Caregiver Burden and Satisfaction in Families of Individuals With Schizophrenia Chiu-Yueh Hsiao ▼ Yun-Fang Tsai
Background: Caring for relatives with schizophrenia is complicated and has been associated with burden. Caregivers may also experience satisfaction, but these outcomes have not been simultaneously studied in family caregivers of relatives with schizophrenia. Objectives: The aim was to investigate the attributes of caregiver burden and satisfaction among individuals and families as well as the association of caregiver burden on caregiver satisfaction in the care of Taiwanese individuals with schizophrenia. Methods: A cross-sectional, descriptive, correlational design was used. A convenience sample of 140 families (243 individual family caregivers) was recruited from two psychiatric hospitals in Taiwan. Participants were interviewed individually to complete questionnaires regarding pileup of demands, sense of coherence, mutuality, caregiver burden, and caregiver satisfaction. Linear mixed modeling was used. Results: Female caregivers, greater family demands, decreased sense of coherence, and lower mutuality were associated with higher levels of caregiver burden, whereas being siblings or close relatives and being friends of the affected individuals were associated with lesser self-reported burden. Satisfaction was positively associated with caregiver age, sense of coherence, and mutuality. Burden and satisfaction were not significantly related. Discussion: Caring for family members with schizophrenia is burdensome but can also be a source of satisfaction. Correlates of caregiver burden appeared to be somewhat distinct from those of caregiver satisfaction. Further research on negative and positive aspects of caregiving is warranted to broaden the understanding of caregiving experiences and design therapeutic interventions to mitigate caregiver burden and enhance the sense of satisfaction with caregiving. Key Words: family caregivers • schizophrenia • Taiwan Nursing Research, July/August 2014, Vol 63, No 4, 260–269
S
chizophrenia, existing as a severe and persistent mental disorder, influences both the affected individuals and their relatives (Aggarwal, Avasthi, Kumar & Grover, 2011; Hsiao & Van Riper, 2009). In Taiwan, the incidence of schizophrenia increased from 3.27‰ of the population in 2006 to 4.87‰ in 2011 (Ministry of Health and Welfare, 2013). Most of the affected individuals either live with or have regular contact with their families, relatives, or significant others after they are discharged from psychiatric hospitals (Hsiao & Van Riper, 2009; Weimand, Hedelin, Sällatröm & Hall-Lord, 2010). It is globally recognized that family members become the primary source of providing care for a loved one with chronic or mental illness (World Federation for Mental Health, 2010). As Chiu-Yueh Hsiao, PhD, RN, is Associate Professor, School of Nursing and College of Medicine, Chung Shan Medical University, and Department of Nursing, Chung Shan Medical University Hospital, Taiwan, Republic of China.
Yun-Fang Tsai, PhD, RN, is Professor and Chair, School of Nursing and College of Medicine, Chang Gung University, and Consultant, Department of Nursing, Chang Gung Memorial Hospital at Keelung, Taiwan, Republic of China. Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s Web site (www.nursingresearchonline.com). DOI: 10.1097/NNR.0000000000000047
260
www.nursingresearchonline.com
the population of individuals with schizophrenia increases, the number of family caregivers is expected to grow substantially, and attention to the impact on family caregivers is likely to increase dramatically as well. Caring for individuals with schizophrenia may affect not only primary family caregivers but also other family members assuming caregiving activities (Hsiao & Van Riper, 2009). The incorporation of multiple intrafamilial correlations is pivotal to reflect outcomes for family units (Knafl et al., 2009). More attention needs to be directed at potential consequences for family caregivers and the family as a whole. The resiliency model of family stress, adjustment, and adaptation (McCubbin, Thompson & McCubbin, 1996) serves as a theoretical framework to guide this study. The resiliency model views the family as a unit for examining cumulative effects of demands, appraisal, resources, problem solving, and communication among members in the family that potentially contribute to family adaptation. On the basis of the resiliency model, the conceptual framework for this study (Figure 1) was primarily designed to address four domains: (a) pile-up of demands, (b) sense of coherence (SOC) as an appraisal, (c) mutuality as a resource, and (d) caregiver burden and satisfaction as an adaptation. Pile-up of demands is defined as the predictable (stressors) and unexpected (strains) life events and changes experienced by Nursing Research • July/August 2014 • Volume 63 • No. 4
Copyright © 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Nursing Research • July/August 2014 • Volume 63 • No. 4
Caregivers in Families With Schizophrenia
261
FIGURE 1. Theorized relationships between pileup of demands, sense of coherence, mutuality, caregiver burden, and caregiver satisfaction.
a family that tax the resources of the individual family members and the family unit (McCubbin et al., 1996). The affected families are not generally subject to one stressor, but rather experience external stressors (Hsiao & Van Riper, 2009). As found in previous caregiving literature, caregiver burden in mental illness is expected to increase with older caregivers (Chien, Chan & Morrissey, 2007; Hadryś, Adamowski & Kiejna, 2011), female caregivers (Weimand et al., 2010), closer kinship ties (Hadryś et al., 2011), limited availability of friends and other family members in caregiving involvement (Chien et al., 2007; Weimand et al., 2010), and prolonged caregiving duration (Hadryś et al., 2011; Huang, Hung, Sun, Lin & Chen, 2009). Most of these demands exclusively address individual-level characteristics. SOC is a belief that the world is structured and predictable; resources are available to meet the demands imposed by stimuli and a perception of demands as challenges worthy of engagement (Antonovsky, 1987). McCubbin et al. (1996) regarded SOC as a cognitive process in which individuals and families called upon their available resources to manage the accumulation of family stressors and strains. SOC has been found to reduce the burden (Andrén & Elmståhl, 2008; Suresky, Zauszniewski & Bekhet, 2008) for family caregivers of relatives with mental illness. Mutuality refers to the positive quality of a caregiver–care recipient relationship (Archbold, Stewart, Greenlick & Harvath, 1990). Mutuality is viewed as an interpersonal resource within the familial caregiving dyad that ameliorates caregiver role strain (Archbold et al., 1990; Shyu et al., 2010) and enhances satisfaction (López, López-Arrieta & Crespo, 2005) when caring for persons with mental illness and dependent elderly people. Caregiver burden refers to the negative connotation of activities involved with providing assistance to dependent relatives or friends that result in an imbalance between the demands and capabilities of the caregivers (Hunt, 2003). Providing care
for a relative with mental illness, such as schizophrenia, has mainly been characterized as an inevitably deleterious caregiving experience that leads to caregiver burden (Chien et al., 2007; Hsiao & Van Riper, 2009; Weimand et al., 2010). Particularly, the caregiving experience perceived by family caregivers of relatives with schizophrenia is more negative compared to other psychiatric illnesses, such as affective disorders (Grover et al., 2012). Caregiver satisfaction is conceptualized as “subjectively perceived gains from desirable aspects of or positive affective returns from caregiving” (Lawton, Moss, Kleban, Glicksman & Rovine, 1991). It is mainly derived from the following sources of satisfaction: (a) interpersonal dynamics between the caregiver and the care recipient, (b) the intrapersonal orientation of the caregiver, and (c) an intention to enhance positive consequences for the care recipient (Nolan, Grant & Keady, 1996). Kate, Grover, Kulhara and Nehra (2013)) asserted that caregiving experiences have the potential for a positive transformation. A study assessing caregiving gains among primary family caregivers of patients with schizophrenia reported that more than half of caregivers become more aware of their inner sense of strengths and rearranged life priorities (Chien & Greenberg, 2004). Studies suggest that family caregivers who were men, older (López et al., 2005), and more educated (Aggarwal et al., 2011) and had greater numbers of friends and other family members’ caregiving involvement (Chen & Greenberg, 2004) reported more positive caregiving experiences. Likewise, other investigators found no significant correlations between sociodemographics and positive aspects of caregiving (Kate et al., 2013). The positive aspect of caregiving has gradually been acknowledged; however, most research has focused on caregiver burden. Some researchers have postulated that caregiving for affected relatives might bring rewards (Huang et al., 2009),
Copyright © 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
262
Caregivers in Families With Schizophrenia
satisfaction (López et al., 2005), and motivation for the caregiver role (Kate et al., 2013). Yet, other studies have indicated that caregivers who reported caregiving experiences more negatively also viewed caregiving experiences more positively (Aggarwal et al., 2011; Grover et al., 2012). The lack of consistent findings with respect to the relationship between positive and negative caregiving experiences remains. Several limitations in relation to caregiving in mental illness have been noted. First, prior research has typically viewed the individual as a unit of analysis, addressing the association between individual characteristics and outcomes of primary family caregivers as opposed to uses of a family-level theoretical framework (e.g., a resiliency model) and family characteristics (e.g., family demands, SOC, and mutuality). Second, despite schizophrenia being a family affair, most caregiving research has ignored the interrelationships existing in a family unit. Little is known about how a family unit reacts to caregiving. Lastly, burden related to caring for relatives with mental illness has been extensively studied. This emphasis might grossly overshadow research that demonstrates positive transformation in caregivers’ lives. Scant attention has been given to the positive aspects of caregiving, particularly both positive and negative experiences of caregiving in Taiwan as well as in other Asian and Western countries. The purpose of this study was to investigate the attributes of caregiver burden and satisfaction among individuals and families caring for individuals with schizophrenia. The three aims were as follows: 1. to identify the degrees of family demands, SOC, mutuality, and caregiver burden and satisfaction at individual and family levels; 2. to investigate the impacts of pile-up of demands, SOC, and mutuality on caregiver burden and satisfaction at individual and family levels; and 3. to assess the effect of caregiver burden on caregiver satisfaction. It was hypothesized that (a) pile-up of demands, SOC, and mutuality significantly accounted for caregiver burden and satisfaction at individual and family levels and (b) caregiver burden significantly affected caregiver satisfaction.
