“Factors Associated with Depressive Symptoms of Carepartners of Stroke Survivors after Discharge from Rehabilitation Therapy”

Abstract

Background:

The long-term consequences of stroke affect both the carepartner (CP) and stroke survivor (SS). Understanding the effects of informal caregiving that may influence the ability of the family to carry over therapeutic activities in the home environment is critical for family-centered care.

Objective:

This study examined the relationship of CP and SS factors associated with CP depressive symptoms to gain insights into CP needs that may occur after formal rehabilitation therapy has ended for SS with upper extremity deficits.

Methods:

This correlational study used baseline data of 32 dyads of family CP and SS with upper extremity impairment who had completed rehabilitation therapy and were enrolled in a pilot study of a web-based CP-integrated rehabilitation program. Data using standard questionnaires for CP factors and SS memory and behavior problems and an objective assessment of SS upper extremity function were obtained. Data analysis included descriptive statistics and Pearson product moment correlations.

Results:

CPs were female (62.5%), White (61.29%), and spouses (68.75%). CPs reported mild-moderate depressive symptoms (M = 9.5±8.3), and a majority had some degree of family conflict. Higher CP depressive symptoms were related to worse life changes (r = −0.41, p = .02), greater fatigue (r=0.50, p=.004), less effective family functioning (r=0.46, p=.01), less autonomy support to SS (r = −0.42, p = .02), and more SS memory and behavior problems (r = 0.45, p = .01). Only CP fatigue was related to SS upper extremity function.

Conclusions:

Negative impacts of caregiving were found in this group of relatively high physically functioning SS which may hinder CP from providing optimal support for SS. Addressing CP needs including education regarding depression, fatigue, SS memory and behavior problems, and family functioning while SS is receiving rehabilitation therapy may be important considerations to help facilitate the CP to support the SS in carrying over therapeutic activities in the home environment.

INTRODUCTION

Stroke recovery is complex and challenging for both the stroke survivor (SS) and the carepartner (CP). Approximately 80% of SSs are discharged home, and a majority rely on the care of family members who frequently feel unprepared for this role.^{,1, 2} Informal caregiving is a critical component in stroke care, yet these responsibilities can result in high levels of burden, stress, and negative effect on CP quality of life.^3–9 Depression is not uncommon, occurring in nearly half (41%) of family CPs,¹⁰ and can adversely affect SS response to rehabilitation.^{5, 11, 12} Individuals who develop depression as they enter the caregiving role often remained depressed,¹³ and experiences of CP stress and perceived burden are worsened by depressive symptoms.^14–16

Recognizing the critical role of CP in stroke recovery, a valuable opportunity for continued rehabilitation is missed if families do not receive support in preparation to address SS needs after formal rehabilitation has ended. The effects of informal caregiving in stroke have been primarily a focus of nursing research^{1, 17–22}, with little attention on the role of the CP in caring for SS with specific stroke deficits such as upper extremity impairment. After traditional inpatient and outpatient therapies are completed, SS are released with a rehabilitation plan to be continued at home. The CP often assumes the role of assisting the SS in continued efforts to improve their functional abilities and must continue therapeutic activities without guidance from rehabilitation professionals for challenges that frequently arise. Taking on these responsibilities may contribute to caregiver burden.

These detrimental effects of caregiving may result from a variety of factors, including inadequate knowledge and skills training to care for the SS, lack of family support, and SS characteristics (such as mood changes, one of the most stressful problems identified¹²). Memory and behavior changes after stroke can negatively affect CP outcomes,^{23, 24} and family conflict worsens the impact of SS memory and behavior problems on CP mental health.^{9, 25, 26} Unmet needs and inadequate training are cited by nearly 75% of CP of SS within the first few months after stroke,^{27, 28} highlighting an important gap in family education that can be addressed by all members of the rehabilitation team.

