Research in Developmental Disabilities 35 (2014) 520–528
Contents lists available at ScienceDirect
Research in Developmental Disabilities
Determinants of quality of life in children with cerebral palsy: A comprehensive biopsychosocial approach Kuan-Lin Chen a,b, Mei-Hui Tseng c,d,*, Jeng-Yi Shieh d, Lu Lu d, Chien-Yu Huang c a
School of Occupational Therapy, College of Medicine, National Cheng Kung University, Tainan, Taiwan Department of Physical Medicine and Rehabilitation, National Cheng Kung University Hospital, Tainan, Taiwan c School of Occupational Therapy, College of Medicine, National Taiwan University, Taipei, Taiwan d Department of Physical Medicine and Rehabilitation, National Taiwan University Hospital, Taipei, Taiwan b
A R T I C L E I N F O
A B S T R A C T
Article history: Received 23 July 2013 Received in revised form 2 December 2013 Accepted 4 December 2013 Available online 27 December 2013
This study investigated the determinants of quality of life (QOL) of children with cerebral palsy (CP) considering possible variables comprehensively from a biopsychosocial perspective by adopting the International Classification of Functioning, Disability and Health (ICF) and using a CP-specific QOL questionnaire. A total of 167 children with CP (mean age 9.06 years, SD 2.61 years) and their caregivers participated in this study. Children’s QOL was measured by the Cerebral Palsy Quality of Life for Children (CP QOLChild) – primary caregiver proxy-report form. The potential determinants of QOL were collected based on all ICF dimensions. Results of seven multiple regression models showed that the determinants of QOL in children with CP were multidimensional and biopsychosocial in nature, i.e., encompassing the domains of health condition, body functions and structures, and contextual factors of the ICF. Children’s behavioral and emotional problems as well as caregiver’s psychological and family-related factors were important determinants of QOL in children with CP. Knowledge of the determinants of QOL could serve as a guide in a holistic approach to evaluation and intervention targeted at these determinants to improve the QOL of children with CP. ß 2013 Elsevier Ltd. All rights reserved.
Keywords: Quality of life Cerebral palsy Determinants ICF
1. Introduction Cerebral palsy (CP) is the most common cause of chronic disability in children (Rosenbaum et al., 2007; Surveillance of Cerebral Palsy in Europe, 2000) with a prevalence of 1.5–2.5 per 1000 live births (Oskoui, Coutinho, Dykeman, Jette, & Pringsheim, 2013). CP denotes a group of non-progressive disorders in development of movement and posture, which are ascribed to disturbances occurring in the developing fetal or infant brain (Rosenbaum et al., 2007). With its accompanying disabilities, CP not only impedes children’s participation in daily activities (Rosenbaum et al., 2007) but compromises their quality of life (QOL) (Maher, Olds, Williams, & Lane, 2008; Pirpiris et al., 2006; Vargus-Adams, 2005; Varni et al., 2005). QOL is defined as ‘‘an individual’s perception of their position in life in the context of the culture and value systems in which they live and in relation to their goals, expectations, standards and concerns (The WHOQOL Group, 1998).’’ In health care, health-related quality of life has been coined to assess how the individual’s well-being may be affected by a disease, disability, or disorder to represent specifically health aspects of well-being (Bullinger, 2003). Compared to typically developing children, children with CP have reduced QOL (Maher et al., 2008; Pirpiris et al., 2006; Vargus-Adams, 2005; Varni
* Corresponding author at: School of Occupational Therapy, College of Medicine, National Taiwan University, No. 17 Xu-Zhou Road, Taipei 10055, Taiwan. Tel.: +886 2 33668175; fax: +886 2 23958960. E-mail address: [email protected] (M.-H. Tseng). 0891-4222/$ – see front matter ß 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ridd.2013.12.002
K.-L. Chen et al. / Research in Developmental Disabilities 35 (2014) 520–528
521
et al., 2005). In 1991, the United Cerebral Palsy Association (1991) adopted the mission statement: ‘‘To affect positively the QOL of persons with cerebral palsy.’’ Improving QOL has then become a treatment goal for children with CP. Children’s functioning and contextual factors have been shown to be correlated with the QOL of children with CP (Arnaud et al., 2008; Beckung et al., 2008; Majnemer, Shevell, Rosenbaum, Law, & Poulin, 2007). The International Classification of Functioning, Disability and Health (ICF), published by the WHO in 2001, is a comprehensive classification system for how health-related conditions affect people’s life. The ICF, encompassing all biopsychosocial aspects of human health and healthrelevant components of QOL, not only offers a well-integrated and standardized approach to health care but also serves as a framework for building a comprehensive picture of possible factors that may influence the QOL of children (WHO, 2001). Although a multitude of studies investigated the determinants of QOL of children with CP, only two multivariate studies took into consideration all dimensions of the ICF (Arnaud et al., 2008; Majnemer et al., 2007). In these studies, determinants of QOL in each dimension of the ICF have been identified such as severity of gross and fine motor (health condition), cognitive functioning (body function and structures), communication (activity and participation), age (personal factor), and education placement (environmental factor). Yet, these studies had some limitations: (1) using generic rather than CP-specific QOL measures; (2) adopting QOL measures focusing on a child’s capability or limitations, i.e., physical functioning, instead of a child’s subjective perception of life as defined by the WHO. Thus, this study was to identify the determinants of QOL of children with CP, considering possible variables comprehensively from a biopsychosocial perspective by adopting the ICF framework and using a CP-specific QOL questionnaire following the WHO’s definition of QOL. Results of this study could inform assessment and guide clinical intervention for improving QOL of children with CP. 2. Method 2.1. Participants The participants of this study were part of a psychometric study of CP QOL-Child. The recruitment of subjects has been described in detail elsewhere (Chen et al., 2013). Briefly, 167 children with CP and their primary caregivers were recruited. These children were included if they were (1) diagnosed with CP; (2) aged between 4 and 12 years; and (3) their primary caregivers gave informed consent. Children were excluded if their primary caregivers were not able to communicate in Mandarin Chinese or Taiwanese. The protocol for this study received approval from the Institutional Review Board of the National Taiwan University Hospital. 2.2. Measure of QOL The QOL of children with CP was measured with the Chinese version of the Cerebral Palsy Quality of Life for Children (CP QOL-Child)-primary caregiver proxy-report form. The CP QOL-Child is the first condition-specific QOL questionnaire designed for children with CP (Waters et al., 2006). Two versions of the CP QOL-Child are available: a primary caregiverproxy report version for children aged 4–12 years, and a self-report version for children aged 9–12 years. The proxy version has seven domains: Social Well-being and Acceptance, Functioning, Participation and Physical Health, Emotional Wellbeing, Access to Services (proxy-report form only), Pain and Feeling about Disability, and Family Health (proxy-report form only). The proxy form contains 66 items. Almost all of the items have the following item stem: ‘How do you think your child feels about. . .’ and a 9-point rating scale, where 1 = very unhappy to 9 = very happy. For example, the domain ‘‘Functioning’’ measures caregiver’s perception of how the child feels about his functioning in the areas of the ability to play on his/her own, the way he/she communicates with people they know well, the way he/she uses his/her arms, and the way he/she uses their hands. Therefore, the domain ‘‘Functioning’’ measures a child’s subjective feelings about his or her functioning via his or her caregiver’s perception, instead of objective capability or limitations. Few items where this stem or rating scale is not appropriate, such as pain, have the following stem and rating scale: ‘How does your child feel about the amount of pain that they have,’ where 1 = not upset at all to 9 = very upset. The reliability and validity of the CP QOL-Child have been established (Waters et al., 2007). The CP QOL-Child was crossculturally translated into Mandarin Chinese, and the reliability and validity of the proxy-report form have been established (Chen et al., 2013; Wang et al., 2010). Internal consistency ranges from .78 to .91and 2-week test–retest reliability ranges from .86 to .97. The factor structure of the proxy-report form has been validated with principal components analysis (Wang et al., 2010) and confirmative factor analysis (Chen et al., 2013). 2.3. Measures of determinants of QOL Measures of the potential determinants of QOL in children with CP were described according to the ICF as follows. 2.3.1. Measures of variables related to health condition Variables related to children’s health condition (disease or disorder) were collected as follows: type of CP (including types of tonal abnormalities and anatomic distribution), severity of gross and fine motor impairment, and other diseases. Type of CP was based on a formal assessment by a pediatrician, pediatric neurologist, or physiatrist. Severity of gross motor
522
K.-L. Chen et al. / Research in Developmental Disabilities 35 (2014) 520–528
impairment was assessed by using the Gross Motor Function Classification System (GMFCS) (Palisano et al., 1997), a fivelevel classification system. Distinctions between the levels, ranging from level I (least limitations) to level V, are based on functional limitations and the need for mobility devices. Severity of fine motor impairment was measured by the Manual Ability Classification System (MACS) (Eliasson et al., 2006), which consists of five levels and its structure was purposely styled on the GMFCS. Information related to other diseases or disorders diagnosed by medical doctors was collected using a demographic questionnaire filled out by primary caregivers. These diseases or disorders included seizure, autism spectrum disorders, attention deficit hyperactivity disorder, premature birth, genetic disorders, heart disease, etc. 2.3.2. Measures of variables related to body functions and structures Variables related to children’s body functions and structures were collected through chart review and parents’ or teachers’ reports based on the child’s medical and school records. These variables included cognitive impairment, emotions and behavior, and other associated impairments. Among them, cognitive impairment was coded as a dichotomous variable and was defined as 0 if it was not impaired and as 1 if there was impairment according to formal cognitive assessments completed within two years, or parents’ or teachers’ reports based on the child’s medical and school records. Children’s behavioral and emotional problems were assessed by the parent-report form of the Chinese version of the Strengths and Difficulties Questionnaire (SDQ-C) (Goodman, 1997). The SDQ (Goodman, 1997) is a brief multidimensional measure of psychological adjustment of children aged 3–16 years. The SDQ contains 25 items divided into four subscales assessing difficulties (hyperactivity/inattention, conduct problems, emotional problems, peer problems) and one subscale assessing strength (prosocial behavior). The reliability and validity of the SDQ-C have been well established (Liu et al., 2007a, 2007b, 2007c; Yao et al., 2009). 2.3.3. Measures of variables related to activity and participation Self-care, mobility, and social function were assessed as factors related to children’s activity and participation using the Chinese version of the Pediatric Evaluation of Disability and Inventory (PEDI-C) (Chen, Hsieh, Sheu, Hu, & Tseng, 2009). That is, self-care, mobility, and social function were assessed with the Functional Skills Scale and the Caregiver Assistance Scale of the PEDI-C, respectively. The PEDI-C was designed for children aged between 6 months and 7.5 years old. It can also be used for older children whose functional performance is below that of 7.5-year-old normally developing children. The PEDI-C contains three scales to evaluate children’s daily function, i.e., a Functional Skills Scale, a Caregiver Assistance Scale, and a Modifications Scale. The three scales can be used together or separately. Each scale is divided into three domains: self-care, mobility, and social function. The PEDI-C used in children with CP showed excellent internal consistency, test–retest reliability, concurrent validity, and discriminative validity (Chen et al., 2009). 2.3.4. Measures of variables related to contextual factors Children’s personal factors, i.e., age, gender, birth order, and hand preference along with children’s environmental factors, i.e., the number of children in the family, educational placement, grade, medication, and past and current rehabilitation services were collected using a demographic questionnaire filled out by primary caregivers. Both standardized measures and a demographic information form were used to collect data on family-related factors, i.e., parents’ ages and socioeconomic status, mother’s nationality, marital status of parents, religion of the family, caregiver’s general mental health, parenting stress, family life impacts, family coping patterns, and domestic helper. The socioeconomic status of parents was computed by educational level and occupational category according to Hollingshead’s index of social position (Hollingshead, 1957). The mental health of the primary caregivers was assessed with the 30-item Chinese Health Questionnaire (CHQ) (Chong & Wilkinson, 1989). The psychometric properties of the CHQ have been established (Cheng, 1988; Cheng, Wu, Chong, & Williams, 1986; Chong & Wilkinson, 1989; Yang, Huang, & Wu, 2003). Parenting stress of caregivers was assessed with Parenting Stress Index-Short Form (PSI-SF), a 36-item scale measuring the level of parental stress with respect to the parent–child relationship and other situational interactions (Abidin, 1990, 1995). The PSI-SF is commonly used in the clinical settings and has been widely used in research. The reliability and validity of the PSI-SF has been established (Haskett, Ahern, Ward, & Allaire, 2006; Reitman, Currier, & Stickle, 2002). Family impacts and family coping patterns were assessed with the Family Life Impacts Questionnaire (FLIQ) and the Family Coping Patterns Questionnaire (FCPQ) (Tang et al., 2005) respectively. The FLIQ measures the impact of a child with special needs on a family. The FLIQ has 47 items divided into 4 subscales of daily routine, affection communication and conflict resolution, family interaction and social life as well as family growth and career development. The FCPQ, originally developed from McCubbin and Thompson’s questionnaire (McCubbin & Huang, 1989) on parents’ health, measures coping patterns used in a family having a child with special needs (Tang et al., 2005). The FCPQ has 34 items divided into three subscales: avoidance-oriented coping strategies, cognitive appraisal and coping strategies, and task-oriented coping strategies. The FLIQ and FCPQ have good reliability and validity (Tang et al., 2005). 2.4. Procedure The Department of Education of local cities or counties in Taiwan was requested first to issue an official document to elementary schools and kindergartens in each city or county for a list of children with CP. The purposes and procedures of the
K.-L. Chen et al. / Research in Developmental Disabilities 35 (2014) 520–528
523
