Social Support, Social Problem-Solving Abilities, and Adjustment of Family Caregivers of Stroke Survivors

Methods 

A descriptive design was used in this study. Data for these analyses were drawn from a 3-group repeated-measures experimental study that examined the impact of social problem-solving telephone partnerships on primary family-caregiver outcomes after stroke survivors were discharged home from a rehabilitation facility. Only sham intervention and control group (but not the intervention group) data were used for this examination because no differences were found between those 2 groups on major study outcomes across the 4 time points analyzed in the current study. A more detailed description of the intervention and corresponding protocols is available.23

Predictor Variables 

Demographic data 

Bivariate associations between each of the criterion variables (depressive symptomatology, well-being, general health) and each of the available demographic variables were examined to identify potential covariates to include within the regression model. Pearson product-moment or Spearman correlations were calculated for interval and ordinal demographic variables, respectively. Analysis of variance (ANOVA) was used to estimate ω2, a measure of relation between nominal and interval variables and test for significance of the relation. Although all demographic variables that had a bivariate relation that was either .20 or greater (using P=.10) were made available for the regression model (table 1), only those that maintained a statistically significant relation or whose removal appreciably degraded model fit were retained within the model. As a result, only stroke survivor age was incorporated into the models, which was consistent with our previous experience.

Table 1. Associations Between Caregiver and Stroke Survivor Demographic and Criterion Variables (n=52)

	Demographic Variable	Criterion Variable
		CES-D	General Well-Being	Mental Health
	Family caregiver’s age†	.21(P=.13)	−.28(P=.05)	−.09(P=.53)
	Family caregiver’s sex⁎	−.14(P=.30)	−.11(P=.42)	−.05(P=.70)
	Family caregiver’s race⁎	−.20(P=.16)	.04(P=.76)	.12(P=.40)
	Family caregiver’s education‡	.05(P=.71)	−.18(P=.19)	.11(P=.44)
	No. of other family members who assisted in caregiving†	−.19(P=.20)	.20(P=.18)	.18(P=.23)
	No. of hours spent (daily) in caregiving by caregiver†	.18(P=.22)	.07(P=.65)	−.18(P=.23)
	No. of hours spent weekly by other family members in caregiving†	−.13(P=.39)	−.23(P=.13)	.01(P=.95)
	Stroke survivors’ age†	.09(P=.54)	−.20(P=.15)	−.01(P=.95)
	Stroke survivors’ sex⁎	−.02(P=.91)	.14(P=.33)	.17(P=.22)
	Stroke survivor’s race⁎	−.22(P=.12)	.06(P=.69)	.12(P=.41)
	Stroke survivors’ educational level‡	.04(P=.80)	−.20(P=.15)	−.03(P=.83)
	Total number of strokes‡	.04(P=.76)	−.04(P=.75)	.11(P=.44)
	Location of stroke⁎	−.06(P=.66)	−.27(P=.07)	.17(P=.24)
	Type of stroke⁎	−.19(P=.17)	−.30(P=.03)	.03(P=.24)
	Stroke survivors’ marital status⁎	.01(P=.88)	.02(P=.99)	.18(P=.47)
	Family caregiver’s relationship to stroke survivor⁎	.21(P=.53)	.09(P=.95)	.27(P=.30)

Abbreviation: CES-D, Center for Epidemiologic Studies Depression Scale.

⁎ Measure of association is ω2. † Measure of association is the Pearson r. ‡ Measure of association is the Spearman ρ.

The reliability of all instruments was examined before beginning the study by using a similar sample of 19 family caregivers of stroke survivors. Reliability coefficients concerning internal consistency and homogeneity were at acceptable levels (ie, α range, .71–.91; r range, .81–.89).

