Psychometric properties of the self-care inventory for children with cerebral palsy in a resource-constrained urban South African context

Abstract

There is a lack of contextually appropriate and standardised assessment tools in South Africa. The Self-Care Inventory for Children with Cerebral Palsy (SCICP) was initially developed in Gauteng as a caregiver report questionnaire to assess the selfcare abilities of children with cerebral palsy (CP) living in poorly resourced and low socio-economic areas in KwaZulu- Natal, South Africa. The tool underwent initial field testing and content validity so as to determine administration and item appropriateness. Burg (2016) recommended that the tool undergo further validity and reliability testing so as to enhance its psychometric properties. Eventually, it is hoped that the tool may be used as a standardised assessment within the South African context. A cross-sectional, non-experimental study design was used that was quantitative in nature. Two hospitals in the Ekurhuleni district in Gauteng, South Africa were used as the sites of data collection. Fifty CP children and fifty typical children’s caregivers were asked to complete a demographic questionnaire as well as the study’s assessment tool, the SCICP. Data obtained from these two sources were converted into numerical data for analysis by descriptive statistics. This data is depicted in the form of tables and graphs. The objectives of this study were as follows: to determine the known-group validity of the SCICP for children between the ages of nought to six-years-old who are typically developing and compare it to children with CP between the ages of nought-to eight-years-old. And to determine the content validity (response process and relationship to other measures) of the SCICP for the two sample groups of children. Additional objectives were to determine the internal consistency, as well as the sensitivity, specificity and accuracy of the tool. The SCICP has been shown to differentiate between children who are typically developing and children with CP who present with developmental delay. The knowngroup validity between the two groups showed significant differences for the test score of typically developing children and the children with CP across all self-care domains for the majority of the different age groups. Further response process was confirmed by the strong correlation of the test scores of the children with CP and their GMFCS levels. Lower correlations were found with the children’s MACS levels. Age was not significantly correlated with the SCICP test scores for children with CP. A moderate correlation was found between the caregivers’ report scores and therapist’s scores after observing selected tasks on the SCICP. The SCICP showed acceptable internal consistency of over 0,70. Overall sensitivity of 87,5% and specificity of 74,4%. Accuracy scores were 80,9% overall and ranged from 69-100% for the different age groups in identifying children with developmental delay in selfcare. Multiple factors are responsible for children with CP’s ability to perform their self-care skills, most notably their gross motor skills and hand functioning. The SCICP shows promising results for use in the clinical setting from the psychometric properties that were tested.