Asking Older Adults with Impaired Cognition and Care Partners About Serious Illness Experiences Can Elicit Goals of Care During Advance Care Planning

Abstract

Background:

Serious illness experiences, i.e. experiencing critical illness personally or through close ones, often influence end-of-life goals of care.¹ Little is known about how older adults with cognitive impairment respond to serious illness experience questions during advance care planning (ACP).

Objectives:

Our aim was to demonstrate the utility of serious illness experience questions as a strategy for eliciting goals of care during ACP among older adults with cognitive impairment and care partners.

Methods:

Thematic analysis of 92 transcribed ACP conversations involving older adults ≥80 years old who screened positive for cognitive impairment and their care partner. Trained facilitators used a structured ACP conversation guide that probed serious illness experiences.

Results:

Participants often described multiple serious illness experiences influencing goals of care, which fell on a spectrum between longevity and comfort. Experiences involving negative clinician interactions provided opportunities for participants to emphasize longevity in their goals of care. Experiences involving severe disability or iatrogenic harm provided opportunities to emphasize comfort in their goals of care.

Conclusions:

Older adults with cognitive impairment and their care partners were able to respond to serious illness experience questions, with responses eliciting a spectrum of goals of care. Integrating such questions into ACP for older adults with cognitive impairment may clarify goals of care and improve decision-making for patients and families.

Registration:

SHARE: Sharing Healthcare Wishes in Primary Care.

ClinicalTrials.gov: NCT04593472.

Introduction

People with dementia are less likely to engage in advance care planning (ACP) than people with cancer or organ failure.^2,3 As dementia progresses, decisions are often made by family. Best practice recommendations highlight the need to elicit goals of care when cognitive impairment is first recognized and engage family early in ACP conversations.⁴ However, the benefit of ACP and how best to encourage participation as cognitive impairment worsens is a point of debate.^2,4–7

ACP in dementia is defined as “a process of communication about future care and treatment preferences, values and goals with the person with dementia, family, and the health care team . . . continued when the person with dementia becomes unable to make their own decisions.”⁸ Typical ACP conversations document end-of-life (EOL) goals of care, a collection of often context-specific preferences falling along a spectrum from maximizing life expectancy (i.e., longevity) to minimizing distress and burden (i.e., comfort). Longevity-focused care includes cardiopulmonary resuscitation (CPR) or interventions whose purpose is to prolong life. Comfort-focused care involves avoidance of interventions which cause discomfort, e.g. CPR or surgery.

Serious illness experiences, e.g. personally experiencing serious illness, surrogate decision-making, or witnessing EOL care in loved ones, are important factors shaping goals of care. However, a systematic review found only 20% of ACP conversation guides prompted facilitators to ask questions about serious illness experiences.⁵ These questions can evoke memories as a point of shared reference for older adults and families when discussing the nature of decisions and useful limits to decisions. For example, reflecting on a relative with anoxic brain injury after surviving cardiac arrest might assist older adults in clarifying a time limit for cardiopulmonary resuscitation efforts. However, people with dementia are often excluded from studies on ACP readiness and communication strategies.^1,9–14

To demonstrate the utility of serious illness experience questions for eliciting goals of care from older adults with cognitive impairment and their families, we thematically analyzed goals of care-related responses to serious illness experience questions during ACP.

Methods

Parent Study, Setting, and Participants

Our study analyzes transcribed audio-recordings of facilitated ACP conversations collected from November 2020 to March 2023 as part of SHARE: Sharing Healthcare Wishes in Primary Care, a single-blind, two-group randomized controlled trial (NCT04593472). SHARE’s primary aim was to test the efficacy of an ACP intervention for improving communication quality between older adults with cognitive impairment and their care partners. The intervention included: (1) an introductory letter from their primary care practice, (2) access to a trained ACP facilitator, (3) person-family agenda setting to align perspectives regarding the family’s role in ACP conversations, and (4) assistance with patient portal registration for the patient and family.¹⁵ Older adults were recruited from eight primary care practices in the Baltimore-Washington area and included English-speaking patients aged 80+ years who screened positive for cognitive impairment, provided written informed consent personally or through a proxy, and regularly attended primary care visits with a family care partner. Cognitive impairment was categorized as mild (32–50), moderate (28–31), or consistent with dementia (i.e., severe, 0–27) based on the Modified Telephone Interview for Cognitive Status (TICS-m).¹⁶ Some older adults could not respond, however their care partner verified their severe impairment. Consent was obtained before data collection. Institutional Review Board approval was obtained from Johns Hopkins University (IRB00242431).

