Brain Injury, February 2009; 23(2): 92–100
Social problem-solving abilities, relationship satisfaction and depression among family caregivers of stroke survivors
KALPANA SHANMUGHAM1, MIGUEL A. CANO2, TIMOTHY R. ELLIOTT2, & MATTHEW DAVIS2 1
Capital H Group, Troy, MI, USA and 2Texas A&M University, College Station, TX, USA
(Received 27 August 2008; revised 26 November 2008; accepted 30 November 2008)
Abstract Primary objective: To examine the prospective relation of caregiver problem-solving abilities and relationship satisfaction to caregiver depression 1 month following care recipient discharge from an inpatient stroke rehabilitation facility. Research design: Correlational, prospective design. Methods and procedures: Participants included 39 women (average age ¼ 51.47, SD ¼ 20.59) and four men (average age ¼ 42.50, SD ¼ 5.57) assuming caregiver roles for persons discharged from stroke rehabilitation. Main outcome measure: The Beck Depression Inventory was the outcome measure at discharge and 1 month later. Results: Caregivers experienced a significant decrease in depression scores between the discharge and 1-month assessments. Lower relationship satisfaction and dysfunctional problem-solving abilities were independently and significantly predictive of depression scores at discharge. Dysfunctional problem-solving abilities were indirectly predictive of depression 1 month later. Conclusions: Dysfunctional problem-solving abilities and low relationship satisfaction are associated with caregiver depression, but dysfunctional problem-solving abilities maintain an indirect association with depression over 1 month. Implications of these findings for developing problem-solving training programmes for caregivers post-discharge are discussed. Keywords: Stroke, caregivers, problem solving, relationship satisfaction, depression
Introduction Strokes are the leading cause of adult disability in the US. Since many survivors are left with mental and physical difficulties after a stroke, these individuals often require assistance with daily needs and functioning. The National Center on Caregiving Responsibilities [1] reports that family caregivers provide 75% of care to stroke survivors above and beyond any assistance that may be provided by a formal agency. Indeed, family caregivers probably constitute the ‘. . . largest group of health providers’ in the US [2] and the market value of their work exceeds that spent on formal health and nursing home care [3]. Nevertheless, family caregivers
experience considerable difficulties with their emotional and physical health [4] and these problems may be pronounced among caregivers of persons with neurological diseases such as stroke [5]. Despite the essential role family caregivers assume in the health and well-being of stroke survivors, effective support programmes for these caregivers are not routinely provided [6]. The needs of family caregivers now represent a major public health concern [7] and it is important to identify effective support services and interventions for community-residing caregivers that address their unique and dynamic concerns [8]. Although several interventions have been studied among caregivers of
Correspondence: Kalpana Shanmugham, PhD, Capital H Group, 3155 West Big Beaver Road, Ste 104, Troy, MI 48084, USA. Tel: 248-729-0314. Fax: 248-729-0301. E-mail:
[email protected] ISSN 0269–9052 print/ISSN 1362–301X online ß 2009 Informa Healthcare Ltd. DOI: 10.1080/02699050802657020
Problem solving and caregiving stroke survivors, this research has been limited by a lack of theory-driven models that take into consideration caregiver characteristics that may influence adjustment and examine testable propositions that can advance available knowledge [9]. There is considerable evidence that some caregivers are at greater risk for distress over time and many of their characteristics may be apparent early in the caregiver career. For example, individuals who report a dissatisfying relationship with the care recipient prior to the disability are more likely to experience greater burden, distress and resentment than caregivers who report a greater satisfaction with their relationship [10–12]. Caregivers of persons with stroke may be at risk for unintentional injuries—such as falls, cuts, scrapes, bruises—that range from minor to potentially serious [13]. There may be emotional consequences for care recipients as well. Some research indicates that the nature of spousal interactions is directly correlated with the functional abilities in stroke survivors [14]. Although these data provide important insights into the interpersonal dynamics of the caregiver experience, they offer little direction for developing theory-based interventions and services for family caregivers, generally. Studies of caregiver adjustment that are guided by the theoretical model of social problem-solving abilities have been particularly encouraging, because the model posits clear directives for cognitive-behavioural interventions that have effectively lowered distress among caregivers of persons with stroke [15], traumatic brain injury [16, 17], spinal cord injuries [18, 19] and among mothers of children with cancer [20]. Caregivers who have ineffective problem-solving styles are at risk for clinical levels of depression [21, 22] and their care recipients are at risk for preventable secondary complications [23]. Effective problem-solvers report greater benefits from social relationships [24] and display more interpersonal skills [25, 26] critical to community integration [27] than ineffective problem-solvers. Preliminary evidence suggests that effective problemsolving abilities may be associated with better family functioning and fewer family conflicts through its beneficial associations with empathy [28]. Unfortunately, one has yet to study caregiver problemsolving abilities and the level of satisfaction with their relationship with a care recipient as these factors pertain to caregiver adjustment over time. The present study examined the prospective associations between caregiver problem-solving abilities, relationship satisfaction with the care recipient and caregiver depression as a family member received services in an inpatient stroke rehabilitation programme and returned for outpatient services 1 month later. This 1-month window following
93
stroke rehabilitation is considered one of the most stressful times for caregivers as they adjust to new roles and routines and as the limitations of the stroke survivor become obvious in the home environment [29]. To control for the probable associations among the predictor variables with the two indicators of caregiver depression, this study relied on statistical procedures that would take into account the contextual relations among these variables [30]. Specifically, this study tested the theoretical assumption that effective problem-solving abilities would be prospectively associated with lower depression, regardless of the quality of relationship satisfaction and the extent of care recipient disability.
