Cross-cultural adaptation, validity and reliability of the Japanese version of ABILHAND-kids for children with cerebral palsy using Rasch measurement model

Purpose: The aim of this study was to develop the Japanese version of the ABILHAND-Kids and to examine its psychometric properties for Japanese children with cerebral palsy (CP). Methods: The experimental version of 75 items was developed using forward-backward translation method. Parents of 137 children with CP answered it. Their responses were analyzed to successive items, and psychometric properties of the final version were investigated through the Rasch measurement model. Results: The Japanese version of the ABILHAND-Kids contained 22 items. It showed valid item-patient targeting, no significant floor and ceiling effects, and no differential item functioning for demographic and clinical subgroups. All items contributed to the definition of one-dimensional measure. For internal consistency, the person separation index was 0.94. For test-retest reliability, the intraclass correlation coefficients were 0.96 (95% CI: 0.92-0.98). The minimal detectable difference was calculated with a logit score of 0.79 and a total raw score of 4.50. The logit score showed a strong correlation with the Manual Ability Classification System level (rho= -0.70) and the Gross Motor Function Classification System level (rho= -0.62). Conclusions: The Japanese version of the ABILHAND-Kids was found to be valid and reliable. It appears to be a good tool for assessing manual abilities in daily activities in children with CP.