“We’re all we got is each other”: Mixed-methods analysis of patient-caregiver dyads’ management of heart failure

Abstract

Background:

Individuals living with heart failure often require informal caregiving assistance for optimal self-care maintenance. The influence of caregiver burden and resilience on dyadic congruence is not well understood.

Objective:

To compare how dyadic congruence is influenced by level of burden and resilience expressed by caregivers of patients with heart failure.

Methods:

Mixed-methods analysis of individuals with heart failure and their caregivers, focusing on measures of caregiver burden (Zarit Burden Interview) and resilience (Brief Resilience Scale). Data were integrated using the Heart Failure Care Dyadic Typology.

Results:

Twelve dyads (n=24 participants) were classified as Type II (n=7) and Type III (n=5) dyads. Among Type II dyads, average caregiver burden was 19.43 (± 13.89) and resilience was 3.16 (± 1.04). For Type III dyads, average caregiver burden was 3.80 (± 4.27) and resilience 4.07 (± 1.36), respectively. Two key themes were derived: 1) caregivers’ tendency to take the lead, and 2) the usefulness of cognitive reframing. Data integration elucidated that theme 1 was more common among Type II dyads and those with higher burden, and theme 2 was more prevalent among Type III dyads and those with higher resilience.

Conclusion:

Findings highlight important variances in how dyads collectively manage heart failure. Future inquiry should involve tailored intervention development to bolster informal caregivers’ quality of life and ability to better support patients throughout their heart failure trajectory.

Introduction

Informal caregivers play a critical role in supporting self-care for individuals living in the community with heart failure (HF). Less than optimal self-care in HF has been linked to adverse health outcomes and diminished quality of life.^1–3 Survival rates have mildly improved in recent years, underscoring the importance of prioritizing self-care, so that individuals can maintain independence, control symptom burden, and remain in their preferred location as their condition worsens.^{4, 5} However, informal caregivers often feel increased burden supporting individuals living with HF in the home, feeling isolated and fully responsible for care.^{6, 7} Further, people caring for individuals with greater symptom burden have poorer physical health-related outcomes and quality of life than those caring for individuals with fewer symptoms.^{6, 8}

Using a dyadic approach to understanding the science of self-care in HF has improved our understanding of how patient-caregiver dyads appraise HF symptoms and prioritize care of the patient.⁹ Dyad types include consideration of 1) responsibility for direct care, and 2) a relational agreement for said responsibility.¹⁰ For instance, more than half of dyads encompass caregivers who report substantial contributions to self-care that patients perceive as insufficient, despite these same patients having lower hospitalization rates.¹¹ Underestimation of informal caregivers’ contributions has significant implications regarding health systems’ ability to accurately budget for indirect costs,¹² clinicians’ ability to effectively coordinate individuals’ care, and most importantly – dyads’ ability to adapt through an individual’s illness trajectory.⁴ Most HF research to date has focused on promoting patients’ self-care in HF. However, the Institute of Medicine reports have stated there is an imperative to ensure the health and well-being of caregivers is prioritized, particularly as illnesses like HF advance and patients’ care needs increase, to decrease risk of caregiver burden.^{13, 14}

The heavy focus on burden in the context of HF functional limitations and caregiving has been criticized as being deficit-oriented and insensitive to the breadth of peoples’ caregiving experiences.^{15, 16} Resilience—a response to stressful stimuli as the ‘bounce back’—has surfaced as a concept which is likely influenced by the dyadic relationship, HF severity, and other contextual factors, but has not been as well explored.^{17, 18} Therefore, the purpose of this study was to compare how dyadic congruence is influenced by level of burden and resilience expressed by HF caregivers.

Methods

Conceptual and organizational framing

The study was initially conceptualized using Szanton and Gill’s Society-to-Cells Resilience Theory.¹⁸ The theory posits six fundamental tenets for resilience; the fifth one states that just as there are particularly vulnerable periods of risk, there are times in which an individual, family, or community are particularly resilient, such as during caregiving. Accordingly, this theory was selected as it takes a multi-level and family-centered view of resilience, which was important for better understanding HF dyads in this light.¹⁸

Buck and colleagues’ Heart Failure Care Dyadic Typology was used to organize and integrate the data.^{10, 19} While not selected a priori, this typology presented a unique opportunity for further interpretation of findings regarding dyadic congruence in HF management. The first two types, ‘Patient Oriented’ (Type I) and ‘Caregiver Oriented’ (Type II), involve one person – patient or caregiver – being primarily involved in decision-making and patient care management while the other person is more passive. ‘Collaborative’ (Type III) and ‘Incongruent’ (Type IV) dyads are team-oriented, where Type III dyads work together in tandem, and Type IV do not necessarily agree on responsibility for HF self-care and management.¹⁰ These previously outlined characteristics of each dyad type were utilized to organize the study sample.

