COMPARING TEST-RETEST RELIABILITY AND RESPONSIVENESS OF EQ-5D-5L WITH FOUR BOLT-ONS, PROMIS-29 AND SKINDEX-16 IN PSORIASIS AND ATOPIC DERMATITIS PATIENTS
Author(s)
Sandra Nael, M.Sc.1, Zhuxin Mao, Ph.D.2, Dina Saadi, Ph.D.1, Aureliano Paolo Finch, Ph.D.3, Maggie Abbassi, Ph.D.1, Samar Farid, PhD1;
1Cairo University, Cairo, Egypt, 2University of Antwerp, Antwerp, Belgium, 3EuroQol Research Foundation, Amsterdam, Netherlands
1Cairo University, Cairo, Egypt, 2University of Antwerp, Antwerp, Belgium, 3EuroQol Research Foundation, Amsterdam, Netherlands
OBJECTIVES: Evidence supporting psychometric validity of EQ bolt-ons has been largely limited to cross-sectional studies. This study evaluated the impact of adding four bolt-ons—Skin Irritation (SI), Self-confidence (SE), Social Relationships (SR), and Sleep (SL)—on test-retest reliability and responsiveness of EQ-5D-5L in patients with Psoriasis (Ps) and Atopic Dermatitis (AD), and compared performance with PROMIS-29 and Skindex-16.
METHODS: This observational, single-center longitudinal study collected patient-reported data at baseline (N=236), 3-7 days later for test-retest reliability (N=180), and after 12 weeks for responsiveness (N=194). Test-retest reliability was assessed using weighted kappa (Kw) and intraclass correlation coefficients (ICC). At week 12, patients were categorized as “stable,” “improved,” or “worsened” based on a combination of two external criteria (Global Rating of Change and clinical endpoints). Responsiveness was evaluated using effect sizes, standardized response mean, relative efficiency (RE), and area under the receiver operating characteristic curve (AUROC).
RESULTS: EQ-VAS and standardized level sum scores showed good reliability with ICCs consistently >0.85 in Ps, and moderate to good reliability in AD (ICC=0.77-0.88). Kw values for SI, SE, SR, and SL were 0.69/0.63, 0.72/0.61, 0.72/0.53, and 0.75/0.75 in Ps/AD, respectively. PROMIS-29 T-scores demonstrated moderate to good reliability (ICC=0.67-0.81 in Ps; 0.71-0.83 in AD). Individual inclusion of SI and SL more efficiently distinguished “improved” subgroups in Ps (RE=1.29 and 1.34), while inclusion of any single bolt-on was more efficient than EQ-5D-5L alone in AD. Skindex-16 showed the highest responsiveness among “improved” patients in Ps and AD (RE= 1.57 and 3.2, respectively). However, in “worsened” groups, EQ-5D-5L alone outperformed bolt-on versions and Skindex-16. All HRQoL measures were responsive (AUROC >0.6).
CONCLUSIONS: Adding SI, SE, SR, and SL bolt-ons improved EQ-5D-5L responsiveness to smaller health changes while maintaining good test-retest reliability. Skindex-16 remained substantially more responsive than generic HRQoL instruments among improved groups.
METHODS: This observational, single-center longitudinal study collected patient-reported data at baseline (N=236), 3-7 days later for test-retest reliability (N=180), and after 12 weeks for responsiveness (N=194). Test-retest reliability was assessed using weighted kappa (Kw) and intraclass correlation coefficients (ICC). At week 12, patients were categorized as “stable,” “improved,” or “worsened” based on a combination of two external criteria (Global Rating of Change and clinical endpoints). Responsiveness was evaluated using effect sizes, standardized response mean, relative efficiency (RE), and area under the receiver operating characteristic curve (AUROC).
RESULTS: EQ-VAS and standardized level sum scores showed good reliability with ICCs consistently >0.85 in Ps, and moderate to good reliability in AD (ICC=0.77-0.88). Kw values for SI, SE, SR, and SL were 0.69/0.63, 0.72/0.61, 0.72/0.53, and 0.75/0.75 in Ps/AD, respectively. PROMIS-29 T-scores demonstrated moderate to good reliability (ICC=0.67-0.81 in Ps; 0.71-0.83 in AD). Individual inclusion of SI and SL more efficiently distinguished “improved” subgroups in Ps (RE=1.29 and 1.34), while inclusion of any single bolt-on was more efficient than EQ-5D-5L alone in AD. Skindex-16 showed the highest responsiveness among “improved” patients in Ps and AD (RE= 1.57 and 3.2, respectively). However, in “worsened” groups, EQ-5D-5L alone outperformed bolt-on versions and Skindex-16. All HRQoL measures were responsive (AUROC >0.6).
CONCLUSIONS: Adding SI, SE, SR, and SL bolt-ons improved EQ-5D-5L responsiveness to smaller health changes while maintaining good test-retest reliability. Skindex-16 remained substantially more responsive than generic HRQoL instruments among improved groups.
Conference/Value in Health Info
2026-05, ISPOR 2026, Philadelphia, PA, USA
Value in Health, Volume 29, Issue S6
Code
PCR69
Topic
Patient-Centered Research
Topic Subcategory
Instrument Development, Validation, & Translation, Patient-reported Outcomes & Quality of Life Outcomes
Disease
No Additional Disease & Conditions/Specialized Treatment Areas, SDC: Sensory System Disorders (Ear, Eye, Dental, Skin)