research project were then explained to the directors of kindergartens and principals of elementary schools, including mainstream and special education schools in Taiwan, as well as clinicians at pediatric rehabilitation clinics at medical centers in the greater Taipei, Taichung, and Kaohsiung areas. Consent forms and cover letters explaining the purposes and procedures of the research project were distributed by school teachers and clinicians to the caregivers. Once the signed consent forms were returned, the following three steps were followed over a two-week period: (1) appointments were made with caregivers for an interview at school, at a clinic, or by telephone. Well-trained therapists interviewed primary caregivers using the PEDI-C, the GMFCS, and the MACS. (2) A package of questionnaires was sent to the caregivers. In the package were a demographic questionnaire, the CP QOL-Child, the SDQ-C, the CHQ, the PSI-SF, the FLIQ, and the FCPQ. (3) Two weeks later, guidance teachers or clinicians collected all the completed questionnaires and returned them along with the scores of intelligence and language tests from the child’s school records. If there were any data missing, unclear, or incomprehensible, the caregiver would be contacted for clarification. After all the aforementioned procedures were completed, reports of all assessments were sent to the caregivers and teachers within one month. 2.5. Statistical analysis Descriptive analysis was conducted to characterize the basic properties of the observed variables. In multivariate analysis, multiple linear regression models were constructed to identify the predictors of the CP QOL-Child of children with CP. Specifically, we calculated the corresponding factor scores based on the 7-factor model confirmed by the confirmatory factor analysis (CFA) of the CP QOL-Child conducted by Chen et al. (2013). In addition, to avoid multicollinearity, a 2-factor CFA for the Functional Skills Scale and the Caregiver Assistance Scale of the PEDI-C was conducted to identify the core of the constructs by linking highly correlated items together (Chen et al., 2009). Next, we fitted seven separate multiple linear regression models to identify the determinants of each domain of the CP QOL-Child. All variables listed in ‘‘Section 2.3’’ were examined as independent variables in the following analyses. The goal of regression analysis was to find one or a few parsimonious regression models that fitted the observed data well for outcome prediction or effect estimation. Therefore, three approaches were adopted to reduce the number of variables. First, univariate correlation analyses were conducted between the independent variables and dependent variables to identify univariate significant variables; second, a stepwise variable selection procedure was performed on the univariate significant variables to obtain a subset of variables as the candidate final multiple regression model. Third, with the aid of substantive knowledge and setting the significant level to 0.05, the best final regression model was identified by manually removing the non-significant variable one at a time. To ensure the quality of the analysis results, basic model-fitting techniques for (1) variable selection, (2) goodness-of-fit (GOF) assessment, and (3) regression diagnostics were used in the regression analyses. The significance levels for entry (SLE) and for stay (SLS) were set to 0.15 or larger. The GOF measure, coefficient of determination (R2), was used to assess the adequacy of the fitted final linear regression model. The statistical tools for regression diagnostics such as residual analysis, detection of influential cases, and variance inflation factor (VIF) were used to discover model or data problems. Statistical analyses were performed using SPSS 15.0 (SPSS Inc., Chicago, IL, USA), LISREL 8.54 (Scientific Software International, Inc., Lincolnwood, IL, USA), and SAS 9.1.3 (SAS Institute Inc., Cary, NC, USA). A two-sided p .05 was considered statistically significant. 3. Results 3.1. Descriptive information of children with cerebral palsy A total of 167 children with CP and their primary caregivers were recruited in our study. Their characteristics are shown in Table 1. The mean age of the children was 9.06 years (SD: 2.61 years), and boys (56.3%) outnumbered girls. Most children (69.5%) were ambulatory (GMFCS levels I–III) with or without hand-held mobility devices. Most caregivers were parents of children with CP (95.8%), with more females (86.2%) than males. The mean raw scores and mean factor scores of the proxy-report form of the CP QOL-Child in the total sample of children with CP and at each level of the GMFCS are shown in Table 2. No significant child gender differences were found in the factor scores of each domain of the CP QOL-Child. As for the relationship with gross motor impairment, no significant correlation was found between factor scores of the CP QOL-Child domains and GMFCS levels, except the factor score of the domain of Pain and Feeling about Disability and GMFCS levels (Spearman’s rho = .403, p < .01). 3.2. Determinants of QOL of children with CP The seven fitted multiple regression models identified predictors of children’s QOL (Table 3). In summary, the model of the Social Well-being and Acceptance domain revealed that the child’s status of prematurity, prosocial behavior, birth order, mother’s nationality, caregiver’s parenting stress, socioeconomic status, presence of domestic helper, and family coping patterns were the predictors and explained 46.41% of the variation. In the domain of Functioning, the predictors of child’s diplegia, genetic disorder, gross motor impairment, emotional problems, prosocial behavior, hand preference, educational
K.-L. Chen et al. / Research in Developmental Disabilities 35 (2014) 520–528
524
Table 1 Descriptive statistics of variables of health condition, body functions and structures, activity and participation, personal factors, and environmental factors from the International Classification of Functioning, Disability, and Health (ICF) in children with cerebral palsy (CP) (N = 167). Characteristics ICF-health condition Type of CP: n (%) Spastic quadriplegia/diplegia/hemiplegia Dyskinetic Hypotonic Mixed GMFCSa: n (%) I/II/III/IV/V MACSb: n (%) I/II/III/IV/V Other diseases: n (%) Seizure/premature birth/others ICF-body functions and structures SDQ-Cc: mean (SD) Hyperactivity/inattention Conduct problems Emotional problems Peer problems Conduct problems Other associated impairments: n (%) Cognitive/hearing/visual impairment ICF-personal factors Hand preference: n (%) Right/left/undeveloped ICF-environmental factors CHQd: mean (SD) PSI-SFe: mean (SD) Parental distress parent–child dysfunctional interaction Difficult Child Father age (years): mean (SD), range Mother age (years): mean (SD), range ICF-environmental factors Marital status of parents: n (%) Married/others Mother’s nationality: n (%) Taiwan/foreign-born Caregiver’s relationship with the child: n (%) Mother/father/others Domestic helper: n (%) No/Taiwanese/foreign domestic helper FLIQf: mean (SD) Daily routine Affection communication and conflict resolution Family interaction and social life Family growth and career development FCPQg: mean (SD) Usage of avoidance-oriented coping strategies Usage of cognitive appraisal coping strategies Usage of task-oriented coping strategies Help of avoidance-oriented coping strategies Help of cognitive appraisal coping strategies Help of task-oriented coping strategies Current rehabilitation services: n (%) Physical/occupational/speech/other therapy School setting: n (%) Not enrolled Regular kindergarten Special kindergarten Regular elementary school Regular class Regular class and resource class Special class Special school Others (e.g., homeschooling service) a b c d e f g
GMFCS: Gross Motor Function Classification System. MACS: Manual Ability Classification System. SDQ-C: Strengths and Difficulties Questionnaire. CHQ: Chinese Health Questionnaire. PSI-SF: Parenting Stress Index-Short Form. FLIQ: Family Life Impacts Questionnaire. FCPQ: Family Coping Patterns Questionnaire.
