Addressing Uncertainty in Informal Familial Caregivers of Stroke Survivors: A Systematic Meta-Ethnography

Background: Informal familial caregivers of stroke survivors experience uncertainty that begins at the time of the stroke event and continues into home-based care. The uncertainty faced by caregivers contributes to poor mental and physical health outcomes. Objective: This review details the factors associated with, impacts of, and coping skills used to manage uncertainty across the caregiving trajectory. By defining uncertainty reduction and tolerance recommendations, this review also builds upon the Stroke Caregiver Readiness Model to improve preparedness following the stroke event. Methods: A meta-ethnographic review was systematically conducted on thirteen qualitative studies with 218 participants from four countries. The Critical Appraisal Skills Programme (CASP) was used to assess study quality. Results: Following the stroke event, caregivers reported a lack of knowing (e.g., about the cause of the stroke event and survivor prognoses) as contributing to post-stroke uncertainty. As a result of this uncertainty, caregivers expressed concerns about their abilities to navigate caregiving responsibilities and how to plan for the future. Longer-term concerns (e.g., managing finances) and feelings of hopelessness occurred after discharge. Still, caregivers identified strategies to manage uncertainty. Caregiver coping skills included present-focused thinking, gratitude, faith, humor, and social support. Conclusions: The uncertainty faced by informal familial caregivers of stroke survivors is pervasive and changes across time. Uncertainty reduction and tolerance interventions can be used to build upon caregiver strengths and promote preparedness across the caregiving trajectory.


Introduction
In the United States alone, more than 795,000 individuals have a stroke each year [1]. Although many individuals now survive a stroke event, strokes represent the leading cause of adult disability [1]. The physical, socioemotional, and functional effects of a stroke are associated with increased personal and public costs [2]. As such, many stroke survivors require at-home caregiving [3], and family members of stroke survivors are often transitioned into caregiving roles. Of the approximately 4.8 million stroke survivors, there are roughly 3.5 million unpaid (i.e., informal) familial caregivers [4].
On average, stroke survivors spend approximately 4.7 days in the hospital following a stroke event [5]. After hospitalization, there are several types of post-acute care options available depending on the rehabilitative needs of the survivor. Inpatient rehabilitation facilities represent a hospital-environment option with continuous care. Such facilities account for approximately 24% of post-acute patients and the typical stay is about one week [6]. Alternatively, skilled nursing facilities, which provide post-acute care to about 30.8% of survivors, allow for a less intensive option that focuses on physical, occupational, and speech rehabilitation. For patients with Medicare, about 45% of individuals are released home immediately after their hospitalization. Home-based rehabilitative care may be available for these stroke survivors; however, options are typically short-term [6]. Regardless of the post-acute care option, informal familial caregivers typically begin providing at-home caregiving within weeks of the stroke event.
The transitory periods from hospital to home-based care have been considered crisis points for stroke survivors and their families [7]. At first, caregivers must face the reality of the stroke event, the unpredictability of survivor prognoses, and plan for rehabilitation services. Then, caregivers must confront the reality that full recovery (e.g., cognitive and functional impairments improved to pre-stroke status) may not be possible for many survivors. Approximately 32-44% of stroke survivors require assistance with activities of daily living three to six months post-stroke [8], and familial caregivers report feeling underprepared to meet patients' basic needs [9]. Once at home, caregivers face the longterm uncertainties associated with the stroke event: worries related to finances, future-plans, and the stability of the survivors and their own health status [7]. As such, informal familial caregivers of stroke survivors must manage both the immediacy of the stroke event and the highly variable nature of patient recovery [10].
The uncertainty endured by informal caregivers of stroke survivors is not without its consequences. Roughly half of post-stroke caregivers experience symptoms of depression and anxiety [11][12][13], for which uncertainty is a transdiagnostic risk factor. Further, heightened levels of distress increase their risks of mortality, strokes, and coronary heart disease [14,15]. As the health of caregivers deteriorates, so does their ability to provide effective care to the stroke survivor [7]. In short, understanding and addressing the uncertainty faced by informal familial caregivers of stroke survivors is necessary to promote the wellbeing of those affected by a stroke event.
To enhance caregiver preparedness, Lutz and colleagues (2016) [16] proposed the Stroke Caregiver Readiness Model. This model first suggests a needs-based assessment for the caregiver and stroke survivor, as well as an evaluation of the caregiver's commitment and capacity to provide care [16]. Then, individualized identification of resources, skills training, home preparation, as well as self-care strategies for the caregiver can be provided. While this framework does not focus on uncertainty as a concept, caregiver uncertainty may undermine readiness through the introduction of doubt and distress. Moreover, understanding caregiver uncertainty as it relates to readiness can provide a deeper understanding of how to best assist caregivers as they transition into their new roles. Accordingly, through a systematic synthesis of previous research, we analyze the factors associated with, impacts of, and coping skills to manage uncertainty for caregivers during the post-stroke crises (i.e., the stroke event and discharge). Conclusions inform intervention recommendations that can further increase caregiver preparedness via uncertainty reduction and management strategies.