Design, sample, and setting

The present study engaged a mixed-methods study design of patient-caregiver dyads. Participants were recruited from the Johns Hopkins Hospital Heart Failure Bridge Clinic, a HF disease management clinic providing multidisciplinary HF care, primarily during the transition from hospitalization to home or episodes of ambulatory worsening HF.^{20, 21} A purposive sampling strategy was used to recruit dyads with a range of caregiving experiences including spousal and non-spousal care dyads and those with unique caregiving needs. Patients had to meet the following inclusion criteria: (1) aged 21 years or older, (2) had a documented diagnosis of HF, (3) indicated functional limitation and at least one other chronic condition, and (4) had a HF hospitalization within the last six months. These inclusion criteria were selected to capture HF dyads at a time when caregiver involvement may be more intense. Caregivers were identified based on patients seen in the clinic and had to meet the following inclusion criteria: (1) aged 21 years or older, (2) spoke English, (3) co-resided with the patient or visited them at least three times per week, and (4) had to provide support for at least one activity of daily living. Institutional Review Board approval was received from Johns Hopkins University. Written informed consent was obtained from patients and caregivers prior to data collection. Recruitment and data collection were stopped once the team established that a vast range of caregiver relationships and experiences were represented in the sample.

Data collection

Demographics including age, education level, and work status were obtained from patients and caregivers at their baseline data collection visit. Study measures relevant to this analysis included the Zarit Burden Interview (ZBI), a 22-item measure which has been widely validated for measuring caregivers’ level of burden, with higher scores indicating greater burden.²² The second measure was the Brief Resilience Scale (BRS), a six-item measure where higher scores indicate an increased ability to bounce back after a stressful or difficult event.²³

Dyads were interviewed for 45–60 min in their home to reduce distress, by minimizing time spent in the health care setting and emphasizing a discussion around home-related caregiving factors. Dyads were encouraged to separate for completion of the interview, yet all of them stayed together for at least a portion of it (i.e., one individual would fill out the survey while the other answered interview questions). Three dyads (2, 9, and 11) opted to complete the patient, caregiver, and dyad questions together. Interviews were conducted by two research team members who are Registered Nurses familiar with the study population, and both received standardized interview facilitator training from the lead investigator (MAS). Facilitators used a semi-structured interview guide containing open-ended questions on topics including caregivers’ roles, the value of their role, met and unmet needs, challenges managing HF, and suggestions for enhancing support for HF dyads.

Data analysis

Descriptive statistics were used to summarize sample characteristics using Stata^® (Version 16.0, College Station, TX). To aid integration of quantitative and qualitative data, reported caregiver burden was characterized as ‘high’ (score ≥ 9) or ‘low’ (score < 9), using the ZBI median as the cutoff. For resilience, dyads were considered ‘high’ (score > 3.6) or ‘low’ (score ≤ 3.6) using the BRS median as the cutoff.²⁴

Qualitative interviews were recorded and transcribed using a professional transcription service and data were managed using f4Analyse^® (version 2.5.2). A qualitative descriptive approach guided the analysis.²⁵ Interviews were read to gain a sense of the content, check the transcripts for accuracy, and note caregiver experiences that were highlighted in the discussion. An open coding method was applied to the transcripts, and an initial codebook was created based on the initial interviews. The codebook was adjusted and applied to all remaining interviews.

Data grouped within the following codes were subsequently reviewed by four study team members (KEN, AA, MF, LD) to assign each dyad within the typology: (1) dyadic relationship qualities, (2) caregiver involvement in care, (3) caregiver burden, and (4) resilience. Dyads were split in half, and teams of two team members met to evaluate the dyads and assign an initial type. Collectively, all five team members (including Principal Investigator, MAS) then met to discuss each case, where both teams presented their findings and evaluation. A final type was assigned via consensus; MAS helped resolve any conflict. Reflexivity and credibility of the data was enhanced by use of written memos and weekly coding meetings to discuss and compare dyads.

Findings were then mixed and organized into two integrative displays. Each display compares qualitative exemplar quotes with quantitative results according to assigned dyad type, which enabled a more realistic and nuanced perspective regarding how dyads manage HF, as opposed to assessing patient- and caregiver-centric factors independently. Two themes were generated from this mixed methods analysis based on the two concepts of interest—caregivers’ reported burden and resilience.

