Home > RNJ > 2007 > March/April > Problem-Solving Early Intervention: A Pilot Study of Stroke Caregivers (CE)

Problem-Solving Early Intervention: A Pilot Study of Stroke Caregivers (CE)
Rosemarie B. King, PhD RN Robert J. Hartke, PhD MPH Florence Denby, MS CRRN

Caregivers (CGs) of stroke survivors assume their role suddenly and with little preparation. Negative emotions are common, persist over time, and are related to other negative outcomes. This pilot study, guided by a coping model, examined the efficacy and durability of a caregiver problem-solving intervention (CPSI) on CG and stroke survivor outcomes. Additional aims included assessment of feasibility issues and reliability and sensitivity of the study measures. The nonrandom sample of 15 stroke CGs was matched on depression and demographics with a comparison group. The CPSI started during acute rehabilitation and continued through 2 months after discharge. Parametric and nonparametric tests were used to assess achievement of the aims. CG depression, anxiety, preparedness, and survivor motor function improved significantly in the intervention group over time. Burden, life changes, and taking care of CG’s own needs did not change significantly. CPSI group CG depression significantly improved compared with the matched group. The improvement in outcomes for the CPSI group supports further testing of the intervention with a large sample.

Caregivers (CGs) of stroke survivors assume their role suddenly and with little preparation. Negative emotions are common, persist over time, and are related to other negative outcomes (King & Shade-Zeldow, 1995; Kotila, Numminen, Waltimo, & Kaste, 1998). Intervention studies that targeted emotional distress have resulted in modest success and rarely included positive outcomes, survivor function, or the durability of treatment effects. Therefore, a pilot study was designed to assess the efficacy and short-term durability of a caregiver problem-solving intervention (CPSI) on positive and negative CG outcomes and survivor function. The aims of the study were to assess the efficacy of the CPSI on CG depression, preparedness, anxiety, life change, burden, healthy caregiving, and survivor motor function immediately after intervention and 2 months later; to assess the feasibility of recruiting 65% of eligible CGs and retaining 75% of participants in the study; to assess the feasibility of achieving treatment helpfulness ratings in 80% of participants; and to assess the reliability and sensitivity of the study measures. The hypothesis was that the severity and rate of depression in CPSI participants immediately after intervention will be lower than those in a matched comparison group, who received usual care.

The crisis of physical illness model was used to guide the study (Moos & Tsu, 1977) (Figure 1). In this stress and coping model, the influence of contextual factors (person, illness, and environmental characteristics) on adaptation outcomes, such as depression, is mediated by coping process variables (appraisal of the situation, coping strategies). The intervention was guided by social problem-solving theory (D’Zurilla, 1986) and cognitive-behavioral theory (CBT; Beck, Rush, Shaw, & Emery, 1979) and targeted coping process factors to affect outcomes.

Background and Significance

Stroke CGs initiate their role at the time of the survivor’s maximum disability, with great uncertainty about the future (King & Shade-Zeldow, 1995). The multiple problems and needs reported by CGs include preparation for caregiving, enhancing the survivor’s function (e.g., dealing with difficult behaviors and emotions), and sustaining the self and family (e.g., dealing with the CG’s own emotions and other responsibilities) (Hartke & King, 2002; King & Semik, 2006; Periard & Ames, 1993).

Reports of high rates of depression, anxiety, and burden are common during acute and chronic caregiving (Anderson, Linto, & Stewart-Wynne, 1995; Dennis, O’Rourke, Lewis, Sharpe, & Warlow, 1998; King, Carlson, Shade-Zeldow, Bares, & Roth, 2001; Schulz, Tompkins, & Rau, 1988). CG depression and anxiety rates as high as 42% and 44%, respectively, have been reported during the first year after stroke (Anderson et al., 1995; Kotila et al., 1998). The impact of CG distress is apparent in findings of a relationship between negative CG outcomes, such as anxiety, and unfavorable survivor outcomes (Dennis et al., 1998). Reports of positive outcomes, such as closer relationships and improved health, are less common (Bakas & Champion, 1999; King, & Shade-Zeldow, 1995) and have not been examined in intervention studies.

Research findings on coping process variables indicate that they may be important targets for interventions. For instance, negative appraisal of the situation, concerns about caregiving, and avoidance coping have been positively related to depressive symptoms or perception of negative life changes in stroke CGs (Bakas & Champion, 1999; King et al., 2001; Schulz et al., 1988).