Design
A meta-ethnographic approach was used to synthesize the findings of qualitative studies [17]. This approach allows for a systematic, re-interpretation of conceptual data (e.g., themes) across studies to draw conclusions and identify directions for future research [17]. Meta-ethnographies involve the development of higher-order themes (i.e., lines of argument) from a comparative synthesis of participant (i.e., first order) and author (i.e., second order) quotes across qualitative research studies (see Sattar et al., 2021 for greater detail) [18]. Meta-ethnographic syntheses are rigorous and the most utilized approach in health-related research [19]. To ensure systematic reporting of the available data, the enhancing transparency in reporting the synthesis of qualitative research (ENTREQ) checklist [20] and the eMERGe meta-ethnography reporting guidelines [21] were also utilized (see below for details).

Criteria for Including Studies
Included studies needed to be empirical, qualitative, and published in English in a peerreviewed journal. All studies must have recruited a sample of informal (i.e., unpaid) familial caregivers of stroke survivors. Further, studies needed to evaluate or draw conclusions about factors associated with, impacts of, and coping skills to manage uncertainty. Studies were not limited by country of origin. All included studies were published prior to March of 2021.

Systematic Search Strategy
A systematic literature search was conducted by S.E.S. and reviewed by G.T.P. This search aimed to be purposeful (e.g., concept saturation) rather than inclusive of all related research, as consistent with meta-ethnography guidelines [16]. To identify key words and abstracts, the following search terms were utilized: "stroke" OR "post-stroke", AND "caregiver" OR "caretaker" OR "carer", AND "uncertainty" OR "anxiety". Searches were conducted using CINAHL, PubMed, PsycINFO, and Google Scholar.

Identification of Studies
Title and abstracts were screened by S.E.S. Full-text reviews were carried out by G.T.P. and S.E.S. to determine their appropriateness for inclusion. There were no discrepancies regarding the included studies.

Data Extraction
Each study was reviewed independently by G.T.P., B.K., and S.E.S. The following data were extracted from each article by S.E.S.: author names, year of publication, country of data collection, sample size, caregiver relation (e.g., spouse), age, gender, and race of caregivers and stroke survivors. Then, G.T.P. extracted study aims, qualitative methods, sampling methods, data collection methods, lengths of interviews, data analysis methods, and main findings.

Quality Appraisal
The included studies were subjected to the Critical Appraisal Skills Programme (CASP) checklist for Qualitative Studies (CASP) [22]. The CASP checklist includes 10 items that assess research clarity, appropriateness of qualitative methods, research design, recruitment strategy, data collection methods, researcher bias, ethical considerations, rigor of data analysis, reporting of findings, and overall value of the research.

Analysis and Synthesis
The analysis was conducted following established meta-ethnographic guidelines [16]. G.T.P. extracted first-order (participant quotes) and second-order (author interpretations of participant data) constructs, which were taken from studies verbatim to be utilized for analysis and synthesis. Then, G.T.P. grouped studies according to factors associated with, impacts of, and coping skills identified to manage uncertainty by caregivers. Next, G.T.P. and B.K. conducted independent comparisons across studies to develop a list of related themes. Themes were agreed upon and clustered based on shared concepts to create third-order constructs (i.e., the reviewers' higher order interpretations based on analysis). Lastly, G.T.P. and B.K. independently re-coded studies as refuting, dissimilar but not disproving the line of argument, or reciprocation of the third-order constructs.