Method Participants Participants included 39 women (average age ¼ 51.47, SD ¼ 20.59) and four men (average age ¼ 42.50, SD ¼ 5.57) who were caregivers of stroke survivors and recruited from a rehabilitation hospital in Philadelphia, PA and another in Institute, WV. Caregivers of individuals diagnosed with CVAs and other neurological illnesses that result in similar, progressive, cognitive and motor deficits were also approached to participate. Individuals were invited to participate if (a) they were new to the caregiving role and (b) were family members who provided services for the care recipient in their home. Of those individuals that were approached to participate, three individuals declined. The sample included 37 Caucasians and six African-Americans. Participants averaged 12.35 years of formal education (SD ¼ 2.76): 79.4% had a high school diploma, 5.9% had bachelor’s degrees, 5.9% had master’s degrees and 8.8% had more than two graduate degrees. Twenty caregivers were spouses of the care recipient (mean age ¼ 64.79, SD ¼ 10.44), 15 were adult children (mean age ¼ 29.20, SD ¼ 6.25), one was a parent (age ¼ 67) and seven were classified as being in ‘other relationships’ (e.g. siblings, cousins (mean age ¼ 55.71, SD ¼ 20.06)). Most caregivers were married (58.1%), 23.3% were single, 7.0% were widowed and 11.6% had other situations (e.g. divorced). Participants averaged 9.44 daily caregiving hours (SD ¼ 7.18) post-discharge. All participation was voluntary and participants were informed that all of their responses would be held in strict confidence. Informed consent was explained to all participants and appropriately obtained before data collection began. Participants were informed of the sensitive nature of some of the questions (e.g. emotional status) prior to giving consent.
94
K. Shanmugham et al.
Measures Social problem-solving abilities. The Social ProblemSolving Inventory-Revised (SPSI-R) [31] is a 52-item self-report measure of social problemsolving abilities. Respondents rate questions on a 5-point Likert-type scale ranging from not very true of me (0) to extremely true of me (4). Higher scores represent more effective problem-solving abilities. The SPSI-R is based on a five-dimensional model of problem-solving and includes five scales [31]. Two of the SPSI-R scales measure problem orientation dimensions: Positive Problem Orientation and Negative Problem Orientation. The remaining three scales are considered problem-solving skills scales: Rational Problem-Solving, Impulsive/ Careless Style and Avoidant Style. Scales on the SPSI–R have shown high reliability (ranging from 0.69–0.93) and they have been moderately correlated in theoretically consistent directions with other external measures of psychological distress and well-being [31]. Criterionreferenced validity has been found in significant correlations with relevant scales on the ProblemSolving Inventory [32] and with other theoretically related constructs as stress, somatic symptoms, anxiety, depression, hopelessness and suicidality [31]. The SPSI–R scales have been predictably associated with self-esteem, life satisfaction, extraversion, social adjustment and social skills [33]. Consonant with contemporary notions of social problem-solving measurement, this study examined separate constructs of constructive and dysfunctional problem-solving styles [34]. The two positive measures (positive orientation, rational problem-solving skills) were summed to obtain an index of a constructive problem-solving style (CPS) and the three negative measures (negative orientation, impulsive/carelessness, avoidant style) were summed to form an index of a dysfunctional problem-solving style (DPS). Confirmatory, exploratory and principal component factor analyses with different samples have supported this conceptualization [35–37]. Higher scores on these two variables (CPS, DPS) indicate greater propensity for each dimension. Relationship satisfaction. Relationship satisfaction was assessed with a 5-point Likert-type scale that was not relationship-specific to accommodate the responses of parents and adult children who might assume the caregiver role. Caregivers were asked to rate the level of satisfaction that most accurately describes their relationship with the care recipient, ranging from 1 (very dissatisfied) to 6 (very satisfied). This item was developed for use in the current study with caregivers who would vary