With few exceptions, a recent review of stroke CG intervention literature indicated disappointing treatment results (Visser-Meily, van Heugten, Post, Schepers, & Lindeman, 2005). The studies often lacked a theoretical base and standardized protocols, used limited treatment intensity and focus, and provided no follow-up. However, two problem-solving intervention studies initiated early and targeting CG emotional and physical health or family functioning and survivor adjustment had strong designs and were effective (Evans, Matlock, Bishop, Stranahan, & Pederson, 1988; Grant, Elliott, Weaver, Bartolucci, & Giger, 2002). Another counseling intervention resulted in improvement in family functioning and stroke survivor function but not in depression and anxiety (Clark, Rubenach, & Winsor, 2003). Two studies reported significant mental health treatment effects, favoring treatment groups over control groups. However, interpretation of the findings and ability to replicate these studies were limited by the lack of baseline group comparisons or the use of unstructured interventions and wide variability in treatment frequency (Dennis, O’Rourke, Slattery, Staniforth, & Warlow, 1997; Mant, Carter, Wade, & Winner, 1998). Another study that used a CBT group intervention resulted in stabilization of burden and significantly greater competence in treatment participants, compared to greater burden and lower competence in control subjects (Hartke & King, 2003). The effects for depression and loneliness were not significant.

The proposed problem-solving intervention is based on D’Zurilla’s (1986) model of social problem-solving therapy (PS). PS influences outcomes by promoting a positive view of problems as normal and manageable, increasing awareness of emotional responses to problems, and improving problem-solving skills. PS has been used effectively to treat depression in CGs and community samples (Grant et al., 2002; Lynch, Tamburrino, & Nagel, 1997).

Findings from CG and community studies of depression treatment provided support for the development of the pilot study to assess the efficacy of PS on the short-term durability of CG and stroke survivor outcomes. The current study contributed to the research on management of CG distress by using a structured intervention that was guided by a strong theoretical foundation.


A single-group repeated-measures design was used to examine the feasibility and efficacy of a PS intervention to promote adaptation in stroke CGs. In addition, depression and functional status findings were compared with a historical matched CG and survivor group. Participants met the following inclusion criteria: age ≥50, primary CG for a stroke survivor in acute rehabilitation, living with the survivor, English speaking, accessible by telephone, not in active psychotherapy or a support group, and a score of ≥10 on the depression scale. The setting was a free-standing rehabilitation hospital. CGs were identified by a research nurse on the stroke service. The nurse contacted family members and verified their role as primary CG.


The study was approved by the university institutional review board. Depression risk was assessed after informed consent was obtained. The nurse offered a list of resources to CGs who were interested but not eligible. Assessment data for the CPSI group were collected by a research assistant at the following times: baseline (preintervention, T1), postintervention (8–10 weeks after discharge, T2), and 2 months later (16–18 weeks after discharge, T3). Data for the matched adaptation study group were available for baseline and 8–12 weeks after discharge (T2).


Twenty-eight CGs consented to participate. Three of them did not meet the screening criteria, so 25 were enrolled. Fifteen of 25 CGs (60%) completed the intervention; 14 completed the final assessment. Reasons for attrition included death of stroke survivor, extreme distress that prevented participation, change in CG, CG too busy, or inability to reach the CG. No significant differences were found in demographic or survivor illness variables between consenting CGs who dropped out or were not eligible and those who completed the study. Table 1 contains demographic statistics for the CPSI and matched adaptation groups. The majority of CGs were female, white, in a spousal relationship with the survivor (spouses and unmarried domestic partners), approximately 63 years of age, and a high school graduate. The majority were caring for survivors who had experienced a first stroke (93%). With the exception of education, which was significantly higher in the CPSI group (p < .01), demographic variables and measures were not significantly different between groups.


The CPSI 10-session intervention combined PS (problem orientation and problem-solving skills) and CBT (relaxation training, reframing negative thoughts). Examples of additional content include information and tips on stress management, self-care, resources, behavior management, and risk factors for recurrent stroke. CGs received a manual that contained the content for each session and included exercises to tailor the content to each CG. In addition, they received a copy of a stress management booklet that reinforced the course content.

A brief introductory meeting and two or three in-person sessions were conducted during the survivor’s hospitalization. The remaining seven or eight weekly postdischarge sessions were conducted primarily by telephone. One CG requested in-person contacts while her husband was in outpatient therapy at the hospital, and another received four in-person sessions because of difficulty participating by telephone. Sessions were 45–60 min long. Participants identified their most difficult problems with the counselor and developed plans to solve them using the PS steps. They were asked to complete home assignments, such as applying PS to problems, rating their mood, and practicing relaxation. The counselors were flexible in fitting the sessions into the lives of busy CGs, often rescheduling sessions and holding sessions late in the evening and on weekends.