Results

Sample characteristics

Twenty-four participants, 12 patients and their respective caregivers, completed questionnaires and interviews (Table 1). Patients’ ages ranged from 44–89, with a mean age of 71 years, and caregivers’ ranged from 33–82, with a mean of 64 years. Several caregivers (n=5, 42%) reported they were still working, and most caregivers (n=9, 75%) identified as female.

Table 1.

Sample characteristics.

	Patient (n=12) n (%), mean (SD)	Caregiver (n=12) n (%), mean (SD)
Age	71.2 (14.6)	63.4 (14.9)
Gender
Male	6 (50)	3 (25)
Female	6 (50)	9 (75)
Race
White	5 (42)	5 (42)
African American	7 (58)	7 (58)
Employment status
Working now	1 (8)	5 (42)
Looking for work, unemployed	1 (8)	1 (8)
Retired	6 (50)	4 (33)
Disabled	4 (34)	2 (17)
Highest education level
Graduated high school or less	3 (25)	3 (25)
Some college, no degree	3 (25)	5 (42)
Bachelor’s or graduate degree	6 (50)	4 (33)
Marital status
Never married	_	1 (8)
Married	10 (83)	9 (75)
Divorced	_	2 (17)
Widowed	2 (17)	_
Type of home
Rented apartment	1 (8)
Single family	10 (84)
Multi-family	1 (8)
Caregiver Burden
Zarit Burden Interview	_	15.6 ± 16.3
Resilience
Brief Resilience Scale	_	3.5 ± 1.2

Quantitative findings

Caregiver burden (ZBI) scores ranged from 0–42; the average score among those classified as ‘low’ was 2.50 (±13.89), and 23.33 (±11.08) among those with ‘high’ caregiver burden. Caregiver resilience (BRS) scores ranged from 1.33–5; the average score among those classified as ‘low’ was 2.22 (±0.81), and 4.53 (±0.54) among those with ‘high’ resilience. Dyadic typologies represented in the study sample were caregiver-oriented (Type II) (n=7, 58%) and collaboratively oriented (Type III) (n=5, 42%). There was no evidence of Type I or IV dyads.

Qualitative findings

Data from this study highlighted two key themes, derived from the concepts of caregivers’ levels of burden and resilience: (1) caregivers’ tendency to lead, and (2) the usefulness of cognitive reframing. Principal exemplars of each theme are outlined below.

Theme 1: Caregivers’ tendency to lead: “I’ve got to do it… we’re all we got is each other.”

Caregiver dominance was a key characteristic of Type II dyads, which were most prevalent in this sample (n=7). In most cases, these caregivers were more heavily involved in patients’ care doing things like managing physical tasks (e.g., bathing, dressing, etc.), coordinating appointments, organizing medications, and providing extra vigilance for safety. One caregiver noted, “ … even putting tags on the car, just stuff that I’m not used to doing, pumping gas. He always took care of all that stuff. Everything he took care of now I have to do. So, it does change your life considerably, very depressing at times…” (Caregiver 8). Taking on additional responsibilities was anecdotally congruent with greater worry and burden compared to Type III caregivers who were part of more collaborative dyads and shared more responsibilities with the patient. Additionally, some patients from Type II dyads reported discussed psychosocial symptoms such as feeling isolated, depressed, and anxious in addition to experiencing physiological symptoms. In these instances, rather than caregivers delegating tasks the patient could handle or assist with, Type II caregivers tended to disregard and step up to lead for both themselves and the patient. As one caregiver stated, “I just do everything, Literally everything. […] And I always only worry that if I can’t do it who will do it?” (Caregiver 1). Some attributed this to stereotypical gender roles and others to long-standing norms within their relationship to the patient. One Type II caregiver went so far as to say, “I’ve got to do it … we’re all we got is each other” (Caregiver 6). This perspective was unlike that of Type III dyads who were more concordant in overall HF management; although, most dyads, in general, noted relying on each other—to differing degrees—for socialization (if nothing else) due to the isolating nature of living with HF. Table 2 further highlights key differences in dyadic congruence by caregivers’ level of burden.

Table 2.

Integrated matrix of experiences of caregiver burden by dyad type.