considerably in formal relationship status (married, adult child, etc.) with a care recipient; therefore, no psychometric data were previously available. Functional disability. Care recipient functional disability of stroke survivors is associated with caregiver distress [38]. The Functional Independence Measure (FIM) [39] was used to assess care recipient disability. Items on this instrument assess self-care (i.e. eating, grooming, bathing, dressingupper, dressing-lower and toileting), sphincter control (bladder, bowel), transfers (bed/chair/wheelchair, toilet, tub/shower), locomotion (walk/wheelchair), communication (i.e. comprehension, expression) and social cognition (social interaction, problem-solving, memory). These items are individually rated on a 7-point Likert type scale and are referred to as FIM Levels. The rating scale is as follows: 1 ¼ total assistance or not testable, 2 ¼ maximal assistance, 3 ¼ moderate assistance, 4 ¼ minimal assistance, 5 ¼ supervision, 6 ¼ maximal assistance, 7 ¼ complete independence. These ratings were completed collaboratively by the care recipient’s multidisciplinary treatment team immediately prior to discharge. No other stroke survivor data were collected. Caregiver depression. The Beck Depression Inventory (BDI) [40] was used to assess caregiver depression. The BDI is a 21-item questionnaire measuring cognitive, somatic and behavioural symptoms of depressive behaviour. Each item is rated on a 4-point scale, with higher scores indicating a more intense display. The BDI has demonstrated adequate internal consistency ( ¼ 0.86) and adequate construct validity with inpatient psychiatric populations [41]. Substantive data indicate that the BDI has an adequate internal consistency (0.84) and its correlations with other depression measures range from 0.54–0.68 [42]. Procedure Caregivers were recruited prior to discharge. At the end of the care recipient’s inpatient stay, caregivers were scheduled to attend all therapies with the patient for the purpose of education and training for their new role. The present study recruited caregivers to fill out the SPSI-R and the BDI within a week of their attendance in an inpatient rehabilitation programme. The Functional Impairment Measure (FIM) was completed by the medical team during this time. Care recipients returned to an outpatient clinic within the same facility for a medical follow-up 1 month after discharge. While waiting for the appointment, participating caregivers
Problem solving and caregiving completed the BDI. Individuals who did not attend this appointment were contacted by telephone and completed the measures while speaking with a research assistant. Data analysis Pearson r correlation coefficients and t-tests were calculated to examine relationships between the selfreport variables and demographic characteristics. Paired t-tests were conducted to examine possible changes in the BDI scores from discharge to the 1-month follow-up. A path model was created and tested with AMOS 7.0 to examine the direct influence of constructive problem-solving, dysfunctional problem-solving, relationship satisfaction and functional impairment of the stroke survivor on caregiver depression at discharge and at the 1-month follow-up assessment. A path model was utilized because it is a confirmatory (i.e. hypothesis testing) analysis that examines direct and indirect influences of measured variables on other measured variables [43, 44] and it was best suited for analysing hypothesized relationships between the variables under investigation. The theoretically-derived a priori model, displayed in Figure 1, was first analysed by treating all variables as observed (measured) variables. This model theoretically assumed that social problem-solving abilities—represented by the constructive and dysfunctional constructs—would be predictably associated with relationship satisfaction and significant paths would flow from each of the predictor variables to depression measured at discharge. This model
included the functional impairment of the stroke survivor to test the relative contributions of the psychological constructs to caregiver depression in the context of disability severity. Recommended statistics were used to determine the fit of the path model for the data, including the root-mean square error of approximation (RMSEA) and the comparative fit index (CFI) [45]. To further examine possible mediation effects between variables a macro for SPSS 15.0 by Preacher and Hayes [46] was utilized. The macro computed bootstrap confidence intervals with 10 000 iterations through multiple regression analysis. Bootstrap confidence intervals were used to determine significance of mediated effects over normal theory with p-values because they lack the power to detect a true mediation effect [47]. The Bootstrap analysis was also used because it is appropriate for testing models with smaller sample sizes [48, 49]. Bootstrapping, also referred to as resampling, removes individual scores in a random fashion, conducts a reanalysis of the remaining dataset and this is repeated in subsequent iterations. Each individual score has an equal possibility of being drawn for the bootstrap sample at each iteration and it is possible for an individual’s score to be drawn multiple times or not at all throughout the process [44, 50]. The bootstrap is used to create confidence intervals around the mean of the variable being resampled [50]. Bootstrapping is a computer intensive procedure because it is logistically difficult to conduct without specialized computer software [51]. Without the benefit of collecting a new
Constructive problem solving Dysfunctional problem solving
Functional impairment
95
Depression at discharge
Depression at follow-up 1 0,
1 0, e8
Relationship satisfaction
Figure 1. A priori model of caregiver problem-solving abilities, relationship satisfaction and depression.