Respecting Choices, an evidence-based ACP conversation guide, formed the basis of facilitator training, supplemented with communication strategies for maximizing participation of older adults with cognitive impairment and family (e.g., providing information in small chunks, soliciting questions and input from both).¹⁷ Facilitators (including a social worker, nurse, and community health workers) were trained to guide ACP discussions virtually or via telephone with approximately 17 questions. All ACP conversations included a standardized serious illness experiences question: “Tell me about any experiences you have had with family, friends, or yourself becoming seriously ill or injured.”

Data Collection and Thematic Analysis

Data included 92 recorded and transcribed ACP conversations with 88 older adults (80+ years), their care partners, and trained facilitators from the intervention arm. First, we deductively coded transcripts into a priori codes aligned with the Respecting Choices conversation guide, e.g. serious illness experience question responses and goals of care-related discussion.^17–19 Codes were refined through weekly consensus meetings, added as patterns emerged (inductive coding), and removed if not reflecting the data. Using the final codebook, two independent coders (M.A.S., V.H., G.T., C.G., or P.Z.) coded remaining transcripts, with coding discrepancies (including when no code was assigned by the other coder) resolved in team consensus meetings for all transcripts. Coding was performed using NVivo v.12, and reporting adhered to COREQ guidelines.^1,13 Thematic analysis was performed on data coded for both serious illness experience question responses and goals of care-related discussion, with iterative generation and refinement of themes and subthemes and frequent referencing of exemplar quotes based on methods described by Braun and Clarke.¹⁸ To consider the effect of disease severity on conversation dynamics, quotes were labeled by theme and subsequently redivided into two groups based on TICS-m score in the mild/moderate versus severe range (i.e., consistent with dementia).

Results

Older adults were 89 years old on average, 62.5% were female, 25% were Black, and 68.1% had scores consistent with dementia (Table 1). Among care partners, 83% were female and 66.2% were adult children. Older adults with mild to moderate cognitive impairment (“mild/moderate group”) often participated in ACP discussions, while those with scores consistent with dementia (“dementia group”) rarely participated.

Table 1:

Characteristics of Older Adults, Care Partners and Advance Care Planning Conversations.

	Normal Cognition (n=15; 17.0%)	Mild Cognitive Impairment (n=13; 14.8%)	Scores consistent with Dementia (n=60; 68.1%)	Overall (N=88)
Older Adult Characteristics
Patient mean age, years (range 80–103)	87	85	90	89
Race/Ethnicity, n (%)
White / Non-Hispanic	13 (86.7)	11 (84.6)	34 (56.7)	58 (65.9)
Black / Non-Hispanic	2 (13.3)	2 (15.4)	18 (30.0)	22 (25.0)
Hispanic, Asian, Other Race	0 (0)	0 (0)	8 (13.3)	8 (9.1)
Gender, n (%)
Female	7 (46.7)	3 (23.1)	45 (75.0)	55 (62.5)
Male	8 (53.3)	10 (76.9)	15 (25.0)	33 (37.5)
TICS-m score (0–50), mean+	36	30	8	16
Care Partner Characteristics
Care partner mean age, years (range 80–103)	65	75	66	67
Relationship to care recipient
Adult child	10 (66.7)	4 (30.8)	45 (75.0)	59 (66.2)
Spouse	4 (26.7)	7 (53.8)	12 (20.0)	23 (26.1)
Other (Friend/ Relative)	1 (6.7)	2 (15.4)	3 (5.0)	6 (6.8)
Race/Ethnicity, n (%)
White / Non-Hispanic	12 (80.0)	11 (84.6)	36 (60.0)	59 (67.0)
Black / Non-Hispanic	2 (13.3)	2 (15.4)	18 (30.0)	22 (25.0)
Hispanic, Asian, Other Race	1 (6.7)	0 (0)	6 (10.0)	7 (8.0)
Gender, n (%)
Female	13 (86.7)	12 (92.3)	48 (80.0)	73 (83.0)
Male	2 (13.3)	1 (7.7)	12 (20.0)	15 (17.0)
Conversation Characteristics (* Values were rated by fidelity auditors)
Meeting length, mean minutes	40.7	30.4	36.4	36.2
Older adults who verbally participated, n (%)	14 (93.3)	12 (92.3)	25 (41.7)	51 (58.0)
ACP mentions dementia/capacity, n (%)	2 (13.3)	4 (30.8)	36 (60.0)	42 (47.7)
Patient participation*, mean	2.2	2.2	0.8	1.3
Care partner participation*, mean	2.2	2.6	2.7	2.6
Patient/care partner relational conflict*, mean	0.3	0.02	0.2	0.2