Comparison Group

The comparison (matched adaptation study) group consisted of 15 of 136 CGs who participated in a descriptive, longitudinal study of stroke CGs and their stroke survivors (King et al., 2001). They were matched with the CPSI sample on baseline Center for Epidemiologic Studies Depression scale (CES-D) scores, gender, age, race, and caregiving relationship. The measures used in both studies, depression and functional status, were compared between the groups. The matched group had received usual care.

Integrity of the Intervention

Two nurses experienced in family caregiving delivered the intervention. The co-principal investigator, a clinical psychologist, met with the nurses biweekly to discuss cases. A record of intervention components was completed after each session to ensure that the treatment structure was maintained. Counselor training included learning PS techniques and relaxation methods and ways to manage negative thoughts. With the exception of two CGs, who participated in 9 meetings instead of 10 to accommodate their schedules, CGs completed 10 sessions. The content of two chapters was combined into one meeting for participants who received fewer meetings.


Clinical efficacy was assessed by using the McNemar test to compare depression rates (scores ≥16) between T1, T2, and T3 in the CPSI group and between T1 and T2 in the matched group. T2 Depression rates were compared between the two groups using the chi-square test. Repeated-measures ANOVAs followed by paired t tests were used to compare scores in the CPSI group over three times. Change in mean scores in the matched group was assessed using paired t tests. Independent t tests were used to examine differences in depression severity and survivor motor function between groups at T2. Feasibility issues were examined using frequencies to compare recruitment, retention, and credibility of the intervention with the study goals. Because of the small sample, a liberal value of p < .10 was selected a priori to determine significance.

Variables and Measures

The level of helpfulness of the intervention components was assessed using an investigator-generated measure. One item asked whether participation in the study was helpful. Other items included rating the helpfulness of each session. Responses for the session items ranged from 0 (not helpful) to 3 (very helpful).

Coping Variables

Burden was measured using the difficulty subscale of the 15-item Caregiving Burden Scale (Carey, Oberst, McCubbin, & Hughes, 1991). The difficulty subscale has evidence of internal consistency reliability and validity (Bakas, Austin, Jessup, Williams, & Oberst, 2004; Carey et al., 1991). The response set ranges from 1 (not difficult) to 5 (extremely difficult). Responses are summed to yield a difficulty score. The potential range of scores is 15–75. Higher scores reflect greater perceived difficulty with tasks. Alpha coefficients in the current study ranged from .65 to .84.

Appraisal of ability to cope was assessed using the Preparation for Caregiving Scale (PCS; Archbold, Stewart, Greenlick, & Harvath, 1990) and the Caregiver Competence and Confidence Scale (CCCS) from the Healthy Caregiving Scale (Edelman & Fulton, 1997). The PCS assesses perceptions of preparedness to manage CG tasks and stresses. The 8-item scale has five response options: 0 (not at all prepared) to 4 (very well prepared). The responses are summed and averaged to yield a potential range of scores from 0 to 4. Evidence of internal consistency reliability and construct validity has been reported (Archbold et al.). Alphas in the current study ranged from .70 to .85. The CCCS is a new 6-item scale, with responses ranging from 1 (not at all) to 5 (very much), that measures caregiving confidence. The summed score can range from 6 to 30, with higher scores indicating greater confidence. Psychometric testing demonstrated an alpha of .84 and significant correlation with a measure of subjective burden (Edelman & Fulton, 1997). This scale was not analyzed further in this study because of strong correlations (r = .70 to .79) with the PCS, which has been used in other stroke CG studies and provided stronger published psychometric support.


Depression was measured using the CES-D, which has established reliability and validity (Radloff, 1977). The CES-D measures the frequency of 20 depressive symptoms experienced during the past week on a scale of 0 (none) to 3 (5 to 7 days). The possible range of scores is 0–60, with higher scores representing greater symptom severity. A score of 16 is the cutoff for depression. Alpha coefficients ranged from .77 to .81 in this study.

Anxiety was measured using the 5-item Tension Anxiety subscale of the Profile of Moods Scale short form (McNair, Lorr, & Droppleman, 1992). The 30-item total scale, comprising six subscales, uses a 5-point response set of 0 (not at all) to 4 (extremely) to respond to descriptors of feelings. The possible range of scores for anxiety is 0–20, with higher scores representing greater anxiety. Construct and concurrent validity and internal consistency ((alpha) = .86 to .88) have been demonstrated (McNair et al.). Alphas ranged from .63 to .88 in the current study.

