A comparison of three instruments (ASCOT, ICECAP, and EQ5D) for the evaluation of outcomes of long-term care services
Florin Vadean | PSSRU, University of Kent
Objectives: The stated purpose of long-term care (LTC) in England is to improve the wellbeing and quality of life of service users. But how might this be measured? The National Institute for Health and Care Excellence (NICE) social care guidance recommends a flexible approach. This should include the EQ-5D measure of health-related quality of life as well as broader preference weighted measures, like the Adult Social Care Outcome Toolkit (ASCOT) care-related quality of life indicator and the Investigating Choice Experiments for the Preferences of Older People/Adults - CAPability (ICECAP-O and ICECAP-A) capability measures.
The main aims of this paper are to assess the construct validity of the ASCOT and ICECAP measures, and to comparatively assess the specificity and sensitivity of these indicators in measuring the impact of services. In particular, we aim to estimate the impact of utilisation of community-based LTC on either health-related quality of life (EQ-5D), care-related quality of life (ASCOT) and capability (ICECAP). While previous research showed that ASCOT is better suited to measuring the impact of long-term care services than EQ5D, we expect ICECAP to also capture (like ASCOT) aspects of peoples’ day-to-day functioning and, therefore, effects of long-term care services.
Methods: Exploratory factor analysis (EFA) was used to assess the dimensionality of ASCOT and ICECAP. The intention was to determine whether the items across the instruments may load onto the same factor(s).
The impact of long-term care services on wellbeing measures was then explored using a ‘production function’ method. The basis of this method is that we ought to see a correlation between intensity of LTC-service utilisation and people’s wellbeing ratings in a survey of service users, and these relationships can be modelled statistically. Instrumental variable (IV) estimation is used with control on observable confounders to tackle selection issues.
Data: A specifically-designed survey was undertaken which sampled people using publicly-funded long-term care services provided by local authorities in England. The survey provided a sample 770 service users.
Results: The EFA indicated that there is some overlap between the components captured by the ASCOT and ICECAP-O but not between ASCOT and ICECAP-A items. The analysis indicated that majority of the ICECAP-O items and the ASCOT ‘higher order’ items load into a single factor, while ASCOT items that are related to ‘basic’ social care needs form a separate factor. By contrast, the ASCOT and ICECAP-A do not overlap on any of the factors identified in the analysis.
We also found that LTC-service use significantly increases the care-related quality of life as measured by ASCOT, but has no significant effect when measured using EQ-5D. The effect of ICECAP is positive and significant only when excluding one capability attribute, indicated by EFA to load on a separate factor.