Statistics
56 (33.5)/43 (25.7)/32 (19.2) 16 (9.6) 15 (9.0) 5 (3.0) 29 (17.4)/46 (27.5)/41 (24.6)/25 (15.0)/26 (15.6) 40 (24.0)/51 (30.5)/31 (18.6)/28 (16.8)/17 (10.2) 50 (29.9)/93 (55.7)/24 (14.4)
4.97 1.78 3.44 3.84 1.78
(2.32) (1.49) (2.22) (2.11) (1.49)
75 (44.9)/8 (4.8)/8 (4.8)
92 (55.1)/49 (29.3)/26 (15.6) 7.23 (7.06) 33.96 27.81 30.21 42.84 39.84
(8.97) (6.85) (8.11) (5.86), 26–69 (5.12), 26–61
145 (86.8)/22 (13.2) 157 (94.0)/10 (6.0) 137 (82.0)/23 (13.8)/23 (13.8) 133 (79.6)/4 (2.4)/30 (18.0) 31.29 23.54 21.63 25.09
(11.33) (9.51) (8.10) (9.96)
40.66 39.37 24.20 41.84 24.60 24.53
(9.50) (6.69) (6.35) (11.09) (6.10) (6.31)
134 (80.0)/122 (73.1)/56 (33.5)/29 (17.4) 8 (4.8) 16 (9.6) 5 (3.0) 36 42 30 18 12
(21.6) (25.1) (18.0) (10.8) (8.3)
K.-L. Chen et al. / Research in Developmental Disabilities 35 (2014) 520–528
525
Table 2 Summary of the mean raw scores and mean factor scores of the Cerebral Palsy Quality of Life for Children (CP QOL-Child) – the proxy-report form in children with cerebral palsy by levels of the Gross Motor Function Classification System (GMFCS) (N = 167). Proxy-report form of the CP QOL-Child
Social well-being and acceptance Functioning Participation and Physical Health Emotional well-being Access to services Pain and Impact of Disability (PAIN) Family health
Level of GMFCS
Raw score Factor score Raw score Factor score Raw score Factor score Raw score Factor score Raw score Factor score Raw score Factor score Raw score Factor score
Total mean (SD)
I
II
III
IV
V
n = 29 mean (SD)
n = 46 mean (SD)
n = 41 mean (SD)
n = 25 mean (SD)
n = 26 mean (SD)
93.45 5.51 122.97 4.70 83.21 1.53 41.03 0.54 41.66 5.45 24.76 2.16 23.48 5.23
94.39 5.51 120.78 4.86 81.37 1.25 41.02 0.60 46.07 5.46 26.91 2.42 22.97 5.23
97.37 5.56 118.59 4.82 81.15 1.21 40.83 0.64 53.83 5.27 34.12 2.89 21.85 5.02
98.04 5.72 114.40 4.58 80.96 1.48 39.76 0.55 56.24 5.17 34.12 2.91 22.48 5.18
92.15 5.34 95.15 4.08 68.42 1.06 37.27 0.67 42.50 4.53 39.77 3.42 20.58 4.49
(16.34) (0.93) (17.27) (0.84) (13.80) (0.72) (8.08) (0.31) (14.14) (1.04) (11.86) (0.81) (5.30) (1.10)
(13.36) (0.86) (21.42) (0.85) (15.68) (0.87) (6.87) (0.29) (15.39) (1.04) (9.40) (0.78) (5.65) (0.99)
(10.50) (0.69) (17.59) (0.76) (13.59) (0.74) (6.87) (0.32) (12.39) (1.00) (11.48) (0.90) (5.48) (1.05)
(12.91) (0.78) (24.22) (0.91) (16.32) (0.72) (7.06) (0.21) (13.57) (0.92) (11.48) (1.08) (4.61) (0.96)
(13.15) (0.85) (19.45) (0.85) (15.40) (1.03) (7.06) (0.36) (13.23) (1.26) (13.07) (0.93) (6.96) (1.14)
95.16 5.53 115.68 4.66 79.56 1.29 40.20 0.60 48.17 5.22 31.39 2.72 22.34 5.09
(13.21) (0.82) (21.82) (0.87) (15.53) (0.83) (7.34) (0.30) (14.81) (1.08) (12.74) (0.97) (5.65) (1.07)
placement, mother’s nationality, socioeconomic status, and family coping patterns can explained 44.44% of the variation of this domain. With regards to the regression model for the domain of Participation and Physical Health, the predictors included child’s visual impairment, prosocial behavior, educational placement, caregiver’s parenting stress and marital status, and caregiver’s relation to the child and accounted for 40.33% of the variation. As for the domain of Emotional Well-being, the child’s fine motor impairment, conduct problems, and hyperactivity/ inattention, father’s age, caregiver’s parenting stress, and family life impacts were the predictors and explained 34.28% of the variance. Predictors of child’s spastic type, fine motor impairment, hearing impairment, emotional problems, mother’s nationality and age, caregiver’s general mental health, and family coping patterns and impact accounted for 47.84% of the variance of the domain of Access to Services. As regards the domain of Pain and Feeling about disability, child’s hemiplegia, gross motor impairment, emotional problems, and current speech therapy, as well as family coping patterns and impact were the predictors and explained 38.95% of the variance. For the model of Family Health, predictors of caregiver’s general mental health and parenting stress as well as family coping patterns and impact explained 48.95% of the variance. 4. Discussion This is the first study investigating the determinants of QOL in children with CP using a CP-specific QOL questionnaire which fits the WHO’s definition of QOL. Furthermore, this study took a holistic approach by using the ICF as a conceptual framework. Results indicated that determinants of QOL in children with CP were multidimensional and biopsychosocial in nature, i.e., encompassing the domains of health condition, body functions and structures, and contextual factors of the ICF. In order to compare our results to previous studies, we classified the CP QOL-Child domains of ‘‘Functioning’’ and ‘‘Participation and Physical Health’’ as physical-related domains and the domains of ‘‘Social Well-being and Acceptance’’ as well as ‘‘Emotional Well-being’’ as psychosocial-related domains. Most determinants of physical-related QOL identified in this study had not been identified in previous multivariate studies such as socioeconomic status and children’s prosocial behavior. Similarly, some determinants of psychosocial-related QOL identified in this study were not identified in previous studies such as birth order and family coping patterns. The discrepancy may be due to two reasons. First, this study took into account the factors from the entire scope of the ICF simultaneously whereas previous studies only considered the potential predictors from part of the dimensions of the ICF. Second, different QOL measures were used. That is, this study used the CP QOL-Child, a condition-specific measure of subjective perception of life following the WHO’s QOL definition, while generic QOL questionnaires focusing on evaluation of a child’s capability or limitations were used in previous studies. Although motor impairment was the main problem of children with CP, the study showed that most of the determinants (40–75%) across all seven domains of the CP QOL-Child were contextual factors such as family impacts, family coping patterns, and caregivers’ parenting stress, indicating an important role of contextual factors in child’s subjective feeling of life as rated by caregivers and lending further support to the assumption that contextual factors may impact a child’s health outcome (WHO, 2001). Children’s behavioral and emotional problems, measured by the SDQ-C, had a pervasive impact on QOL of children with CP for they were the determinants in four of seven domains of CP QOL-Child, including both physical and psychosocialrelated aspects of QOL. For example, emotional problems were the determinant of QOL in the domains of Functioning, Access to Service, and Pain and Impact of Disability, a finding lending support to Majnemer et al.’s (2007) study in school-aged children with CP. In addition, conduct problems were determinants of QOL in the domain of Emotional Well-being. Namely,
K.-L. Chen et al. / Research in Developmental Disabilities 35 (2014) 520–528
526
Table 3 Multiple regression models of the 7 domains of the Cerebral Palsy Quality of Life for Children (CP QOL-Child) in children with cerebral palsy (N = 167). Determinants
Estimate
Standard error
Standardized estimate
p value
Social well-being and acceptance: R2 = 0.4641h Intercept The status of being the second child Mother’s nationality: foreign-born Taiwanese domestic helper Premature birth Socioeconomic status SDQ-Cc: prosocial behavior PSI-SFe: parent–child dysfunctional interaction FCPQg: usage of avoidance-oriented coping strategies
2.209 0.235 0.502 0.872 0.171 0.023 0.097 0.037 0.015
0.408 0.100 0.203 0.321 0.098 0.007 0.019 0.008 0.005
0.000 0.142 0.146 0.163 0.104 0.216 0.332 0.311 0.175
Contents lists available at ScienceDirect
Research in Developmental Disabilities
Determinants of quality of life in children with cerebral palsy: A comprehensive biopsychosocial approach Kuan-Lin Chen a,b, Mei-Hui Tseng c,d,*, Jeng-Yi Shieh d, Lu Lu d, Chien-Yu Huang c a
School of Occupational Therapy, College of Medicine, National Cheng Kung University, Tainan, Taiwan Department of Physical Medicine and Rehabilitation, National Cheng Kung University Hospital, Tainan, Taiwan c School of Occupational Therapy, College of Medicine, National Taiwan University, Taipei, Taiwan d Department of Physical Medicine and Rehabilitation, National Taiwan University Hospital, Taipei, Taiwan b
A R T I C L E I N F O
A B S T R A C T
Article history: Received 23 July 2013 Received in revised form 2 December 2013 Accepted 4 December 2013 Available online 27 December 2013
This study investigated the determinants of quality of life (QOL) of children with cerebral palsy (CP) considering possible variables comprehensively from a biopsychosocial perspective by adopting the International Classification of Functioning, Disability and Health (ICF) and using a CP-specific QOL questionnaire. A total of 167 children with CP (mean age 9.06 years, SD 2.61 years) and their caregivers participated in this study. Children’s QOL was measured by the Cerebral Palsy Quality of Life for Children (CP QOLChild) – primary caregiver proxy-report form. The potential determinants of QOL were collected based on all ICF dimensions. Results of seven multiple regression models showed that the determinants of QOL in children with CP were multidimensional and biopsychosocial in nature, i.e., encompassing the domains of health condition, body functions and structures, and contextual factors of the ICF. Children’s behavioral and emotional problems as well as caregiver’s psychological and family-related factors were important determinants of QOL in children with CP. Knowledge of the determinants of QOL could serve as a guide in a holistic approach to evaluation and intervention targeted at these determinants to improve the QOL of children with CP. ß 2013 Elsevier Ltd. All rights reserved.
