Response Performance of General Chinese Children Aged 6-11 to EQ-5D-Y-3L: A Multicenter Study
Author(s)
Yukun Wei, PhD candidate, Lei Chen, MD, Xiaoyu Xi, PhD.
The Research Center of National Drug Policy & Ecosystem, China Pharmaceutical University, Nanjing, China.
The Research Center of National Drug Policy & Ecosystem, China Pharmaceutical University, Nanjing, China.
Presentation Documents
OBJECTIVES: This study aimed to examine the EQ-5D-Y-3L self-report validity of general Chinese children of varying chronological age, explore their response characteristics and discuss the age-appropriateness of EQ-5D-Y-3L for them.
METHODS: A sample of general Chinese children aged 6-11 years were recruited from 4 schools in Wuzhong city. Children self-completed EQ-5D-Y-3L, Cantril scales and PedsQLTM4.0 accompanied by researchers in random order. The researchers randomly selected children for cognitive interviews using CI guide for verbal probing developed referring to the Tourangeau four-stage response model. Feasibility and acceptability were assessed by scale completion rates and researcher-rated children’s self-reported difficulty, respectively. Construct validity (convergent and divergent) were examined using Spearman’s rank correlation. Content validity (comprehensiveness, relevance, and comprehensibility) was assessed by analyzing the CI‘s texts.
RESULTS: A total of 906 children (mean age, 8.5 years; boy, 53.5%) completed the survey. 653 children completed cognitive interview. Feasibility is acceptable in all age groups. The acceptability of children aged 8 and below is uncertain. Except for the 8-year-old group, moderate to high correlations (|r|>0.3) were presented between EQ-5D-Y-3L utility scores and PedsQLTM4.0 total score in other age groups. In 10-year-old group, most conceptually similar dimensions and items showed significant correlations (P<0.01) as hypothesized. The connotations of health that children of all ages focus on and those covered by EQ-5D-Y-3L are partially different. In each age group, there were children who confused capacity limitations and health factor limitations, resulting in inappropriate relevance judgments. This problem was more prominent in the 6-8-year-old age group. Children showed a good understanding of the questions exploring internal manifestations.
CONCLUSIONS: Children's response performance to EQ-5D-Y-3L shows a non-monotonic relationship with age. It is necessary to select applicable populations by combining specific application scenarios with some individual factors, rather than simply relying on age. The factors influencing children's response performance and their mechanism require further research.
METHODS: A sample of general Chinese children aged 6-11 years were recruited from 4 schools in Wuzhong city. Children self-completed EQ-5D-Y-3L, Cantril scales and PedsQLTM4.0 accompanied by researchers in random order. The researchers randomly selected children for cognitive interviews using CI guide for verbal probing developed referring to the Tourangeau four-stage response model. Feasibility and acceptability were assessed by scale completion rates and researcher-rated children’s self-reported difficulty, respectively. Construct validity (convergent and divergent) were examined using Spearman’s rank correlation. Content validity (comprehensiveness, relevance, and comprehensibility) was assessed by analyzing the CI‘s texts.
RESULTS: A total of 906 children (mean age, 8.5 years; boy, 53.5%) completed the survey. 653 children completed cognitive interview. Feasibility is acceptable in all age groups. The acceptability of children aged 8 and below is uncertain. Except for the 8-year-old group, moderate to high correlations (|r|>0.3) were presented between EQ-5D-Y-3L utility scores and PedsQLTM4.0 total score in other age groups. In 10-year-old group, most conceptually similar dimensions and items showed significant correlations (P<0.01) as hypothesized. The connotations of health that children of all ages focus on and those covered by EQ-5D-Y-3L are partially different. In each age group, there were children who confused capacity limitations and health factor limitations, resulting in inappropriate relevance judgments. This problem was more prominent in the 6-8-year-old age group. Children showed a good understanding of the questions exploring internal manifestations.
CONCLUSIONS: Children's response performance to EQ-5D-Y-3L shows a non-monotonic relationship with age. It is necessary to select applicable populations by combining specific application scenarios with some individual factors, rather than simply relying on age. The factors influencing children's response performance and their mechanism require further research.
Conference/Value in Health Info
2025-05, ISPOR 2025, Montréal, Quebec, CA
Value in Health, Volume 28, Issue S1
Code
PT34
Topic
Patient-Centered Research
Topic Subcategory
Patient-reported Outcomes & Quality of Life Outcomes
Disease
No Additional Disease & Conditions/Specialized Treatment Areas