Validation and test-retest reliability of the Japanese version of the EQ-TIPS tool for health-related quality of life assessment

Assessing health-related quality of life in children under 48 months of age is useful for both diagnostic, prognostic, and therapeutic purposes. The EuroQoL Toddler and Infant Populations (EQ-TIPS) survey is designed for administration to adults, as proxies. This study aims to validate the Japanese version of this tool. Guardians of children aged 0-47 months were administered the EQ-TIPS questionnaire, including a six-dimension score and a visual analogue scale (VAS) for general health status, as well as the already-validated PedsQL survey. Validity was assessed via regression analysis, Cronbach’s alpha, and Spearman’s ρ . In order to assess test-retest reliability of the tool, a second instance of the same questionnaire was administered one month after the first. Reliability was investigated via Cohen’s kappa and intraclass correlation. Validation analysis was carried out on 215 participants, while only 168 answered both instances of the study questionnaire and could be involved in reliability testing. Cronbach’s alpha highlighted good internal consistency (α =0.82), but regression analysis identified significant association with VAS only for the EQ-TIPS score’s pain component. In multivariable regression, the communication component also was significantly associated with the VAS. Weak correlation among the EQ-TIPS components and VAS was identified by Spearman’s ρ test. Reliability was significant, but fair, for both EQ-TIPS ( k : 0.26; p-value <0.001) and VAS ( k : 0.23; p-value <0.001). However, good reliability was not identified by intraclass correlation. Although valid, the Japanese version of the EQ-TIPS has only limited reliability. Additional analysis should explore its functioning in different populations from Japanese settings. JEL Classification code: I10