Positionality Statements
To promote reflexivity and transparency with regards to data interpretation, the authors have self-identified their areas of expertise, relationship to stroke survivorship and/or caregiving, and their demographic identities.
G.T.P. is a PhD Candidate in clinical psychology with expertise in the research and treatment of anxiety disorders and obsessive-compulsive disorder. She was a predoctoral fellow on the 'Stroke and Its Comorbidities' NIH T32 grant at West Virginia University. During her time as a fellow, she evaluated the role of post-stroke anxiety on the well-being of long-term survivors and their families. She is quantitatively trained but has spent the last two years receiving mentorship on qualitative methods. Although she has never been a caregiver, she has had a close family member survive a stroke. B.K. is a PhD Candidate with over seven years of experience in qualitative health research. She does not have research expertise related to stroke but has acted as an informal caregiver for relatives who have experienced stroke, dementia, and physical disability related to limb amputation. S.E.S. is Psychometrist for an outpatient neuropsychology department. She does not have research expertise related to stroke but has had experience assessing stroke patients and caregivers. E.A.C is an assistant professor in public health with six years of qualitative research experience. Her research is primarily focused on mental health and public health interventions. She does not have research expertise related to stroke but has had family members act as informal caregivers of relatives. E.B.E.-C. is a behavioral neuroscientist with expertise in preclinical models of stroke and its comorbidities. She has a close family member who suffered from a stroke. S.A.S. is a clinical psychologist that specializes in anxiety and obsessive-compulsive disorder and has published on the intolerance of uncertainty. Although she has some qualitative research experience, she typically uses quantitative methodology. All authors identify as white, heterosexual, cis-gender females.

Searches
A total of 1894 articles were identified from the initial searches. Duplicates across searches (n = 83) were removed. The remaining 1770 articles were screened for relevant titles and abstracts. Of these articles, 41 were deemed appropriate for full-text reads. Following article review, 13 studies remained for the present review. See Figure 1 for details.  Table 1 shows demographic characteristics and Table 2 shows methodological characteristics for the included studies.   Table 1 shows demographic characteristics and Table 2 shows methodological characteristics for the included studies.  [32] 13 Note. CG = caregiver; SS = stroke survivor; M = mean; SD = standard deviation; N/A = data are not available or not appropriate given study methods.  Note. Post-discharge refers to at-home care.

Quality Appraisal (CASP)
For each study, G.T.P. and B.K. independently coded each item of the checklist as YES (i.e., manuscript allows for a conclusion that criteria are met), NO (i.e., manuscript allows for a conclusion that criteria are not met), or UNSURE (i.e., manuscript does not allow for a conclusion to be drawn). Then, G.T.P. and B.K. reviewed coding. Agreement between coders prior to review was 90% and was 100% following discussions. Across the included studies, quality was strong. All studies provided enough information to draw conclusions on at least seven of the ten CASP items. Studies most often did not provide enough detail to determine the justification of their research design or address research bias and the relationship between researchers and participants. Still, the overall quality of the studies allowed for all to be included and weighed equally in the present review. See Table 3 for agreed upon quality appraisal ratings.
Note. Q1-Q10 indicates CASP question number. Coding is as follows: Y = Yes, U = Unsure (cannot be determined by information presented), N = No. Table 4 shows the key concepts derived from constant comparison. Provided the significant overlap across study findings, a subset of representative quotes (chosen for their descriptiveness) is used to demonstrate constructs.  [27] Availability of Support "The lack of certainty about the availability of a home care package created additional anxiety about securing the best possible help. Some carers felt that care needs were not adequately assessed, often after numerous calls to social services, but there was also concern about when promised home help would materialize." "We were meant to have a care assistant to help with personal care but nothing. It was stressful not knowing if home care services would be provided. For the first few weeks there was no grab rail. His (spouse) left side is quite bad so he could not wash or get dressed himself unaided. They shouldn't offer services that they couldn't deliver. It's very depressing." Katbamna    Note. First order = Participant quote from article; Second order = Author interpretation of participant quote from article; Third order = Theme identified in current review.

Factors Associated with Uncertainty
Across studies, the unexpected nature of the stroke event served as a catalyst for caregiver uncertainty. Following the event, caregivers detailed their lack of knowledge as maintaining their uncertainty. Specifically, they reported an inability to "know" the cause of the event [26,30,34], the stroke survivors' prognosis [25,27,29,[31][32][33][34][35], the safety of the survivor if left alone [24,27,29,30,36], the support options available [27,29,35,36], and how to provide care to the survivor [27,30,[34][35][36]. While only expressed by some caregivers in one study [35], the loss of intimacy with and lack of empathy from survivors also contributed to uncertainty.
Of the factors associated with uncertainty, none were associated with caregiver experience (i.e., prior caregiving experience or novice caregivers). Still, uncertainty about the cause of the stroke event and survivor prognoses persisted from acute settings to postdischarge. Longer-term worries (e.g., concerns about the safety of the survivor), however, were only expressed post-discharge.

