Structural validation and dyadic child-parent measurement invariance of the celiac disease quality of life questionnaire.

Abstract

The celiac disease quality of life questionnaire (CDDUX) is used widely in numerous languages worldwide. However, it's structural and construct validity and child-parent invariance had not been thoroughly examined. The objective of this study was to examine the psychometric properties of the 12-item CDDUX and the extent to which it meets the acceptable requirements of reliability and structural and convergent validity, as well as its child-parent invariance. In this cross-sectional study, 126 dyads of children aged 8-18 years and their parents completed the Hebrew version self-report and parent-proxy report CDDUX. Recently developed methods to examine psychometric properties and to measure invariance of dyadic samples were used while properly accounting for nonindependence in measurement patterns. A three-factor structure, each with sufficient internal consistency, is confirmed for both children and parents. Removing a single indicator of the diet subscale resulted in full configural (χ2(181) = 202.277, P > 0.05, RMSEA = 0.026) and metric (χ2(189) = 209.543, P > 0.05, RMSEA = 0.043) invariance of the measure between children and parents. However, this occurred only in partial-scalar (χ2(198) = 229.813, P > 0.05, RMSEA = 0.031) and uniqueness invariance, which is nevertheless sufficient for meaningful comparison between the groups. Overall, with minor modifications, the Hebrew version of the CDDUX was found to be a valid measure of children's celiac-related quality of life when measured across children's self-reports and parent-proxy reports. The CDDUX provides meaningful measurement and allows child-parent comparison.