Psychometric evaluation of the Disability Wellbeing Index (DWI) against the Adult Social Care Outcomes Toolkit (ASCOT) among people with disability.

Abstract

Purpose: The Disability Wellbeing Index (DWI) is a preference-based wellbeing measure developed among people with disabilities. This study examined the psychometric properties of the DWI in both its Standard and Easy English versions and compared them with two Adult Social Care Outcomes Toolkit (ASCOT) instruments, the ASCOT-SCT4 and ASCOT-ER.

Methods: We conducted an online survey among National Disability Insurance Scheme (NDIS) participants aged ≥ 15 years in 2024. DWI was scored using an Australian preference-based algorithm, and ASCOT was scored using UK preference weights. Convergent validity was assessed using Spearman's correlations between total utility scores and corresponding domain items. We examined underlying constructs between DWI and ASCOT using exploratory factor analyses (EFAs), and known-groups validity using regression analyses comparing scores across groups based on self-rated health, self-advocacy, participation in meaningful leisure activities, and autonomy in decision-making.

Results: A total of 1871 participants were studied; among them, 39% completed the Easy English/Read version, and 30% were aged 15-24 years. Correlations between DWI and ASCOT were moderate for young people (ρ = 0.62, Easy English/Read; ρ = 0.66, Standard) and strong for adults (ρ = 0.74, Easy English/Read; ρ = 0.75, Standard). Convergence was strongest for safety items, whereas convergence was weaker for dignity items. EFA indicated partial overlap in underlying constructs. Known-groups validity was supported for both measures.

Conclusions: The findings support the construct validity of DWI. DWI and ASCOT focus on different but overlapping aspects. The DWI places a stronger emphasis on general wellbeing, while the ASCOT focuses on social care related quality of life.