While family involvement can positively impact SS recovery,^29–33 family relationships can be negatively impacted after stroke.³⁴ Family functioning can impact how CPs adjust to caregiving roles,^{15, 35} and CPs report worse mental health in families with ineffective family functioning.^{9, 34} These findings primarily have been identified earlier post stroke with less focus on whether family function changes as CP and SS adjust to life after formal rehabilitation support has ended. CP adverse effects may be related to tensions, miscommunications and frustrations regarding loss of SS’s ability following stroke. In particular, CP frequently struggle with knowing how to balance encouragement of the SS to perform rehabilitation activities and feelings of “nagging”, identifying family conflict when trying to promote adherence to the prescribed home therapy exercise programs.³⁶ Skills training in effective communication strategies is one opportunity for family education during rehabilitation therapy. An autonomy supportive environment that emphasizes problem solving, choice, empathy and reduction of controlling language (Self-Determination Theory³⁷), can be motivating, support individual autonomy and confidence, and has been shown to influence SS motor skill acquisition^{38, 39} Autonomy supportive language is a critical component motor learning theory in stroke rehabilitation;⁴⁰ however, we are unaware of any studies to date that have explored CP characteristics that may be associated with the CP’s ability to provide autonomy supportive environment for SS to continue rehabilitation activities at home.

Recognizing the interdependence of CP and SS health during recovery is an important step to develop effective family centered treatments that may benefit dyads after discharge from therapy and over the long-term stroke recovery. Identifying ways rehabilitation therapists can provide anticipatory support to guide caregivers’ adjustments to their roles can be a key component to improve recovery and quality of life for CPs and SSs.

The purpose of this study was to examine the influence of CP factors (life changes related to stroke, caregiving strain, family conflict surrounding stroke recovery, fatigue, and autonomy support), family functioning, and SS characteristics (upper extremity function, CP reported memory and behavior problems) on depressive symptoms in CPs to gain insights into CP needs that may occur after formal rehabilitation therapy has ended.

METHODS

Sample and setting

Individuals with upper extremity impairment after stroke who had completed rehabilitation (inpatient and outpatient therapies) were consecutively recruited with their CPs from a database of individuals from a large urban rehabilitation center. Baseline data were used from SS and CP recruited to be part of a randomized clinical trial piloting a web-based CP-integrated home-based rehabilitation program. Inclusion criteria for SS included the ability to initiate wrist and finger extension of the weaker hand, a Mini-mental⁴¹ test score ≥24, completion of traditional rehabilitation therapy and the presence of a CP (defined as a family member dwelling in the household and self-identified as the primary caregiver). CPs were required to be greater than 18 years of age, and able to read and write English. There were 168 dyads screened, with 104 excluded due to not meeting criteria. The primary reasons for exclusion (n=69 of 104, 66%) being either too little upper extremity movement or too high - no upper extremity impairments. Of the 64 eligible, 50% (n=32) declined to participate, resulting in 32 dyads consented and enrolled.

Ethical approval was obtained by the primary author’s University Institutional Review Board (IRB00070957), and all participants provided written consent. The full-randomized study protocol for the main study is registered on clinicaltrials.gov (NCT02703532). This manuscript conforms to STROBE Guidelines.

Study Design

A correlational, cross-sectional design was used to examine factors associated with CP outcomes. Data from this non-random sample are from baseline data and therefore not affected by the study intervention. Study questionnaires were administered in-person to the CP and the SS in the clinic and the SS upper extremity measure was administered in the clinic laboratory by a standardized research evaluator.

Measures

Carepartner assessment

Depressive symptoms were measured using the Center for Epidemiologic Studies Depression Scale (CES-D)⁴², a 20-item, Likert-type scale extensively used to measure depression in numerous healthy⁴³ and physically ill populations, including persons with stroke,^{44, 45} and caregivers.⁴⁶ Scores range from 0–60, with scores of ≥16 an indicator of potential clinical depression. This scale has been used in the general population, has documented acceptable validity and internal consistency reliability with alpha coefficient of .86,⁴³ and has good inter-rater reliability and validity in patients with stroke.⁴⁵

CP life changes related to caregiving were measured by the revised Bakas Caregiver Outcome Scale (BCOS),⁴⁷ a 15-items scale (total BCOS score ranging from 15–105) with higher scores representing more positive caregiving outcomes since the stroke. The dimensions of life change included physical and emotional health, family relationships, and financial well-being. The revised BOCS has evidence of adequate reliability and construct validity.⁴⁷