Perception of life changes was measured using the Bakas Caregiving Outcomes Scale, which measures changes in social functioning, subjective well-being, and somatic health (Bakas & Champion, 1999). The 15 items, with a 7-point response scale of 1 (changed for the worst) to 7 (changed for the best), are summed for an overall score that reflects the extent of negative and positive life change. The possible range of scores is 15–105, with higher scores reflecting more positive life change. Alpha coefficients ((alpha) = .77 to .90) and content and construct validity were strong (Bakas & Champion). Alphas in this study ranged from .66 to .77.

Healthy caregiving was measured using two scales from the Healthy Caregiving Scale (Edelman & Fulton, 1997). The Caregiver Cares for Own Needs Scale is a 3-item scale that assesses perceptions about health self-care using a 5-point response set from 1 (strongly disagree) to 5 (strongly agree). The potential range of scores is 3–15; lower scores indicate greater attention to the CG’s own needs. Internal consistency reliability has been reported to be .85; alphas ranged from .82 to .94 in the current study. The 4-item Makes Time for Self Scale used a 5-point response set, 1 (almost never) to 5 (almost always), with higher scores representing perceptions of making more time for self. The Makes Time for Self Scale was reported to have an alpha of .88. A low T1 alpha (.55) in this study precluded further analysis of the measure.

Stroke Survivor Functional Status

The CGs completed the Functional Independence Measure (FIM; Uniform Management Service, 1993). The FIM measures severity of disability using an 18-item, 7-point scale ranging from 1 (maximum dependence) to 7 (independent function). The scale consists of a 13-item motor scale and a 5-item cognitive scale. The possible ranges of scores are 13–91 and 5–35, respectively, for the motor and cognitive scales; higher scores reflect greater independence. Adequate internal consistency has been reported for motor and cognitive scales for stroke survivors (Hartke & King, 2003). In this study, alpha coefficients for cognitive and motor scales ranged from .91 to .99. Only the motor scale was used in between-group comparisons because the cognitive scale was not collected in the matched group study.


The CPSI resulted in significant improvement over three times for depression, F(2, 26) = 13.74, p < .001; preparedness, F(2, 26) = 20.20, p < .001; anxiety, F(2, 26) = 3.67, p <.05; and survivor motor function, F(2, 26) = 17.38, p < .001 and cognitive function F(2, 26) = 13.40, p < .001. The life change scale, F(2, 26) = 1.45; burden, F(2, 26) = .83; and taking care of own needs, F(2, 26) = 1.13, did not change significantly, p > .10. Effect sizes ranged from d = .16 (burden) to 1.33 (preparedness). Table 2 shows descriptive statistics and p values for change over time and group differences.

The hypothesis that the severity and rate of depression will be lower in the CPSI group after intervention compared with the matched group, assessed at 2 months after discharge, was supported. Independent t tests indicated that the T2 CES-D severity score was significantly lower for the CPSI group, t(28) = 2.42, p < .05. The T1 to T2 change in severity was greater in the pilot group, t(14) = 4.31, p <.001, than the adaptation group, t(14) = 1.97, p < .10. Because education was not equivocal between groups, we examined the relationship between depression severity and education level. The relationship was not significant for the combined groups at T1, r = –.03, n = 30, p >.05, and T2, r = –.05, n = 30, p >.05; or for the matched group only at T1 and T2, r = –.08, n = 15, p >.05; r = .27, n = 15, p >.05, respectively. In the matched sample, the relationship converted over time from negative to positive, such that greater education was related to greater depression.

A significant reduction was found in the CPSI depression rate, from 53% to 13% and 21% at T2 and T3, respectively (McNemar, p < .10). In contrast, the percentage depressed in the matched group was unchanged at T2, 53% (McNemar, p = 1.0). Significant improvements in anxiety (p < .05), from a little or moderate to not at all or a little, and in preparedness (p < .01), from somewhat well prepared to pretty well prepared (T2) and very well prepared (T3), were maintained in the CPSI group (see Table 2). The baseline motor function score was not significantly different between groups. Motor function improved for both groups; however, the change was significantly greater for the CPSI group, t(14) = (-)4.19, p < .01, than for the adaptation group, t(14) = (-)1.53, p > .10. Cognitive function for the CPSI survivors improved significantly over three times F(2, 12) = 8.04, p < .01.

