Validity, reliability, and feasibility of EQ-5D-3L, VAS, and time trade-off among Jordanians.

Abstract

Introduction and objectives: The EQ-5D VAS is a widely used measure for health-related quality of life (HRQol). The EQ-5D health states often weighted using the time trade-off (TTO). The assessment of the EQ-5D, VAS, TTO psychometric properties is necessary to inform adaption across countries and populations due to cross-cultural and language diversity. The objective of the study is to evaluate validity, reliability, and feasibility of the Arabic version of the EQ-5D-3L, VAS, and TTO among a broad sample of Arabic-speaking Jordanians.

Methods: A cross-sectional study was conducted using structured face-to-face interviews with 1007 participants attending outpatient clinics at a central university affiliated hospital. The participants completed the EQ-5D-3L, described own health via VAS and in words (excellent, very good, good, fair, and poor). Additionally, participants requested to answer a TTO question. Validity was assessed using correlation analysis, while test-retest methods were used to evaluate reliability. Perspectives of participants on how easy to complete or understand the study measures were investigated.

Results: Completion rates were 100% for the EQ-5D, 98% for VAS and 77% for TTO. The construct validity of EQ-5D, VAS, and TTO was supported by significant correlation across each other and participants' characteristics. Higher severity of EQ-5D health states was significantly associated with lower VAS scores (r_s=-0.38, p<0.01). Usual activity and pain/discomfort had the strongest correlations with VAS scores (-0.40 and -0.39, respectively, p<0.001). A higher willingness to trade years was observed with more severe EQ-5D states (r_s=0.32, p<0.001), with usual activity, and pain/discomfort showing the strongest associations. Self-reported health via VAS and words was highly correlated (r_s=0.76, p<0.001). Reliability of the weighted Cohen's k for EQ-5D dimension ranged between 0.3 and .72. Intraclass correlation coefficients for VAS were 0.80 and 0.92 for TTO, all p value <0.001. VAS showed substantial test-retest reliability (ICC=0.80), and TTO showed excellent reliability (ICC=0.92). However, only 62.3% and 54.6% found EQ-5D and TTO easy to understand, respectively, while 89.4% found the VAS easy to use.

Conclusion: The Arabic EQ-5D-3L, VAS, and TTO demonstrated validity and reliability for measuring HRQoL among Jordanians. The study highlights the importance of cultural and linguistic contexts in HRQoL measures adaption. Future research should focus on refining TTO to improve cultural appropriateness and understanding among Arabic-speaking populations.