e9
96
K. Shanmugham et al.
data set, bootstrapping may be the closet approximation to external replicability when limited to one data set [44]. Results Caregiver age was not significantly correlated with either dysfunctional or constructive problem-solving style, nor did it significantly correlate with caregiver depression scores at discharge or at the follow-up visit. There were no significant differences in BDI scores (at discharge and follow-up), relationship satisfaction or the problem-solving scores as a function of caregiver gender. Years of formal education among caregivers was significantly correlated with depression at discharge (0.33; p < 0.05) but not with depression at the follow-up visit. Relationship status was unrelated to depression at discharge and at the follow-up visit. Caregiver depression scores decreased significantly from discharge to the 1-month follow-up assessment, t(42) ¼ 5.08, p < 0.05. Table I presents the means, standard deviations and correlations for all variables used in subsequent analyses. Fit statistics for the theoretically-derived a priori model indicate a poor fit for the data, 2(8) ¼ 55.43, p < 0.01 (RMSEA ¼ 0.38, CFI ¼ 0.44). Consequently, the paths of the initial model were refitted to develop and test an empirically-derived model. The corrected model (see Table II) excluded path coefficients between the exogenous variables that were not statistically significant. The corrected model—depicted in Figure 2—was constructed with six observed variables including four exogenous variables (dysfunctional problem-solving, constructive problem-solving, functional impairment, relationship satisfaction) and two endogenous variables of depression (discharge, 1-month follow-up). This empirically-derived model provided a good fit to the data, 2(9) ¼ 3.96, p ¼ 0.91 (RMSEA ¼ 0.0, CFI ¼ 1.0). Dysfunctional problem-solving and
greater functional impairment were significantly predictive of greater depression at discharge ( p < 0.01). Additionally, lower relationship satisfaction was significantly predictive of higher caregiver depression at discharge ( p < 0.01). Although dysfunctional problem-solving and constructive problem-solving were significantly related as expected (0.84, p < 0.01), constructive problemsolving was not significantly associated with relationship satisfaction. Therefore, the path between constructive problem-solving and relationship was eliminated from the final model. Finally, caregiver depression at discharge was significantly associated with depression scores at the 1-month follow-up assessment, as initially expected (0.51, p < 0.01). All possible indirect effects on depression 1 month later were tested with a bias-corrected bootstrap method with 10 000 iterations on SPSS 15.0 with a macro program developed by Preacher and Hayes [46]. Procedures for evaluating mediational effects were observed [52]. Bootstrapping results computed a bias-corrected 95% confidence interval (CI) to indicate a statistically significant indirect effect between dysfunctional -olving on follow-up depression, mediated by depression at discharge, F(9, 42) ¼ 0.42, CI(0.020, 0.265). Consequently, a dysfunctional problem-solving style was indirectly predictive of caregiver depression at the 1-month follow-up assessment. Higher dysfunctional Table II. Maximum likelihood estimates of corrected model. Unstandardized Standardized estimate estimate
Path FIM ! BDI1*** DPS ! BDI1*** RELSAT ! BDI1*** BDI1 ! BDI2***
0.10 0.17 2.30 0.56
0.44 0.63 0.28 0.49
1
1. 2. 3. 4. 5. 6.
–
CPS DFS Depression discharge Relation satisfaction Depression follow-up FIM
2 0.84* –
3 0.33* 0.49* –
Critical ratio
0.023 0.030 0.919 0.152
4.30 5.74 2.51 3.68
DPS ¼ Dysfunctional problem solving; FIM ¼ Functional impairment; RELSAT ¼ Relationship Satisfaction; BDI1 ¼ Depression at discharge; BDI2 ¼ Depression at follow-up. *p < 0.05, **p < 0.01, ***p < 0.001.
