Measurement properties of the EQ-5D-Y administered through a smartphone app in children with asthma: a longitudinal questionnaire study

Abstract

Background: Asthma impacts children’s physical, emotional, and psychosocial Health-Related Quality of Life (HRQL). The EQ-5D-Y is a generic econometric instrument developed to measure HRQL in children. Objective: Evaluation of feasibility, validity, reliability, and responsiveness of EQ-5D-Y descriptive system and util‑ ity index to allow the assessment of HRQL in children with asthma, aged 8–11 years (self-response version) or under 8 years old (proxy-response version). Methods: We used data from baseline to 10 months of follow-up of an observational, prospective study of children with persistent asthma recruited by pediatricians in Spain (2018–2020). HRQL instruments were administered through a smartphone application: ARCA app. The EQ-5D-Y is composed of a 5-dimension descriptive system, a utility index ranging from 1 to − 0.5392, and a general health visual analogue scale (EQ-VAS). The Pediatric Asthma Impact Scale (PROMIS-PAIS) includes 8 items, providing a raw score. Construct validity hypotheses were stated a priori, and evalu‑ ated following two approaches, multitrait–multimethod matrix and known groups’ comparisons. Reliability and responsiveness subsamples were defned by stability or change in EQ-VAS and the Asthma Control Questionnaire (ACQ), to estimate the intraclass correlation coefcient (ICC) and the magnitude of change over time. Results: The EQ-5D-Y was completed at baseline for 119 children (81 self-responded and 38 through proxy response), with a mean age of 9.1 (1.7) years. Mean (SD) of the EQ-5D-Y utility index was 0.93 (0.11), with ceiling and foor efects of 60.3% and 0%, respectively. Multitrait–multimethod matrix confrmed the associations previously hypothesized for the EQ-5D-Y utility index [moderate with PROMIS-PAIS (0.38) and weak with ACQ (0.28)], and for the EQ-5D-Y dimension “problems doing usual activities” [moderate with the ACQ item (0.35) and weak with the PROMIS PAIS item (0.17)]. Statistically signifcant diferences were found in the EQ-5D-Y between groups defned by asthma control, reliever inhalers use, and second-hand smoke exposure, with mostly moderate efect sizes (0.45–0.75). The ICC of the EQ-5D-Y utility index in the stable subsamples was high (0.81 and 0.79); and responsiveness subsamples presented a moderate to large magnitude of change (0.68 and 0.78), though without statistical signifcance. Conclusions: These results support the use of the EQ-5D-Y as a feasible, valid, and reliable instrument for evaluating HRQL in children with persistent asthma. Further studies are needed on the responsiveness of the EQ-5D-Y in this population.