PROMIS PEDIATRIC MEASURES IN PEDIATRIC ONCOLOGY: A RELIABLE AND VALID HEALTH-RELATED QUALITY OF LIFE ASSESSMENT FOR CHILDREN WITH CANCER

Back to Page Authors: S. W. W. Chan, M. Y. C. Pang, A. K. P. Leung, Y. B. Ho, N. C. Chan, A. Y. L. Wong

Keywords: quality of life, childhood cancer, questionnaire, oncology

Abstract: Introduction: Paediatric-25 Profile version 2.0 (PROMIS-25) is a validated self-reported questionnaire for measuring Health-Related Quality of Life (HRQOL) in children with chronic diseases (e.g., asthma) and cancers (e.g., leukemia). It comprises physical functions, anxiety, depressive symptoms, fatigue, peer relationships and pain interference domains. However, the direct translation of PROMIS-25 is unlikely to be valid for assessing HRQOL of children with cancer. Objectives: This study aimed to translate and culturally adapt PROMIS-25 for Chinese children in Hong Kong, and to evaluate the psychometric properties of this instrument. Methodology: An established translation procedure (including forward and backward translation, and a pilot testing through cognitive debriefing interviews) was adopted to translate PROMIS-25. Three panel members evaluated the semantic and content equivalence scores. The translated questionnaire was validated on 103 Chinese children undergoing cancer treatment (10 to 18 years) in a local public hospital. Structural validity was assessed by Principal Component Analysis (PCA) and item fit statistics. Convergent and divergent validities were assessed by the correlations between Chinese PROMIS-25 domains, Patient Health Questionnaire-9 (PHQ-9) and Paediatric Quality of Life InventoryTM 4.0 (PedsQL 4.0) domains. The internal consistency and test-retest reliability were measured by Cronbach’s alpha and Intraclass Correlation Coefficient (ICC), respectively. The responsiveness of Chinese PROMIS-25 was tested against 24 predefined hypotheses with anticipated magnitudes and directions of correlation between Chinese PROMIS-25 domains, PHQ-9 and PedsQL 4.0 domains following the cancer treatment. The Minimal Important Difference (MID) of each domain was determined by an anchor-based method (Numeric Global Rating of Change Scale as an anchor). Results and Outcomes: The semantic equivalence and content validity scores of Chinese PROMS-25 ranged from 96% to 100%. The PCA supported the unidimensionality of each PROMIS-25 domain (each domain variance and eigenvalue: > 55.5% and < 2.0). All items showed acceptable item fit statistics (between 0.6 and 1.4). There were high convergent and divergent validity between Chinese PROMIS-25 domains and PHQ-9/PedsQL 4.0 domains (related domains: r ≥ 0.68; unrelated domains: r ≤ 0.39). All Chinese PROMIS-25 domains demonstrated good internal consistency (Cronbach’s alpha: 0.82-0.85), test-retest reliability (ICC: 0.90-0.97) and responsiveness (confirmed 22 of the 24 predefined hypotheses). All Chinese PROMIS-25 domains (except peer relationships) indicated deteriorated functioning after two days of cancer treatment in responsiveness analyses. The MIDs were determined (out of 100): Physical Function: 12.3; Anxiety: 17.2; Depression: 15.4; Fatigue: 13.6; Peer Relationships: 6.5; and Pain Interference: 13. Conclusion: The findings substantiate that traditional Chinese PROMIS-25 is a reliable and valid instrument for assessing HRQOL among Chinese children with cancer. With this validated self-reported instrument, the changes of health-related functioning can be well documented so that supplementary service to address the HRQOL can be considered for children undergoing cancer treatment.