	Type II dyads: Caregiver-oriented Dyads: 2, 4, 6, 7, 8, 10, 12 Burden Mean = 19.43 (± 13.89) Score range = 2–42	Type III dyads: Collaborative Dyads: 1, 3, 5, 9, 11 Burden Mean = 3.80 (± 4.27) Score range = 0–10
Low caregiver burden Dyads: 1, 3, 4, 7, 9, 11 Mean = 2.50 (± 13.89) Score range = 0–6	“Maybe I, sometimes I think I’m putting so much on her. She needs a knee, replacement and I know it’s bad for her pushing me and she’s in pain, and if she don’t push me I’m in pain, so, you know, I feel for her at most, of the time, but she does it with no complaint.” (Patient 4) “But like I say, I have really learned a lot being a care provider for her., Things that I didn’t know existed have really come along and helped me, to learn about her conditions and stuff.” (Caregiver 7)	“Patient: Most of the time we’re together. We do all the shopping, together. And we do the shopping for her antique business together.”, (Patient 2) “I’ve always felt it’s a balance for your emotional health as well because, you learn, if you’re on a team, you’re not always going to win and you, have to come up with ways of, “I lost and what am I going to do with, that loss?” So, we’ve talked a lot about that and with eating too… there are a lot of things that he doesn’t like that I do, and I don’t cook with any salt although I did before we got together so I’ve tried to blend or connect with my eating habits and what I value with the social emotional and physical … I think we’re trying to complement each other with having those understandings.” (Caregiver 3)
High caregiver burden Dyads: 2, 5, 6, 8, 10, 12 Mean = 23.33 (± 11.08) Score range = 10–42	“…even to have someone come in once a week and, say, bathe her or, something. They said that’s available, but I haven’t found it, you know, all of that. Just to give me a break, you know, because it gets depressing, sometimes, you know, if you’re doing the same tasks over and over and you don’t feel as though you have an outlet.” (Caregiver 2) “It’s been challenging because your whole life changes, and you go from having a social life to basically you don’t have any, and with the day in, day out that we really don’t see anybody, that’s very difficult. As far as caregivinggoes, sometimes you feel like you’re in a world all by yourself, that you’re forgotten about.” (Caregiver 8)	“I notice that when he’s in the overload I have to do a little extra, and then, sometimes I just feel tired. I do it, but I still feel tired.” (Caregiver 5)

Theme 2: Usefulness of cognitive reframing: “The one thing I can control is what I think and feel.”

Results of our analysis revealed a variety of coping strategies utilized by patients and caregivers, across both types, to “bounce back” throughout the unpredictable trajectory that comes with a HF diagnosis. Where dyads diverged was in the degree to which maladaptive coping strategies were discussed, such as lack of sleep, eating unhealthy foods, shopping and avoiding difficult discussions. “Every day is just so stressful that if I started thinking about what the future would be I don’t think I would be able to do it. So, I just don’t really think about it” (Caregiver 10). This avoidant behavior was similarly shared among some Type II dyads, where caregivers’ threshold to promote resilience was likely lower in the setting of having greater demands and responsibilities. Whereas caregivers who had higher reported resilience, of which three were in Type III dyads, were able to position feelings of stress within their broader values to maintain realistic expectations. I go placidly amid the noise and the haste, and I remember what peace there is in silence. Sometimes I feel like everything is out of my locus of control and the one thing I can control is what I think and feel” (Caregiver 3). Discussing stressors in relation to attitudinal values was not uncommon and seemed to help them put things in perspective. One patient shared: “I realized that I had to step up to the plate and make some quick adjustments, even though, you know, it’s hard at the same time. But I had to make it a positive to go ahead and do it, because if I don’t do it, then it’s not going to make the situation better” (Patient 11). Each dyad with ‘high’ resilience discussed previous life experiences where they had to collectively bounce back from difficult challenges, which in some ways may have better positioned them for enduring and successfully managing the patients’ HF than those without those types of experiences and/or lower reported caregiver resilience. Table 3 further highlights key differences in dyadic congruence by caregivers’ level of resilience.

Table 3.

Integrated matrix of experiences of resilience by dyad type.

