Measuring quality of life in care homes when self-report is challenging: the construct validity, structural characteristics and internal consistency of the mixed-methods adult social care outcomes toolkit

Abstract

Abstract Introduction The adult social care outcomes toolkit (ASCOT) measures social care-related quality of life (SCRQoL) using self-completion questionnaires and interviews. Many care home residents find such methods inaccessible, leading to a reliance on proxy-reporting. This study aimed to establish the psychometric properties of the mixed-methods toolkit [ASCOT-Care Homes, 4 outcome (CH4)] for measuring SCRQoL when residents cannot self-report. Methods Two cross-sectional, mixed-methods studies were undertaken in care homes for older people in England between 2015 and 2020. We used the ASCOT-CH4 (observation, and interviews with residents and proxies) to collect information about SCRQoL and collected additional data on residents’ needs and characteristics, and variables hypothesised to be related to SCRQoL. Hypothesis testing was applied to establish construct validity, Cronbach’s alpha for internal consistency and exploratory factor analysis for structural validity. Results The combined dataset included 475 residents from 54 care homes (34 nursing, 20 residential). Half had a diagnosis of dementia. Less than a third of residents were able to complete an ASCOT interview. Observations and proxy interviews informed researcher ratings, meaning there were no missing ASCOT-CH4 scores. ASCOT-CH4 was found to be a weak unidimensional scale, consistent with other ASCOT measures, with acceptable internal consistency (α = 0.77, 8 items). Construct validity was supported by the findings. Conclusions The ASCOT-CH4 is an alternative to conventional proxy-questionnaires for measuring the SCRQoL of care home residents, with good psychometric properties. A limitation is that users need a range of data collection skills. Future research should explore whether findings are replicable when data are collected by other researchers.