Comparison of the self and the proxy versions of the KIDSCREEN quality of life scale: To what extend can parents predict quality of life of their children?

No Thumbnail Available

Date

Journal Title

Journal ISSN

Volume Title

Abstract

Objective: The purpose of this study is to compare the agreement between self and proxy responses of the KIDSCREEN health-related quality of life (HRQOL). Methods: This cross sectional study was carried out on 662 children/adolescents aged between 8 to 18 years and 543 parents in 6 schools of Manisa city center, Turkey. In this study sociodemographic characteristics of the participants were determined, and KIDSCREEN quality of life questionnair forms were applied. Following descriptive findings, Pearson correlation analyses, Intraclass Correlation Coefficients and Bland-Altman graphs were performed in the statistical analyses so as to determine the association between basic descriptive findings, self-, and proxy responses. In addition Linear Regression analysis was performed in order to show the extend of proxy/parents' evaluations in the explanation of children's, and adolescents! HRQOL. Results: 49.4 % of the children were male; and mean age was 13.1 +/- 2.4 years. Correlation coefficients and ICCs of self and proxy HRQOL evaluations were between 0.27 to 0.52 ranging between 0.43 to 0.68 respectively for the various dimensions of the KIDSCREEN. The range of Determination coefficients (R2) for the prediction of self-ratings by proxy ratings were between 0.10 to 0.41 for different dimensions. Bland-Altman graph showed a sound distribution (majority of the values are between +/- 2 standard deviation limits) around zero line of the graph devoid of systematic error for both KIDSCREEN-52 physical well-being scale and KIDSCREEN-10. Conclusion: KIDSCREEN self-and proxy ratings showed a moderate level of correlation, and agreement. A significant disagreement was detected for bullying dimension between self-and proxy ratings. Parents overestimated HRQOL in the moods and emotions dimension compared to their children whereas children better perceived their physical well-being than their parents for the long version (KIDSCREEN-52). A good agreement was detected between self-and proxy ratings for the short version (KIDSCREEN-10).

Description

Citation