Among the scales that did not change significantly, the life change mean scores changed in the expected direction from negative to slightly positive. Both burden and taking care of own needs worsened over time. The Caregiver Burden Scale may not have been appropriate for use during hospitalization. The T1 means for 14 of 15 items were less than 2 (not difficult). Items with the lowest T1 means included medical and nursing treatments, personal care, and watching for and reporting the patient’s symptoms. The only item with a mean score in the slightly difficult range was managing behavior problems.

Both the recruitment and retention rates were lower than expected. Of 63 probably eligible CGs, only 44% consented to participate. Of 25 participants, 15 (60%) participated through T2 (postintervention assessment) and 14 (56%) completed T3.

All intervention components were rated as somewhat to very helpful by 92% of the CGs. Whereas the depression and coping with stress sessions received the highest percentages of very helpful ratings (73% and 80%, respectively), stroke risk factors received the least very helpful ratings (53%). Yet three CGs viewed the risk factor information as most helpful. The median score for all sessions was 3.0 (very helpful). In response to the question “Overall, did you find it helpful to participate in the study,” 93% of the caregivers reported that the intervention was helpful. A 76-year-old male CG indicated that the program was not helpful because “It didn’t fit for me. I’m kind of bull headed; I just do what I have to do as it comes.” Four participants indicated that the amount of time needed for the postdischarge sessions was problematic. When asked what was most helpful, 80% of participants identified learning how to relax and cope with stress; 50% stated that analyzing problems was most helpful. The sum of these ratings is more than 100% because several CGs rated two topics as most helpful. A CG stated that the program “was a gift that helped [her] feel more secure and confident” as a CG. Another CG recommended that we make the program “mandatory” for the self-esteem, moral support, and help it gave.

Internal consistency reliability was adequate ((alpha) >.70), with few exceptions, (T2 burden, .65, and life change, .66; T1 anxiety, .63; and T1 time for self, .55). In addition, the measures were sensitive to change over time.


Similar to the findings reported by Evans et al. (1988) and Grant et al. (2002), significant improvements were found for most outcomes. Unlike other intervention studies, this study examined the intervention effect on both percentage depressed and depression severity. Change in percentage depressed is important because it indicates that CGs at risk for clinical depression benefit from the intervention. Compared with the matched adaptation group, the CPSI CGs experienced a greater reduction in depressive symptoms and a lower rate of depression. The CPSI CGs reported significantly greater education than the adaptation group. In the current study, level of education was not related significantly to depression at T1 or T2. Other investigators have reported both a negative relationship between education and depression in CGs of older adults with disabilities (Clark, 2002) and no significant relationship between CG depression and education or socioeconomic status, a proxy for education (Grant, Bartolucci, Elliot, & Newman Giger, 2000; King et al., 2001). The nonequivalence between groups is a concern because it is not known how CGs who do not have a high school education would respond to the intervention or whether they would participate. Randomization to group is necessary to avoid such differences between groups.

Other clinically important findings are the significant improvements in anxiety, perception of preparedness, and stroke survivor motor function. Decreased anxiety may be clinically significant, not only because of benefits to the CG but also because anxiety has been related to stroke survivor emotions (Dennis et al., 1998). The baseline anxiety score indicates greater anxiety than in reports of people with chronic obstructive lung disease, older adults, and adult and college samples (Kapella, Larson, Patel, Covey, & Berry, 2006; McNair et al., 1992; Nyenhuis, Yamamoto, Luchetta, Terrien, & Parmentier, 1999). After the intervention, the T2 and T3 scores were equal to or lower than those of these samples. Improved preparedness for caregiving after a problem-solving treatment also was reported by Grant et al. (2002). Feeling better prepared for suddenly becoming a CG may help to reduce the reports of unmet needs. Motor function improved significantly in both groups between T1 and T2, with no difference between groups in T2 scores. FIM cognition improved significantly in the CPSI group in each time interval. Unfortunately, differences in measurement of cognition in the groups precluded comparisons between groups. A larger study that includes a randomized, concurrent comparison group and accounts for factors such as continued therapy, stroke severity, and comorbidity is needed to test the CPSI effects on function.

The success of the CPSI in achieving moderate to strong effect sizes for several outcomes may be related to the use of a structured treatment that allowed tailoring of content to the individual CG’s needs. Schulz et al. (2003) recently recommended this approach to achieve clinically important outcomes. The maintenance of gains in most outcomes for 2 months after the intervention ended is encouraging, but a longer follow-up is needed to test durability effectively.

