BACKGROUND: Assessment of wellbeing is essential in the context of chronic diseases. The Investigating Choice Experiments for the Preferences of Older People-Capability Index (ICECAP-O) is a generic wellbeing measure beyond health. We aimed to assess whether ICECAP-O is a valid measure of capability-wellbeing in people with mild dementia by assessing its discriminative and convergent validity. Further, we assessed its feasibility, internal consistency, and test-retest reliability. METHODS: Four quality-of-life (QoL) and wellbeing measures were completed by 123 community-dwelling people with mild dementia. Discriminative validity was assessed by the ability of ICECAP-O to distinguish between levels of QoL, wellbeing, and related variables. Convergent validity was assessed by correlations between ICECAP-O index and domain scores, and the 5-level EQ-5D version (EQ-5D-5L), European Quality of Life visual analog scale (EQ VAS), Engagement and Independence in Dementia Questionnaire (EID-Q), and the World Health Organization Wellbeing Index (WHO-5). The feasibility of ICECAP-O in people with mild dementia was evaluated by missing data analysis. Internal consistency was assessed with Cronbach's α, and test-retest reliability was assessed using intraclass correlation coefficient (ICC). RESULTS: ICECAP-O significantly distinguished between different levels of QoL and wellbeing. Further, ICECAP-O was not affected by sociodemographic characteristics. ICECAP-O index scores were strongly correlated with EQ-5D-5L (ρ = 0.61) and EID-Q (ρ = 0.62), and moderately with WHO-5 (ρ = 0.58) and EQ-VAS (ρ = 0.58). Missing data on ICECAP-O dimensions, ranging from 2% to 3%, indicate good feasibility in people with mild dementia. ICECAP-O demonstrated acceptable internal consistency (α = 0.74) and good test-retest reliability (ICC = 0.86). CONCLUSION: The Danish version of ICECAP-O is a valid and reliable measure of wellbeing in people with mild dementia, and demonstrated acceptable psychometric properties, providing further evidence for its use in this population.
Pedersen et al. (Thu,) studied this question.