"The validity of the EQ-5D-5L in measuring quality of life benefits of breast reconstruction". Issue 1 (January 2019)

• Record Type: Journal Article
• Authors: Kouwenberg, Casimir A.E.; Kranenburg, Leonieke W.; Visser, Martijn S.; Busschbach, Jan J.; Mureau, Marc A.M.
• Is Part Of: Journal of plastic, reconstructive & aesthetic surgery. Volume 72:Issue 1(2019)
• Page Start: 52
• Page End: 61
• Publication Date: 2019-01
• Subjects: Breast reconstruction -- EQ-5D -- Outcome assessment (health care) -- Quality of life -- Mastectomy -- Cost-benefit analysis; Surgery, Plastic -- Great Britain -- Periodicals; Reconstructive Surgical Procedures -- Periodicals; Surgery, Plastic -- Great Britain -- Periodicals; 617.9505
• DOI: 10.1016/j.bjps.2018.08.023
• Languages: English
• ISSNs: 1748-6815

Abstract

Summary: Background: The EuroQol EQ-5D-5L instrument is the most widely used quality of life (QoL) measure in health economic evaluations. It is unclear whether such a generic instrument is valid enough to estimate the benefits of breast reconstruction (BR), given the specific changes observed in QoL after BR. Hence, we aimed to evaluate the validity of the EQ-5D-5L in patients who had undergone postmastectomy BR. Methods: In a 10-year cross-sectional cohort study, 463 mastectomy patients completed an online survey: 202 patients with autologous-BR (A-BR), 103 with implant-based-BR (I-BR), and 158 without BR (MAS). The results were used to evaluate the psychometric performance of the EQ-5D-5L with respect to the ceiling effect and to known-group, convergent, and discriminant validity, by comparing it with the Breast-Q, the cancer-specific (EORTC-QLQ-C30), and breast cancer-specific (EORTC-QLQ-BR23) questionnaires. Results: The EQ-5D-5L was able to discriminate between patients with and without complications, MAS with or without BR and MAS versus the general population. It was, however, not able to discriminate between A-BR vs . I-BR as well as BR vs . general population . It is not clear whether this was due to the insensitivity of the instrument, insufficient sample sizes, or because there were no actual differences in QoL between these groups. Good convergent and discriminant validity of both the EQ-5D-5L and its individual dimensions were demonstrated. Additional support for the instrument's validity was revealed by moderate correlations between the generic EQ-5D-5L and specific QoL aspects of BR such as sexuality and body image. Conclusions: The results of this study support the validity of the EQ-5D-5L as an outcome measure in health economic evaluations of BR.