METHODS Design A cross-sectional correlational design was used.
Sample A convenience sample of 140 families (243 individual family caregivers) was recruited from the psychiatric outpatient clinics at a medical center and mental teaching hospital in middle Taiwan. Most of the families who participated in the study had
www.nursingresearchonline.com
more than two family caregivers within each family (66.4%). In 47 families (33.6%), only one family caregiver participated. For this study, a caregiver was defined as the one who provides medical, social, or financial support for dependent individuals (Anderson, Keith, Novak & Elliot, 2002), such as individuals with schizophrenia. The inclusion criteria of family caregivers for this study were as follows: (a) age of 18 years or older, (b) having a relative diagnosed with schizophrenia according to the criteria in the Diagnostic and Statistical Manual of Mental Disorders-IV-TR, (c) living in the same household or contacting with the patient weekly, (d) ability to read Mandarin or communicate either Taiwanese dialect or Mandarin (the official Chinese language), and (e) willingness to participate in the study. The sample size was determined according to a medium effect size of .15, an alpha of .05, and a power of .80 (Cohen, 1992). Considering the characteristics or behaviors of individuals within the same family to be similar, the target sample size would be multiplied by the design effect (DEFF). The formula for the DEFF is DEFF = 1 + (m − 1) × ICC (Donner & Klar, 2000), where m is the average number of caregivers within a family, and intraclass correlation (ICC) was set as .25 for the family data (Kenny & Kashy, 1991). Another 5% was added to adjust for potential missing survey data. As suggested by Dillman (2007), a size of 200–300 was expected for a survey study.
Data Collection Following approval from institutional review boards in two Taiwanese hospitals, a pilot testing was conducted with 15 families who volunteered to be interviewed to ascertain the extent to which the questionnaires were clear; minor amendments were then made. Participants who enrolled in a pilot study were excluded from the main analysis. The researcher provided mental healthcare providers (psychiatrists and nurses) with detailed information about the purpose and the recruitment process and asked them to identify eligible participants of this study. Caregivers who met the inclusion criteria were approached by the researcher with a full explanation of the study; confidentiality was assured. After obtaining written informed consent, an appointment was made. Information was collected through an approximately 50-minute, face-to-face structured interview.
Variables and Measurement Demographic characteristics of family caregivers included age, gender, marital status, education, coresidence, current employment status, relationship to the individual with schizophrenia, the total number of additional co-caregivers, and years of family caregiving involvement. Family demands were measured by the Family Stressors Index (McCubbin & Patterson, 1996a) and Family Strains Index (McCubbin & Patterson, 1996b), a 20-item survey developed from a modification of the entire 71-item Family Inventory of Life Events (FILE), designed to assess both normative (stressor) and nonnormative (strain) life events and changes experienced
Copyright © 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Nursing Research • July/August 2014 • Volume 63 • No. 4
by individuals in a family unit during the past year. Each individual family caregiver was asked to indicate whether or not an item had occurred in their family during the past year. Each item score is assigned a standard weight according to its magnitude and the intensity of the life situations or strains. Scores are summed and then divided by the total numbers of items. The final score is the sum of scores of the Family Stressors Index and Family Strains Index, with higher scores reflecting greater proportion of demands experienced in relation to the number of possible demands. The possible scores range from a low of 6 to a high of 92. Family Stressors Index possessed a validity coefficient of .60 with the original FILE (McCubbin & Patterson, 1996a), and the reliability of scores on the Family Strains Index was .69 and a correlation of .87 with the original FILE was reported (McCubbin & Patterson, 1996b). The Family Stressors Index and Family Strains Index were translated into Mandarin Chinese by Hsiao and Van Riper (2011). Cronbach’s alpha was .82 for the Family Stressors Index and .81 for the Family Strains Index in Taiwanese parents of children with Down syndrome. Cronbach’s alpha in this study was .76 for the Family Stressors Index and .84 for the Family Strains Index. SOC was assessed with the 13-item SOC Scale developed by Antonovsky (1987) to measure three components: (a) comprehensibility, (b) manageability, and (c) meaningfulness. The items are measured on a 7-point scale, with total scores ranging from 13 (weakest SOC) to 91 (strongest SOC). Eriksson and Lindström (2005) found that Cronbach’s alphas of scores on the SOC were .70–.92 in 127 studies, and the SOC was significantly correlated with constructs measured by Kobasa’s Hardiness Scale, HealthRelated Hardiness Scale, and Colby’s Adaptive Potential Scale. The Chinese version of the 13-item SOC was translated by Tang and Dixon (2002). Cronbach’s alpha was .89 for the total scale, and the total scale of the SOC-13 and its each subscale were significantly correlated with perceived overall coping capability. In the current study, Cronbach’s alpha was .92 for the entire scale and .86, .84, and .85 for the subscales of comprehensibility, manageability, and meaningfulness, respectively. Mutuality was measured using the 15-item Mutuality Scale to assess the relationship quality between family caregivers and care recipients (Archbold et al., 1990). Total scores were obtained by summing the mean across all items, with scores ranging from 0 = not at all to 4 = a great deal. Higher scores reflect more positive quality of the relationship. Cronbach’s alphas of .91 for family caregivers of older person (Archbold et al., 1990) have been reported. The Mutuality Scale was translated into Mandarin Chinese by Shyu et al. (2010), and its Cronbach’s alpha was .94 for primary family caregiver of older people with dementia. In the current study, Cronbach’s alpha was .95. Caregiver burden was measured with the Caregiver Burden Scale-Brief (Song, 2002) developed to assess objective and subjective burden. It consists of family disruption, stigma, guild, caregiver strain, and client dependency subscales. The Caregiver Burden Scale-Brief consists of 18 items with responses
Caregivers in Families With Schizophrenia
263
reflecting a 5-point scale, ranging from 0 (never) to 4 (almost always). Total scale scores are obtained by summing across 16 items, excluding two positive items. Scores of 0–7 are generally interpreted to represent no burden; 8–20, mild to moderate burden; 21–32, moderate burden; 33–44, moderately severe burden; and 45–64, severe burden. Cronbach’s alpha for scores on the entire scale ranged from .88 to .94, and subscales ranged from .65 to .95 (Hou, Ke, Su, Lung & Huang, 2008; Hsiao & Van Riper, 2009; Song, 2002) have been reported. In this study, Cronbach’s alpha for the overall score was .93, and the family disruption, stigma, guilt, caregiver strain, and client dependency subscales were .89, .92, .83, .81, and .75, respectively. Caregiver satisfaction was measured using the Carers’ Assessment of Satisfaction Index (CASI) designed to explore family caregivers’ subjective experiences of satisfaction (Nolan et al., 1996). The 30-item CASI is classified into three categories corresponding to the main potential beneficiary, namely satisfaction experienced by family caregivers related to (a) the person being cared for, (b) the family caregiver, and (c) the interpersonal dynamics. Four alternative responses for each item range from 0 = does not apply to them to 3 = a great deal of satisfaction. Scores range from 0 to 90, with higher scores indicating greater satisfaction. Cronbach’s alpha for the CASI was .94 (Kuuppelomäki, Sasaki, Yamada, Asakawa & Shimanouchi, 2004). For this study, the self-reported of CASI was translated into Madarin Chinese by the principal investigator, back-translated into English by a second bilingual translator, and then a comparison of the back-translated and original versions was made. The expert panel was asked to rate the relevance of the content of the Chinese version of the CASI with a content validity index of .87. In this study, Cronbach’s alphas were .95.
