[Pubmed] Patient-reported health utility in adults with myasthenia gravis: a systematic review and meta-analysis of EQ-5
Posté : 03 juin 2026 12:00
BMC Neurol. 2026 Jun 2. doi: 10.1186/s12883-026-05033-x. Online ahead of print.
ABSTRACT
BACKGROUND: Myasthenia gravis (MG) is a chronic autoimmune neuromuscular disorder associated with impaired health-related quality of life. EQ-5D utility and EQ-VAS scores are used in health economic evaluations, but reported estimates vary across MG populations, instruments, tariffs, and study designs.
METHODS: We systematically searched PubMed, Embase, Scopus, and CENTRAL from inception to February 2026. Observational studies and randomized controlled trials reporting baseline or pre-intervention EQ-5D utility or EQ-VAS data in adults with MG were included. Random-effects meta-analyses used restricted maximum likelihood estimation. Subgroup, sensitivity, GRADE-informed certainty, and exploratory aggregate-level meta-regression analyses were performed.
RESULTS: Ten studies were included. Eight studies (n = 1,514) contributed to EQ-5D utility synthesis, yielding a descriptive pooled mean of 0.74 (95% CI 0.68-0.80), with very high heterogeneity (I² = 95.4%; Cochran Q p < 0.001) and a wide prediction interval (0.52-0.96). Seven studies (n = 2,760) contributed to EQ-VAS synthesis, with a descriptive pooled mean of 66.49 (95% CI 60.99-71.99), very high heterogeneity (I² = 96.3%; Cochran Q p < 0.001), and a wide prediction interval (47.70-85.28). Sensitivity analysis excluding the RCT baseline dataset did not materially change the pooled EQ-5D estimate. In an exploratory analysis based on two studies, utility declined across Myasthenia Gravis Foundation of America (MGFA) classes I-IV. Exploratory meta-regression suggested a possible negative association between mean disease duration and utility, but certainty was very low.
CONCLUSION: EQ-5D and EQ-VAS estimates indicate impaired patient-reported health status in MG, but pooled values should be interpreted as low-certainty descriptive summaries, not definitive utility inputs.
PMID:42231266 | DOI:10.1186/s12883-026-05033-x
Source: https://pubmed.ncbi.nlm.nih.gov/4223126 ... 0&v=2.20.0