Fatigue was measured using 22 of the Likert-type items of the Piper Fatigue Scale (PFS)^{48, 49} to address four subjective dimensions of fatigue (behavior, affective, cognitive/mood, and sensory); higher scores indicated higher fatigue.⁴⁹ The PFS has evidence of reliability and validity in a caregiving sample.⁴⁹

CP strain was measured by the Caregiver Strain Index (CSI).^{50, 51} This 13-item dichotomously scored (yes/no) questionnaire has total scores range from 0–13 with higher scores indicating more strain. A total score of ≥ 7 represents high level of strain. The CSI has good reproducibility (0.93; 95% confidence interval 0.84–0.97).⁵² However, during the administration of this study, only 12 items were used (omitted question: “There have been other demands on my time”) resulting in potential total scores ranging from 0–12.

Perceived autonomy support provided by the CP to the person with stroke was measured by the Family Care Climate Questionnaire-Carepartner (FCCQ-CP) version.⁵³ Clark & Dunbar adapted this measure to use in persons with heart failure from the Health Care Climate Questionnaire originally developed to measure autonomy support with health care providers and patients.⁵⁴ Item stems were adapted to reflect autonomy support with rehabilitation activities. An example of an adapted item is “I listen to how my family member would like to do things related to rehabilitation activities.” Rehabilitation activities were defined in the instructions as the CP’s experience with the SS performing recommended exercises and activities for recovery from stroke. The FCCQ-CP includes 14 items with a response scale of 1 “not true at all” to 7 “very true.” Item responses are averaged with total possible scores of 1 to 7 with higher scores representing more autonomy support. The FFCQ-CP has evidence of adequate internal consistency reliability and construct validity with factor analysis.⁵³

Family conflict was measured by the Family Caregiver Conflict Scale (FCCS),⁵⁵ which addresses conflict about stroke recovery. The FCCS is a 15-item, Likert-type scale with a response format of 1 “not true at all” to 7 “very true”. For a total score, item responses are summed (possible total scores 15–105) with higher scores indicating higher family conflict. The FCCS scale has demonstrated content validity, construct validity with factor analysis, and adequate reliability with Cronbach’s alpha coefficient of .93.⁵⁵

Family functioning was measured using the McMaster Family Assessment Device (FAD)⁵⁶ - General Family Functioning (GF) 12-item subscale. Using 1- “strongly agree” to 4 - “strongly disagree” response format, mean scores are calculated for total scores ranging from 1 to 4. Higher scores (≥2) indicate ineffective family functioning. Concurrent and predictive validity, internal consistency reliability, sensitivity and specificity have been demonstrated in various samples for the FAD⁵⁶ and the measure has been used in stroke studies.^{25, 57}

Stroke survivor assessment

The SS’s memory and behavior changes were measured by CP report with the revised Memory and Behavior Problems Checklist (MBPS),^{25, 58} a 19-item scale that measured the number and frequency of memory and behavior changes and how bothersome/upsetting the behavior was to the CP. Only the number of memory and behavior changes experienced using a dichotomous scale (0/1) with a possible total score of 0 to 19 are reported here. Reported reliability has been adequate with a Cronbach’s alpha of .73.²⁵

The Wolf Motor Function Test (WMFT)⁵⁹ is a functional assessment of the UE, using 15 speed and 2 strength measures. Frequently used in CIMT research studies, the WMFT has been shown to be both reliable (internal consistency Cronbach’s alpha of .92, inter-rater reliability ICC=.97) and valid in the stroke population.^{59, 60}

Statistical Analyses

All statistical analyses were performed using SAS (version 9.4). To characterize the study sample and major study variables, descriptive statistics were used. Data were examined for outliers and distributional characteristics using scatter plots and descriptive statistics. Associations were examined among variables using Pearson product moment correlation coefficients, with significance level set at p < .05. No adjustments were made for multiple testing. Sample size was fixed from the sample size of the clinical trial.

