Background: This study was designed to evaluate the feasibility, reliability, and validity of use of caregivers' ratings of two health preference measures as outcomes for cost-effectiveness analyses in persons with very mild to moderate Alzheimer's disease (AD).
Methods: Caregivers completed ratings of preference for AD patients' health by use of the EuroQol-5D system (EQ-5D) and the Health Utilities Index Mark 2 (HUI2). They also rated patients' cognition, mood, burden, AD-specific and generic health-related quality of life (QOL), and activities of daily living.
Results: Caregivers' HUI2 scores were reliable. Neither the caregiver ratings of the patients' health by use of the EQ-5D nor the HUI2 had a relationship with severity of cognitive impairment. Both the EQ-5D and the HUI2 had expected relationships with caregivers' assessments of patients' function, AD-specific QOL, and physical and mental health and selected subscales of the measures of AD-specific QOL and overall health. In addition, caregiver scores showed relationships with patient self-rated function, mood, and physical health but not AD-specific QOL. Caregiver burden was associated with caregivers' scores.
Conclusions: Caregiver-completed ratings of preference for patients' health made by use of the EQ-5D and the HUI2 have many of the characteristics of valid preference measures. However, the lack of association with patient Mini-Mental Status Exam scores and patient self-rated AD-specific QOL and the associations with caregiver subjective burden might present limitations to their use as proxy measures for cost-effectiveness analyses.