Keywords: Quality of life Cerebral palsy Determinants ICF
1. Introduction Cerebral palsy (CP) is the most common cause of chronic disability in children (Rosenbaum et al., 2007; Surveillance of Cerebral Palsy in Europe, 2000) with a prevalence of 1.5–2.5 per 1000 live births (Oskoui, Coutinho, Dykeman, Jette, & Pringsheim, 2013). CP denotes a group of non-progressive disorders in development of movement and posture, which are ascribed to disturbances occurring in the developing fetal or infant brain (Rosenbaum et al., 2007). With its accompanying disabilities, CP not only impedes children’s participation in daily activities (Rosenbaum et al., 2007) but compromises their quality of life (QOL) (Maher, Olds, Williams, & Lane, 2008; Pirpiris et al., 2006; Vargus-Adams, 2005; Varni et al., 2005). QOL is defined as ‘‘an individual’s perception of their position in life in the context of the culture and value systems in which they live and in relation to their goals, expectations, standards and concerns (The WHOQOL Group, 1998).’’ In health care, health-related quality of life has been coined to assess how the individual’s well-being may be affected by a disease, disability, or disorder to represent specifically health aspects of well-being (Bullinger, 2003). Compared to typically developing children, children with CP have reduced QOL (Maher et al., 2008; Pirpiris et al., 2006; Vargus-Adams, 2005; Varni
* Corresponding author at: School of Occupational Therapy, College of Medicine, National Taiwan University, No. 17 Xu-Zhou Road, Taipei 10055, Taiwan. Tel.: +886 2 33668175; fax: +886 2 23958960. E-mail address: [email protected] (M.-H. Tseng). 0891-4222/$ – see front matter ß 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ridd.2013.12.002
K.-L. Chen et al. / Research in Developmental Disabilities 35 (2014) 520–528
521
et al., 2005). In 1991, the United Cerebral Palsy Association (1991) adopted the mission statement: ‘‘To affect positively the QOL of persons with cerebral palsy.’’ Improving QOL has then become a treatment goal for children with CP. Children’s functioning and contextual factors have been shown to be correlated with the QOL of children with CP (Arnaud et al., 2008; Beckung et al., 2008; Majnemer, Shevell, Rosenbaum, Law, & Poulin, 2007). The International Classification of Functioning, Disability and Health (ICF), published by the WHO in 2001, is a comprehensive classification system for how health-related conditions affect people’s life. The ICF, encompassing all biopsychosocial aspects of human health and healthrelevant components of QOL, not only offers a well-integrated and standardized approach to health care but also serves as a framework for building a comprehensive picture of possible factors that may influence the QOL of children (WHO, 2001). Although a multitude of studies investigated the determinants of QOL of children with CP, only two multivariate studies took into consideration all dimensions of the ICF (Arnaud et al., 2008; Majnemer et al., 2007). In these studies, determinants of QOL in each dimension of the ICF have been identified such as severity of gross and fine motor (health condition), cognitive functioning (body function and structures), communication (activity and participation), age (personal factor), and education placement (environmental factor). Yet, these studies had some limitations: (1) using generic rather than CP-specific QOL measures; (2) adopting QOL measures focusing on a child’s capability or limitations, i.e., physical functioning, instead of a child’s subjective perception of life as defined by the WHO. Thus, this study was to identify the determinants of QOL of children with CP, considering possible variables comprehensively from a biopsychosocial perspective by adopting the ICF framework and using a CP-specific QOL questionnaire following the WHO’s definition of QOL. Results of this study could inform assessment and guide clinical intervention for improving QOL of children with CP. 2. Method 2.1. Participants The participants of this study were part of a psychometric study of CP QOL-Child. The recruitment of subjects has been described in detail elsewhere (Chen et al., 2013). Briefly, 167 children with CP and their primary caregivers were recruited. These children were included if they were (1) diagnosed with CP; (2) aged between 4 and 12 years; and (3) their primary caregivers gave informed consent. Children were excluded if their primary caregivers were not able to communicate in Mandarin Chinese or Taiwanese. The protocol for this study received approval from the Institutional Review Board of the National Taiwan University Hospital. 2.2. Measure of QOL The QOL of children with CP was measured with the Chinese version of the Cerebral Palsy Quality of Life for Children (CP QOL-Child)-primary caregiver proxy-report form. The CP QOL-Child is the first condition-specific QOL questionnaire designed for children with CP (Waters et al., 2006). Two versions of the CP QOL-Child are available: a primary caregiverproxy report version for children aged 4–12 years, and a self-report version for children aged 9–12 years. The proxy version has seven domains: Social Well-being and Acceptance, Functioning, Participation and Physical Health, Emotional Wellbeing, Access to Services (proxy-report form only), Pain and Feeling about Disability, and Family Health (proxy-report form only). The proxy form contains 66 items. Almost all of the items have the following item stem: ‘How do you think your child feels about. . .’ and a 9-point rating scale, where 1 = very unhappy to 9 = very happy. For example, the domain ‘‘Functioning’’ measures caregiver’s perception of how the child feels about his functioning in the areas of the ability to play on his/her own, the way he/she communicates with people they know well, the way he/she uses his/her arms, and the way he/she uses their hands. Therefore, the domain ‘‘Functioning’’ measures a child’s subjective feelings about his or her functioning via his or her caregiver’s perception, instead of objective capability or limitations. Few items where this stem or rating scale is not appropriate, such as pain, have the following stem and rating scale: ‘How does your child feel about the amount of pain that they have,’ where 1 = not upset at all to 9 = very upset. The reliability and validity of the CP QOL-Child have been established (Waters et al., 2007). The CP QOL-Child was crossculturally translated into Mandarin Chinese, and the reliability and validity of the proxy-report form have been established (Chen et al., 2013; Wang et al., 2010). Internal consistency ranges from .78 to .91and 2-week test–retest reliability ranges from .86 to .97. The factor structure of the proxy-report form has been validated with principal components analysis (Wang et al., 2010) and confirmative factor analysis (Chen et al., 2013). 2.3. Measures of determinants of QOL Measures of the potential determinants of QOL in children with CP were described according to the ICF as follows. 2.3.1. Measures of variables related to health condition Variables related to children’s health condition (disease or disorder) were collected as follows: type of CP (including types of tonal abnormalities and anatomic distribution), severity of gross and fine motor impairment, and other diseases. Type of CP was based on a formal assessment by a pediatrician, pediatric neurologist, or physiatrist. Severity of gross motor