The change in burden was not significant. The short follow-up interval may have precluded finding a benefit on burden because CGs were still adjusting to their new roles and were adding new responsibilities beyond those acquired during hospitalization. Although the T3 follow-up showed a slight decrease from T2 in the expected direction, it is unknown whether this was the beginning of a trend toward lower burden. Using the same measure Bakas et al. (2004) recently reported higher levels of burden in CGs who had been caregiving an average of 18 months, compared with our postintervention findings (M = 30.3 vs. 22.6). With the exception of the study by Grant et al. (2002), for which scores were not provided, the burden measure has been used only with outpatient samples. If this measure is used in intervention studies, careful thought should be given to the study design and hypothesis because many tasks are not pertinent during the survivor’s hospitalization. Therefore, it is difficult to effect positive change between baseline and the subsequent data collection time. Future studies should examine durability of treatment effects on burden during an extended follow-up. Although the life change scores showed improvement in the expected direction, the change was not significant. There are no reports of Bakas Caregiving Outcomes Scale findings in studies of new CGs, so comparisons are not possible. The ability to view changes after a stroke as positive may take a longer period of adjustment than provided in this study. Scores for the outcome “taking care of their own needs” worsened for CGs by T2 and remained worse 2 months later. The intervention may have increased sensitivity to their needs without sufficient time for participants to feel confident in meeting them.

The improvements in CG preparedness and emotional well-being may reduce the distress and unmet needs that occur during the early phase of caregiving. The potential benefit to stroke survivor function is important because it can reduce CG burden and has potential to improve the survivor’s quality of life. Quality of life and other survivor outcomes, such as depression, changes in the dyadic relationship, and family functioning, could be examined in future studies that target stroke survivor benefits from a CG intervention. To date, stroke CG intervention studies that include survivor outcomes are uncommon (Clark et al., 2003; Dennis et al., 1997; Evans et al., 1988).

Although attrition was high, this is not unusual in stroke CG studies (King et al., 2001; Rodgers et al., 1999; Teel, Duncan, & MinLai, 2001; Wright, Hickey, Buckwalter, Hendrix, & Kelechi, 1999). Unlike Alzheimer’s CGs, these CGs are faced with a sudden crisis, resulting in destabilization in their ways of functioning and the acquisition of multiple new responsibilities (Moos & Tsu, 1977). The crisis and the vulnerability of survivors to mortality and morbidity may increase attrition. In the current study, several CGs who completed the intervention experienced stressful events, such as the deaths of a stroke survivor (n = 1) or a best friend (n = 1) and one or more hospitalizations of the stroke survivor (n = 4). These CGs completed the intervention despite the added stressors, but similar stressors may have contributed to distress and attrition. These events were reported anecdotally in the current study, so the relationship between such events and attrition and outcomes could not be assessed. Such events should be monitored and controlled statistically in future intervention studies to increase validity. It is also possible that the time commitment required to participate became too great during this difficult time. High attrition may be an unavoidable risk that adds to the complexity of stroke CG intervention studies during the first months of caregiving.

With one exception, the participants found the intervention helpful. Median ratings for all session topics were in the very helpful range. However, the differences in ratings of topics, such as stroke risk factors, highlight the need to tailor information and counseling to the needs of the CG. The CGs’ comments about time constraints must be considered in future interventions. Their comments indicated that scheduling sessions was problematic after discharge when the demands on their time increased.

For the most part, the study measures had adequate psychometrics, were sensitive to change, and could be considered for inclusion in future studies.

The findings must be viewed cautiously because of the small sample size, high attrition rate, use of a nonrandomized sample, and short duration of follow-up. Despite the limitations, the strength of improvement in CG outcomes holds promise for future testing of the intervention.

Implications for Practice

The hospitalization period and first months after discharge have been reported to be the most difficult for CGs (King & Semik, 2006). Aspects of the intervention, which targeted this time period, can readily be applied in practice. Nurses see the CG more often than other professionals during hospitalization and can play an important role in promoting their adjustment during this time and during clinic visits. They can assess the CG’s concerns and how effective he or she is in managing them, watch for evidence of negative thinking, attend to responses during teaching, and make referrals to assist in building skills (e.g., stress management and relaxation training or counseling) to reduce distress during the early stage of coping with caregiving.

