Abstract

Objectives

The Modified EQ Health and Wellbeing (EQ-HWB) and its shorter version (EQ-HWB-S) are new health and well-being measures nearing finalization. Since the 2022 version, the development group has proposed revisions to the item order, response levels, and wording. This study aimed to assess the psychometric properties of the modified Simplified Chinese EQ-HWB and EQ-HWB-S and compare their performance with other measures.

Methods

The data were collected through a web-based survey of the Chinese general population (n = 1053). Outcome measures included the modified EQ-HWB, EQ-5D-5L, SF-6Dv2, ICEpop CAPability measure for Adults, short version of Warwick-Edinburgh Mental Wellbeing Scale (SWEMWBS), and World Health Organization Well-Being Index 5-Item. We examined psychometric properties, such as distributional characteristics (ceiling and floor), convergent (correlations), known-group (effect sizes), and structural validity (principal component analysis).

Results

At the instrument level, neither the modified EQ-HWB nor EQ-HWB-S showed ceiling effect. The response distribution of the 3 items (enjoyable activities, excluded by others, and negative feelings about oneself) modified from positively to negatively framed was in the range of the other items. The EQ-HWB “getting around inside or outside,” “day-to-day activities,” and “enjoyable activities” items showed only moderate correlations with the corresponding EQ-5D-5L and SF-6Dv2 dimensions (rho = 0.31-0.36). The EQ-HWB items mainly loaded on separate factors from the ICEpop CAPability measure for Adults, short version of Warwick-Edinburgh Mental Wellbeing Scale, and World Health Organization Well-Being Index 5-Item items. The EQ-HWB effectively differentiated between sociodemographic and health-related known groups with larger effect sizes than the EQ-HWB-S.

Conclusions

The modified EQ-HWB and EQ-HWB-S demonstrated mixed psychometric performance. Several modified items showed improved distributional characteristics compared with earlier findings with the original EQ-HWB. However, item-level correlations were somewhat weaker than those reported in previous studies using the original version.