Data Analysis Descriptive statistics was examined to describe the characteristics of individual family caregivers and the family unit. The ICC coefficient was computed to indicate within-family homogeneity in caregiver burden and satisfaction. Residual analyses were undertaken to perform the regression diagnoses, and the results revealed that the assumptions of normality, linearity, and homoscedasticity were generally met. Preliminary correlation analyses were undertaken to determine the correlation of sociodemographic characteristics (i.e., age, gender, marital status, education, coresidence, occupation, relationship to the individuals with schizophrenia, the total number of additional co-caregivers, and caregiving duration), family demands, SOC, and mutuality to caregiver burden and satisfaction. Only significant attributes were selected for subsequent analyses. Linear mixed modeling was applied to estimate the effects of determinants of caregiver burden and satisfaction from individual family caregivers (Level 1) within the nested structure of the families (Level 2; Knafl et al., 2009; Singer, 1998). It was proposed that each family has a different intercept and slope that aims to predicting the Level 1 dependent variable. It was
Copyright © 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
264
Caregivers in Families With Schizophrenia
stipulated that an individual’s caregiver burden and satisfaction is related to his or her caregiver burden and satisfaction, but also that the relationship between family demands, SOC, and mutuality vary across families. The effects of independent variables on dependent variables were undertaken with a twolevel random coefficient model correcting for an intrafamilial correlations within the family. Level 1 examines the variation of dependent variables (i.e., caregiver burden and satisfaction) among individual family caregivers within each family. The regression coefficient of each family’s equation differs randomly. Level 2 represents variations in regression coefficients across families by each independent variable (i.e., family demands, SOC, and mutuality). In this model, individual characteristics included age, gender, the total number of additional co-caregivers, relationship to individuals with schizophrenia, and caregiving duration. Family characteristics included family demands, SOC, and mutuality. An unstructured covariance matrix variance– covariance matrix structure was selected with the lowest Bayesian Information Criterion to select the best model. The analytic model is shown in Document, Supplemental Digital Content 1, http://links.lww.com/NRES/A118. A nominal type I error rate with an alpha level of .05 was set for all statistical analyses.
The analytic model is shown in Document, Supplemental Digital Content 1 RESULTS The characteristics of family caregivers and the means, standard deviations, and ranges of study variables of individual family caregivers and the families are presented in Tables 1 and 2, respectively. Table 1 describes the caregiver characteristics; the mean age of the caregivers was 56.24 years (SD = 14.33). Most of family caregivers were female (57.2%, n = 139), married (81.5%, n = 198), junior school graduate or above (63.4%, n = 154), lived with the patients (75.7%, n = 184), employed (76.1%, n = 185), and parents (54.7%, n = 133). Two hundred twenty-three family caregivers (91.8%) reported that they had at least one co-caregiver, and duration of caregiving ranged from 1 to 46 years (M = 12.67, SD = 8.52). Table 2 depicts a summary of study variables for individuals and families, respectively. The mean scores for family demands, SOC, mutuality, caregiver burden, and satisfaction at an individual level and a family level were slightly different, reflecting individual family members within the family unit caring for relatives with schizophrenia might respond differently. The ICC was .25 and .13 for caregiver burden and satisfaction, respectively.
www.nursingresearchonline.com
TABLE 1. Characteristics of Family Caregivers Characteristic Gender (male) Marital status Single Married Divorced Widowed Education Elementary school or below Junior school High school Junior college or college Graduate school Coresidence (yes) Employed (yes) Relationshipa Parents Spouse Sibling Child Significant others Additional co-caregivers 0 1 >1
n
%
104
42.8
17 198 9 19
7.0 81.5 3.7 7.8
89 37 51 59 7 184 185
36.6 15.3 20.9 24.3 2.9 75.7 76.1
133 19 56 19 16
54.7 7.9 23 7.8 6.6
20 171 52
8.2 70.4 21.4
Note. n = 243. Average age was 56.24 years (SD = 14.33). Average caregiving duration was 12.67 years (SD = 8.52). aTo person with schizophrenia.
Satisfaction experienced by family caregivers derived from the cared-for person, the family caregiver, and interpersonal dynamics are illustrated in Table 3. For all items, the majority endorsed “a great deal of satisfaction or quite a lot of satisfaction.” Among sources of satisfaction associated with the person being cared for, most of the family caregivers (90.5%) derived a great deal of satisfaction or quite a lot of satisfaction from “seeing the person I care for clean, comfortable, and well turned out.” Approximately 87.2% of family caregivers reported a great deal of satisfaction or quite a lot of satisfaction from “caring enables me to fulfill my sense of duty.” The most common source of satisfaction, expressed as a great deal of satisfaction or quite a lot of satisfaction, relating to interpersonal dynamics, was “caring is one way of expressing my love for the person I care for” (87.2%). In addition, the proportion of sources of satisfaction rated by family caregivers as no real satisfaction ranged from 9.1% to 29.2%. Table 4 shows that female caregivers, greater family demands, decreased SOC, and lower mutuality increased caregiver burden; however, being siblings or close relatives and friends of the affected individuals were found to decrease caregiver burden. Older caregivers, greater SOC, and higher mutuality enhanced caregiver satisfaction at an individual level. With respect to the random effects of estimates at family levels, family demands and mutuality influenced caregiver burden across families differed notably. The impact of family demands, SOC, and mutuality on caregiver satisfaction significantly varied across families.