RESULTS

Participant Characteristics

The SS (N=32) were primarily male, White, with a mean age 60.6 (SD=14.2) years, and median time post stroke of more than a year ranging from 1.8–50 months post stroke with the majority having the dominant hand affected (Table 1). The CP were primarily female, White, mean age was 57.8 (SD=13.0) years. Most CPs were spouses/significant others, 43.8% were partially or fully employed, all had some college experience, and 67.7% had an annual household income greater than $50,000. Only two CP were receiving anti-depressant medication at baseline.

Table 1:

Sociodemographic Profile for Carepartners (N=32) and Stroke Survivors (N=32)

	Carepartner	Stroke Survivor
Age, y M (SD)	57.8 (13.0), range 27–75	60.6 (14.2), range 19–76
Sex: n (%)
Female	20 (62.5)	14 (43.75)
Male	12 (37.5)	18 (56.25)
Relationship, n (%)*
Spouse or significant other	22 (68.75)
Child	7 (21.88)
Parent	2 (6.25)
Other	1 (3.13)
Work Status, n (%)*
Full	13 (40.63)
part-time	1 (3.13)
unemployed	10 (31.25)
Retired	1 (3.13)
Homemaker	7 (21.88)
Education level, n (%)†
Post-graduate	14 (46.67)	7 (21.88)
4 yrs college	8 (26.67)	8 (25.00)
1–3 yrs college	8 (26.67)	11(34.38)
High school grad	0 (0)	6(18.75)
Income, n (%)‡
10,000–20,000	---------------	1 (3.13)
21,000–30,000	3 (9.86)	1 (3.13)
31,000–40,000	-------------	3 (9.38)
41,000–50,000	2 (6.45)	3 (9.38)
>50,000	21 (67.74)	13 (40.63)
Disability Supplement	---------------	1 (3.13)
Prefer Not to Answer	5 (16.13)	10 (31.25)
Race/Ethnicity, n (%)
Black or African American	12 (38.71)	13 (40.63)
White, Non-Hispanic	19 (61.29)	18 (56.25)
Asian, Pacific Islander	-------------	1 (3.13)
Hispanic, Latino	1 (3.13)
Time post stroke, mo Mean, (SD), Median	-------------	15.8 (13.3), 9.5, range 1.8–50
Hand affected-dominant, n (%)	-------------	18 (56.25)

^*Caregiver only variables;

^†missing data (n=30);

^‡missing data (n=31)

Descriptive statistics and Cronbach’s alpha values for instruments for CP and SS variables are provided in Table 2. All instruments had adequate reliability coefficients. Mild to moderate levels of CP depressive symptoms were found (9.5 ± 8.3), although 16% scored higher than the cut-off score of 16 indicating possible clinical depression. On average, CP caregiving outcomes were below cut off score of 60 on BCOS (56.7 ± 10.5), indicating a worsening of life changes as a result of providing care. The majority of CPs indicated some level of family conflict, and 47% were experiencing moderate to severe fatigue (PFS levels >3). Overall, autonomy support was high, and average FAD-GF scores indicated healthy family functioning; although, 28% of CPs had scores above the cut-off point (FAD-GF >2) for ineffective family functioning. SS upper extremity function (WMFT) indicate relatively high upper extremity function and experienced on average 7 different memory and behavior problems (MBPS; Table 2).

Table 2.

Descriptive Statistics of Major Study Variables for Carepartners (CP) and Stroke Survivors (SS) at Baseline (N=32).

Variable	Baseline Mean ± SD	Actual Minimum/Maximum	Possible Minimum/Maximum	Cronbach’s alpha
CP Depressive Symptoms (CES-D)	9.5 ± 8.3	0–40.0	0–60	.89
CP Life Change (BCOS)	56.7 ± 10.5	32.0–81.0	15–105	.88
CP Fatigue (PFS)	3.6 ± 1.9	0.2–7.0	0–10	.96
CP Strain (CSI)	5.3 ± 3.1	0–11.0	0–12	.79
CP Autonomy Support (FCCQ-CP)	6.1 ± 0.7	4.2–7.0	1–7	.80
Family Conflict about Recovery (FCCS)	25.6 ± 10.8	15.0–57.0	15–105	.85
Family Function (FAD-GF)	1.7 ± 0.4	1.0–2.6	1–4	.92
SS Memory & Behavior Problems (MBPS)	7.3 ± 4.1	1–16	0–19	.86
Upper Extremity Motor Function (WMFT, sec)	11.6 ± 13.3	2.2–59.3	-----	.76