522
K.-L. Chen et al. / Research in Developmental Disabilities 35 (2014) 520–528
impairment was assessed by using the Gross Motor Function Classification System (GMFCS) (Palisano et al., 1997), a fivelevel classification system. Distinctions between the levels, ranging from level I (least limitations) to level V, are based on functional limitations and the need for mobility devices. Severity of fine motor impairment was measured by the Manual Ability Classification System (MACS) (Eliasson et al., 2006), which consists of five levels and its structure was purposely styled on the GMFCS. Information related to other diseases or disorders diagnosed by medical doctors was collected using a demographic questionnaire filled out by primary caregivers. These diseases or disorders included seizure, autism spectrum disorders, attention deficit hyperactivity disorder, premature birth, genetic disorders, heart disease, etc. 2.3.2. Measures of variables related to body functions and structures Variables related to children’s body functions and structures were collected through chart review and parents’ or teachers’ reports based on the child’s medical and school records. These variables included cognitive impairment, emotions and behavior, and other associated impairments. Among them, cognitive impairment was coded as a dichotomous variable and was defined as 0 if it was not impaired and as 1 if there was impairment according to formal cognitive assessments completed within two years, or parents’ or teachers’ reports based on the child’s medical and school records. Children’s behavioral and emotional problems were assessed by the parent-report form of the Chinese version of the Strengths and Difficulties Questionnaire (SDQ-C) (Goodman, 1997). The SDQ (Goodman, 1997) is a brief multidimensional measure of psychological adjustment of children aged 3–16 years. The SDQ contains 25 items divided into four subscales assessing difficulties (hyperactivity/inattention, conduct problems, emotional problems, peer problems) and one subscale assessing strength (prosocial behavior). The reliability and validity of the SDQ-C have been well established (Liu et al., 2007a, 2007b, 2007c; Yao et al., 2009). 2.3.3. Measures of variables related to activity and participation Self-care, mobility, and social function were assessed as factors related to children’s activity and participation using the Chinese version of the Pediatric Evaluation of Disability and Inventory (PEDI-C) (Chen, Hsieh, Sheu, Hu, & Tseng, 2009). That is, self-care, mobility, and social function were assessed with the Functional Skills Scale and the Caregiver Assistance Scale of the PEDI-C, respectively. The PEDI-C was designed for children aged between 6 months and 7.5 years old. It can also be used for older children whose functional performance is below that of 7.5-year-old normally developing children. The PEDI-C contains three scales to evaluate children’s daily function, i.e., a Functional Skills Scale, a Caregiver Assistance Scale, and a Modifications Scale. The three scales can be used together or separately. Each scale is divided into three domains: self-care, mobility, and social function. The PEDI-C used in children with CP showed excellent internal consistency, test–retest reliability, concurrent validity, and discriminative validity (Chen et al., 2009). 2.3.4. Measures of variables related to contextual factors Children’s personal factors, i.e., age, gender, birth order, and hand preference along with children’s environmental factors, i.e., the number of children in the family, educational placement, grade, medication, and past and current rehabilitation services were collected using a demographic questionnaire filled out by primary caregivers. Both standardized measures and a demographic information form were used to collect data on family-related factors, i.e., parents’ ages and socioeconomic status, mother’s nationality, marital status of parents, religion of the family, caregiver’s general mental health, parenting stress, family life impacts, family coping patterns, and domestic helper. The socioeconomic status of parents was computed by educational level and occupational category according to Hollingshead’s index of social position (Hollingshead, 1957). The mental health of the primary caregivers was assessed with the 30-item Chinese Health Questionnaire (CHQ) (Chong & Wilkinson, 1989). The psychometric properties of the CHQ have been established (Cheng, 1988; Cheng, Wu, Chong, & Williams, 1986; Chong & Wilkinson, 1989; Yang, Huang, & Wu, 2003). Parenting stress of caregivers was assessed with Parenting Stress Index-Short Form (PSI-SF), a 36-item scale measuring the level of parental stress with respect to the parent–child relationship and other situational interactions (Abidin, 1990, 1995). The PSI-SF is commonly used in the clinical settings and has been widely used in research. The reliability and validity of the PSI-SF has been established (Haskett, Ahern, Ward, & Allaire, 2006; Reitman, Currier, & Stickle, 2002). Family impacts and family coping patterns were assessed with the Family Life Impacts Questionnaire (FLIQ) and the Family Coping Patterns Questionnaire (FCPQ) (Tang et al., 2005) respectively. The FLIQ measures the impact of a child with special needs on a family. The FLIQ has 47 items divided into 4 subscales of daily routine, affection communication and conflict resolution, family interaction and social life as well as family growth and career development. The FCPQ, originally developed from McCubbin and Thompson’s questionnaire (McCubbin & Huang, 1989) on parents’ health, measures coping patterns used in a family having a child with special needs (Tang et al., 2005). The FCPQ has 34 items divided into three subscales: avoidance-oriented coping strategies, cognitive appraisal and coping strategies, and task-oriented coping strategies. The FLIQ and FCPQ have good reliability and validity (Tang et al., 2005). 2.4. Procedure The Department of Education of local cities or counties in Taiwan was requested first to issue an official document to elementary schools and kindergartens in each city or county for a list of children with CP. The purposes and procedures of the
K.-L. Chen et al. / Research in Developmental Disabilities 35 (2014) 520–528
523
