Head-to-Head Comparison of the Psychometric Properties of 3 Carer-Related Preference-Based Instruments.

Abstract

To compare the psychometric properties of the Adult Social Care Outcomes Toolkit for carers (ASCOT-Carer), the Carer Experience Scale (CES), and the Care-related Quality of Life (CarerQol) to inform the choice of instrument in future studies. Data were derived from a 2018 online survey of informal carers in Australia. Reliability was assessed via internal consistency (Cronbach alpha, α) and test-retest reliability (intraclass correlation coefficient, ICC) for respondents who self-reported no change in their quality of life as a carer over 2 weeks. Convergent validity was evaluated via predetermined hypotheses about associations (Spearman's rank correlation) with existing, validated measures. Discriminative validity was assessed based on the ability of the carer-related scores to distinguish between different informal care situations (Mann-Whitney U, Kruskal-Wallis one-way analysis of variance). Data from 500 carers were analyzed. The ASCOT-Carer demonstrated a higher degree of internal consistency, possibly due to a unidimensional structure, and test-retest reliability than the CarerQol and CES (α= 0.87, 0.65, 0.59; ICC, 0.87, 0.67, 0.81, respectively). All 3 instruments exhibited convergent validity and detected statistically significant associations between carer-related scores and different informal care situations, except for the CarerQol-7D and sole carer status. The ASCOT-Carer, CarerQol, and CES performed reasonably well psychometrically; the ASCOT-Carer exhibited the best psychometric properties overall in this sample of Australian informal carers. Findings should be used in conjunction with consideration of research goals, carer population, targeted carer-related constructs, and prevailing perspectives on the economic evaluation to inform choice of instrument in future studies.