CES-D = Center for Epidemiological Studies Depression Scale; BCOS = Bakas Outcomes of Caregiving Scale; PFS = Piper Fatigue Scale, CSI = Caregiver Strain Index (12-item); FCCQ-PC = Carepartner version of the Family Care Climate Questionnaire; FCCS = Family Conflict Surrounding Stroke; FAD-GF – General Functioning scale of the Family Assessment Device; MBPS = Revised Memory and Behavior Scale (Carepartners’ report of Stroke Survivors’ Memory and Behavior Problems); WMFT = Wolf Motor Function Test

Analyses were conducted to assess which factors demonstrated relationships with the CP depressive symptoms (CES-D). All correlation coefficients are shown in Table 3. Higher CP depressive symptoms were related to worse life changes since stroke, higher fatigue, less effective family functioning, less autonomy support provided to SS for rehabilitation activities, and more SS memory and behavior problems. Higher family conflict around stroke recovery was related to less CP autonomy support provided to the SS. The only CP variable related to SS upper extremity function was fatigue, with higher fatigue correlated with lower (faster) WMFT scores.

Table 3:

Relationships Among CarePartner (CP) and Stroke Survivor (SS) Variables N=32)

Variable	1	2	3	4	5	6	7	8
1. CP Depression (CES-D)
2. CP Life Changes (BOCS)	−0.41
3. CP Fatigue (PFS)	0.50	−0.46
4. CP Strain (CSI)	0.31	−0.45	0.57
5. CP Autonomy support (FCCQ-CP)	−0.42	0.05	−0.25	−0.23
6. CP Conflict about stroke recovery (FCCS)	0.06	0.07	0.38	0.43	−0.52
7. CP Family function (FAD-GF)	0.46	−0.21	0.39	0.47	−0.59	0.41
8. SS Memory & Behavior problems (MBPS)	0.45	−0.33	0.14	0.45	−0.14	0.21	0.41
9. SS Wolf Motor Function Test (WMFT)	−0.30	0.24	−0.50	−0.06	0.23	0.04	−0.17	0.24

Note: Pearson Correlation Coefficients; bolded correlation coefficients significant at p < .05;

CES-D = Center for Epidemiological Studies Depression Scale; BCOS = Bakas Outcomes of Caregiving Scale; PFS = Piper Fatigue Scale, CSI = Caregiver Strain Index; FCCQ-PC = Carepartner version of the Family Care Climate Questionnaire; FCCS = Family Conflict Surrounding Stroke; FAD-GF – General Functioning scale of the Family Assessment Device; MBPS = Revised Memory and Behavior Scale (Carepartners’ report of Stroke Survivors’ Memory and Behavior Problems); WMFT = Wolf Motor Function Test

DISCUSSION

To gain a greater understanding of CP factors that may influence the ability of the CP to support the SS in carrying over therapeutic activities in the home environment, this study examined the impacts of caregiving of SSs with persistent upper extremity impairment during the stroke recovery period when formal rehabilitation therapy has ended. On average, CPs in our study reported mild-moderate depressive symptoms. Despite the relatively mild to moderate levels, worse CP depressive symptoms were found to be associated with multiple variables, including higher strain and fatigue, less effective family functioning, and worse life changes after stroke, supporting the importance of addressing family CP needs and concerns.^{2, 61} It is unclear if depressive symptoms result from the higher strain and fatigue or if depressive symptoms affect CPs’ perception of the situation. Regardless, the intersections of CP and SS health are important factors to consider as potential moderators in long-term effects of caregiving post-stroke.

