The Psychometric Performance of Generic Preference-Based Measures in Informal Carers: A Systematic Review of Validation Studies

Abstract

Background and objective: A growing number of health technology assessment agencies recommend inclusion of informal carer outcomes in health economic evaluations. While generic preference-based measures (GPBMs) are favoured, the evidence regarding their performance in measuring the health-related quality of life of informal carers has not been synthesised. The aim of this systematic review was to synthesise the psychometric evidence of GPBMs in informal carers.

Methods: Following Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, a literature search (indexed through October 2024) was conducted in CINAHL, PsycInfo, Embase and MEDLINE databases, supplemented with forward and backward citation searches. Publications were included that reported the psychometric performance of GPBMs in informal carers, regardless of care recipients' condition. Narrative synthesis was used to summarise the evidence. Quality of studies was evaluated using the COSMIN risk of bias checklist. International Prospective Register of Systematic Reviews (PROSPERO) registration is CRD42023434651.

Results: Twenty-one studies (published between 2001 and 2024) were identified, with nine evaluating multiple GPBMs (head-to-head comparisons). The EQ-5D 3-level (EQ-5D-3L) [n = 9] and EQ-5D 5-level (EQ-5D-5L) [n = 7] were the most frequently evaluated, followed by the Short-form 6-Dimension version 1 (SF-6Dv1) [n = 4], EuroQol Health and Wellbeing Short Form (EQ-HWB-9) [n = 4], Health Utilities Index (HUI) marks 2/3 (n = 3), Health-related Quality of Life Instrument with 8 Items (HINT-8) [n = 1] and Quality of Well Being Self-Administered (QWB-SA) [n = 1]. Studies were conducted in the USA (n = 6), UK (n = 4), China (n = 4), Australia (n = 3), Italy (n = 1), Iran (n = 1) and South Korea (n = 1), including a multi-country study (UK, Germany and France) study (n = 1). Care recipient conditions included carers of unspecified conditions, adults using long-term care, Alzheimer's disease or dementia, autism, cancer, leukaemia, craniofacial malformations, meningitis and multiple sclerosis. The EQ-5D-3L and EQ-5D-5L had evidence of ceiling effects at the index level. The EQ-5D-3L, EQ-5D-5L and EQ-HWB-9 demonstrated at least 'good' (intraclass correlation coefficient > 0.60) test-retest reliability. Known-group validity evidence was available for the EQ-5D-3L, EQ-5D-5L, EQ-HWB-9, HUI3 and SF-6Dv1 where each GPBM was able to discriminate over 60% of the groups (known or exploratory). Convergent validity studies reported that the EQ-5D-3L, EQ-5D-5L, EQ-HWB-9, HUI3, SF-6Dv1 and QWB-SA had moderate correlations with at least one care-specific preference-based measure (Adult Social Care Outcomes Toolkit for Carers [ASCOT-Carer], Care-Related Quality of Life [CarerQol] and Carer Experience Scale [CES]). Responsiveness was evaluated for the EQ-5D-5L, EQ-HWB-9 and SF-6Dv1 where mixed evidence was reported for the two EuroQol instruments, whereas the SF-6Dv1 was not found to be responsive. The studies identified were generally of adequate quality.

Conclusions: Current literature supports the use of GPBMs for informal carers; however, evidence on individual psychometric indicators is still limited. Further research is still needed, preferably involving head-to-head comparison and content validity studies in carers of people with various conditions and across countries that utilise cost-effectiveness evidence in resource allocation decisions, ideally employing longitudinal study designs.