Table I. Means, standard deviations and correlations used in the path models. Variable
SE
4 0.29 0.34* 0.04 –
5 0.23 0.28 0.47* 0.09 –
6 0.10 0.17 0.34* 0.06 0.11 –
M
SD
60.07 31.14 8.51 5.72 5.02 82.19
16.29 15.55 3.98 0.50 4.71 19.02
*p < 0.05. n ¼ 43. CPS ¼ Constructive problem-solving style (observed range 28–90), DPS ¼ Dysfunctional problemsolving style (observed range 8–71), Depression Discharge ¼ Beck Depression Inventory at Discharge (observed range 0–19), Relation Satisfaction ¼ Relationship Satisfaction (observed range 4–6), Depression Follow-up ¼ Beck Depression Inventory at Follow-up visit (observed range 0–14), FIM ¼ Functional Independence Measure (observed range 46–111).
Problem solving and caregiving problem-solving styles were associated with higher depression scores at discharge and they were significantly associated with higher depression scores at the follow-up assessment through its association with depression at discharge. The model fit statistics provide a good estimation of effect sizes [53]. Post-hoc analyses The significant decrease in scores on the Beck Depression Inventory from the discharge to 1-month follow-up assessment raised some concerns about the rates of clinical levels of depression in this sample. Following recommendations for determining clinical significance for determining depression severity from the BDI [54], persons with scores 10 and higher were classified as having a greater likelihood of mild-to-severe depression. Using this scheme, 44% of the sample at discharge could be classified as having symptoms associated with mild-to -severe symptoms of depression (19 out of 43). However, only 16% of the sample at the 1-month assessment met operational criteria for determining clinically significant levels of depression (seven out of 43).
Discussion These results are generally consistent with other prospective studies of family caregivers of persons with
97
spinal cord injury [55] and of stroke survivors [56]. A dysfunction problem-solving style was directly associated with caregiver depression at the time the care recipient was discharged from inpatient rehabilitation and it maintained a significant association with depression scores 1 month after their return to the community. Unique to the current study is the additional insight into the contributions of these associations in the context of the caregiver– care-recipient relationship. Problem-solving abilities were not meaningfully associated with relationship satisfaction and dysfunctional problem-solving styles remained independently predictive of caregiver depression. Although lower relationship satisfaction was significantly predictive of higher caregiver depression at discharge, there were no apparent indirect effects from relationship satisfaction to caregiver depression 1 month later. The predictive associations of dysfunctional problem-solving styles and relationship satisfaction to caregiver depression were independent of the degree of stroke survivor functional impairment. Social problem-solving abilities may be associated with a variety of interpersonal and relationship skills (e.g. empathy, assertiveness), but they remain important predictors of caregiver adjustment over time regardless of the quality of the caregiver–carerecipient relationship. It should be noted that the current caregiver sample reported a significant
Constructive problem solving
−0.84
Dysfunctional problem solving
0.33 0.84
0.26
0.62 Functional impairment
0.45
0.27
Depression at discharge
e1
0.51
Depression at follow-up
e2
Relationship satisfaction Figure 2. Empirically-derived model of caregiver problem-solving abilities, relationship satisfaction and depression.
98
K. Shanmugham et al.
decline in depression in the month following discharge of the care recipient from the inpatient rehabilitation programme. This decline imposed a more conservative test for the a priori model, in contrast to the observed experiences of other caregivers in other disability scenarios (e.g. spinal cord injury) [55]. The restricted range in the BDI scores from discharge to the 1-month assessment may have attenuated the predictive potential of the relationship satisfaction variable. The functional disability scores for the stroke survivors in the current sample were in the mild impairment range. Many were cognitively intact and experienced few physical limitations. Therefore, it is likely that the resulting caregiver depression may not be as overwhelming compared to those living with someone requiring more continuous care and monitoring. This may account in part for the low sub-clinical rates of depression among caregivers 1 month following discharge. At discharge, the rate of depression in the current sample of stroke caregivers was comparable with that observed in other studies of family members assuming care for a stroke survivor (44% in the present study compared to 37% and 38%) [21, 57]. Interestingly, dysfunctional problem-solving styles were particularly important in the prediction of depression, as measured by the BDI. Constructive problem-solving styles—deemed essential in problem-solving training [33] and theoretically pivotal as a motivational factor in effective problem-solving [34]—did not significantly contribute to either model. Dysfunctional styles are consistently associated with self-report measures of adjustment and constructive styles have emerged as significant predictors in studies involving larger samples and in studies using observational measures of criterion variables [58]. The exact implications of these patterns are uncertain, but collectively they suggest that we have yet to fully appreciate how problemsolving styles influence optimal outcomes [58]. There are several limitations to the present study that temper the implications of the results. The sample size was rather small and the majority of participants were Caucasian. Generalizations from the results to other samples, generally, and to people from ethnic and minority backgrounds, specifically, are limited. A brief, one-item measure of relationship satisfaction was used in this study. Future work should examine the problem-solving and relationship satisfaction association with established, psychometrically-sound instruments of relationship satisfaction. The Beck Depression Inventory is sensitive to mood state and distress in general, so it is imprudent to equate elevated Beck scores with clinical depressive syndromes.
