Psychometric Properties of the EQ-5D-5L in Patients with Alopecia Areata

Andrew Lloyd,Daniel Aggio,Caleb Dixon,Ernest H Law,Thomas Price
DOI: https://doi.org/10.1007/s41669-024-00504-8
Abstract:Background: For many decision makers in Health Technology Assessment the EQ-5D-5L is the standard measure of health-related quality of life (HRQL). However, evidence has shown the limitations of the measure in certain disease areas, including dermatology. Alopecia areata (AA) is associated with a significant HRQL impact, partly due to the emotional impact of hair loss. Objectives: This study explores the psychometric properties of the EQ-5D-5L in people with AA in reference to the short-form 36 version 2 (SF-36v2), the Alopecia Areata Patient Priority Outcomes (AAPPO), the Severity of Alopecia Tool (SALT) and the Patient Global Impressions of Change (PGI-C). Methods: Data from participants with AA enrolled in the ALLEGRO-2b/3 trial (NCT03732807) of ritlecitinib were analysed. Participants completed the AAPPO measure (an AA-specific measure assessing emotional symptoms and activity limitations), PGI-C, EQ-5D-5L and SF-36v2 across 48-weeks of follow up. Extent of scalp hair loss was assessed using the SALT. Ceiling effects, known groups validity, convergent validity and responsiveness were examined. Known groups were defined by SALT score and a PGI-C defined response from baseline. Exploratory factor analysis was also performed. Results: Data were available from 612 adult participants. Ceiling effects were observed for the EQ-5D-5L (55.3-61.2%) and analyses suggested that the EQ-5D did not capture important differences between patients that the SF-36v2 did. The EQ-5D-5L very weakly correlated with SALT score, whereas the AAPPO correlated more strongly with the extent of hair loss. Compared with the EQ-5D-5L, the AAPPO was better able to discriminate between known groups defined by SALT and PGI-C. An exploratory factor analysis suggested that the EQ-5D-5L had limitations in content validity compared with the AAPPO. Conclusions: The EQ-5D-5L may not adequately measure the burden of AA on patients' HRQL. Insensitivity to the burden of AA suggests that the EQ-5D-5L may not measure treatment-related benefit with hair regrowth. Data from other measures could be considered if they are shown to be more relevant.
What problem does this paper attempt to address?