Future Studies

Findings support the effectiveness of the CPSI intervention in a small sample of stroke CGs. Practice in solving difficult problems and teaching relaxation techniques and methods to reduce negative thinking facilitated development of these skills for participants. These skills may have contributed to the durability of findings for the outcomes. The high ratings given to content such as problem solving, relaxation, stress management, and reducing negative thinking support the inclusion of such techniques in future studies to manage CG stress. Future studies with large samples and long-term follow-up to determine durability of the intervention are necessary. The cycle of chronic depression seen in stroke CGs may be broken through a tailored, early, multicomponent intervention that arms the CG with skills to manage stress and prevent chronic negative outcomes.


Despite the limitations, results suggest that a problem-solving intervention can help both CGs and stroke survivors adapt to a stroke in the family. The treatment was accessible to CGs, including those in rural communities, and was cost-effective because the telephone intervention eliminates the need for travel by participants and counselors. Such interventions may be even more important in the current environment of ever shorter lengths of stay for inpatient rehabilitation.


This work was supported by the Buehler Center on Aging and the Ralph and Marian C. Falk Research Trust (Rosemarie B. King, PI). We wish to thank the caregivers who graciously gave of their time and energy to participate in this study. We also thank Maria Bergman, PhD, and Patrick Semik, BA, for their assistance with this study.

About the Authors

Rosemarie B. King, PhD RN, is a research associate professor at the Northwestern University Feinberg School of Medicine and adjunct senior research associate at the Rehabilitation Institute of Chicago. Address correspondence to her at 345 E. Superior Street, Chicago, IL 60611, or rbking@northwestern.edu.

Robert J. Hartke, PhD MPH, is a staff psychologist at the Rehabilitation Institute of Chicago and an assistant professor at the Northwestern University Feinberg School of Medicine.

Florence Denby, MS CRRN, is a nurse practitioner at the Stroke Rehabilitation Center, Rehabilitation Institute of Chicago, and an instructor at Northwestern University Feinberg School of Medicine.


Anderson, C. S., Linto, J., & Stewart-Wynne, E. G. (1995). A population based assessment of the impact and burden of caregiving for long-term stroke survivors. Stroke, 26, 843–849.

Archbold, P. G., Stewart, B. J., Greenlick, M. R., & Harvath, T. (1990). Mutuality and preparedness as predictors of caregiver role strain. Research in Nursing and Health, 13, 315–384.

Bakas, T., Austin, J. K., Jessup, S. L., Williams, L. S., & Oberst, M. T. (2004). Time and difficulty of tasks provided by family caregivers of stroke survivors. Journal of Neuroscience Nursing, 36(2), 95–106.

Bakas, T., & Champion, V. (1999). Development and psychometric testing of the Bakas Caregiving Outcome Scale. Nursing Research, 48, 250–259.

Beck, A. T., Rush, A. J., Shaw, B. F., & Emery, G. (1979). Cognitive therapy of depression. New York: Guilford.

Carey, P. J., Oberst, M. T., McCubbin, M. A., & Hughes, S. H. (1991). Appraisal and caregiving burden in family members caring for patients receiving chemotherapy. Oncology Nursing Forum, 18(8), 1341–1348.

Clark, M. S., Rubenach, S., & Winsor, A. (2003). A randomized controlled trial of an education and counseling intervention for families after stroke. Clinical Rehabilitation, 17, 703–712.

Clark, P. C. (2002). Effects of individual and family hardiness on caregiver depression and fatigue. Research in Nursing & Health, 25, 37–48.

Dennis, M., O’Rourke, S., Lewis, S., Sharpe, M., & Warlow, C. (1998). A quantitative study of the emotional outcome of people caring for stroke survivors. Stroke, 29, 1867–1872.

Dennis, M., O’Rourke, S., Slattery, J., Staniforth, T., & Warlow, C. (1997). Evaluation of a stroke family care worker: Results of a randomised controlled trial. British Medical Journal, 314 (7087), 1071–1076.

D’Zurilla, T. J. (1986). Problem-solving therapy: A social competence approach to clinical intervention. New York: Springer.

Edelman, P., & Fulton, B. (1997). Development and testing of a measure of healthy caregiving. Presented at the Sixth National Alzheimer’s Disease Education Conference, Chicago, IL, July 1997.

Evans, R. L., Matlock, A., Bishop, D. S., Stranahan, S., & Pederson, C. (1988). Family intervention after stroke: Does counseling or education help? Stroke, 19, 1243–1249.

Grant, J. S., Bartolucci, A. A., Elliot, T. R., & Newman Giger, J. (2000). Sociodemographic, physical, and psychosocial characteristics of depressed and non-depressed family caregivers of stroke survivors. Brain Injury, 14(12), 1089–1100.

