementia-Specific ACP:

Patients and caregivers brought up a number of barriers to doing ACP. Several of these were not specific to patients with cognitive impairment, including procrastination and a desire to avoid thinking about the end of life, as well as the belief by patients that their loved ones already knew their wishes. Other barriers had a more direct connection to the patient’s diagnosis of dementia, and are listed, along with representative quotations, in Table 2. These included: a) lack of knowledge regarding the health trajectory of individuals with dementia and an understanding of physical health as completely separate from cognitive health; b) a strong need to stay focused on the present in order to avoid acknowledging the certainty of progressive impairment; c) the patient’s belief that this acknowledgment would be upsetting to loved ones; d) the expectation that family members who are closely involved will take care of issues as they arise; and e) lack of interest because the patient will not have awareness of the decisions being made.

Table 2:

Barriers to ACP/Reasons Why ACP Not Necessary

Barrier	Representative Quotation(s)
Lack of knowledge of dementia trajectory	Interviewer: For people who do have dementia, what sorts of medical problems they develop as the dementia gets worse, has that come up? Caregiver #1 : I’ve never heard that. Caregiver #2: No. Interviewer: For example, many people with dementia, eventually, develop trouble swallowing. Has that come up? Caregiver #1: I’ve never heard that.
Understanding of physical health as separate from cognitive health.	Caregiver: He is very healthy. He’s physically very able. He’s still very strong. He’s deteriorated some. I would say his balance isn’t what it used to be, but he still works out three mornings a week at a workout place. Patient: I’ll be 74 next month. I think I’m physically strong because I make an effort to be so. I still climb a mountain on my birthday.
Need to live in the moment to avoid acknowledging certainty of future impairment	Caregiver: When you’re in this position, when you’re the one who has the Alzheimer’s, you try to live from day to day the best you can. You don’t want to be looking down a road that I’m going to turn into this helpless thing, because if you start thinking about that, that’s very debilitating. Caregiver: I’m not sure what we would do if he needed more care. I’m hoping he won’t need more care. That we’ll be able to be together for a long time. Caregiver: I know she would probably share that, it’s just talking about the fact that it’s going come to that, that is upsetting. Patient: I don’t know what it’ll be like next year or whatever. I just try to keep focused on the good things that I have. That probably sees me through in difficult times.
No need because patient will not be aware	Patient: I thought about it somewhat, and I’ve come to the conclusion that I really don’t care. Interviewer :You don’t care what kind of decision they make? Patient: If I’m incapable of making’ it, I got no say in it anyway.
Patient does not want to upset family	Interviewer: Have you talked to your family about what your wishes would be if you couldn’t express them? Patient: No. I have not. I don’t think it’s because I don’t want to go there for myself. I don’t want to go there for them. They all have children that are from ages 2 to 16 and they have everything to look forward to. I try not to make them worry or to think about—I’m just so grateful.
Expectation that involved family will take care of issues as they arise	Caregiver: We have five children. I have a daughter she’s very close to, and we have four sons. I would say two of them are—my daughter and two of my sons are very close by. They’re all aware, and they’re all very concerned and committed to taking care of her if it becomes necessary in doing whatever has to be done. While we don’t know exactly what is going to transpire, we have some comfort in the fact that everybody is on board.

Notably, some patients had difficulty in a discussion about ACP in the context of a focus group discussion. Their tangential responses suggested they were not fully understanding the questions.

Views on future selves and on surrogate decision making as they relate to dementia-specific planning

Despite a lack of knowledge about health trajectories in dementia, many patients and caregivers recognized the likelihood of functional and cognitive decline. While they had not engaged in any sort of planning to avoid severely impaired states, several caregivers were aware of their loved ones’ views on these states:

Caregiver: What [my wife] says is, I don’t want to get to the stage where I’m out of control of everything, basically, which is really I guess what we’re talking about.

Caregiver: Long term, I think particularly for [my wife], she really does not want to be in a Alzheimer’s care facility or a skilled nursing facility. This is really a horrible fate in her mind.

For the second patient/caregiver pair, this desire had led to a conversation about physician-assisted suicide as their dementia-specific planning:

[My wife] was really thinking we ought to—we really ought to have a plan, so that there would be a drawer and at some point she could take whatever potion and end it.

In contrast to dementia-specific planning based on limiting care at a point in the dementia trajectory representing an unacceptable quality of life, another patient expressed the impossibility of knowing what her quality of life would be with more significant cognitive deficits:

Patient: My mother was in a nursing home a number of stages, and, sometimes, when I went to see her, it was really hard … Sometimes it was hard and laughable. Do I want to [get to the stage she was in]? No way. Am I terrified of it? No, because she was quite a happy person at the moment. I guess I’m very ambivalent about it … Sometimes she’d be sad because she’d be remembering the things that were gone or people that she had lost or stuff like—her personality and her disease, I guess is really what you’d say—she was an even-keeled person, orderly-in-her-head person … She was not, most of the time, unhappy.

Uncertainty about the future led participants to reflect on what they saw as the challenges of surrogate decision making in dementia. Once participants understood that patients might require multiple healthcare decisions made on their behalf, they questioned the extent to which pre-specified wishes should be binding when an actual decision needed to be made and how well these pre-specified statements would inform those decisions:

Caregiver: I don’t think there is, and there should be, a clear policy that if somebody says clearly when they are in good enough shape to say it that they want a certain outcome, that ought to be honored. They shouldn’t even turn to me, right?

Caregiver: If that person has it listed…—I hate to bring up my aunt again, but it was where she had to go back in [to the hospital]. It clearly stated she didn’t want anything. I said well, if it’s something mild and simple or something like this to keep her going, great, but if it’s anything major, no.