research project were then explained to the directors of kindergartens and principals of elementary schools, including mainstream and special education schools in Taiwan, as well as clinicians at pediatric rehabilitation clinics at medical centers in the greater Taipei, Taichung, and Kaohsiung areas. Consent forms and cover letters explaining the purposes and procedures of the research project were distributed by school teachers and clinicians to the caregivers. Once the signed consent forms were returned, the following three steps were followed over a two-week period: (1) appointments were made with caregivers for an interview at school, at a clinic, or by telephone. Well-trained therapists interviewed primary caregivers using the PEDI-C, the GMFCS, and the MACS. (2) A package of questionnaires was sent to the caregivers. In the package were a demographic questionnaire, the CP QOL-Child, the SDQ-C, the CHQ, the PSI-SF, the FLIQ, and the FCPQ. (3) Two weeks later, guidance teachers or clinicians collected all the completed questionnaires and returned them along with the scores of intelligence and language tests from the child’s school records. If there were any data missing, unclear, or incomprehensible, the caregiver would be contacted for clarification. After all the aforementioned procedures were completed, reports of all assessments were sent to the caregivers and teachers within one month. 2.5. Statistical analysis Descriptive analysis was conducted to characterize the basic properties of the observed variables. In multivariate analysis, multiple linear regression models were constructed to identify the predictors of the CP QOL-Child of children with CP. Specifically, we calculated the corresponding factor scores based on the 7-factor model confirmed by the confirmatory factor analysis (CFA) of the CP QOL-Child conducted by Chen et al. (2013). In addition, to avoid multicollinearity, a 2-factor CFA for the Functional Skills Scale and the Caregiver Assistance Scale of the PEDI-C was conducted to identify the core of the constructs by linking highly correlated items together (Chen et al., 2009). Next, we fitted seven separate multiple linear regression models to identify the determinants of each domain of the CP QOL-Child. All variables listed in ‘‘Section 2.3’’ were examined as independent variables in the following analyses. The goal of regression analysis was to find one or a few parsimonious regression models that fitted the observed data well for outcome prediction or effect estimation. Therefore, three approaches were adopted to reduce the number of variables. First, univariate correlation analyses were conducted between the independent variables and dependent variables to identify univariate significant variables; second, a stepwise variable selection procedure was performed on the univariate significant variables to obtain a subset of variables as the candidate final multiple regression model. Third, with the aid of substantive knowledge and setting the significant level to 0.05, the best final regression model was identified by manually removing the non-significant variable one at a time. To ensure the quality of the analysis results, basic model-fitting techniques for (1) variable selection, (2) goodness-of-fit (GOF) assessment, and (3) regression diagnostics were used in the regression analyses. The significance levels for entry (SLE) and for stay (SLS) were set to 0.15 or larger. The GOF measure, coefficient of determination (R2), was used to assess the adequacy of the fitted final linear regression model. The statistical tools for regression diagnostics such as residual analysis, detection of influential cases, and variance inflation factor (VIF) were used to discover model or data problems. Statistical analyses were performed using SPSS 15.0 (SPSS Inc., Chicago, IL, USA), LISREL 8.54 (Scientific Software International, Inc., Lincolnwood, IL, USA), and SAS 9.1.3 (SAS Institute Inc., Cary, NC, USA). A two-sided p .05 was considered statistically significant. 3. Results 3.1. Descriptive information of children with cerebral palsy A total of 167 children with CP and their primary caregivers were recruited in our study. Their characteristics are shown in Table 1. The mean age of the children was 9.06 years (SD: 2.61 years), and boys (56.3%) outnumbered girls. Most children (69.5%) were ambulatory (GMFCS levels I–III) with or without hand-held mobility devices. Most caregivers were parents of children with CP (95.8%), with more females (86.2%) than males. The mean raw scores and mean factor scores of the proxy-report form of the CP QOL-Child in the total sample of children with CP and at each level of the GMFCS are shown in Table 2. No significant child gender differences were found in the factor scores of each domain of the CP QOL-Child. As for the relationship with gross motor impairment, no significant correlation was found between factor scores of the CP QOL-Child domains and GMFCS levels, except the factor score of the domain of Pain and Feeling about Disability and GMFCS levels (Spearman’s rho = .403, p < .01). 3.2. Determinants of QOL of children with CP The seven fitted multiple regression models identified predictors of children’s QOL (Table 3). In summary, the model of the Social Well-being and Acceptance domain revealed that the child’s status of prematurity, prosocial behavior, birth order, mother’s nationality, caregiver’s parenting stress, socioeconomic status, presence of domestic helper, and family coping patterns were the predictors and explained 46.41% of the variation. In the domain of Functioning, the predictors of child’s diplegia, genetic disorder, gross motor impairment, emotional problems, prosocial behavior, hand preference, educational
K.-L. Chen et al. / Research in Developmental Disabilities 35 (2014) 520–528
524
Table 1 Descriptive statistics of variables of health condition, body functions and structures, activity and participation, personal factors, and environmental factors from the International Classification of Functioning, Disability, and Health (ICF) in children with cerebral palsy (CP) (N = 167). Characteristics ICF-health condition Type of CP: n (%) Spastic quadriplegia/diplegia/hemiplegia Dyskinetic Hypotonic Mixed GMFCSa: n (%) I/II/III/IV/V MACSb: n (%) I/II/III/IV/V Other diseases: n (%) Seizure/premature birth/others ICF-body functions and structures SDQ-Cc: mean (SD) Hyperactivity/inattention Conduct problems Emotional problems Peer problems Conduct problems Other associated impairments: n (%) Cognitive/hearing/visual impairment ICF-personal factors Hand preference: n (%) Right/left/undeveloped ICF-environmental factors CHQd: mean (SD) PSI-SFe: mean (SD) Parental distress parent–child dysfunctional interaction Difficult Child Father age (years): mean (SD), range Mother age (years): mean (SD), range ICF-environmental factors Marital status of parents: n (%) Married/others Mother’s nationality: n (%) Taiwan/foreign-born Caregiver’s relationship with the child: n (%) Mother/father/others Domestic helper: n (%) No/Taiwanese/foreign domestic helper FLIQf: mean (SD) Daily routine Affection communication and conflict resolution Family interaction and social life Family growth and career development FCPQg: mean (SD) Usage of avoidance-oriented coping strategies Usage of cognitive appraisal coping strategies Usage of task-oriented coping strategies Help of avoidance-oriented coping strategies Help of cognitive appraisal coping strategies Help of task-oriented coping strategies Current rehabilitation services: n (%) Physical/occupational/speech/other therapy School setting: n (%) Not enrolled Regular kindergarten Special kindergarten Regular elementary school Regular class Regular class and resource class Special class Special school Others (e.g., homeschooling service) a b c d e f g
GMFCS: Gross Motor Function Classification System. MACS: Manual Ability Classification System. SDQ-C: Strengths and Difficulties Questionnaire. CHQ: Chinese Health Questionnaire. PSI-SF: Parenting Stress Index-Short Form. FLIQ: Family Life Impacts Questionnaire. FCPQ: Family Coping Patterns Questionnaire.
