How does EQ-5D-5L perform in asthma patients?

Background: EQ-5D-5L is a widely used generic health related quality of life (HRQL) tool, for evaluating interventions. Its performance in asthma patients has been sparsely investigated. Methods: We used data from 371 patients participating in the EPRA randomized controlled trial with waiting group design for pulmonary rehabilitation (PR). We assessed the EQ-5D and the visual analogue scale (VAS) at randomization (T0), at start (T1) and end of PR (T2), and three months post PR (T3). We calculated floor and ceiling effects (T0, T1, T2, T3), intraclass correlation (ICC) (T0-T1) and responsiveness to changes measured with the asthma control test (ACT) (T0-T1, T1-T2, T2-T3) through regression analysis. Our outcome variable was HRQL change adjusted for change on ACT (minimally important difference=3) in 5 categories (ACT change ≥3, 0 ACT change> -3, ACT change ≤ - 3), age, sex, BMI, smoking status, group, employed before PR (yes/no) and previous HRQL. Results: EQ-5D showed no floor effects but ceiling effects at T2 and T3 (32%) whereas VAS showed none of them. ICC was 0.85 for VAS and 0.82 for EQ-5D. EQ-5D was not responsible to changes in any of the waves. VAS detected ACT-changes bigger equal |MID| (T1: Δ VAS 5.67 (p=0.01) vs -7.05 (p=0.05), T2: Δ VAS 5.83 (p=0.02) vs -7.77 (p=0.01) and T3: Δ VAS 5.96 (p=0.01) vs -8.97 (p Discussion: EQ-5D showed good reliability but unsatisfactory responsiveness to changes. This could be an issue for the health economic evaluation of interventions. Considering an additional domain or using additional outcome variables (e.g. VAS) might mitigate the problem.