Item-Level Convergent and Structural Validity of the EQ-HWB-9 in US Informal Caregivers Compared With ASCOT-Carer,...

To examine convergent validity and factorial structure of the EQ Health and Well-being 9-items (EQ-HWB-9) among US caregivers relative to caregiver-specific (Care-Related Quality of Life-7 Dimension [CarerQoL], Adult Social Care Outcomes Toolkit for Carers) and generic (EuroQol 5-Dimension 5-Level) measures.

Cross-sectional data from 504 informal caregivers were collected via online survey (2022-2023). Convergent validity was assessed using Spearman correlations with a priori hypotheses of strong correlations (r_s > 0.50) between conceptually similar items. Exploratory factor analysis using unweighted least-squares extraction and oblique rotation with parallel analysis examined dimensional structure across all instruments.

Convergent validity hypotheses were supported. Strong correlations emerged for physical dimensions (EQ-HWB mobility with EQ-5D mobility, r_s = 0.57; EQ-HWB daily activities with EQ-5D usual activities, r_s = 0.55) and emotional dimensions (EQ-HWB sadness/depression with EQ-5D anxiety/depression, r_s = 0.67). For caregiver-specific measures, strong correlations included EQ-HWB sadness/depression with CarerQoL mental health (r_s = 0.56) and EQ-HWB control with both ASCOT-Carer control (r_s = 0.51) and ASCOT-Carer occupation/spend time (r_s = 0.51). Weaker correlations between theoretically distinct constructs supported discriminant validity (r_s = 0.02-0.45). A 4-factor solution explained 100% of common variance with minimal cross-loading: social care-related quality of life (ASCOT-Carer items exclusively), mental/emotional well-being (primarily EQ-HWB and EQ-5D emotional items), physical functioning (EQ-5D and EQ-HWB physical items), and care-related burden (CarerQoL items). The 100% of common variance explained reflects the common factor model's extraction of shared variance only, excluding unique variance.

EQ-HWB-9 demonstrated strong convergent validity with clear factorial distinction from caregiver-specific measures, capturing general health constructs complementary to unique care-related domains assessed by ASCOT-Carer and CarerQoL. These findings support EQ-HWB-9 as a valid measure of general health and well-being in caregiver populations, while indicating that comprehensive caregiver assessment may benefit from combining generic and caregiver-specific instruments depending on study objectives.