Copyright © 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Caregivers in Families With Schizophrenia
Nursing Research • July/August 2014 • Volume 63 • No. 4
265
TABLE 2. Descriptive Statistics of Variables for Individual Family Caregivers Individual caregivers Variable Family demands Sense of coherence Mutuality Caregiver burden Caregiver satisfaction
Families a
M
SD
Range
20.01 64.88 2.09 24.32 57.21
15.32 13.10 0.69 10.78 14.24
0–68 26–91 0–4 3–64 5–90
M
SD
Rangea
ICC
20.05 64.54 2.1 24.03 57.84
14.14 11.33 0.58 9.40 11.69
0–68 30–91 1–4 3–64 17–89
.25 .13
Note. Level 1 (caregivers): n = 243; Level 2 (families): n = 140. ICC = intraclass correlation. aRounded to nearest integer.
The absence of significant influence of caregiver burden on caregiver satisfaction (β = −.15; p = .09) indicated that caregiver burden and satisfaction might be different aspects of caregiving.
DISCUSSION The findings suggest that caring for people with schizophrenia not only is associated with the burden perceived by caregivers
but can also be a source of satisfaction in caregivers’ daily lives. Gender, kinship, and family demands were associated only with caregiver burden, whereas SOC and mutuality were related to both caregiver burden and satisfaction at individual levels. The influences of family demands and mutuality on caregiver burden as well as the impact of family demands, SOC, and mutuality on caregiver satisfaction significantly differ across
TABLE 3. Means, SD, and Corrected Item-Total Correlations for the Carers’ Assessment of Satisfaction Index Item Cared-For Person as Main Beneficiary 15. Knowing the person means I can give better care than anyone else 5. I am able to help the person I care for reach their potential 22. I am able to ensure that the person I care for has their needs tended to 4. It’s good to see small improvements in their condition 20. I am able to keep the person I care for out of an institution 13. It’s good to help the person I care for overcome difficulties and problems 24. Maintaining the dignity of the person I care for is important to me 14. It’s nice when something I do gives the person I care for pleasure 12. I get pleasure from seeing the person I care for happy 9. It is nice to see the person I care for clean, comfortable and well turned out Caregiver as Main Beneficiary 23. Caring has given me the chance to widen my interests and contacts 27. Caring has provide a purpose in my life that I did not have before 7. Caring provides a challenge 19. It helps to stop me from feeling guilty 8. Despite all their problems the person I care for does not grumble or moan 1. Caring has allowed me to develop new skills and abilities 16. Caring has helped me to grow and develop as a person 25. I am able to test myself and overcome difficulties 2. The person I care for is appreciative of what I do 17. It’s nice to feel appreciated by those family and friends I value 30. Caring makes me feel needed and wanted 10. Caring enables me to fulfill my sense of duty Interpersonal Dynamics as Main Beneficiary 26. Caring is one way of showing my faith 6. I am able to repay their past acts of kindness 3. Caring has brought me closer to the person I care for 21. If the situation were reversed, the person I care for would do the same for me 18. Caring has strengthened close family ties and relationships 11. I am the sort of person who enjoys helping people 28. At the end of the day, I know I will have done the best I could 29. Caring is one way of expressing my love for the person I care for
M
SD
ra
1.79 1.79 1.98 2.01 2.01 2.14 2.17 2.19 2.21 2.26
0.76 0.79 0.70 0.66 0.88 0.65 0.58 0.61 0.63 0.63
.61 .70 .67 .65 .61 .64 .50 .62 .62 .60
1.47 1.51 1.55 1.62 1.67 1.70 1.71 1.78 1.86 2.00 2.12 2.14
0.92 0.97 0.88 0.84 0.80 0.85 0.88 0.74 0.72 0.68 0.73 0.69
.62 .57 .57 .45 .60 .55 .51 .63 .57 .57 .63 .51
1.80 1.81 1.86 1.88 1.90 2.04 2.06 2.20
0.78 0.88 0.82 .81 .82 .65 .67 .72
.62 .49 .68 .57 .65 .57 .62 .63
Note. Within subscales, items are ordered by mean values. Items reprinted with permission of the instrument developer and copyright holder, Dr. Mike Nolan. aCorrected item-total correlation.
Copyright © 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
266
Caregivers in Families With Schizophrenia
www.nursingresearchonline.com
families. These results are in harmony with an application of the resiliency model (McCubbin et al., 1996); particularly, increased pileup of demands, negative appraisal, and inadequate resources resulted in maladaptation among family members within the family units. Given that caregiver burden and satisfaction have some different attributes and no significant association was noted between caregiver burden and satisfaction, the study lends support to the idea that caregiver burden and satisfaction are two variables reflecting different reactions in caregiving circumstances. The mean score on the caregiver burden scale for individual family caregivers in this study was similar to those found in a study of primary family caregivers of people with schizophrenia (Hou et al., 2008) and slightly higher than those in relatives of people with severe and mental illness (Hsiao & Van Riper, 2009). Overall, family caregivers experienced more moderate degree of burden. The primary source of caregiver satisfaction was the cared-for person, especially keeping the individual clean, comfortable, and well turned out as well as maintaining the dignity of the cared-for person. In interpersonal dynamics, family caregivers mostly rooted their source of satisfaction in their way of expressing love to the cared-for person. These results are similar to other studies of family caregivers of patients with older relatives (Kuuppelomäki et al., 2004). In this study, most of family caregivers reported being satisfied with fulfilling their sense of duty—a finding that was not consistent with those reported by Kuuppelomäki et al. (2004). In contrast to previous findings (Aggarwal et al., 2011), the absence of a significant relationship between burden and satisfaction in family caregivers suggested that caregiver burden may coexist with
caregiver satisfaction in the care of family members with schizophrenia. As Folkman and Moskowitz (2000) postulated, individuals may generate or reinforce positive values for their efforts during stressful periods of caregiving, which enables them to sustain being a caregiver. Thus, it is noteworthy to gain a better understanding of how family caregivers identify positive transformation as a sense of satisfaction in the face of adverse caregiving circumstances. Similar to findings of earlier research (López et al., 2005; Weimand et al., 2010), female caregivers appeared to report increased caregiver burden whereas older caregivers experienced greater satisfaction. With traditional Confucian values and social expectations, female caregivers tend to regard taking care of their affected offspring as their inborn responsibility; therefore, it places tremendous burden on them when caring for relatives with long-term schizophrenia (Huang et al., 2009). In addition, older caregivers may become experienced and resilient in response to demands of care over time, which, in turn, generates a greater sense of inner strength and experienced greater satisfaction. Comparing individualistic Western populations, the interpersonal harmony is particularly salient in Chinese societies that address mandatory responsibilities toward others. The perception of losing face associated with stigma attached to mental illness may lead families to rely on the immediate or significant relatives and friends (Chien et al., 2007). Family caregivers with social desires may express greater satisfaction with caregiving and even being reluctant to access healthcare services. Continued investigations into the potential attributes of burden and satisfaction in diverse ethnic families of persons with schizophrenia are recommended.