Despite these limitations, results of the present study are consistent with prior research of social problem-solving abilities among family caregivers. The first month following discharge from stroke rehabilitation is a time of considerable stress for families [29]. There is little evidence that basic educational programmes for caregivers are effective [6], but these may be supplemented by problemsolving principles. Home-based problem-solving interventions have been effective in lower distress among caregivers of children with brain injuries [17, 59] and to adults with brain injuries [16]. Problem-solving training programmes can be successfully provided to family caregivers of stroke survivors in the first month following rehabilitation [15]. However, the present study indicates that caregivers of stroke survivors may experience considerable distress prior to discharge from inpatient rehabilitation. Therefore, caregiver preparation programmes that provide training in problem-solving strategies might benefit caregivers during the inpatient stay. Follow-up sessions may then be provided to caregivers via telephone following their return to the community. Developing a partnership with the caregiver prior to discharge from inpatient rehabilitation has been a feature of randomized clinical trials that have supported the use of problem-solving training [15, 18]. In this model, preparation programmes may help alleviate caregiver concerns during this time of transition and equip them with useful cognitive-behavioural skills that may generalize to a variety of everyday problems.
Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.
References 1. National Center on Caregiving. Do stroke caregivers receive adequate support from our healthcare system? [On-line]. 2009. Available online at: http://www.caregiver.org/caregiver/ jsp/conten_node.jsp?nodeid ¼ 1758 Accessed 6 January 2009. 2. Parish SL, Pomeranz-Essley A, Braddock D. Family support in the United States: Financing trends and emerging initiatives. Mental Retardation 2003;41:174–187. 3. Vitaliano PP, Young H, Zhang J. Is caregiving a risk factor for illness? Current Directions in Psychological Science 2004;13: 13–16. 4. Vitaliano PP, Zhang J, Scanlan J. Is caregiving hazardous to one’s physical health? A meta-analysis. Psychological Bulletin 2003;129:946–972. 5. Glozman JM. Quality of life of caregivers. Neuropsychological Review 2004;14:183–196. 6. Han B, Haley WE. Family caregiving for patients with stroke: Review and analysis. Stroke 1999;30:1478–1485.
Problem solving and caregiving 7. Talley RC, Crews JE. Framing the public health of caregiving. American Journal of Public Health 2007;97: 224–228. 8. Elliott T, Pezent G. Family caregivers of older persons in rehabilitation. NeuroRehabilitation 2008;23:439–446. 9. Lui M, Ross F, Thompson D. Supporting family caregivers in stroke care: A review of the evidence for problem solving. Stroke 2005;36:2514–2522. 10. Williamson GM, Schulz R. Relationship orientation, quality of prior relationship, and distress among caregivers of Alzheimer’s patients. Psychology and Aging 1990;5: 502–509. 11. Williamson GM, Schulz R. Caring for a family member with cancer: Past communal behavior and affective reaction. Journal of Applied Social Psychology 1995;25:93–116. 12. Williamson GM, Shaffer DR, Schulz R. Activity restriction and prior relationship history as contributors to mental health outcomes among middle-aged and older spousal caregivers. Health Psychology 1998;17:152–162. 13. Hartke RJ, King R, Heinemann A, Semik P. Accidents in older caregivers of persons surviving stroke and their relation to caregiver stress. Rehabilitation Psychology 2006;51: 150–156. 14. Wright LK, Hickey JV, Buckwalter KC, Kelechi T, Hendrix S. Spousal interactions in Alzheimer’s disease and stroke caregiving: Relationships to care recipient’s functional abilities and physical and emotional health. Journal of the American Psychiatric Nurses Association 1998;4:169–181. 15. Grant J, Elliott T, Weaver M, Bartolucci A, Giger J. A telephone intervention with family caregivers of stroke survivors after hospital discharge. Stroke 2002;33: 2060–2065. 16. Rivera P, Elliott T, Berry J, Grant J. Problem-solving training for family caregivers of persons with traumatic brain injuries: A randomized controlled trial. Archives of Physical Medicine and Rehabilitation 2008;89:931–941. 17. Wade SL, Carey J, Wolfe CR. An online family intervention to reduce parental distress following pediatric brain injury. Journal of Consulting and Clinical Psychology 2006;74: 445–454. 