Social support 

The 30-item, Interpersonal Support Evaluation List (ISEL) was used to measure social support and its subscales, appraisal, belonging, and tangible support.27 Empirical evidence supports its validity and reliability, with α coefficients ranging from .88 to .91. The total ISEL score was used in the present study; lower scores indicate greater social support.28, 29

Social problem-solving abilities 

The sixth-grade version of the Social Problem Solving Inventory–Revised,30 a 52-item, Likert-type scale, was used to measure social problem solving in this study. Positive and negative problem orientation scales assess a general orientation toward problem solving, whereas the other 3 scales, rational problem solving, impulsivity and carelessness, and avoidance, assess general problem-solving styles.30

People who have higher scores on the problem-orientation scale are more optimistic, have greater confidence in themselves, are more motivated toward problem solving, and experience less frustration when confronted with problems. Furthermore, persons with positive orientations believe problems are a part of everyday life and tend to be more objective in analyzing factors that contribute to problems. In contrast, people who have higher scores on the negative orientation scale tend to be less optimistic and more pessimistic, have less confidence in themselves, are less motivated toward problem solving, and experience more frustration when confronted with problems. Persons with negative orientations doubt their ability to effectively solve problems and try to avoid solving them. Persons with negative orientations also personalize problems, commonly attributing them to “less than ideal behaviors” they showed.30

Rational problem solving represents a constructive problem-solving style of effective problem-solving abilities and how people use them. Persons who have higher scores on rational problem solving systematically identify problems, generate potential solutions, and implement and evaluate best solutions to problems. Reflective of their name, impulsivity and carelessness and avoidance styles are dysfunctional patterns in which people use impetuous, imprecise, and evasive methods for managing problems The impulsivity and carelessness scale assesses dispositions toward solving problems in impulsive and haphazard manners, whereas the avoidance scale assesses dysfunctional problem solving characterized by putting problems off. Internal consistency and test-retest coefficients range from .76 to .92 and .72 to .88, respectively.30

Results 

Research participants were 52 family caregivers (46 women, 6 men) who ranged in age from 25 to 74 years, with a mean age ± standard deviation of 55.6±11.6 and 57.2±13.6 years, respectively. The caregivers primarily were women (88%) who were spouses (n=21 [40.4%]) or daughters (n=18; 35%), were related by blood or marriage, and provided unpaid assistance to stroke survivors at least 6 hours a day in ADLs and instrumental ADLs after discharge from a rehabilitation facility. Most were married (n=37 [73%]) and either completed high school (n=28 [54%]) or college (n=17 [33%]); there were more whites (n=37 [71%]) than blacks (n=15 [29%]).

Stroke survivors ranged in age from 37 to 92 years, with nearly equal numbers of women (n=28) and men (n=24), with a mean age of 74.2±12.6 and 74.0±9.4 years, respectively. Most were either married (n=27 [52%]) or widowed (n=21 [40%]), and their ethnicity was identical to that of the caregivers. The majority either completed the eighth grade (n=14 [27%]) or high school (n=20 [39%]). Their median income was less than $20,000 per year. The number of strokes for each person ranged from 1 to 5 (average, 2) strokes. More stroke survivors had left-sided (n=31 [60%]) than right-sided (n=17 [32%]) hemispheric strokes, whereas 2 (4%) had bilateral strokes and 2 (4%) were unknown.

Caregiver Depressive Symptomatology 

Table 2 provides a model predicting risk for caregiver depressive symptomatology, well-being, and general health. Significant variance of random time coefficients (σ2=.225, z=2.03, P=.021) supported modeling the time slope as a random effect. The fixed effect for time was not significant, indicating (on average) no change in caregiver depressive symptomatology over time.

Table 2. Model Predicting Risk for Caregiver Depressive Symptomatology, Well-Being, and General Health

	Variable Effect	β	df	t	P
	Depression
	Intercept	27.910	50	3.20	.002†
	Positive orientation	−0.278	72	−1.08	.285
	Negative orientation	0.537	72	4.30	<.001†
	Rational problem-solving	−0.032	72	−0.35	.728
	Impulsivity/carelessness	−0.002	72	−0.02	.986
	Avoidance	−0.043	72	−0.20	.840
	Social support	−0.764	72	−5.13	<.001†
	Care recipient age	0.093	72	1.24	.220
	Time	−0.099	46	−0.88	.382
	Well-being
	Intercept	13.567	50	1.18	.242
	Positive orientation	0.486	72	1.18	.242
	Negative orientation	−0.883	72	−3.83	<.001†
	Rational problem-solving	0.198	72	1.46	.148
	Impulsivity/carelessness	0.111	72	0.57	.568
	Avoidance	0.281	72	0.79	.434
	Social support	0.619	72	2.51	.014†
	Care recipient age	0.143	72	1.11	.270
	Time	2.853	45	2.93	.005†
	Care recipient age by time	−0.039	72	−2.81	.006†
	General health
	Intercept	62.102	50	3.37	.002†
	Positive orientation	1.015	70	1.54	.128
	Negative orientation	0.259	70	1.01	.314
	Rational problem-solving	0.128	70	0.68	.400
	Impulsivity/carelessness	0.121	70	0.42	.679
	Avoidance	−0.651	70	−1.39	.169
	Social support	0.671	70	2.70	.009†
	Care recipient age	−0.408	70	−2.01	.048†
	Time	0.083	46	0.15	.885
	Positive orientation by time	−0.148	70	−2.36	.021†
	Rational problem-solving by time⁎	0.032	70	1.85	.069