TABLE 4. Mixed Models for Caregiver Burden and Caregiver Satisfaction Caregiver burden Effects Fixed, individual
Fixed, family
Random
Predictors Age Gendera Relationshipb Close relative/friend Child Spouse Sibling Additional co-caregiversc 0 >1 Caregiving duration Family demands Sense of coherence Mutuality Family demands Sense of coherence Mutuality Level 1
Estimate
Caregiver satisfaction
SE
p
SE
p
0.00 2.75
0.07 0.94
ns .01
Estimate 0.22 2.16
0.10 1.34
.05 ns
−7.94 −2.02 −2.26 −4.09
2.47 2.71 2.12 1.9
.01 ns ns .05
−1.34 6.68 3.37 1.39
3.68 3.61 3.10 2.81
ns ns ns ns
−1.32 1.20 0.08 0.21 −0.14 −0.22 0.05 0.06 0.30 26.16
1.86 1.26 0.08 0.05 0.06 0.08 0.03 0.05 0.11 6.83
ns ns ns
Background: Caring for relatives with schizophrenia is complicated and has been associated with burden. Caregivers may also experience satisfaction, but these outcomes have not been simultaneously studied in family caregivers of relatives with schizophrenia. Objectives: The aim was to investigate the attributes of caregiver burden and satisfaction among individuals and families as well as the association of caregiver burden on caregiver satisfaction in the care of Taiwanese individuals with schizophrenia. Methods: A cross-sectional, descriptive, correlational design was used. A convenience sample of 140 families (243 individual family caregivers) was recruited from two psychiatric hospitals in Taiwan. Participants were interviewed individually to complete questionnaires regarding pileup of demands, sense of coherence, mutuality, caregiver burden, and caregiver satisfaction. Linear mixed modeling was used. Results: Female caregivers, greater family demands, decreased sense of coherence, and lower mutuality were associated with higher levels of caregiver burden, whereas being siblings or close relatives and being friends of the affected individuals were associated with lesser self-reported burden. Satisfaction was positively associated with caregiver age, sense of coherence, and mutuality. Burden and satisfaction were not significantly related. Discussion: Caring for family members with schizophrenia is burdensome but can also be a source of satisfaction. Correlates of caregiver burden appeared to be somewhat distinct from those of caregiver satisfaction. Further research on negative and positive aspects of caregiving is warranted to broaden the understanding of caregiving experiences and design therapeutic interventions to mitigate caregiver burden and enhance the sense of satisfaction with caregiving. Key Words: family caregivers • schizophrenia • Taiwan Nursing Research, July/August 2014, Vol 63, No 4, 260–269
S
chizophrenia, existing as a severe and persistent mental disorder, influences both the affected individuals and their relatives (Aggarwal, Avasthi, Kumar & Grover, 2011; Hsiao & Van Riper, 2009). In Taiwan, the incidence of schizophrenia increased from 3.27‰ of the population in 2006 to 4.87‰ in 2011 (Ministry of Health and Welfare, 2013). Most of the affected individuals either live with or have regular contact with their families, relatives, or significant others after they are discharged from psychiatric hospitals (Hsiao & Van Riper, 2009; Weimand, Hedelin, Sällatröm & Hall-Lord, 2010). It is globally recognized that family members become the primary source of providing care for a loved one with chronic or mental illness (World Federation for Mental Health, 2010). As Chiu-Yueh Hsiao, PhD, RN, is Associate Professor, School of Nursing and College of Medicine, Chung Shan Medical University, and Department of Nursing, Chung Shan Medical University Hospital, Taiwan, Republic of China.
Yun-Fang Tsai, PhD, RN, is Professor and Chair, School of Nursing and College of Medicine, Chang Gung University, and Consultant, Department of Nursing, Chang Gung Memorial Hospital at Keelung, Taiwan, Republic of China. Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s Web site (www.nursingresearchonline.com). DOI: 10.1097/NNR.0000000000000047
260
www.nursingresearchonline.com
the population of individuals with schizophrenia increases, the number of family caregivers is expected to grow substantially, and attention to the impact on family caregivers is likely to increase dramatically as well. Caring for individuals with schizophrenia may affect not only primary family caregivers but also other family members assuming caregiving activities (Hsiao & Van Riper, 2009). The incorporation of multiple intrafamilial correlations is pivotal to reflect outcomes for family units (Knafl et al., 2009). More attention needs to be directed at potential consequences for family caregivers and the family as a whole. The resiliency model of family stress, adjustment, and adaptation (McCubbin, Thompson & McCubbin, 1996) serves as a theoretical framework to guide this study. The resiliency model views the family as a unit for examining cumulative effects of demands, appraisal, resources, problem solving, and communication among members in the family that potentially contribute to family adaptation. On the basis of the resiliency model, the conceptual framework for this study (Figure 1) was primarily designed to address four domains: (a) pile-up of demands, (b) sense of coherence (SOC) as an appraisal, (c) mutuality as a resource, and (d) caregiver burden and satisfaction as an adaptation. Pile-up of demands is defined as the predictable (stressors) and unexpected (strains) life events and changes experienced by Nursing Research • July/August 2014 • Volume 63 • No. 4
Copyright © 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Nursing Research • July/August 2014 • Volume 63 • No. 4
Caregivers in Families With Schizophrenia
261
FIGURE 1. Theorized relationships between pileup of demands, sense of coherence, mutuality, caregiver burden, and caregiver satisfaction.
a family that tax the resources of the individual family members and the family unit (McCubbin et al., 1996). The affected families are not generally subject to one stressor, but rather experience external stressors (Hsiao & Van Riper, 2009). As found in previous caregiving literature, caregiver burden in mental illness is expected to increase with older caregivers (Chien, Chan & Morrissey, 2007; Hadryś, Adamowski & Kiejna, 2011), female caregivers (Weimand et al., 2010), closer kinship ties (Hadryś et al., 2011), limited availability of friends and other family members in caregiving involvement (Chien et al., 2007; Weimand et al., 2010), and prolonged caregiving duration (Hadryś et al., 2011; Huang, Hung, Sun, Lin & Chen, 2009). Most of these demands exclusively address individual-level characteristics. SOC is a belief that the world is structured and predictable; resources are available to meet the demands imposed by stimuli and a perception of demands as challenges worthy of engagement (Antonovsky, 1987). McCubbin et al. (1996) regarded SOC as a cognitive process in which individuals and families called upon their available resources to manage the accumulation of family stressors and strains. SOC has been found to reduce the burden (Andrén & Elmståhl, 2008; Suresky, Zauszniewski & Bekhet, 2008) for family caregivers of relatives with mental illness. Mutuality refers to the positive quality of a caregiver–care recipient relationship (Archbold, Stewart, Greenlick & Harvath, 1990). Mutuality is viewed as an interpersonal resource within the familial caregiving dyad that ameliorates caregiver role strain (Archbold et al., 1990; Shyu et al., 2010) and enhances satisfaction (López, López-Arrieta & Crespo, 2005) when caring for persons with mental illness and dependent elderly people. Caregiver burden refers to the negative connotation of activities involved with providing assistance to dependent relatives or friends that result in an imbalance between the demands and capabilities of the caregivers (Hunt, 2003). Providing care
for a relative with mental illness, such as schizophrenia, has mainly been characterized as an inevitably deleterious caregiving experience that leads to caregiver burden (Chien et al., 2007; Hsiao & Van Riper, 2009; Weimand et al., 2010). Particularly, the caregiving experience perceived by family caregivers of relatives with schizophrenia is more negative compared to other psychiatric illnesses, such as affective disorders (Grover et al., 2012). Caregiver satisfaction is conceptualized as “subjectively perceived gains from desirable aspects of or positive affective returns from caregiving” (Lawton, Moss, Kleban, Glicksman & Rovine, 1991). It is mainly derived from the following sources of satisfaction: (a) interpersonal dynamics between the caregiver and the care recipient, (b) the intrapersonal orientation of the caregiver, and (c) an intention to enhance positive consequences for the care recipient (Nolan, Grant & Keady, 1996). Kate, Grover, Kulhara and Nehra (2013)) asserted that caregiving experiences have the potential for a positive transformation. A study assessing caregiving gains among primary family caregivers of patients with schizophrenia reported that more than half of caregivers become more aware of their inner sense of strengths and rearranged life priorities (Chien & Greenberg, 2004). Studies suggest that family caregivers who were men, older (López et al., 2005), and more educated (Aggarwal et al., 2011) and had greater numbers of friends and other family members’ caregiving involvement (Chen & Greenberg, 2004) reported more positive caregiving experiences. Likewise, other investigators found no significant correlations between sociodemographics and positive aspects of caregiving (Kate et al., 2013). The positive aspect of caregiving has gradually been acknowledged; however, most research has focused on caregiver burden. Some researchers have postulated that caregiving for affected relatives might bring rewards (Huang et al., 2009),
Copyright © 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
262
Caregivers in Families With Schizophrenia
satisfaction (López et al., 2005), and motivation for the caregiver role (Kate et al., 2013). Yet, other studies have indicated that caregivers who reported caregiving experiences more negatively also viewed caregiving experiences more positively (Aggarwal et al., 2011; Grover et al., 2012). The lack of consistent findings with respect to the relationship between positive and negative caregiving experiences remains. Several limitations in relation to caregiving in mental illness have been noted. First, prior research has typically viewed the individual as a unit of analysis, addressing the association between individual characteristics and outcomes of primary family caregivers as opposed to uses of a family-level theoretical framework (e.g., a resiliency model) and family characteristics (e.g., family demands, SOC, and mutuality). Second, despite schizophrenia being a family affair, most caregiving research has ignored the interrelationships existing in a family unit. Little is known about how a family unit reacts to caregiving. Lastly, burden related to caring for relatives with mental illness has been extensively studied. This emphasis might grossly overshadow research that demonstrates positive transformation in caregivers’ lives. Scant attention has been given to the positive aspects of caregiving, particularly both positive and negative experiences of caregiving in Taiwan as well as in other Asian and Western countries. The purpose of this study was to investigate the attributes of caregiver burden and satisfaction among individuals and families caring for individuals with schizophrenia. The three aims were as follows: 1. to identify the degrees of family demands, SOC, mutuality, and caregiver burden and satisfaction at individual and family levels; 2. to investigate the impacts of pile-up of demands, SOC, and mutuality on caregiver burden and satisfaction at individual and family levels; and 3. to assess the effect of caregiver burden on caregiver satisfaction. It was hypothesized that (a) pile-up of demands, SOC, and mutuality significantly accounted for caregiver burden and satisfaction at individual and family levels and (b) caregiver burden significantly affected caregiver satisfaction.