Statistics
56 (33.5)/43 (25.7)/32 (19.2) 16 (9.6) 15 (9.0) 5 (3.0) 29 (17.4)/46 (27.5)/41 (24.6)/25 (15.0)/26 (15.6) 40 (24.0)/51 (30.5)/31 (18.6)/28 (16.8)/17 (10.2) 50 (29.9)/93 (55.7)/24 (14.4)
4.97 1.78 3.44 3.84 1.78
(2.32) (1.49) (2.22) (2.11) (1.49)
75 (44.9)/8 (4.8)/8 (4.8)
92 (55.1)/49 (29.3)/26 (15.6) 7.23 (7.06) 33.96 27.81 30.21 42.84 39.84
(8.97) (6.85) (8.11) (5.86), 26–69 (5.12), 26–61
145 (86.8)/22 (13.2) 157 (94.0)/10 (6.0) 137 (82.0)/23 (13.8)/23 (13.8) 133 (79.6)/4 (2.4)/30 (18.0) 31.29 23.54 21.63 25.09
(11.33) (9.51) (8.10) (9.96)
40.66 39.37 24.20 41.84 24.60 24.53
(9.50) (6.69) (6.35) (11.09) (6.10) (6.31)
134 (80.0)/122 (73.1)/56 (33.5)/29 (17.4) 8 (4.8) 16 (9.6) 5 (3.0) 36 42 30 18 12
(21.6) (25.1) (18.0) (10.8) (8.3)
K.-L. Chen et al. / Research in Developmental Disabilities 35 (2014) 520–528
525
Table 2 Summary of the mean raw scores and mean factor scores of the Cerebral Palsy Quality of Life for Children (CP QOL-Child) – the proxy-report form in children with cerebral palsy by levels of the Gross Motor Function Classification System (GMFCS) (N = 167). Proxy-report form of the CP QOL-Child
Social well-being and acceptance Functioning Participation and Physical Health Emotional well-being Access to services Pain and Impact of Disability (PAIN) Family health
Level of GMFCS
Raw score Factor score Raw score Factor score Raw score Factor score Raw score Factor score Raw score Factor score Raw score Factor score Raw score Factor score
Total mean (SD)
I
II
III
IV
V
n = 29 mean (SD)
n = 46 mean (SD)
n = 41 mean (SD)
n = 25 mean (SD)
n = 26 mean (SD)
93.45 5.51 122.97 4.70 83.21 1.53 41.03 0.54 41.66 5.45 24.76 2.16 23.48 5.23
94.39 5.51 120.78 4.86 81.37 1.25 41.02 0.60 46.07 5.46 26.91 2.42 22.97 5.23
97.37 5.56 118.59 4.82 81.15 1.21 40.83 0.64 53.83 5.27 34.12 2.89 21.85 5.02
98.04 5.72 114.40 4.58 80.96 1.48 39.76 0.55 56.24 5.17 34.12 2.91 22.48 5.18
92.15 5.34 95.15 4.08 68.42 1.06 37.27 0.67 42.50 4.53 39.77 3.42 20.58 4.49
(16.34) (0.93) (17.27) (0.84) (13.80) (0.72) (8.08) (0.31) (14.14) (1.04) (11.86) (0.81) (5.30) (1.10)
(13.36) (0.86) (21.42) (0.85) (15.68) (0.87) (6.87) (0.29) (15.39) (1.04) (9.40) (0.78) (5.65) (0.99)
(10.50) (0.69) (17.59) (0.76) (13.59) (0.74) (6.87) (0.32) (12.39) (1.00) (11.48) (0.90) (5.48) (1.05)
(12.91) (0.78) (24.22) (0.91) (16.32) (0.72) (7.06) (0.21) (13.57) (0.92) (11.48) (1.08) (4.61) (0.96)
(13.15) (0.85) (19.45) (0.85) (15.40) (1.03) (7.06) (0.36) (13.23) (1.26) (13.07) (0.93) (6.96) (1.14)
95.16 5.53 115.68 4.66 79.56 1.29 40.20 0.60 48.17 5.22 31.39 2.72 22.34 5.09
(13.21) (0.82) (21.82) (0.87) (15.53) (0.83) (7.34) (0.30) (14.81) (1.08) (12.74) (0.97) (5.65) (1.07)
placement, mother’s nationality, socioeconomic status, and family coping patterns can explained 44.44% of the variation of this domain. With regards to the regression model for the domain of Participation and Physical Health, the predictors included child’s visual impairment, prosocial behavior, educational placement, caregiver’s parenting stress and marital status, and caregiver’s relation to the child and accounted for 40.33% of the variation. As for the domain of Emotional Well-being, the child’s fine motor impairment, conduct problems, and hyperactivity/ inattention, father’s age, caregiver’s parenting stress, and family life impacts were the predictors and explained 34.28% of the variance. Predictors of child’s spastic type, fine motor impairment, hearing impairment, emotional problems, mother’s nationality and age, caregiver’s general mental health, and family coping patterns and impact accounted for 47.84% of the variance of the domain of Access to Services. As regards the domain of Pain and Feeling about disability, child’s hemiplegia, gross motor impairment, emotional problems, and current speech therapy, as well as family coping patterns and impact were the predictors and explained 38.95% of the variance. For the model of Family Health, predictors of caregiver’s general mental health and parenting stress as well as family coping patterns and impact explained 48.95% of the variance. 4. Discussion This is the first study investigating the determinants of QOL in children with CP using a CP-specific QOL questionnaire which fits the WHO’s definition of QOL. Furthermore, this study took a holistic approach by using the ICF as a conceptual framework. Results indicated that determinants of QOL in children with CP were multidimensional and biopsychosocial in nature, i.e., encompassing the domains of health condition, body functions and structures, and contextual factors of the ICF. In order to compare our results to previous studies, we classified the CP QOL-Child domains of ‘‘Functioning’’ and ‘‘Participation and Physical Health’’ as physical-related domains and the domains of ‘‘Social Well-being and Acceptance’’ as well as ‘‘Emotional Well-being’’ as psychosocial-related domains. Most determinants of physical-related QOL identified in this study had not been identified in previous multivariate studies such as socioeconomic status and children’s prosocial behavior. Similarly, some determinants of psychosocial-related QOL identified in this study were not identified in previous studies such as birth order and family coping patterns. The discrepancy may be due to two reasons. First, this study took into account the factors from the entire scope of the ICF simultaneously whereas previous studies only considered the potential predictors from part of the dimensions of the ICF. Second, different QOL measures were used. That is, this study used the CP QOL-Child, a condition-specific measure of subjective perception of life following the WHO’s QOL definition, while generic QOL questionnaires focusing on evaluation of a child’s capability or limitations were used in previous studies. Although motor impairment was the main problem of children with CP, the study showed that most of the determinants (40–75%) across all seven domains of the CP QOL-Child were contextual factors such as family impacts, family coping patterns, and caregivers’ parenting stress, indicating an important role of contextual factors in child’s subjective feeling of life as rated by caregivers and lending further support to the assumption that contextual factors may impact a child’s health outcome (WHO, 2001). Children’s behavioral and emotional problems, measured by the SDQ-C, had a pervasive impact on QOL of children with CP for they were the determinants in four of seven domains of CP QOL-Child, including both physical and psychosocialrelated aspects of QOL. For example, emotional problems were the determinant of QOL in the domains of Functioning, Access to Service, and Pain and Impact of Disability, a finding lending support to Majnemer et al.’s (2007) study in school-aged children with CP. In addition, conduct problems were determinants of QOL in the domain of Emotional Well-being. Namely,
K.-L. Chen et al. / Research in Developmental Disabilities 35 (2014) 520–528
526
Table 3 Multiple regression models of the 7 domains of the Cerebral Palsy Quality of Life for Children (CP QOL-Child) in children with cerebral palsy (N = 167). Determinants
Estimate
Standard error
Standardized estimate
p value
Social well-being and acceptance: R2 = 0.4641h Intercept The status of being the second child Mother’s nationality: foreign-born Taiwanese domestic helper Premature birth Socioeconomic status SDQ-Cc: prosocial behavior PSI-SFe: parent–child dysfunctional interaction FCPQg: usage of avoidance-oriented coping strategies
2.209 0.235 0.502 0.872 0.171 0.023 0.097 0.037 0.015
0.408 0.100 0.203 0.321 0.098 0.007 0.019 0.008 0.005
0.000 0.142 0.146 0.163 0.104 0.216 0.332 0.311 0.175