	Type II dyads: Caregiver-oriented Dyads: 2,4, 6, 7,8,10,12 Resilience Mean = 3.16 (± 1.04), Score range = 1.33 – 4.00	Type III dyads: Collaborative Dyads: 1,3,5,9,11 Resilience, Mean = 4.07 (± 1.36) Score range = 2.00 – 5.00
Low resilience Dyads: 1, 4, 5, 6, 7, 10 Mean = 2.22 (± 0.81) Score range = 1.33 – 3.33	“She needs help with a lot of things, and I try to be there for her. Personally, it really gets hard sometimes, but that’s the way it is.” (Caregiver 6) “Unfortunately, he’s sick and he also is my only outlet. So, I kind of- and I, have been flying off the handle with him so I’m not really handling it very well at all.” (Caregiver 10)	“I do everything. Literally everything. He never took the trash out all his, life. Now, he sits when I’m pumping gas. He’s in the car. So those are, things that I’m used to. And I always only worry that if I can’t do it who, will do it?” (Caregiver 1)
High resilience Dyads: 2, 3, 8, 9, 11, 12 Mean = 4.52 (± 0.54) Score range = 3.83 – 5.00	“I don’t know, we just do it. When I’m faced with something you just do it as, best you can. Some things are harder to get through than other things, and you just, you basically do what you have to do.” (Caregiver 8) “You have to stay calm. Stay focused. Sweat it out and think things out.”, (Patient 12)	“And sometimes you have to go through - well I believe things happen for a, reason and I tell [patient], I wouldn’t be here today had I not gone, through all the discomforts… “ (Caregiver 3) “The problems and stress that I have today really are minimal compared, to the stresses I’ve had in the past, so I just think back to what real problems are, and this is too certainly a real problem, but it’s something that I guess I try to deal with the best way I can.” (Caregiver 9) “You know, we don’t ever see eye-to-eye all the time, but just being able to just listen. You know, sometimes that’s all it takes. You don’t have to respond or say anything.” (CarWEegiver 11)

Discussion

This study was performed to better understand and compare how dyadic congruence is influenced by level of caregiver burden and resilience among HF patient-caregiver dyads. We integrated quantitative scores of caregivers’ burden and resilience with qualitative data to derive two themes which highlighted some nuance in dyads’ ability to manage HF. We also added an additional layer of analysis to these findings by categorizing the sample within an established dyadic typology to support existing gaps in the HF self-care and caregiving literature.

Our findings corroborate that caregiver burden is dependent on a variety of factors, including 1) the nature of the relationship between patient and caregiver, and (2) time elapsed since initial HF diagnosis. Dyads who were diagnosed more recently and/or had more advanced disease, tended to require further assistance, often making caregivers feel the need to be more invested and potentially over-compensate in patients’ care. This can negatively impact the overall dyads’ illness appraisal and care management, resulting in greater stress and burden—particularly for caregivers.^{24, 25}

Further, synthesizing these data using the Heart Failure Care Dyadic Typology revealed unique differences in caregiver burden based by dyad type but proved difficult to apply post-hoc. The codes used to group each dyad—aside from the resilience code itself—had a deficit-based undertone, making it too convenient to overlook important strengths-based aspects of caregiving and HF management that were evident in the data, such as gratification and relationship-building that can come from caring for someone with a chronic illness.

Accordingly, we did identify cognitive reframing as a common strategy used among those dyads where caregivers had higher reported levels of resilience. Reframing is a component of cognitive behavioral therapy and has been proven to be an effective strategy for enhancing emotional regulation.²⁶ Related caregiver interventions have shown reframing can be easily taught and reduces depression and anxiety.^{26, 27} This practice was not highlighted by all caregivers, underscoring variation in peoples’ ability to display resilience depending on multitude of individual, familial, cultural, and systemic factors.^{2, 6, 14} We cannot begin to develop interventions until we fully understand the structural factors that facilitate or inhibit peoples’ ability to overcome obstacles so that we can thoughtfully tailor the support patients and families need to successfully manage chronic illnesses.

Limitations

This study should be interpreted within the context of its limitations. Its impact is ultimately inhibited by our sample demographics. We were also unable to clearly identify dyads which matched Types I and IV, which may indicate a restriction in range of the dyadic experience and should be considered as an avenue for future inquiry. Additionally, saturation is the typical standard for pausing data collection.²⁸ However, we ended data collection based on representation of varied relationships and experiences, so there is potential that people and/or perspectives were not holistically captured in the data-set. Finally, the interviews were completed following a recent hospitalization, which may bias the degree to which dyads’ level of resilience was truly captured. However, the study offers insight on differences between dyad types in terms of challenges faced and overall HF management. Our typing process may also provide a helpful approach for enhancing dyadic understanding in the context of HF for improved intervention development in this field.

Conclusion

Shifting demographics in aging and comorbidity burden require increasing informal caregiver involvement in HF management. Findings from this study highlight the influence of stress, resilience, and relationship factors on dyadic congruence and management of HF. Many dyads showed cognitive reframing can be a promising protective factor for bolstering resilience in the face of uncertainty and challenges. These results have important implications for future intervention development, as researchers, clinicians, and policy-makers envision new strategies to amplify the important role of informal caregivers and as well as their quality of life in contributing to patient care.