METHODS Design A cross-sectional correlational design was used.
Sample A convenience sample of 140 families (243 individual family caregivers) was recruited from the psychiatric outpatient clinics at a medical center and mental teaching hospital in middle Taiwan. Most of the families who participated in the study had
www.nursingresearchonline.com
more than two family caregivers within each family (66.4%). In 47 families (33.6%), only one family caregiver participated. For this study, a caregiver was defined as the one who provides medical, social, or financial support for dependent individuals (Anderson, Keith, Novak & Elliot, 2002), such as individuals with schizophrenia. The inclusion criteria of family caregivers for this study were as follows: (a) age of 18 years or older, (b) having a relative diagnosed with schizophrenia according to the criteria in the Diagnostic and Statistical Manual of Mental Disorders-IV-TR, (c) living in the same household or contacting with the patient weekly, (d) ability to read Mandarin or communicate either Taiwanese dialect or Mandarin (the official Chinese language), and (e) willingness to participate in the study. The sample size was determined according to a medium effect size of .15, an alpha of .05, and a power of .80 (Cohen, 1992). Considering the characteristics or behaviors of individuals within the same family to be similar, the target sample size would be multiplied by the design effect (DEFF). The formula for the DEFF is DEFF = 1 + (m − 1) × ICC (Donner & Klar, 2000), where m is the average number of caregivers within a family, and intraclass correlation (ICC) was set as .25 for the family data (Kenny & Kashy, 1991). Another 5% was added to adjust for potential missing survey data. As suggested by Dillman (2007), a size of 200–300 was expected for a survey study.
Data Collection Following approval from institutional review boards in two Taiwanese hospitals, a pilot testing was conducted with 15 families who volunteered to be interviewed to ascertain the extent to which the questionnaires were clear; minor amendments were then made. Participants who enrolled in a pilot study were excluded from the main analysis. The researcher provided mental healthcare providers (psychiatrists and nurses) with detailed information about the purpose and the recruitment process and asked them to identify eligible participants of this study. Caregivers who met the inclusion criteria were approached by the researcher with a full explanation of the study; confidentiality was assured. After obtaining written informed consent, an appointment was made. Information was collected through an approximately 50-minute, face-to-face structured interview.
Variables and Measurement Demographic characteristics of family caregivers included age, gender, marital status, education, coresidence, current employment status, relationship to the individual with schizophrenia, the total number of additional co-caregivers, and years of family caregiving involvement. Family demands were measured by the Family Stressors Index (McCubbin & Patterson, 1996a) and Family Strains Index (McCubbin & Patterson, 1996b), a 20-item survey developed from a modification of the entire 71-item Family Inventory of Life Events (FILE), designed to assess both normative (stressor) and nonnormative (strain) life events and changes experienced
Copyright © 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Nursing Research • July/August 2014 • Volume 63 • No. 4
by individuals in a family unit during the past year. Each individual family caregiver was asked to indicate whether or not an item had occurred in their family during the past year. Each item score is assigned a standard weight according to its magnitude and the intensity of the life situations or strains. Scores are summed and then divided by the total numbers of items. The final score is the sum of scores of the Family Stressors Index and Family Strains Index, with higher scores reflecting greater proportion of demands experienced in relation to the number of possible demands. The possible scores range from a low of 6 to a high of 92. Family Stressors Index possessed a validity coefficient of .60 with the original FILE (McCubbin & Patterson, 1996a), and the reliability of scores on the Family Strains Index was .69 and a correlation of .87 with the original FILE was reported (McCubbin & Patterson, 1996b). The Family Stressors Index and Family Strains Index were translated into Mandarin Chinese by Hsiao and Van Riper (2011). Cronbach’s alpha was .82 for the Family Stressors Index and .81 for the Family Strains Index in Taiwanese parents of children with Down syndrome. Cronbach’s alpha in this study was .76 for the Family Stressors Index and .84 for the Family Strains Index. SOC was assessed with the 13-item SOC Scale developed by Antonovsky (1987) to measure three components: (a) comprehensibility, (b) manageability, and (c) meaningfulness. The items are measured on a 7-point scale, with total scores ranging from 13 (weakest SOC) to 91 (strongest SOC). Eriksson and Lindström (2005) found that Cronbach’s alphas of scores on the SOC were .70–.92 in 127 studies, and the SOC was significantly correlated with constructs measured by Kobasa’s Hardiness Scale, HealthRelated Hardiness Scale, and Colby’s Adaptive Potential Scale. The Chinese version of the 13-item SOC was translated by Tang and Dixon (2002). Cronbach’s alpha was .89 for the total scale, and the total scale of the SOC-13 and its each subscale were significantly correlated with perceived overall coping capability. In the current study, Cronbach’s alpha was .92 for the entire scale and .86, .84, and .85 for the subscales of comprehensibility, manageability, and meaningfulness, respectively. Mutuality was measured using the 15-item Mutuality Scale to assess the relationship quality between family caregivers and care recipients (Archbold et al., 1990). Total scores were obtained by summing the mean across all items, with scores ranging from 0 = not at all to 4 = a great deal. Higher scores reflect more positive quality of the relationship. Cronbach’s alphas of .91 for family caregivers of older person (Archbold et al., 1990) have been reported. The Mutuality Scale was translated into Mandarin Chinese by Shyu et al. (2010), and its Cronbach’s alpha was .94 for primary family caregiver of older people with dementia. In the current study, Cronbach’s alpha was .95. Caregiver burden was measured with the Caregiver Burden Scale-Brief (Song, 2002) developed to assess objective and subjective burden. It consists of family disruption, stigma, guild, caregiver strain, and client dependency subscales. The Caregiver Burden Scale-Brief consists of 18 items with responses
Caregivers in Families With Schizophrenia
263
reflecting a 5-point scale, ranging from 0 (never) to 4 (almost always). Total scale scores are obtained by summing across 16 items, excluding two positive items. Scores of 0–7 are generally interpreted to represent no burden; 8–20, mild to moderate burden; 21–32, moderate burden; 33–44, moderately severe burden; and 45–64, severe burden. Cronbach’s alpha for scores on the entire scale ranged from .88 to .94, and subscales ranged from .65 to .95 (Hou, Ke, Su, Lung & Huang, 2008; Hsiao & Van Riper, 2009; Song, 2002) have been reported. In this study, Cronbach’s alpha for the overall score was .93, and the family disruption, stigma, guilt, caregiver strain, and client dependency subscales were .89, .92, .83, .81, and .75, respectively. Caregiver satisfaction was measured using the Carers’ Assessment of Satisfaction Index (CASI) designed to explore family caregivers’ subjective experiences of satisfaction (Nolan et al., 1996). The 30-item CASI is classified into three categories corresponding to the main potential beneficiary, namely satisfaction experienced by family caregivers related to (a) the person being cared for, (b) the family caregiver, and (c) the interpersonal dynamics. Four alternative responses for each item range from 0 = does not apply to them to 3 = a great deal of satisfaction. Scores range from 0 to 90, with higher scores indicating greater satisfaction. Cronbach’s alpha for the CASI was .94 (Kuuppelomäki, Sasaki, Yamada, Asakawa & Shimanouchi, 2004). For this study, the self-reported of CASI was translated into Madarin Chinese by the principal investigator, back-translated into English by a second bilingual translator, and then a comparison of the back-translated and original versions was made. The expert panel was asked to rate the relevance of the content of the Chinese version of the CASI with a content validity index of .87. In this study, Cronbach’s alphas were .95.