^*Data from fidelity audit, range 0–3, higher score= higher participation

Nineteen of 88 older adults and their care partners (21.6%) related serious illness experiences to goals of care. These experiences included witnessing EOL care personally or through relatives, surrogate decision-making, negative clinician interactions, among others. Two themes described goals of care-related discussion raised in response to the serious illness experience question: “Serious Illness Experiences and Comfort-focused Goals of Care” and “Serious Illness Experiences and Longevity-focused Goals of Care.” Subthemes are shown with exemplar quotes (Table 2).

Table 2:

Themes, Subthemes, and Selected Quotes by Effect on End-of-Life Goals of Care

Theme 1 Serious Illness Experiences and Comfort-focused Goals of Care
Subtheme 1.1 Witnessing severe cognitive and physical disability	Example 1 Patient: And I can remember her living like that. And I thought, well, she wasn’t living. She was existing… I don’t want to be like that. I don’t want to just exist… I’m sorry, that sounds terrible, but that’s what I would tell her. Just put a pillow over my head and leave me go… Care Partner: I would say yeah, because I saw what they went through. Like, my grandmother and my great-grandmother, that they ended up just sitting there. My great-grandmother, all I can remember is her sitting in a dark room. Patient: In the little room. Care Partner: And then listening to the radio. She didn’t have a lot.. Patient: She didn’t do anything. (record 31) Example 2 Care Partner: She doesn’t want any of that… We have-- my dad’s experience is not the only experience. We have relatives in Taiwan. They have been going through the hell of prolonging death, actually. It’s not prolonging life… She knows all this and she doesn’t want it. ACP Facilitator: …So, it sounds like you pretty much are clear and know what you want. Care Partner: I’m really clear because we have like-- unlike here, our relatives in Taiwan, they have a different set of the value of the-- the value for caring your elderly people and I think their goal is so miserable because they want their parents to live as long as they can live… My sister-in-law’s father in 90s and he has been on respirator for about a year now, more than a year now and he doesn’t even recognize his daughter. They are being kept alive on the respirator. (record 24)
Subtheme 1.2 Witnessing healthcare systems fail to protect patients from burdensome care	Example 1 Care Partner: Having been in healthcare myself, some things I've seen, sometimes you wonder, are you pretty much next to abusing a corpse? But it’s very difficult sometimes for relatives to accept that this person is gone. And that this person-- and not knowing what the person would have wanted it, sometimes they feel like erring on the side of caution is to continue to keep them alive, and I totally disagree. (record 84) Example 2 Patient: She really wanted to die, and the nursing home was, well, both trying to kill her and keep her alive. I don’t know, but it seemed like they were more concerned about… Care Partner: Income. Patient: ...income than they were about the… Care Partner: Her wishes. Patient: ...about what she wanted. (record 36) Example 3 Care Partner: What I appreciate about going to a geriatrician and someone who specializes in the aging population like that-- and I share this with friends-- is there's that perspective of what do you do, how aggressive are you when things do arise that require attention?... I will tell you when I took my father-- he was in an emergency room at one point-- and this young guy came in and he's like, well, you know, we might have to go in and crack his chest open and see about doing a bypass. And I'm like whoa, whoa, whoa, you know, he's 90 years old almost. We're not doing that… Keep [my mother] comfortable. She doesn’t want any, you know, she's a DNR, she's a lot of stuff. So, yeah, and she knows what that means. Like I said, she would just prefer not to wake up. (record 85)
Theme 2 Serious Illness Experiences and Longevity-focused Goals of Care
Subtheme 2.1 Advance directives are not as nuanced as the situation and can introduce confusion	Example 1 Patient: That worries me because there must be situations where-- I've been through it, that, you know, somebody else has to kind of help with the decision and yeah, I don’t know. And I don’t want to be murdered. <laughs> I don’t want to be-- I don’t want to be killed. I don’t want to be-- I mean, I'm [age], so I look at it, I'm frightened by the fact that I'm, you know, all this is happening. I could take care of my parents. I was at a different position, I guess… but I wouldn’t want somebody to say, "Oh, well, just kill her." <laughs> "Let’s just let her go." Even though, you know, I have said I don’t want any life support, but I would want the doctors' advice as to what to do. There might be an alternative. (record 41) Example 2 ACP Facilitator: Yeah. From what I see it’s no CPR… Care Partner: Yeah, that’s what I see here. No artificial ventilation… It was interesting because he had a rectal bleed… And he would tell them, “Just do whatever you can do to save my life.” … I feel like, well, we need to do whatever we can to help him, in fact, because that’s what he wants… He does say, “I want to live to 100.” (record 60) Example 3 Care Partner: I was just concerned because it says about not having a feeding tube or anything, but he has one now because he's recovering from the Whipple surgery for pancreatic cancer, and he hasn’t gotten to the point yet where he could eat a regular amount of food every day, so I didn’t want that taken away, because on this form we filled-out-- but this is if he can’t respond himself, right? ACP Facilitator: Right. We can go over that in a little bit. Care Partner: Just so they don’t take this away from him. (record 87)
Subtheme 2.2 Reasons for distrusting clinicians	Example 1 Care Partner: Doctors are only people, right? So, they told her with this cancer here-- what did they tell you in DC? Patient: I was going to die, get my will made because I was going to die within… Care Partner: Three to six months. Like, it’s terminal. It’s done and then we go to [health system 1] and they’re like “This isn’t terminal. We can address this and we can probably completely cure this,” …So, it’s like this whole idea of terminal, I’ve just learned myself having cancer three times… I’m just not this believer anymore, that just because one doctor says “Oh, it’s terminal,” that that’s the case. So, I mean, terminal sort of becomes in quotes… But different doctors have different opinions, right? …I mean, these doctors are really just people. They’ve gotten degrees. They’re trying to help, but that’s why you go and get different [opinions] and I think that’s why they call it medical opinions because it is just an opinion. (record 11) Example 2 Care Partner: She was asking about whether the hospitalization taught us anything about for the next time. Well, it’s to stay on top of things… You have to call them and you have to--you do have to make your expectations known. So, where the physical therapists were saying “Well, we did this, this, this,” and I’m like “Well, that’s not good enough. Let’s try a more aggressive therapy. Let’s try to do more,” because my expectations are that-- of course, I had higher expectations than-- I said she was independently living before the stroke… So, I pushed for more therapy and not to let up on her… I said “I don’t care if she’s 101. Do it anyway because that will get her over that complacency.” ACP Facilitator: That’s the advocacy. Care Partner: Yeah. That was the advocacy. So, you can’t let up at any time… Care Partner: Is there anything that you learned in the hospital? Patient: No, they were terrible to me. They were terrible. Care Partner: They were terrible. They were terrible. She would say something and they would ignore it… again, she’s 101, senile, dementia, and all of that, what people think of someone who is 101 and in fact, she really had a lot of her faculties and she knew exactly what was going on. (record 13) Example 3 Care Partner: It wouldn’t be entirely up to the care team because that’s where we were the last time. She was at the point where she thought that should have happened and I disagreed. I think that he-- I had talked to other doctors, and they had told me that he wasn’t there yet, but the following day she thought that could have happened. So if we had gone along with her idea I don’t think he would be here today, I think I should have some input as long as I'm capable of helping decide and vice versa. (record 82)

