Comparing the psychometric properties of the EQ-5D-3L and EQ-5D-5L descriptive systems and utilities in atopic dermatitis

BACKGROUND

Atopic dermatitis (AD) is a common chronic inflammatory skin disorder affecting up to 10% of adults. The EQ-5D is the most commonly used generic preference-accompanied measure to generate quality-adjusted life years (QALYs) for economic evaluations.

OBJECTIVES

We aimed to compare psychometric properties of the three-level and five-level EQ-5D (EQ-5D-3L and EQ-5D-5L) in adult patients with AD.

METHODS

In a multicentre cross-sectional study, 218 AD patients with a broad range of severity completed the EQ-5D-3L, EQ-5D-5L, Dermatology Life Quality Index (DLQI) and Skindex-16. Disease severity outcomes included the Investigator Global Assessment, Eczema Area and Severity Index and the objective SCORing Atopic Dermatitis.

RESULTS

A good agreement was established between the two EQ-5D versions with an intraclass correlation coefficient of 0.815 (95% CI 0.758-0.859, p < 0.001). Overall, 33 different health state profiles occurred in the EQ-5D-3L and 84 in the EQ-5D-5L. Compared to the EQ-5D-3L, ceiling effect was reduced for the mobility, self-care, usual activities and pain/discomfort dimensions by 4.6-11.5%. EQ-5D-5L showed higher average relative informativity (Shannon's evenness index: 0.64 vs. 0.59). EQ-5D-5L demonstrated better convergent validity with EQ VAS, DLQI and Skindex-16. The two measures were similar in distinguishing between groups of patients based on disease severity and skin-specific quality of life with a moderate or large effect size (η² = 0.083-0.489).

CONCLUSION

Both instruments exhibited good psychometric properties in AD; however, the EQ-5D-5L was superior in terms of ceiling effects, informativity and convergent validity. We recommend the use of the EQ-5D-5L to measure health outcomes in clinical settings and for QALY calculations in AD.

The measurement performance of the EQ-5D-5L versus EQ-5D-3L in patients with hidradenitis suppurativa

PURPOSE

Hidradenitis suppurativa (HS) is a chronic inflammatory skin disease that affects up to 1% of the population in Europe. The EQ-5D is the most commonly used generic instrument for measuring health-related quality of life among HS patients. This study aims to compare the measurement properties of the two adult versions of EQ-5D (EQ-5D-3L and EQ-5D-5L) in patients with HS.

METHODS

We recruited 200 consecutive patients with HS (mean age 37 years, 38% severe or very severe HS) to participate in a multicentre cross-sectional survey. Patients completed the EQ-5D-3L, EQ-5D-5L, Dermatology Life Quality Index (DLQI) and Skindex-16 questionnaires.

RESULTS

More than twice as many different health state profiles occurred in the EQ-5D-5L compared to the EQ-5D-3L (101 vs. 43). A significant reduction in ceiling effect was found for the mobility, self-care and usual activities dimensions. A good agreement was established between the EQ-5D-3L and EQ-5D-5L with an intraclass correlation coefficient of 0.872 (95% CI 0.830-0.903; p < 0.001) that was confirmed by a Bland-Altman plot. EQ-5D-5L improved both the absolute and relative informativity in all dimensions except for anxiety/depression. EQ-5D-3L and EQ-5D-5L demonstrated similar convergent validity with DLQI and Skindex-16. EQ-5D-5L was able to better discriminate between known groups of patients based on the number of comorbidities and disease severity (HS-Physician's Global Assessment).

CONCLUSION

In patients with HS, the EQ-5D-5L outperformed the EQ-5D-3L in feasibility, ceiling effects, informativity and known-groups validity for many important clinical characteristics. We recommend using the EQ-5D-5L in HS patients across various settings, including clinical care, research and economic evaluations.

Validity of EQ-5D-5L, Skindex-16, DLQI and DLQI-R in patients with hidradenitis suppurativa

BACKGROUND

Numerous generic, skin- and disease-specific health-related quality of life (HRQoL) measures are available for patients with hidradenitis suppurativa (HS). Yet, robust psychometric evidence is lacking in many aspects of these outcome measures.

OBJECTIVES

We sought to determine convergent and known-groups validity of multiple generic and skin-specific HRQoL measures and to identify predictors of impaired HRQoL in patients with HS.

METHODS

Between 2017 and 2019, a multicentre cross-sectional study was carried out involving 200 consecutive HS patients. HRQoL outcomes included the EQ-5D-5L, EQ visual analogue scale (EQ VAS), Skindex-16, Dermatology Life Quality Index (DLQI) and DLQI-Relevant (DLQI-R). Disease severity was graded by HS-Physician's Global Assessment (HS-PGA) scale and the Modified Sartorius scale (MSS).

RESULTS

Overall, 77%, 56%, 51%, 46% and 28% reported problems in the pain/discomfort, usual activities, anxiety/depression, mobility and self-care dimensions of EQ-5D-5L. Mean ± SD EQ VAS, DLQI and DLQI-R scores were 64.29 ± 22.68, 11.75 ± 8.11 and 12.19 ± 8.33, respectively. Skindex-16 responses indicated that the emotional burden of HS (64.55 ± 29.28) far exceeded those of functioning (49.40 ± 34.70) and physical symptoms (46.74 ± 29.36). EQ-5D-5L, EQ VAS, DLQI, DLQI-R and Skindex-16 total scores had moderate or strong correlations with each other (range: |0.487| to |0.993|), weak or moderate correlations with HS-PGA (|0.350| to |0.433|) and weak correlations with MSS (|0.324| to |0.389|). DLQI-R slightly outperformed DLQI both in terms of convergent and known-groups validity. Being female, lower education level, more severe disease and genital involvement were associated with worse HRQoL (P < 0.05).

CONCLUSION

This study provides high-quality evidence that among skin-specific outcomes, the DLQI, DLQI-R and Skindex-16, and among generic instruments, the EQ-5D-5L are suitable to be used in HS patients. In future research, we recommend the use of existing well-validated HRQoL tools instead of developing new measures for each study. The development of composite measures that combine physician- and patient-reported outcomes is not supported by evidence in HS. [Correction added on 25 July 2020, after first online publication: in the Abstract section, the ± signs were missing and have been added to this version.].