Levels of CP depressive symptoms in our study are similar to findings by Haley, et. al.,¹² in a cohort of individuals caring for community dwelling chronic stroke survivors; however, are lower than previously reported studies that have estimated the incidence of CP depression to reach levels of 30–52%^{2, 13, 62} and a recent study⁶³ of family CPs five years post stroke. Previous research has shown that CPs with depression can have a negative impact on SS response to rehabilitation.^{5, 12, 17, 64} Our results indicated that higher CP depressive symptoms were related to worse scores of the CP’s ability to provide an autonomy supportive environment for the SS in performing rehabilitation activities at home. Autonomy-supportive contexts in rehabilitation activities are those in which the CP provides the SS with choice, reduces controlling language and acknowledges the others feelings and perspectives to promote collaborative goal setting, improve SS motivation and exercise adherence.⁶⁵ A limited ability to create an autonomy supportive environment for the SS may be a potential mechanism responsible for the poorer SS outcomes found when CP experience depression. Preliminary data from a small feasibility study from our lab suggested a CP focused intervention designed to improve CP skills and promote an autonomy-supportive home environment paired with intensive upper extremity rehabilitation may help reduce CP depressive symptoms while improving SS upper extremity function.²⁹ Clinicians may need to provide more anticipatory guidance to families as formal rehabilitation ends and CPs and SSs continue their efforts in the home. In particular, preparing CPs with ways they can be a postive motivator by using autonomy supportive strategies, such as giving examples or scenerios that CPs may face when at home, may be helpful.

Memory and behavior changes (such as sadness, agitation or interrupting the CP) are common after stroke, and can negatively impact both the SS and CP. Clark and colleagues²⁵ suggested that family members often struggle to address these symptoms, with limited understanding of why they occur after stroke, or how they can impair the SS rehabilitation process. Our results showed a relationship between a higher frequency of SS memory and behavior changes and higher levels of CP depressive symptoms, similar to findings by Gonzalez and Bakas.⁶⁶ Furthermore, this sample of CPs reported worsening of life changes, despite caring for a higher functioning SS well after discharge from therapy, and had levels similar to studies in acute stroke,^{3, 15, 67} indicating that over the course of recovery, SS and CPs may continue to feel a negative impact that does not improve with time. Worse CP life changes were related to more SS memory and behavior problems, as well as higher fatigue and higher strain. Given this potential influence of SS memory and behavior changes on CP mental health, this presents another opportunity for rehabilitation specialists to prepare CPs for the continued stroke recovery. As part of family education, clinicians can provide list of common challenges CPs face as the SS recovers, such as fatigue and strain that may occur with the role of caregiving, even if the SS has good physical function. Reviewing a list of potential SS memory and behavior problems that may become apparent when at home along with potential strategies to address these problems may be helpful. Additional interventions include providing educational information describing signs and symptoms of depression for SSs and CPs with community referral options; encouraging the CP to keep a written record of the SS memory and behavior problems to take for a discussion with the SS health care provider; or providing memory notebook of rehabilitation activities for the SS.

Although most of the SSs were months past the initial stroke event and had completed formal rehabilitation, the majority of CPs may have continued to experience family conflict surrounding stroke recovery, potentially influencing motivation and adherence to therapy. Higher levels of CP strain and fatigue were related to higher levels of conflict around stroke recovery, but not CP depressive symptoms. Although the importance of following a home exercise program for continuing therapeutic activities after discharge from therapy is heavily emphasized in rehabilitation, frequently family members are unsure how much to support or encourage their loved one to adhere to their home program, citing their efforts as being a source of family conflict.³⁶ However, Harris and colleagues³³ found SS with CP support were more likely to spend more time participating in exercise and the presence of a CP improved upper extremity function as compared to those without CP support. Consequently, teaching CPs how to provide positive support to reduce conflict and improve psychological and health outcomes for both CP and SS may enhance current neurorehabilitation interventions. This approach may include providing tips to address potential SS frustration in practicing home tasks and improve adherence to a home exercise plan. Sample scenarios for CPs to learn how to use concepts of autonomy support, such as ways of fostering empathy (e.g., the CPs’ using their non-dominant hand during activities to understand some of SS challenges), problem solving (guidance for adaptation of functional activities at home to increase success or challenge), instruction in the use of non-controlling language with role playing situations and the importance of creating choice in activities may be helpful.