Theme 1: Serious Illness Experiences and Comfort-focused Goals of Care

Subtheme 1.1 Witnessing severe cognitive and physical disability

Given the rarity of participation by older adults in the dementia group, their care partners provided most accounts witnessing severe disability (e.g., akinetic mutism or ventilator dependence). However, two older adults in the dementia group engaged deeply with the serious illness experience question, with one stating “She wasn’t living. She was existing… I don’t want to just exist” (Subtheme 1.1, Example 1).

Subtheme 1.2 Witnessing healthcare systems fail to protect patients from burdensome care

Some participants described longevity-focused care as ‘prolonging death,’ expressing a strong value for quality over quantity of life. Several care partners indicated in remarks that patients needed advocacy to protect them from this care. One care partner expressed moral distress for enabling medically futile care as a healthcare worker (Subtheme 1.2, Example 1). Another patient and their care partner remarked that a relative “really wanted to die, and the nursing home was, well, both trying to kill her and keep her alive,” going on to state that financial interests rather than honoring the relative’s wishes were driving decisions (Subtheme 1.2, Example 2).

Theme 2: Serious Illness Experiences and Longevity-focused Goals of Care

Serious illness experiences which introduced longevity-focused nuance in participants’ goals of care were more heterogeneous and more often reported by care partners or older adults with less impaired cognition.

