Psychometric Properties of the EQ-5D-5L in Post-COVID-19 Condition: Results From the Long CORFU Study

The EQ-5D-5L is commonly used to assess health-related quality of life in patients with post-COVID-19 condition (PCC), but evidence on its psychometric properties and responsiveness to change in this population is limited. This study assessed psychometric properties of the EQ-5D-5L in long-term health-related quality of life assessments of patients with PCC.

Participants from multiple Dutch COVID-19 cohorts completed a self-administered questionnaire 2 and 3 years after acute COVID-19. Measures included 13 PCC-related symptoms, the EQ-5D-5L, self-rated health (SRH), and (at 3-year follow-up) a Global Rating of Change (GRC) scale. Psychometric properties were assessed via hypothesis testing of the following: distributional characteristics (ceiling/floor), divergent validity, convergent validity, explanatory power, known-group validity, and responsiveness.

Data from 237 participants were analyzed. A ceiling was observed in 7.2% of participants. Correlations between EQ-5D-5L dimensions ranged from weak to strong (r_s: 0.24-0.62). EQ-5D-5L index and EQ visual analog scale showed a strong correlation and EQ-5D-5L index and SRH showed moderate correlation. Usual activities and mobility explained most of the EQ visual analog scale variance (R²) (39.0% of total 41.3%). Median EQ-5D-5L index increased with higher symptom severity (P <.001). Weak correlations were observed between EQ-5D-5L index change scores and changes in SRH and symptom severity, and GRC, with low discrimination between change subgroups (area under the receiver operating characteristic curve < 0.7).

Most psychometric properties of the EQ-5D-5L in PCC populations were sufficient, with some limitations in capturing health in people with more severe symptoms, and an inability to detect change over time.