When asked directly, patients varied in terms of the amount of leeway they would want their loved ones to have in decision making:

Patient 1: Well, I think they should respect my wishes and do what they agreed to do when I made them the person that would be responsible for making my decisions for me. If they couldn’t do that, then they need to pass on that. I believe that they have an obligation to do what’s best for me if they’ve agreed to do that, and I’ve trusted them to do it. Patient 2: I’m more on the other side on that. I would trust the girls. If they decided to do something that was not what I had thought would be best, they wouldn’t do it easily or lightly. By that time, the conditions might be something I wouldn’t even understand. I just think it would be—they’d really, at that point, have to bite the bullet and decide what’s—might be different than I had suggested.

Insights into Dementia-Specific ACP from Focus Group Participation:

By the end of several groups, at which point caregivers had a clearer sense of the kinds of healthcare decisions they would need to make, some caregivers realized they needed additional discussions with their loved ones about their wishes. One caregiver who had engaged in traditional ACP with her husband commented:

Caregiver: You want to be really clear with your spouse on what they want, because you want to make the decision that they want. It’s difficult. It’s not easy. You look to the doctor to help you, but again, the doctor has some hands tied, too. Be really clear. I think I’m not really clear—I’m just sitting here thinking I’m not really clear on what [my husband] wants.

Patients also recognized the need for additional discussions, but also expressed some concerns about how this might be done:

Patient 1: I think it would be hard. I think it would be hard if you said to me we want to have a family meeting. We want you to bring your kids in while we discuss what’s going to happen when you die. That’s awkward … On the other hand, if you got them thinking about it, they could talk about it themselves. Patient 2: … A brochure, a well written brochure, which was not threatening to say things to think about, you probably could write something that would be intelligently written, not threatening, and just the logical explanation of things that you might want to consider.

DISCUSSION

In this study, we asked patients with early cognitive impairment and their caregivers to discuss their understanding of dementia and the planning they had done for the future. Those who had engaged in medical planning had thought about acute and sudden end-of-life events but few participants had prepared for a long trajectory of incapacity. The discussions revealed a number of dementia-specific barriers to ACP, including the difficulty of asking people to consider how to plan for a loss of identity and the desire of many patients and caregivers to live purely in the moment; the tendency to think about bodily and cognitive health separately; the loss of ability, even in patients with mild cognitive impairment, to understand and engage in the process; and a lack of interest from patients who recognized that they would be unaware of the decisions being made. Caregivers weighed patients’ pre-specified wishes and the role they would need to play as the decision maker, and patients held variable views on the leeway caregivers should have in decision making. While some patients were so distressed by the prospect of future severe disability that they wanted physician-assisted suicide, others believed it was possible they could be happy or content in the future.

By asking patients and caregivers to consider ACP in the context of cognitive impairment, our study provides a detailed examination of attitudes toward dementia-specific ACP among those with early-stage cognitive impairment. Several of the themes that emerged in the current study echo findings from a systematic review of qualitative studies primarily involving caregivers of patients with advanced dementia who were looking back at the experiences they had.²⁰ This review found that some caregivers regretted decisions they had made to treat acute health conditions based on a mistaken understanding that the conditions were unrelated to dementia. Other caregivers realized that there had been insufficient consideration earlier in the patient’s course about the decisions they would face. The study also found a struggle with finding the “right” time to engage in ACP, with some caregivers feeling no urgency and others regretful that they had waited too long and the patient could not participate. These findings highlight that many of the themes emerging from the current study continue to be relevant to caregivers as dementia progresses.

We enrolled patients who were receiving care in specialized clinics, and the absence of minority representation, as well as the high level of education among participants, substantially limit the generalizability of our results. Our study can be thought of as a “best case scenario” in terms of the knowledge of persons with early cognitive impairment and their caregivers. It is striking, therefore, how many concerns the results raise about the use of dementia-specific advance directive documents. Our findings suggest that most patients and caregivers are not prepared to complete such documents. Few participants in our study had even discussed medical planning in detail. This issue is likely to be magnified in more representative populations, in which underdiagnosis and lack of awareness or acceptance of dementia diagnoses are highly prevalent.²¹ Moreover, dementia-specific directives do not address the uncertainty of some patients and caregivers about the quality of the patient’s future life or role of pre-specified wishes versus the contingencies of actual decisions. The latter issue has demonstrable effects on caregivers, whose ability, for example, to cope with the burdens associated with placing their loved ones in a nursing home is negatively affected when they believe they are breaking a promise to avoid placement.²²

Interventions to promote engagement in ACP by individuals with chronic illnesses other than dementia have evolved from a focus on completion of documents to promotion of communication. The National Academy of Medicine supports a lifecycle approach, with ACP beginning prior to the development of chronic illness. The finding that some patients in the focus groups could not engage in the kind of abstract thinking necessary to engage in ACP supports the need for individuals to consider a health trajectory involving dementia before receiving a diagnosis of dementia. Rather than focusing on dementia-specific directives among persons with dementia, the goals of ACP may be better met by helping to prepare all individuals for different end-of-life trajectories, including those that involve progressive cognitive impairment.

While our study population is not representative of most older adults with early-stage cognitive impairment and their caregivers, it yields an in-depth look at issues relevant to the broader population. The results highlight the challenges of performing dementia-specific ACP and demonstrate that further innovation is needed to devise an approach to planning that fully meets the needs of patients and their caregivers.

Supplementary Material

supinfo

Supplemental File S1: Final version of patient discussion guide