Subtheme 2.1 Advance directives are not as nuanced as the situation and can introduce confusion

Some care partners described conflicting statements of preferences made by the older adult, so without clarity, they defaulted to longevity-focused care (Subtheme 2.1, Example 2). One older adult reflected on surrogate decision-making experiences as a source of anxiety, fearing that expressing comfort-focused preferences could bias surrogates to prematurely withdraw care (Subtheme 2.1, Example 1). Experiences of successful intensive treatments made older adults and care partners hesitant to limit future treatments, which they perceived advance directives to do. Care partners pointed to ongoing interventions (e.g., feeding tube) as reasons to continue maximizing treatment options (Subtheme 2.1, Example 3).

Subtheme 2.2 Reasons for distrusting clinicians

Serious illness experiences involving negative clinician interactions included receiving an inaccurate cancer prognosis and age-related discrimination. Care partners expressed concerns that clinicians had been dismissive or overly pessimistic about the life expectancy of themselves or a relative, complicating their ability to trust medical counsel (Subtheme 2.2, Examples 1–3).

Discussion

This is the first study to explore serious illness experience questions as a strategy for eliciting goals of care during ACP conversations for older adults with cognitive impairment and their care partners. Many people have an unrealistic or distorted understanding of EOL care options, often based on cinema where immediate survival rates after CPR are almost twice that of actual rates.^20,21 Personal experiences with serious illness allow people to observe the realities and limitations of medical care. Inviting discussion of these experiences in ACP provides an opportunity to discuss risks and benefits of real-world medical care using examples from people’s lives.

Both positive and negative experiences of EOL care are known to increase readiness to engage with ACP.^22,23 Several patients with more severe cognitive impairment conveyed goals of care by describing relatives with severe disability, suggesting that serious illness experience questions may, in select cases, help older adults with dementia during ACP. However, dementia can impair ability to conceptualize one’s future self, a vital skill in ACP.²⁴ Care partners in mild/moderate and dementia groups emphasized caution when interpreting ACP conversations, particularly when older adults voiced contradicting preferences over time. More investigation is needed into strategies for navigating conflicting statements of preference from patients with cognitive impairment, e.g. utilization of Medical Ethics consultation services.

Previous literature notes that surrogate decision-making experience can heighten awareness of medicine’s limits and facilitate transition to comfort-focused care.¹ Interestingly, surrogate decision-making experiences made transitions more difficult for some participants. Fear that putting preferences in writing can lead to premature withdrawal of care is a known barrier to ACP completion.¹¹ Clinicians can support those hesitant about documentation by facilitating high quality conversations between patients and family.^25,26

Distrust towards clinicians has been documented among older adults with cognitive impairment and their family.² Having received an inaccurate prognosis caused care partners to hang on to hope and struggle with accepting medical counsel that EOL decision-making may soon be necessary. Though this impact has been shown in those affected by lung cancer, it has not previously been documented among those affected by dementia.²⁷ Serious illness experiences involving negative healthcare interactions are important for clinicians to broach in a non-defensive, non-judgmental manner. By listening to patients’ formative medical traumas, clinicians can offer talk therapy, gain insights into microaggressions, and potentially heal rifts between patients and the healthcare system.

Strengths/Limitations

Being a qualitative analysis, the purpose of this study is to generate hypotheses. Therefore, associations between goals of care and disease severity cannot be made. In addition, due to low levels of engagement by older adults with severe cognitive impairment, themes likely overrepresent the perspective of care partners and older adults with mild/moderate cognitive impairment. Many responses prioritizing comfort-focused goals of care (Theme 1) included descriptions of relatives rather than reflections on one’s own illness, most likely because most quotes in this analysis come from the dementia group, of which care partner voices predominate. Also, all participants were English-speaking and 65.9% were Non-Hispanic White, limiting generalizability to the diverse U.S. population. Despite this, our study had notable strengths, using rigorous qualitative methods on a large dataset of ACP conversations. Such empirically derived findings have increased applicability to clinical settings.

Conclusion

ACP provides opportunities to communicate goals of care, but such conversations may be difficult for older adults with cognitive impairment. We demonstrate that serious illness experience questions during ACP can elicit goals of care from older adults with cognitive impairment and family.

Figure 1: Impact of Serious Illness Experiences on End-of-Life Goals of Care.

An overview of common serious illness experiences and their impact on end-of-life goals of care. Blue arrows pointing to the right represent subthemes under Theme 1, “Serious Illness Experiences and Comfort-focused Goals of Care.” Pink arrows pointing to the left represent subthemes under Theme 2, “Serious Illness Experiences and Longevity-focused Goals of Care.”