Agreement between Patient and Proxy Assessments of Quality of Life among Older Adults with Vascular Cognitive Impairment Using the EQ-5D-3L and ICECAP-O

Background The assessment of quality of life is critical in ascertaining the benefit of interventions aimed to reduce morbidity among individuals with cognitive impairment. However, the assessment of quality of life is challenging in this population due to the uncertain validity of patient responses as cognitive function declines. Hence, we examined the level of agreement between patient and proxy assessments of health related quality of life (HRQoL) and wellbeing based on the domains that comprise each of these constructs. Methods Analysis of baseline data from 71 community-dwelling older adults with mild Vascular Cognitive Impairment (VCI) who participated in a six-month proof-of-concept single-blinded randomized trial. Level of agreement between patient and caregiver ratings of HRQoL (EQ-5D-3L) and wellbeing (ICECAP-O) were compared using raw agreement (%), intraclass correlation coefficient (ICC) and weighted Cohen's kappa statistic. Results Self-care (assessed via the EQ-5D-3L) demonstrated almost perfect raw agreement between the patient and caregiver ratings. Three domains (mobility, pain and anxiety) of the EQ-5D-3L demonstrated fair agreement between the patient and caregiver ratings. Two (attachment and control) of the five ICECAP-O domains demonstrated slight agreement. The ICC indicated good agreement for the EQ-5D-3L and poor agreement for the ICECAP-O. Conclusion There is better patient-proxy agreement for the EQ-5D-3L compared with the ICECAP-O among individuals with mild VCI. These findings imply that the ICECAP-O may have limited clinical, research and policy related utility among individuals with mild VCI.