Review of the Minimal Clinically Important Difference of EQ-5D-3L Based on Anchor-Based Method

Zheng Yu; Dou Lei; Li Shunping

Abstract：Objective:To review the Minimal Clinically Important Difference(MCID)of EQ-5D-3L calculated by anchor-based method,evaluate the reliability of existing MCID,and provide reference thresholds for the interpretation of clinical research results.Methods:CNKI,PubMed,Embase and others were searched from inception to June 30,2023.The nonparametric tests were used to analyze the impact of statistical methods and sample size on MCID.Results:A total of 51 studies were eventually included.Studies were primarily prospective in design,32 studies focused on musculoskeletal disorders,37 studies used a single anchor,45 studies adopted patient-perspective anchor,and the Receiver Operating Characteristic(ROC)curve was the main statistical analysis method.A total of 160 MCID were obtained from 51 studies,ranging from 0.003 to 0.540,and only 3 MCID from 2 studies were rated as"high reliability"across all core items of the evaluation instrument.It is found that statistical analysis methods,sample size,and measurement time interval all significantly affect MCID.Conclusion:The disease type in EQ-5D-3L MCID studies were not abundant and unbalanced,and the heterogeneity of MCID measured in the included studies was high.The reliability evaluation results showed that the existing MCID has poor reliability in all items except the first and second items of the evaluation instrument.