Psychometric properties of instruments to measure the quality of end-of-life care and dying for long-term care residents with dementia

Purpose Quality of care for long-term care (LTC) residents with dementia at the end-of-life is often evaluated using standardized instruments that were not developed for or thoroughly tested in this population. Given the importance of using appropriate instruments to evaluate the quality of care (QOC) and quality of dying (QOD) in LTC, we compared the validity and reliability of ten available instruments commonly used for these purposes. Methods We performed prospective observations and retrospective interviews and surveys of family (n = 70) and professionals (n = 103) of LTC decedents with dementia in the Netherlands. Results Instruments within the constructs QOC and QOD were highly correlated, and showed moderate to high correlation with overall assessments of QOC and QOD. Prospective and retrospective ratings using the same instruments differed little. Concordance between family and professional scores was low. Cronbach's alpha was mostly adequate. The EOLD-CAD showed good fit with pre-assumed factor structures. The EOLD-SWC and FPCS appear most valid and reliable for measuring QOC, and the EOLD-CAD and MSSE for measuring QOD. The POS performed worst in this population. Conclusions Our comparative study of psychometric properties of instruments allows for informed selection of QOC and QOD measures for LTC residents with dementia.