NOTE. Error degrees of freedom computed by the containment method,61 which results in error degrees of freedom for linear mixed models that are different from traditional fixed linear models N–k–1 (independent subjects – number of predictors – 1).

⁎ Not statistically significant, but removal degraded model fit. † Statistically significant.

Furthermore, removing the set of problem-solving variables from the full model significantly degraded model performance (F5,72=8.30, P<.01), as did removing the social support variable (F1,72=26.35, P<.01), indicating that problem solving and social support were independently related to depressive symptomatology. However, removing the set of 4 nonsignificant problem-solving variables (ie, positive orientation, rational problem solving, impulsivity and carelessness, avoidance styles) did not degrade model explanatory power (F4,72=.66, not significant [NS]). Therefore, negative orientation was primarily responsible for the association between social problem solving and caregiver depressive symptomatology.

Caregiver well-being 

Significant variability of the random time coefficient (σ2=.844, z=2.75, P=.003) supported modeling the time slope as a random effect. The fixed effect for time was statistically significant, indicating (on average) an increase in caregiver well-being over time. Caregiver well-being declined faster over time when caring for older stroke survivors than when caring for younger stroke survivors.

Removing the set of problem-solving variables from the full model degraded model performance significantly (F5,72=9.34, P<.01) as did removing the social support variable (F1,72=6.32, P<.05), indicating that problem solving and social support were independently related to well-being. However, removing the set of 4 nonsignificant problem-solving variables (ie, positive orientation, rational problem solving, impulsivity and carelessness, avoidance styles) did not degrade model explanatory power (F4,72=2.08, NS). Therefore, negative orientation was primarily responsible for the association between social problem-solving abilities and caregiver well-being.

Discussion 

Our results support social support, and the emotion-focused component of social problem solving, problem orientation, independently contributes to caregiver adjustment. Trajectory analysis indicated that higher levels of social support were associated with lower levels of caregiver depressive symptomatology and higher levels of well-being and general health, independent of social problem solving. A greater negative problem orientation was related to higher levels of depressive symptomatology and lower levels of well-being. A more positive problem orientation was associated with smaller decreases in general health. The strength or slope of this positive relation lessened over time.

Consistent with prior research, declines in social support are linked intricately with family-caregiver adjustment, and these associations are evident soon after their return to the community.35 The present study shows the dynamic association between social support and caregiver adjustment the first 3 months after stroke, affecting depressive symptomatology, well-being, and general health.

Social support often erodes over time for many caregivers who assist family members with chronic health conditions.36 Caregivers who are at risk for complicated adjustment have social and individual characteristics reflective of their constellation of supportive networks under routine and stressful situations. Distressed persons tend to seek support and alienate supportive people over time.37 Conversely, persons who are more likely to thrive in the caregiver role may be buffered by social support. We suspect this trend would continue throughout the year, based on prior modeling of the social support−caregiver adjustment relation after the onset of severe disability in family members.35

A negative orientation toward solving problems can disrupt individual abilities to regulate mood, think rationally,21 and effectively perform complex learning tasks.38 Therefore, caregivers with a lower negative orientation tend to be less pessimistic and more open to positive aspects of caregiving. More succinctly, caregivers with a lower negative orientation are more likely to interpret stressful aspects of caregiving in a more benign manner than those prone to a higher negative orientation. A greater negative orientation is associated with a low sense of preparation for caregiving roles and duties23 and may contribute to higher levels of caregiver depressive symptomatology and lower levels of well-being in the first months of caring for a stroke survivor.