Impacts of Uncertainty
As a result of this uncertainty, caregivers reported worries about their abilities to successfully navigate their newfound responsibilities. Specifically, they noted concerns about their abilities to manage finances [24,35,36], effectively plan for the future [24,26,29], identify others who would assume caregiving responsibilities for the survivor if they themselves fell ill [25,27,35], and ways to cope with the stroke event [26,35]. The lack of clarity surrounding these concerns resulted in feelings of hopelessness [24,25,27].
Of the impacts of uncertainty, none were associated with caregiver experience. Concerns about caregivers' abilities to plan for the future were noted specifically in caregivers whose stroke survivors were in rehabilitative settings. Additionally, worries associated with finances, caring for the survivor if they themselves fell ill, and feelings of hopelessness were specific to caregivers providing care-post-discharge. Across both acute and home-based settings, caregivers reported concerns about how to cope with the stroke event, pointing to the longstanding effects of uncertainty.
While several caregivers reported using humor to cope with distress [29,31,34], they also found comfort in reliving memories from the past [32] and continuing with prior hobbies [31]. Further, while most caregivers reported social support as a coping skill [26,29,31,32], a few caregivers noted the need to "turn inward", and minimize outside contact [32]. Although caregiver experience was not found to be associated with any notable benefits across studies, one study found that experienced caregivers expressed less uncertainty and greater confidence in their abilities to cope with the stroke event compared with newer caregivers [29].
Lastly, effective coping skills were identified by caregivers across the caregiving trajectory (acute hospitalization through post-discharge). Across all stages of post-stroke caregiving, present-focused thinking, gratitude, faith, humor, and social support were noted as effective coping strategies by caregivers. For caregivers whose stroke survivors were still in acute care, continuation of prior hobbies was identified as especially important. Some strategies, however, appeared to be more useful following discharge. That is, acceptance, increased self-reliance, information seeking, routine establishment, watchful monitoring of the stroke survivor, embracing their duty, and reliving memories were all specific to caregivers providing home-based care.

Discussion
The present meta-ethnographic review details factors related to, impacts of, and coping strategies to manage uncertainty in informal familial caregivers of stroke survivors. For caregivers, the stroke event served as the catalyst for uncertainty. Then, in accordance with the stroke crises [16], caregivers reported concerns about coping with the stroke event and survivors' abilities to recover while in inpatient settings. Concerns about how to cope with the stroke event also persisted into the delivery of home-based care. Our findings suggest that caregiver concerns about the safety of the survivor, available support, and abilities to provide care, as well as worries about finances, caring for the survivor if they themselves fell ill, and feelings of hopelessness were specific to providing care post-discharge. Of note, as caregivers progress from the crisis of the stroke event to the crisis of discharge, their focus shifted more towards attending to the needs of survivors and away from caring for themselves [7]. Encouragingly, however, caregivers were able to identify coping strategies they perceived as helpful to minimize the negative effects of uncertainty.
In all, the course of uncertainty experienced by informal familial caregivers appears to parallel the path to recovery for stroke survivors: both trajectories are influenced by various and sometimes unpredictable factors which change across time. This long-term unpredictability and the constantly changing nature of caregiving may explain why prior caregiving experience was not associated with uncertainty reduction or improved uncertainty tolerance across studies. Moreover, although factors contributing to uncertainty appeared to change across time (e.g., concerns about how to provide care for the stroke survivor to concerns about who would care for the stroke survivor if the caregiver fell ill), the experience of uncertainty itself remained stable throughout the caregiving trajectory. Thus, in addition to supporting the changing needs of stroke survivors, there is also a need to improve the perceived preparedness of informal familial caregivers to reduce the associated uncertainty and distress of caregiving.
In the present review, we provide a strong foundation for understanding informal familial caregiver uncertainty as it relates to stroke survivors. The quality of the included studies (as evidenced by the CASP review) and the rigor with which the review was conducted (use of eMERGe guidelines and ENTREQ checklist) enhance the trustworthiness of the abovementioned conclusions. Additionally, study results were largely parsimonious, suggesting a shared experience of uncertainty across countries of origin, levels of post-acute care, and experiences of caregiving during the decades which the included research was conducted. Accordingly, to assist caregivers in the management of their uncertainty across the caregiving trajectory, we propose the use of interventions grounded in research that build upon caregiver strengths and can be implemented shortly following a stroke event. Such interventions also expand on recommendations by Lutz and colleagues' Stroke Caregiver Readiness Model (2016; e.g., targeted skills-based trainings, as well as family counseling) to further support caregiver readiness through uncertainty reduction and tolerance. Educational and skills-based interventions have aimed at improving caregiver knowledge and perceived competence by providing information on stroke events, practical skills for caregiving, and general coping skills (e.g., emotion regulation) [37]. These interventions, which are often delivered by nurses, social workers, or psychologists, have demonstrated beneficial effects on stroke caregivers' mental health [37][38][39][40][41]. While caregivers in the current review reported a desire to "know" (e.g., about the stroke event, caregiving skills, and survivor prognoses), they also identified concerns about the accessibility of information provided to them by hospital staff. To address this concern and better support the needs of the caregivers, we suggest the addition of education and skills-focused peer-support groups. Peer-support groups permit those with similar experiences to provide practical and emotional support during the management of health conditions [42]. Peer support groups for caregivers of stroke survivors have led to an increased sense of community and awareness, as well as perceived empowerment for caregivers [43,44]. Accordingly, trained peer-support specialists (i.e., group leaders with personal experience) [45] may also be able to increase the accessibility of educational materials and skills-based training. By delivering such interventions across the caregiving trajectory (e.g., in-person groups during acute care and video groups during home-based care), these interventions may help to reduce the negative impact of prognostic expectations regarding the stroke survivor's recovery. That is, by providing accessible information via a community-driven format, caregivers may be able to better prepare for the longer-term outcomes associated with unpredictable course of this illness.