18. Elliott T, Berry JW. Brief problem-solving training for family caregivers of persons with recent-onset spinal cord injuries: A randomized controlled trial. Journal of Clinical Psychology; in press. 19. Elliott T, Brossart D, Berry, JW, Fine PR. Problem-solving training via videoconferencing for family caregivers of persons with spinal cord injuries: A randomized clinical trial. Behaviour Research and Therapy 2008;46:1220–1229. 20. Sahler O, Fairclough D, Phipps S, Mulhern R, Dolgin M, Noll R, Katz E, Varni JW, Copeland D, Butler D. Using problem-solving skills training to reduce negative affectivity in mothers of children with newly diagnosed cancer: Report of a multisite randomized trial. Journal of Consulting and Clinical Psychology 2005;73:272–283. 21. Grant J, Weaver M, Elliott T, Bartolucci A, Newman-Giger J. Family caregivers of stroke survivors: Characteristics of caregivers at risk for depression. Rehabilitation Psychology 2004;49:172–179. 22. Dreer L, Elliott T, Shewchuk R, Berry J, Rivera P. Family caregivers of persons with spinal cord injury: Predicting caregivers at risk for depression. Rehabilitation Psychology 2007;52:351–357. 23. Elliott T, Shewchuck R, Richards JS. Family caregiver problem solving abilities and adjustment during the initial year of the caregiving role. Journal of Counseling Psychology 2001;48:223–232.
99
24. Elliott T, Herrick S, Witty T. Problem solving appraisal and the effects of social support among college students and persons with physical disabilities. Journal of Counseling Psychology 1992;39:219–226. 25. Heppner PP, Hibel J, Neal GW, Weinstein CL, Rabinowitz FE. Personal problem solving: A descriptive study of individual differences. Journal of Counseling Psychology 1982;29:580–590. 26. Neal GW, Heppner PP. Problem solving self-appraisal, awareness and utilization of campus helping resources. Journal of Counseling Psychology 1986;33:39–44. 27. Rath JF, Hennessy JJ, Diller L. Social problem solving and community integration in postacute rehabilitation outpatients with traumatic brain injury. Rehabilitation Psychology 2003;48:137–144. 28. Siu AMH, Shek DTL. Relations between social problemsolving and family well-being among Chinese adolescents in Hong Kong. Social Indicators Research 2005;71:517–539. 29. Grant JS, Glandon GL, Elliott T, Giger JN, Weaver M. Caregiving problems and feelings experienced by family caregivers of stroke survivors the first month post-discharge. International Journal of Rehabilitation Research 2004;27: 105–111. 30. Weston R, Gore Jr PA, Chan F, Catalano D. An introduction to using structural equation models in rehabilitation psychology. Rehabilitation Psychology 2008;53:340–356. 31. D’Zurilla TJ, Nezu AM, Maydeu-Olivares A. Social problem-solving inventory-revised (SPSI-R): Technical manual. North Tonawanda, NY: Multi-health Systems; 2002. 32. Heppner PP. Theproblem solving inventory: Manual. Palo Alto, CA: Consulting Psychologists Press; 1988. 33. Nezu A. Problem solving and behavior therapy revisited. Behavior Therapy 2004;35:1–33. 34. D’Zurilla TJ, Nezu AM, Maydeu-Olivares A. Social problem solving: theory and assessment. In: Chang E, D’Zurilla TJ, Sanna LJ, editors. Social problem solving: Theory, research, and training. Washington, DC: American Psychological Association; 2004. pp 11–27. 35. Berry J, Elliott T, Rivera P. Resilient, undercontrolled, and overcontrolled personality prototypes among persons with spinal cord injury. Journal of Personality Assessment 2007;89:292–302. 36. Johnson M, Elliott T, Neilands T, Morin SF, Chesney MA. A social problem-solving model of adherence to HIV medications. Health Psychology 2006;25:355–363. 37. Rivera P, Elliott T, Berry J, Oswald K, Grant J. Predictors of caregiver depression among community-residing families living with traumatic brain injury. NeuroRehabilitation 2007;22:3–8. 38. Perrin P, Heesacker M, Stidham BS, Rittman M, GonzalezRothi LJ. Structural equation modeling of the relationship between caregiver psychosocial variables and functioning of individuals with stroke. Rehabilitation Psychology 2008;53: 54–62. 39. Uniform Data Set for Medical Rehabilitation. Guide for the use of the Uniform Data Set for Medical Rehabilitation (Version 5.0). Buffalo, NY: State University of New York at Buffalo Research Foundation; 1996. 40. Beck AT, Ward CH, Mendelsohn MJ, Erbaugh J. An inventory for measuring depression. Archives of General Psychiatry 1961;4:561–571. 41. Beck A. Depression: Causes and treatment. Philadelphia: University of Pennsylvania Press; 1972. 42. Tanaka-Matsuma J, Kameoka VA. Reliabilities and concurrent validities of popular measures of depression, anxiety, and social desirability. Journal of Consulting and Clinical Psychology 1986;54:328–333.