Grant, J. S., Elliott, T. R., Weaver, M., Bartolucci, A. A., & Giger, J. N. (2002). Telephone intervention with family caregivers of stroke survivors after rehabilitation. Stroke, 33, 2060–2065.

Hartke, R. J., & King, R. B. (2002). Analysis of problem types and difficulty among older stroke caregivers. Topics in Stroke Rehabilitation, 9(1), 16–33.

Hartke, R. J., & King, R. B. (2003). Telephone group intervention for older stroke caregivers. Topics in Stroke Rehabilitation, 9(4), 65–81.

Kapella, M. C., Larson, J. L., Patel, M. K., Covey, M. K., & Berry, J. K. (2006). Subjective fatigue, influencing variables, and consequences in chronic obstructive pulmonary disease. Nursing Research, 55, 10–17.

King, R. B., Carlson, C. E., Shade-Zeldow, Y., Bares, K. K., Roth, E. J. (2001). Transition to home care after stroke: Depression, physical health, and adaptive processes in support persons. Research in Nursing & Health, 24, 307–323.

King, R. B., & Semik, P. E. (2006). Difficult times, service use, and needs during the first two years of stroke caregiving. Journal of Gerontological Nursing, 32(4), 37–44.

King, R. B., & Shade-Zeldow, Y. (1995). Adaptation to stroke: Comparison of spouses of stroke survivors with and without aphasia. Topics in Stroke Rehabilitation, 2, 20–32.

Kotila, M., Numminen, H., Waltimo, O., & Kaste, M. (1998). Depression after stroke: Results of the FINNSTROKE Study. Stroke, 29, 368–372.

Lynch, D. J., Tamburrino, M. B., & Nagel, R. (1997). Telephone counseling for patients with minor depression: Preliminary findings in a family practice setting. Journal of Family Practice, 44, 293–298.

Mant, J., Carter, J, Wade, D. T., & Winner, S. (1998). The impact of an information pack on patients with stroke and their carers: A randomized controlled trial. Clinical Rehabilitation, 12, 465–476.

McNair, D. M., Lorr, M., & Droppleman, L. F. (1992). EDITS manual for the Profile of Mood States (revised). San Diego: Educational and Industrial Testing Service.

Moos, R. H., & Tsu, V. D. (1977). The crisis of physical illness: An overview. In R. H. Moos (Ed.), Coping with physical illness (pp. 3–22). New York: Plenum.

Nyenhuis, D. L., Yamamoto, C., Luchetta, T., Terrien, A, & Parmentier, A. (1999). Adult and geriatric normative data and validation of the profile of mood states. Journal of Clinical Psychology, 55(1), 79–86.

Periard, M. E., & Ames, B. D. (1993). Lifestyle changes and coping patterns among caregivers of stroke survivors. Public Health Nursing, 10(4), 252–256.

Radloff, L. S. (1977). The CES-D scale: A self-report depression scale for research in the general population. Applied Psychological Measurement, 3, 385–401.

Rodgers, H., Atkinson, C., Bond, S., Suddes, M., Dobson, R., & Curless, R. (1999). Randomized controlled trial of a comprehensive stroke education program for patients and caregivers. Stroke, 30, 2585–2591.

Schulz, R., Burgio, L., Burns, R., Eisdorfer, C., Gallagher-Thompson, D., Gitlin, L. N., et al. (2003). Resources for enhancing Alzheimer’s caregiver health (REACH): Overview, site-specific outcomes, and future directions. The Gerontologist, 43(4):514–520.

Schulz, R., Tompkins, C. A., & Rau, M. T. (1988). A longitudinal study of the psychosocial impact of stroke on primary support persons. Psychology and Aging, 3(2), 131–141.

Teel, C. S., Duncan, P., & MinLai, S. (2001). Caregiving experiences after stroke. Nursing Research, 50, 53–60.

Uniform Data Management Service. (1993). Guide for the use of the Uniform Data Set for Medical Rehabilitation, Version 3.0. Buffalo: State University of New York at Buffalo Research Foundation.

Visser-Meily, A., van Heugten, C., Post, M., Schepers, V., & Lindeman, E. (2005). Intervention studies for caregivers of stroke survivors: A critical review. Patient Education and Counseling, 56, 257–267.

Wright, L. K., Hickey, J. V., Buckwalter, K. C., Hendrix, S. A., & Kelechi, T. (1999). Emotional and physical health of spouse caregivers of persons with Alzheimer’s disease and stroke. Journal of Advanced Nursing, 30(3), 552–563.

Earn nursing contact hours

Nursing contact hours for this activity are no longer available.