People with positive orientations perceive they will be successful in their problem-solving efforts and manage problems effectively, experiencing improved health over time.39 Other empirical data support the importance of a positive problem orientation in general health. For example, structural equation modeling techniques support an indirect association between a positive constructive problem-solving style and psychologic adjustment, with greater adherence to human immunodeficiency virus antiretroviral medication regimens.40 Similarly, a positive problem orientation is associated with greater life satisfaction in patients with spinal cord injury41 and other chronic diseases.42 A challenge in examining the theoretical link between a positive problem orientation and general health in this and other studies is a pragmatic one. Changes in general health occur over time. The current study was limited to 3 months. However, longitudinal studies examining this relation are essential.

In this study, caregivers of older stroke survivors had declines in their well-being faster over time than caregivers of younger stroke survivors. This finding is in contrast to other studies that report no significant differences between stroke survivor age and caregiver well-being, after controlling for survivor functional deficits,43 as in this study. Stroke survivors in this analysis, however, had moderate levels of disability. The higher degree of homogeneity regarding functional deficits in this sample may contribute to differences in findings. The sample size also limits examining the potentially moderating or mediating effect of other variables on family adjustment as well as other study findings.

These results suggest that social support and positive problem orientation and the interaction with time are independent predictors of general health. Our findings suggest a moderating effect of positive problem orientation on change in general health over time so that caregivers with higher levels of social support and positive problem orientation experienced smaller decreases in their general health over time than did caregivers with lower levels of social support and positive problem orientation.

These results also indicate that both social support and problem orientation may be important in family caregiver adjustment. Therefore, health care professionals should explore strategies to provide these components. Although empirical literature examining the effect of problem orientation, itself, is absent, problem-solving training that incorporates both problem orientation and systematic problem-solving skills is used in home family-education programs44 by videophones45 and telephone.23, 46, 47 Although these studies are still limited in number, these problem-solving training programs provide support while decreasing caregiver depressive symptomatology and increasing wellness and perceptions of preparedness. These training programs also improve other aspects of caregiver health by increasing vitality and social functioning and lessening role limitations related to emotional problems.23

Health professionals face many challenges and opportunities in meeting the full array of unmet needs of stroke survivors and their caregivers. Although our research reveals the importance of social support and problem orientation, health care providers can assist caregivers to obtain and maintain social support to sustain them in challenges they face. Health care professionals can provide this assistance by developing a proactive approach as they interact with caregivers.

Assessment of caregivers’ networks of social support includes determining availability of family and friends who are willing and able to provide support. Other important considerations are caregivers’ comfort and ability to specifically identify needed assistance in meeting long-term needs of family members poststroke, the ability to ask others for assistance, and the investigation of potential resources of support. Health care professionals, working in tandem with medical social workers, can assist caregivers by providing them with a list of available local, regional, and national resources. Although many caregivers are savvy in using and accessing resources via certain modes (eg, face-to-face and telephone contacts), other caregivers may admit they are uncomfortable accessing resources using other modes (eg, accessing resources via electronic devices).

Through appropriate social support referrals, health care professionals can assist caregivers to adjust positively to this role. The following organizations may be useful in caregivers’ quests for social support outside immediate family and network of friends: Stroke Awareness for Everyone Inc,48 the Stroke Information Directory,49 American Stroke Association,50 National Stroke Association,51 Pediatric Stroke Network,52 Lexi-Comp Inc Stroke Explained,53 Different Strokes,54 Strokesurvivors International,55 and the Internet Stroke Center.56

The Support Team Network57 is an example of a nationally available resource that uses an empirically based team approach to create social support for care recipients and their caregivers in the community. In the past 55 years, more than 10,000 teams provided support to over 25,000 people and their families across the United States and southern Africa through this organization’s efforts. Volunteers, typically associated with churches and community organizations, use a coordinated, built-in support system to sustain their efforts. Support-team volunteers offer practical, tangible, emotional, and spiritual support and assistance based on their unique skills. Although there are many ways to organize a team, the “basic model” usually meets a variety of needs for 1 person or family and the team approach allows that flexibility. Although support teams cannot do everything for stroke survivors and their caregivers, they can improve their quality of life.