In general, this cohort reported relatively overall healthy family functioning, with levels slightly better than reported by Clark and Smith⁶⁸ at 12 months after hospital discharge. Family functioning has been identified as an important variable post stroke impacting satisfaction with stroke recovery^{68, 69}, CP depression³⁴, and SS quality of life⁷⁰; however, rarely is this type of family dynamics assessment included in physical rehabilitation intervention studies, despite being recognized as a fundamental component of post-stroke care and part of clinical practice guidelines². Rochette, et. al.,⁷¹ investigated family-related rehabilitation practice post-stroke in Canadian occupational, physical and speech therapists and found only 12/1755 individuals surveyed used a standardized assessment of family functioning. Of particular importance to therapeutic interventions, family functioning refers to dimensions of family dynamics related to communication, collaboration and problem solving.³⁴ Results from this investigation found ineffective family functioning in 28% of CPs, levels comparable to Clark, et al.,⁹ in a similar sample of chronic stroke dyads. Less effective family functioning was significantly related to most of the CP variables, including higher depressive symptoms, higher family conflict surrounding stroke, and less autonomy support to the SS. This may be because demands on the family after stroke can have a negative impact on how family members interact and work together, worsening the risk for depression in both the CP and SS. An alternative explanation is that families who generally have ineffective functioning have more difficulty in responding when a family member has an illness with a long recovery. Assessing a family’s general function, may indicate to clinicians how CP and SS may need more education and support during rehabilitation. Development of therapy interventions in the context of the family environment is important to address this critical intersection of the CP and SS.²⁹

Although studies have found a relationship between CP strain and mental health with SS levels of functioning,^{5, 13} our results only found a significant association between CP fatigue and SS upper extremity function. The difference may be related to the fact that these previous studies were inclusive of SSs with a broad range of functional ability after stroke, as compared to the relatively independent levels of function for the SS in this study who had to meet the inclusion criteria for constraint induced movement therapy (ability to extend the wrist and fingers). The counter-intuitive finding that higher levels of CP fatigue were related to higher levels of SS upper extremity function is unclear. The WMFT measures capacity of the upper extremity (speed of movement) in a laboratory setting, but not performance (actual use of the limb in functional activities), thus the CP may still be assuming more responsibilities at home despite reductions in SS upper extremity deficits. Developing family-centered goals for functional activities and including approaches like constraint-induced movement therapy^{65, 72–74} as part of a home exercise program may improve real-world use of the upper extremity in the home environment by reminding the SS to use the more affected limb by restraining the less affected limb.

Limitations

The cross-sectional design limits interpretation of the data, in particular, determining if current family functioning is in response to the stroke or congruent with historical family dynamics. Generalizability of the findings is limited beyond White, well-educated, female CP of mildly impaired chronic SS. The data were part of a pilot clinical study so formal sample size calculations were not appropriate; however, the small sample size creates a greater probability of a Type II error.⁷⁵ As such, we may have missed a possible effect that existed in the data. However, even with this fixed sample size of 30 (n=30 is used to account for missing data), there was 80% power to detect a correlation (ρ) = 0.50, (or a ρ² = 25%). A ceiling effect of the FCCQ-CP scale may have occurred, as data were skewed toward higher autonomy support. The current study administration of the CSI inadvertently omitted one of scale questions (using 12-items instead of the complete 13-item version).

CONCLUSION

Providing care after stroke can be a challenging task and long-term task, placing the CP at risk for depression, stress, and conflict around stroke recovery. Importantly for rehabilitation therapists, failure to provide care that addresses family needs as formal rehabilitation ends may result in negative outcomes for both the CP and SS. This study supports previous findings^{9, 25, 34, 66} showing higher levels of CP depressive symptoms are related higher levels of strain and fatigue, and less effective family functioning. We believe this study is the first to find associations of CP depressive symptoms in relation to the CP’s ability to provide an autonomy supportive environment for the SS rehabilitation activities, which may lend insights into areas to explore further when assessing the impact of rehabilitation interventions and the role of the CP during stroke recovery.