The Eq Health and Wellbeing (EQ-HWB) and the Eq Health and Wellbeing Short (EQ-HWB-S) As an Outcome Measure in Mental Health and Well-Being: A Comparison to the Eq-5D-5L, Eq-5D-3L and the Short Warwick-Edinburgh Mental Well-Being Scale

OBJECTIVES: The extent to which generic measures are sensitive to changes in symptoms, functioning, and quality of life that are of relevance to capturing mental health and well-being has been the subject of debate. The EQ-HWB was developed to capture aspects of health and well-being that may be missed by existing generic measures. This study aims to examine content overlap and compare measurement properties of EQ-HWB/EQ-HWB-S in relation to the EQ-5D-5L, EQ-5D-3L, and Short Warwick-Edinburgh Mental Well-Being Scale (S-WEMWBS) with respect to mental health conditions.

METHODS: An online panel of US-based respondents completed a survey that included the EQ-HWB, EQ-5D-5L, EQ-5D-3L, and S-WEMWBS. We examined content overlap between measures qualitatively, the strength of correlation between related items/constructs and discriminative ability of measures based on any self-reported mental health problem, clinical depression, and general anxiety disorder (GAD).

RESULTS: The dataset included 172 participants who self-reported any mental health problems, 113 clinical depression, and 98 GAD. Most content overlap measured with Jaccard Index was found between EQ-HWB-S and S-WEMWBS (33%). Strong associations (r_s > 0.5) were found between conceptually overlapping/related items of S-WEMWBS and EQ-HWB. The EQ-HWB-psychosocial level summary score (LSS) and EQ-HWB-S LSS tended to better discriminate than S-WEMWBS summary score based on any MH problem (F-ratio: 2.22, 95% Cl 1.61-2.89 and 2.13, 95% Cl 1.51-3.00 respectively), clinical depression (F-ratio: 2.24, 95% Cl 1.64-3.02 and 2.08, 95% Cl 1.48-2.92) and GAD (F-ratio: 2.13, 95% Cl 1.51-2.99 and 2.23, 95% Cl 1.53-3.28).

CONCLUSIONS: Initial evidence supports the validity of the EQ-HWB and EQ-HWB-S as outcome measures in mental health and well-being. All measures demonstrated discriminative ability, with the EQ-HWB psychosocial LSS and EQ-HWB-S tending to outperform both the SWEBMS and EQ-5D-5L/EQ-5D-3L in terms of known groups based on mental health, highlighting its future potential as a measure of mental health using a psychometrically derived summary score.