Uncertainty Tolerance: Cognitive Behavioral Strategies
Caregivers in the current review identified numerous strategies to cope with the uncertainty brought about by the stroke event. In addition to these strategies, psychotherapeutic interventions such as cognitive behavioral therapy (CBT) have been effective in reducing symptoms of anxiety and depression in caregivers of stroke survivors [46]. Specifically, skills such as cognitive restructuring (i.e., modifying existing beliefs to be more realistic and helpful) can help challenge caregivers' beliefs about their ability to tolerate uncertainty (e.g., "If I don't know their prognosis, I won't be able to manage my distress") or their interpretation of uncertain events as negative (e.g., "If I don't know their prognosis, then it must be poor"). Likewise, cognitive restructuring can be used to formulate more neutral or positive beliefs about caregiver preparedness during transitory periods of care (e.g., "Although I have not had to be a caretaker before, I have skills that will allow me to be successful in this role"). In addition, behavioral strategies based in acceptance and commitment therapy (ACT), such as values-identification and goal setting, may help caregivers feel greater control over their caregiving experience [47]. Given that caregivers identified ACT-based skills (i.e., present-focused thinking and engaging in meaningful activities) as strategies to cope with uncertainty, additional acceptance-based skills may enhance caregivers' abilities to shift their focus from what they cannot control to what they can control [7,16]. By attending individual psychotherapy or incorporating CBT and ACT-based skills into the abovementioned peer support groups, caregivers may experience enhanced readiness and preparedness in navigating their caregiving trajectories.

Limitations of the Review
While this review had numerous strengths, limitations must also be considered. First, within the studies reviewed, researcher bias was poorly assessed and addressed by study authors. Accordingly, the interpretation of participants' data may have been influenced by study authors' beliefs. Further, the authors of this review want to acknowledge their own potential bias (including personal experiences of informal familial caregiving) on their interpretation and synthesis of the data. Moreover, the search strategy utilized was purposeful and aimed to reach concept saturation as it related to informal familial caregiver uncertainty and related coping strategies to manage uncertainty. We acknowledge, however, that studies on relevant topics (e.g., caregiver burden) may have been missed due to the specificity of our search terms. Lastly, the authors want to acknowledge that the recommendations put forth in this paper may not be inherently accessible to all informal familial caregivers of stroke survivors. Historically, individuals who have been marginalized within the healthcare system (e.g., racial/ethnic minorities, low-income individuals, and those in LGBTQIA+ communities) have been burdened with limited access to resources and poorer quality care. Furthermore, given that the samples in the studies reviewed were largely white, the abovementioned findings about uncertainty as well as the intervention recommendations may not be generalizable. It is our strong recommendation that any future interventions related to caregiver uncertainty are developed with, evaluated by, and targeted to address the specific needs of marginalized communities.

Conclusions
This review provides a detailed account of uncertainty experienced by informal familial caregivers of stroke survivors to demonstrate how such uncertainty undermines readiness and preparedness across the caregiving trajectory. Findings from this review demonstrate the impacts of and coping strategies to manage uncertainty across the caregiving trajectory, as well as intervention recommendations that build upon caregiver strengths and the Stroke Caregiver Readiness Model to help caregivers better reduce and tolerate uncertainty. Subsequent assessments of these recommendations are warranted to determine their effectiveness in promoting caregiver well-being.
Funding: This work was supported by the following grants: National Institute of Aging (T32 AG052375), National Institute of Mental Health (K01 MH117343), West Virginia Stroke CoBRE (P20 GM109098), National Institute of Neurological Disorders and Stroke (R01 NS099918-03), National Institute of General Medical Sciences (U54 GM104942), and the West Virginia University Research Office.