Hackert, M., van Exel, J. and Brouwer, W. (2017). Valid Outcome Measures in Care for Older People: Comparing the ASCOT and ICECAP-O Value in Health, 20(7):936--944.

Background: Limited health care budgets and population aging result in a need to compare care services on their benefits and costs. Because services for older people often aim to improve multiple life aspects, valid measures are needed to examine their benefits on individuals{\textquoteright} health and well-being simultaneously. Two measures may meet this end: the Adult Social Care Outcomes Toolkit (ASCOT) and the ICEpop CAPability measure for Older people (ICECAP-O). Objectives To compare the validity of both measures, the ASCOT and the ICECAP-O. Methods: A sampling agency gathered cross-sectional data in May 2015. Using exploratory factor analysis, the underlying factor structure of the ASCOT and the ICECAP-O was examined for the first time. Convergent and discriminant validity in relation to health measures (five-level EuroQol five-dimensional questionnaire, EuroQol Visual Analogue Scale, Barthel Index, and Geriatric Depression Scale-15) and well-being measures (Older People{\textquoteright}s Quality of Life Questionnaire-13, Satisfaction with Life Scale, and Cantril{\textquoteright}s Ladder) were tested using Spearman rank correlations and variance analysis. Results: The ASCOT and the ICECAP-O tapped into a shared factor, whereas both measures also loaded on two separate factors. The ASCOT and the ICECAP-O correlated highly with the health and well-being measures, but the correlation with the physical health measure Barthel Index was moderate. Both measures discriminated between subgroups of respondents. Conclusions: The ASCOT and the ICECAP-O seem promising measures to evaluate well-being among older people, whereby the ASCOT seems more specific to social care–related outcomes. The performance of both measures in other respondent groups and countries, and their relation to physical health, need to be further examined before their use in economic evaluations can be recommended.