100
K. Shanmugham et al.
43. Bryne B. Structural equation modeling with AMOS: Basic concepts, applications, and programming (Multivariate Applications Series). Florence, KY: Lawrence Erlbaum Associates; 2001. 44. Thompson B. Foundations of behavioral statistics: An insightbased approach. New York: The Guilford Press; 2006. 45. Kline RB. Principles and practice of structural equation modeling. 2nd ed. New York: The Guilford Press; 2005. 46. Preacher KJ, Hayes AF. SPSS and SAS macros for estimating and comparing indirect effects in multiple mediator models. 2007. Available online at: http://www.comm.ohio-state.edu/ ahayes/SPSS%20programs/indirect.htm, accessed 28 June 2007. 47. Mallinckrodt B, Abraham WT, Wei M, Russell DW. Advances in testing the statistical significance of mediation effects. Journal of Counseling Psychology 2006;53:372–387. 48. Mackinnon DP, Lockwood CM, Williams J. Confidence limits for the indirect effect: Distribution of the product and resampling methods. Multivariate Behavioral Research 2004;39:99–128. 49. Shrout PE, Bolger N. Mediation in experimental and nonexperimental studies: New procedures and recommendations. Psychological Methods 2004;7:422–445. 50. Erceg-Hurn DM, Mirosevivh VM. Modern robust statistical methods: An easy way to maximize the accuracy and power of your research. American Psychologist 2008;63:591–601. 51. Thompson B. DISCSTRA: A computer program that computes bootstrap resampling estimates of descriptive discriminant analysis function and structure coefficients and group centriods. Educational and Psychological Measurement 1992;52:905–911.
52. Hoyt WT, Imel ZE, Chan F. Multiple regression and correlation techniques: Recent controversies and best practices. Rehabilitation Psychology 2008;53: 321–339. 53. Thompson B. A suggested revision to the forthcoming 5th edition of the APA Publication Manual; 2000. Available online at: http://www.coe.tamu.edu/bthompson/apaeffec. htm, accessed 3 July 2007. 54. Groth-Marnat G. Handbook of psychological assessment. 3rd ed. New York: John Wiley & Sons; 1997. 55. Elliott T, Shewchuck R, Richards JS. Family caregiver problem solving abilities and adjustment during the initial year of the caregiving role. Journal of Counseling Psychology 2001;48:223–232. 56. Grant JS, Elliott T, Weaver M, Glandon G, Giger J. Social problem-solving abilities, social support, and adjustment of family caregivers of stroke survivors. Archives of Physical Medicine and Rehabilitation 2006;87: 343–350. 57. Grant JS, Bartolucci A, Elliott T, Giger JN. Sociodemographic, physical, and psychosocial characteristics of depressed and non-depressed family caregivers of stroke survivors. Brain Injury 2000;14:1089–1100. 58. Elliott T, Hurst M. Social problem solving and health. In: Walsh WB, editor. Biennial review of counseling psychology. New York: Lawrence Erlbaum; 2008. pp 295–314. 59. Wade SL, Carey J, Wolfe CR. The efficacy of an online cognitive behavioral family intervention in improving child behavior and social competence in pediatric brain injury. Rehabilitation Psychology 2006;51: 179–189.
