Psychometric properties of the EQ-5D-5L in patients with major depression: factor analysis and Rasch analysis

Abstract

Background

The EQ-5D is one of the most recommended questionnaires for cost-effectiveness studies.

Aims

To study the psychometric properties of the EQ-5D-5L in patients with major depression.

Methods

This prospective observational study included 433 patients with major depression who completed the EQ-5D-5L and Patient Health Questionnaire-9 (PHQ-9) questionnaires at baseline, of whom 310 also did six months later. The structural validity was assessed by confirmatory factor analysis, the item functioning by item response analysis, and reliability by Cronbach’s alpha. Convergent validity and known-groups validity was studied using the PHQ-9 and a general health question. To assess responsiveness effect sizes were calculated.

Results

The results supported the unidimensionality and showed adequate item functioning, with somewhat age-related item differential functioning for the mobility dimension. Cronbach’s alpha was 0.77. The EQ-5D-5L showed a high correlation with the PHQ-9 and general health. The more severe the depression level and the poorer the general health, the lower the EQ-5D-5L scores (p < 0.001). Responsiveness parameters showed moderate changes among “improved” patients.

Conclusions

These findings support the adequate psychometric properties of the EQ-5D-5L in patients with major depression. It could be very useful for clinicians and researchers as an outcome measure and for use in economic evaluation.

Keywords:

• EQ-5D-5L
• depression
• utility index
• psychometric properties
• item response theory
• economic evaluation

Acknowledgements

We are grateful to colleagues in the participating hospitals, primary care centers and mental health centers for their support and to all patients for their collaboration. The authors also acknowledge editorial assistance provided by Punto Mas Traducción y Servicios, S.L.u., funded by the Bilbao-Basurto Integrated Health Organisation.

Disclosure statement

The authors declare that they have no conflict of interest.

Additional information

Funding

This study was supported in part by grants from the Instituto de Salud Carlos III [PI13/00560, PI13/00518 and PI13/00648], and the European Regional Development Fund.