Translation and psychometric validation of the traditional Chinese version of patient-reported outcomes measurement information system Pediatric-25 Profile version 2.0 (PROMIS-25) in Chinese children with cancer in Hong Kong

Abstract

Purpose: To translate and cross-cultural validate the PROMIS Pediatric-25 Profile 2.0 (PROMIS-25) into traditional Chinese, and to investigate its psychometric properties in children with cancer in Hong Kong. Methods: The Functional Assessment of Chronic Illness Therapy translation methodology was adopted in this study. Three panel members evaluated the semantic equivalence and content validity. The psychometric properties were tested with 103 children with cancer (10–18 years). Internal consistency and structural validity were examined by Cronbach’s alpha and Rasch analysis. Convergent and divergent validity were assessed by correlating it with traditional Chinese pediatric quality of life inventory™ 4.0 domains (traditional Chinese PedsQL™ 4.0), traditional Chinese Health Questionnaire-9 (C-PHQ-9), and the numeric pain rating scale (NPRS). Results: The semantic equivalence score and content validity index were both 100%. All domains indicated good internal consistency (α = 0.83–0.88) and unidimensionality (variance explained > 55.5% and 1st contrast eigenvalues < 2.0). All items showed good item fit (0.6–1.4). For convergent validity, the traditional Chinese PROMIS-25 domains demonstrated moderate-to-large correlations with traditional Chinese PedsQL™ 4.0 domains (r ≥ ± 0.69), C-PHQ-9 Item-4 and total score (r = 0.75–0.80), except NPRS (r = 0.44). For divergent validity, traditional Chinese PROMIS-25 had low correlations with traditional Chinese PedsQL™ 4.0 domains (r < ± 0.21), C-PHQ-9 item-4 (r = 0.3), and NPRS (r = − 0.12). The traditional Chinese PROMIS-25 fatigue domain was weakly correlated with NPRS (r = 0.39). Conclusion: The traditional Chinese PROMIS-25 is semantically and conceptually like the original PROMIS-25 with satisfactory internal consistency, structural validity, and construct validity.