Preference-based measures in people with dementia: a systematic review of psychometric properties

PURPOSE: Accurate measurement of health and well-being outcomes is essential for economic evaluations of dementia care interventions. However, the suitability of existing preference-based measures (PBMs) for people with dementia remains uncertain. This review aims to provide guidance in measurement choice by assessing the psychometric properties of PBMs in people with dementia. METHODS: A systematic review was conducted of studies indexed in PubMed, Embase, PsycINFO and Web of Science up to July 2025. Studies were eligible if they evaluated psychometric properties of PBMs in people with dementia. Title and abstract screening were performed using a validated AI-powered tool, ASReview. The quality of the included studies and psychometric properties of the PBMs were assessed using the COSMIN methodology. RESULTS: A total of 21,347 records were identified, of which 32 studies covering fourteen different PBMs were included. The EQ-5D-3L was used in most studies, and studies focused mainly on construct validity and reliability. Study quality was predominantly high, while most psychometric evidence was inconsistent or insufficient. The ICECAP-O and HUI2 demonstrated the most appropriate psychometric evidence. CONCLUSION: Evidence of psychometric properties of PBMs is insufficient for some measures and limited for other measures. Therefore, none can yet be recommended as fully appropriate for use in people with dementia. This underscores the need for further validation studies to improve preference-based outcome measurement, and to strengthen economic evaluations of dementia care interventions.