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.
Alopecia areata (AA) is a disease that causes hair loss on the scalp and, in some cases, other parts of the body. It affects 18.4 million people worldwide. We know that AA can have a significant impact on a person's health-related quality of life (HRQL). Understanding the impact of AA on HRQL is important, but frequently used questionnaires to assess HRQL may not accurately measure the impact of the condition. This study uses data from a clinical trial (ALLEGRO-2b/3 trial) conducted in patients with AA from multiple countries to examine whether frequently used HRQL questionnaires can measure the impact of AA. We compared the HRQL of people with different levels of hair loss to see how well these questionnaires measure the impact of AA. We used data from 612 participants who took part in the trial. We found that some of the frequently used questionnaires did not detect differences between people with different levels of hair loss or those who thought their condition had improved compared with those who did not, suggesting that they may not accurately measure the impact of AA. Overall, some frequently used questionnaires to assess HRQL may not be appropriate for use in people with AA. Other ways of measuring HRQL may be more appropriate for understanding the full impact of AA.
ABSTRACT
INTRODUCTION: Patients with alopecia areata (AA) report high levels of dissatisfaction with commonly used treatments. Patient-reported outcomes are essential to understanding patients' experiences with AA treatments. The objective of this study was to evaluate patient-reported satisfaction with hair growth among patients with AA receiving ritlecitinib or placebo and the correlation between clinician-assessed efficacy and patient-reported satisfaction. METHODS: In the ALLEGRO-2b/3 (NCT03732807) trial, patients with AA and ≥50% scalp hair loss were randomized to daily ritlecitinib or placebo for 24 weeks, with a 24-week extension of continued ritlecitinib or switch from placebo to ritlecitinib. The Patient Satisfaction with Hair Growth (P-Sat) measure evaluated patients' satisfaction with hair growth in 3 domains: amount, quality, and overall satisfaction with hair growth. The prespecified analysis evaluated the proportion of patients who were slightly, moderately, or very satisfied with hair growth. Several post hoc analyses assessed the proportion of patients who were moderately/very satisfied and moderately/very dissatisfied and calculated polyserial correlations between change from baseline (CFB) in Severity of Alopecia Tool (SALT) and P-Sat scores at weeks 24 and 48. RESULTS: At week 24, the proportion of patients (N = 718) reporting satisfaction (slightly, moderately, or very satisfied) overall with their hair growth ranged from 36.4% in the ritlecitinib 10-mg group (evaluated for dose ranging only) to 67.5% in the 200/50-mg group versus 22.6% in the placebo groups. In patients randomized to ritlecitinib, the proportion who were satisfied increased or was maintained at week 48. A substantially greater proportion of placebo patients who switched to ritlecitinib reported satisfaction at week 48 than at week 24. Similar results were observed for patient satisfaction with the amount and quality of hair growth. In the post hoc analyses defining satisfaction as moderately/very satisfied and dissatisfaction as moderately/very dissatisfied, the benefit of ritlecitinib was also observed. All P-Sat domain scores strongly correlated with CFB-SALT scores at weeks 24 (range 0.73-0.76; p < 0.05) and 48 (0.74-0.77; p < 0.05). CONCLUSIONS: Patients receiving active ritlecitinib doses reported favorable results versus placebo in satisfaction with hair growth up to week 48. High concordance was observed between improvement in scalp hair growth evaluated by clinicians and patient-reported satisfaction.
ABSTRACT
PURPOSE: Alopecia areata (AA) is an autoimmune-mediated inflammatory dermatological disease characterised by non-scarring hair loss affecting the scalp and sometimes other hair-bearing sites. This study aimed to elicit health state utility values (HSUVs) from the UK general population for AA using time trade off (TTO) interviews. METHODS: Vignette descriptions of health states defined by the extent of hair loss were developed (as well as one describing caregiver burden). These were developed using data from standardised patient reported outcome (PRO) measures, a literature review and qualitative interviews. Health states were defined based on the severity of alopecia tool (SALT), which assesses extensiveness of scalp hair loss. HSUVs were then elicited for each health state in TTO interviews with the UK public. RESULTS: One caregiver and five patient health states were developed based on the literature review findings, clinical trial PRO (Hospital Anxiety and Depression Scale and Alopecia Areata Patient Priority Outcomes Questionnaire) data and qualitative interviews with patients (N = 11), clinical experts (N = 4) and caregivers of adolescents with AA (N = 10). These data showed a more severe impact among patients with more extensive hair loss. One hundred and twenty participants evaluated the vignettes in TTO interviews. Patient HSUVs ranged from 0.502 for the most extensive hair loss health state (SALT 50-100 + eyebrow and eyelash loss) to 0.919 (SALT 0-10) for the mildest health state. The caregiver HSUV was 0.882. CONCLUSION: Quantitative and qualitative data sources were used to develop and validate vignettes describing different AA health states. Patient and caregiver HSUVs demonstrate a large impact associated with AA, especially for states defined by more extensive hair loss.