Typically, support-team activities include providing rides to health care providers or grocery stores, helping with household or yard chores, running errands, cooking or delivering meals, visiting or making telephone calls, or offering caregivers a brief time away or break from their daily routine through respite. Some support-team volunteers also share prayer or communion, based on the mutual needs and desires of a person or caregiver. A support team generally has 3 major boundaries: money, medicine, and medical care. The support team does not loan or give money, dispense prescription medications, or offer medical care. In addition to referring persons to existing support teams, health care professionals can serve as the impetus for the creation of local support teams. An initiative typically consists of a 1-year partnership between a community (eg, 1–10 congregations, agencies, and organizations) and the support team network to teach volunteers how to develop support teams. The 5 phases of development of the partnership result in an ability to develop support teams.

There are also useful techniques we, as clinicians, can use with family caregivers to strengthen social problem solving. During hospitalization, provide both oral and written information to reinforce components and steps of social problem solving. In introducing social problem solving, share how emotions affect how caregivers view problems. Furthermore, emphasize knowing a systematic process for managing problems also influences emotions. Initially, assist families to identify a variety of potential solutions for problems that commonly occur on discharge home.47 For example, problems may include assuring safety and managing ADLs and cognitive, behavioral, and emotional changes of stroke survivors.58

Provide laminated handouts for use after discharge and posters for display on walls to remind caregivers of key problem orientation components and social problem-solving steps. In problem orientation, emphasize the following: (1) problems are a part of everyday life; (2) they are capable of solving these problems by using appropriate resources; (3) some problems will be solved rather quickly, whereas others will be solved more slowly; and (4) they choose how to view problems, which will influence their feelings toward and effectiveness in solving problems. In the problem-solving component, accentuate the following key steps: identify and define the problem, identify possible solutions to the problem, choose and test the best solution, and evaluate outcomes of problem solving. In addition, provide examples that show how to apply problem orientation and problem-solving steps. Initially, make examples simple and only address 1 problem. Later, provide situations that show how to apply problem orientation and problem-solving steps in more complex situations (eg, assure safety in managing physical, cognitive, and behavioral problems).47

Some investigators59, 60 suggest people tend to view situations from 3 possible explanatory styles: assigning an internal or external cause to a problem, assessing whether the problem has affected specific (ie, some) or global (ie, all) areas of their lives, and examining whether the problem has transient or enduring effects on their lives. Persons with negative orientations tend to personalize problems, ascribing internal, enduring, and global attributes to situations, while ignoring other perspectives.

Although it seems almost automatic to enhance correct explanatory styles in problem orientation, successful programs strengthen these skills in both problem orientation and problem solving. Problem orientation and problem solving overlap so that orientation influences problem solving and vice versa. For example, in identifying problems, their associated feelings, and possible causes of problems, caregivers enhance their positive problem orientation and lessen their negative problem orientation while strengthening systematic problem solving.

Health care practitioners should encourage caregivers to identify specific problems (rather than global problems that often are unwieldy and unmanageable), external causes of problems they can realistically change, and practical solutions. Internal feelings also should be explored, assisting them to recognize situations they have limited control over, except how they are viewed and managed. Positive self-reinforcement also should be inherent in the process (eg, throw rugs on floors and weakness of my _______’s leg make me afraid of falls, safety issues that occur from throw rugs on floors are common after a stroke, removing the throw rugs will solve this problem quickly and also make me feel less anxious, I feel good I solved this problem). Over time, applying these skills enhances viewing problems less negatively and more positively, which our study findings suggest are both essential in improving outcomes related to depressive symptomatology, well-being, and general health.

Conclusions 

Social support and the emotion-focused component of social problem solving, problem orientation, independently contribute to caregivers’ adjustment after their return to the community with stroke survivors. Higher levels of social support were associated with lower levels of caregiver depressive symptomatology and higher levels of well-being and general health, independent of social problem solving. A greater negative problem orientation was associated with higher levels of depressive symptomatology and lower levels of well-being. A more positive problem orientation was associated with greater increases in general health. The strength or slope of this positive relation lessened over time. Although social support is independently associated with general health, the influence of a positive problem orientation on this outcome, over time, is tenuous and should be explored further. Interventions that provide social support and assist caregivers to develop a more adaptive orientation toward problem solving may benefit family caregivers after their return to the community.