Data Analysis Descriptive statistics was examined to describe the characteristics of individual family caregivers and the family unit. The ICC coefficient was computed to indicate within-family homogeneity in caregiver burden and satisfaction. Residual analyses were undertaken to perform the regression diagnoses, and the results revealed that the assumptions of normality, linearity, and homoscedasticity were generally met. Preliminary correlation analyses were undertaken to determine the correlation of sociodemographic characteristics (i.e., age, gender, marital status, education, coresidence, occupation, relationship to the individuals with schizophrenia, the total number of additional co-caregivers, and caregiving duration), family demands, SOC, and mutuality to caregiver burden and satisfaction. Only significant attributes were selected for subsequent analyses. Linear mixed modeling was applied to estimate the effects of determinants of caregiver burden and satisfaction from individual family caregivers (Level 1) within the nested structure of the families (Level 2; Knafl et al., 2009; Singer, 1998). It was proposed that each family has a different intercept and slope that aims to predicting the Level 1 dependent variable. It was
Copyright © 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
264
Caregivers in Families With Schizophrenia
stipulated that an individual’s caregiver burden and satisfaction is related to his or her caregiver burden and satisfaction, but also that the relationship between family demands, SOC, and mutuality vary across families. The effects of independent variables on dependent variables were undertaken with a twolevel random coefficient model correcting for an intrafamilial correlations within the family. Level 1 examines the variation of dependent variables (i.e., caregiver burden and satisfaction) among individual family caregivers within each family. The regression coefficient of each family’s equation differs randomly. Level 2 represents variations in regression coefficients across families by each independent variable (i.e., family demands, SOC, and mutuality). In this model, individual characteristics included age, gender, the total number of additional co-caregivers, relationship to individuals with schizophrenia, and caregiving duration. Family characteristics included family demands, SOC, and mutuality. An unstructured covariance matrix variance– covariance matrix structure was selected with the lowest Bayesian Information Criterion to select the best model. The analytic model is shown in Document, Supplemental Digital Content 1, http://links.lww.com/NRES/A118. A nominal type I error rate with an alpha level of .05 was set for all statistical analyses.
The analytic model is shown in Document, Supplemental Digital Content 1 RESULTS The characteristics of family caregivers and the means, standard deviations, and ranges of study variables of individual family caregivers and the families are presented in Tables 1 and 2, respectively. Table 1 describes the caregiver characteristics; the mean age of the caregivers was 56.24 years (SD = 14.33). Most of family caregivers were female (57.2%, n = 139), married (81.5%, n = 198), junior school graduate or above (63.4%, n = 154), lived with the patients (75.7%, n = 184), employed (76.1%, n = 185), and parents (54.7%, n = 133). Two hundred twenty-three family caregivers (91.8%) reported that they had at least one co-caregiver, and duration of caregiving ranged from 1 to 46 years (M = 12.67, SD = 8.52). Table 2 depicts a summary of study variables for individuals and families, respectively. The mean scores for family demands, SOC, mutuality, caregiver burden, and satisfaction at an individual level and a family level were slightly different, reflecting individual family members within the family unit caring for relatives with schizophrenia might respond differently. The ICC was .25 and .13 for caregiver burden and satisfaction, respectively.
www.nursingresearchonline.com
TABLE 1. Characteristics of Family Caregivers Characteristic Gender (male) Marital status Single Married Divorced Widowed Education Elementary school or below Junior school High school Junior college or college Graduate school Coresidence (yes) Employed (yes) Relationshipa Parents Spouse Sibling Child Significant others Additional co-caregivers 0 1 >1
n
%
104
42.8
17 198 9 19
7.0 81.5 3.7 7.8
89 37 51 59 7 184 185
36.6 15.3 20.9 24.3 2.9 75.7 76.1
133 19 56 19 16
54.7 7.9 23 7.8 6.6
20 171 52
8.2 70.4 21.4
Note. n = 243. Average age was 56.24 years (SD = 14.33). Average caregiving duration was 12.67 years (SD = 8.52). aTo person with schizophrenia.
Satisfaction experienced by family caregivers derived from the cared-for person, the family caregiver, and interpersonal dynamics are illustrated in Table 3. For all items, the majority endorsed “a great deal of satisfaction or quite a lot of satisfaction.” Among sources of satisfaction associated with the person being cared for, most of the family caregivers (90.5%) derived a great deal of satisfaction or quite a lot of satisfaction from “seeing the person I care for clean, comfortable, and well turned out.” Approximately 87.2% of family caregivers reported a great deal of satisfaction or quite a lot of satisfaction from “caring enables me to fulfill my sense of duty.” The most common source of satisfaction, expressed as a great deal of satisfaction or quite a lot of satisfaction, relating to interpersonal dynamics, was “caring is one way of expressing my love for the person I care for” (87.2%). In addition, the proportion of sources of satisfaction rated by family caregivers as no real satisfaction ranged from 9.1% to 29.2%. Table 4 shows that female caregivers, greater family demands, decreased SOC, and lower mutuality increased caregiver burden; however, being siblings or close relatives and friends of the affected individuals were found to decrease caregiver burden. Older caregivers, greater SOC, and higher mutuality enhanced caregiver satisfaction at an individual level. With respect to the random effects of estimates at family levels, family demands and mutuality influenced caregiver burden across families differed notably. The impact of family demands, SOC, and mutuality on caregiver satisfaction significantly varied across families.
Copyright © 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Caregivers in Families With Schizophrenia
Nursing Research • July/August 2014 • Volume 63 • No. 4
265
TABLE 2. Descriptive Statistics of Variables for Individual Family Caregivers Individual caregivers Variable Family demands Sense of coherence Mutuality Caregiver burden Caregiver satisfaction
Families a
M
SD
Range
20.01 64.88 2.09 24.32 57.21
15.32 13.10 0.69 10.78 14.24
0–68 26–91 0–4 3–64 5–90
M
SD
Rangea
ICC
20.05 64.54 2.1 24.03 57.84
14.14 11.33 0.58 9.40 11.69
0–68 30–91 1–4 3–64 17–89
.25 .13
Note. Level 1 (caregivers): n = 243; Level 2 (families): n = 140. ICC = intraclass correlation. aRounded to nearest integer.
The absence of significant influence of caregiver burden on caregiver satisfaction (β = −.15; p = .09) indicated that caregiver burden and satisfaction might be different aspects of caregiving.
DISCUSSION The findings suggest that caring for people with schizophrenia not only is associated with the burden perceived by caregivers
but can also be a source of satisfaction in caregivers’ daily lives. Gender, kinship, and family demands were associated only with caregiver burden, whereas SOC and mutuality were related to both caregiver burden and satisfaction at individual levels. The influences of family demands and mutuality on caregiver burden as well as the impact of family demands, SOC, and mutuality on caregiver satisfaction significantly differ across
TABLE 3. Means, SD, and Corrected Item-Total Correlations for the Carers’ Assessment of Satisfaction Index Item Cared-For Person as Main Beneficiary 15. Knowing the person means I can give better care than anyone else 5. I am able to help the person I care for reach their potential 22. I am able to ensure that the person I care for has their needs tended to 4. It’s good to see small improvements in their condition 20. I am able to keep the person I care for out of an institution 13. It’s good to help the person I care for overcome difficulties and problems 24. Maintaining the dignity of the person I care for is important to me 14. It’s nice when something I do gives the person I care for pleasure 12. I get pleasure from seeing the person I care for happy 9. It is nice to see the person I care for clean, comfortable and well turned out Caregiver as Main Beneficiary 23. Caring has given me the chance to widen my interests and contacts 27. Caring has provide a purpose in my life that I did not have before 7. Caring provides a challenge 19. It helps to stop me from feeling guilty 8. Despite all their problems the person I care for does not grumble or moan 1. Caring has allowed me to develop new skills and abilities 16. Caring has helped me to grow and develop as a person 25. I am able to test myself and overcome difficulties 2. The person I care for is appreciative of what I do 17. It’s nice to feel appreciated by those family and friends I value 30. Caring makes me feel needed and wanted 10. Caring enables me to fulfill my sense of duty Interpersonal Dynamics as Main Beneficiary 26. Caring is one way of showing my faith 6. I am able to repay their past acts of kindness 3. Caring has brought me closer to the person I care for 21. If the situation were reversed, the person I care for would do the same for me 18. Caring has strengthened close family ties and relationships 11. I am the sort of person who enjoys helping people 28. At the end of the day, I know I will have done the best I could 29. Caring is one way of expressing my love for the person I care for
M
SD
ra
1.79 1.79 1.98 2.01 2.01 2.14 2.17 2.19 2.21 2.26
0.76 0.79 0.70 0.66 0.88 0.65 0.58 0.61 0.63 0.63
.61 .70 .67 .65 .61 .64 .50 .62 .62 .60
1.47 1.51 1.55 1.62 1.67 1.70 1.71 1.78 1.86 2.00 2.12 2.14
0.92 0.97 0.88 0.84 0.80 0.85 0.88 0.74 0.72 0.68 0.73 0.69
.62 .57 .57 .45 .60 .55 .51 .63 .57 .57 .63 .51
1.80 1.81 1.86 1.88 1.90 2.04 2.06 2.20
0.78 0.88 0.82 .81 .82 .65 .67 .72
.62 .49 .68 .57 .65 .57 .62 .63
Note. Within subscales, items are ordered by mean values. Items reprinted with permission of the instrument developer and copyright holder, Dr. Mike Nolan. aCorrected item-total correlation.
Copyright © 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
266
Caregivers in Families With Schizophrenia
www.nursingresearchonline.com
families. These results are in harmony with an application of the resiliency model (McCubbin et al., 1996); particularly, increased pileup of demands, negative appraisal, and inadequate resources resulted in maladaptation among family members within the family units. Given that caregiver burden and satisfaction have some different attributes and no significant association was noted between caregiver burden and satisfaction, the study lends support to the idea that caregiver burden and satisfaction are two variables reflecting different reactions in caregiving circumstances. The mean score on the caregiver burden scale for individual family caregivers in this study was similar to those found in a study of primary family caregivers of people with schizophrenia (Hou et al., 2008) and slightly higher than those in relatives of people with severe and mental illness (Hsiao & Van Riper, 2009). Overall, family caregivers experienced more moderate degree of burden. The primary source of caregiver satisfaction was the cared-for person, especially keeping the individual clean, comfortable, and well turned out as well as maintaining the dignity of the cared-for person. In interpersonal dynamics, family caregivers mostly rooted their source of satisfaction in their way of expressing love to the cared-for person. These results are similar to other studies of family caregivers of patients with older relatives (Kuuppelomäki et al., 2004). In this study, most of family caregivers reported being satisfied with fulfilling their sense of duty—a finding that was not consistent with those reported by Kuuppelomäki et al. (2004). In contrast to previous findings (Aggarwal et al., 2011), the absence of a significant relationship between burden and satisfaction in family caregivers suggested that caregiver burden may coexist with
caregiver satisfaction in the care of family members with schizophrenia. As Folkman and Moskowitz (2000) postulated, individuals may generate or reinforce positive values for their efforts during stressful periods of caregiving, which enables them to sustain being a caregiver. Thus, it is noteworthy to gain a better understanding of how family caregivers identify positive transformation as a sense of satisfaction in the face of adverse caregiving circumstances. Similar to findings of earlier research (López et al., 2005; Weimand et al., 2010), female caregivers appeared to report increased caregiver burden whereas older caregivers experienced greater satisfaction. With traditional Confucian values and social expectations, female caregivers tend to regard taking care of their affected offspring as their inborn responsibility; therefore, it places tremendous burden on them when caring for relatives with long-term schizophrenia (Huang et al., 2009). In addition, older caregivers may become experienced and resilient in response to demands of care over time, which, in turn, generates a greater sense of inner strength and experienced greater satisfaction. Comparing individualistic Western populations, the interpersonal harmony is particularly salient in Chinese societies that address mandatory responsibilities toward others. The perception of losing face associated with stigma attached to mental illness may lead families to rely on the immediate or significant relatives and friends (Chien et al., 2007). Family caregivers with social desires may express greater satisfaction with caregiving and even being reluctant to access healthcare services. Continued investigations into the potential attributes of burden and satisfaction in diverse ethnic families of persons with schizophrenia are recommended.
TABLE 4. Mixed Models for Caregiver Burden and Caregiver Satisfaction Caregiver burden Effects Fixed, individual
Fixed, family
Random
Predictors Age Gendera Relationshipb Close relative/friend Child Spouse Sibling Additional co-caregiversc 0 >1 Caregiving duration Family demands Sense of coherence Mutuality Family demands Sense of coherence Mutuality Level 1
Estimate
Caregiver satisfaction
SE
p
SE
p
0.00 2.75
0.07 0.94
ns .01
Estimate 0.22 2.16
0.10 1.34
.05 ns
−7.94 −2.02 −2.26 −4.09
2.47 2.71 2.12 1.9
.01 ns ns .05
−1.34 6.68 3.37 1.39
3.68 3.61 3.10 2.81
ns ns ns ns
−1.32 1.20 0.08 0.21 −0.14 −0.22 0.05 0.06 0.30 26.16
1.86 1.26 0.08 0.05 0.06 0.08 0.03 0.05 0.11 6.83
ns ns ns
