A head-to-head comparison of EQ-5D-5 L and SF-6D in Chinese patients with low back pain

Abstract

Background: The comparative performance of the 3-level EuroQol 5-dimension and Short Form 6-dimension (SF-6D) has been investigated in patients with low back pain (LBP). The aim of this study was to explore the performance including agreement, convergent validity as well as known-groups validity of the 5-level EuroQol 5-dimension (EQ-5D-5 L) and SF-6D in Chinese patients with LBP. Methods: Individuals with LBP were recruited from a large tertiary hospital in China. All subjects were interviewed using a standardized questionnaire including the EQ-5D-5 L, 36-item Short Form Health Survey (SF-36), the Oswestry questionnaire and socio-demographic questions from June 2017 to October 2017. Agreement was evaluated by intra-class correlation coefficients (ICCs) and Bland-Altman plots. Spearman's rank correlation coefficients were applied to assess convergent validity. For known-groups validity, the Mann-Whitney U test or Kruskal-Wallis H test were used, effect size (ES) and relative efficiency (RE) were also reported. The efficiency of detecting clinically relevant differences was measured by receiver operating characteristic (ROC) curves between pre-specified groups based on Oswestry disability index (ODI), ES and RE statistics were also reported. Results: Two hundred seventy-two LBP patients (age 38.1, 38% female) took part in the study. Agreement between the EQ-5D-5 L and the SF-6D was good (ICC 0.661) but with systematic discrepancy in the Bland-Altman plots. In terms of convergent validity, most priori assumptions were more related to EQ-5D-5 L than SF-6D, but MCS derived from SF-36 was more associated with SF-6D. EQ-5D-5 L demonstrated better performance for most groups except location and general health grouped by the general assessment of health item from SF-36. Furthermore, when we applied ODI as external indicator of health status, the area under the ROC curve for EQ-5D-5 L was larger than that for the SF-6D (0.892, 95% CI 0.853 to 0.931 versus 0.822, 95% CI 0.771 to 0.873), the effect size was 0.63 for EQ-5D-5 L and 0.44 for SF-6D, and it was proved that EQ-5D-5 L was 42% more efficient than SF-6D at detecting differences measured by ODI. Conclusions: Both EQ-5D-5 L and SF-6D are valid measures for LBP patients. Even though these two measures had good agreement, they cannot be used interchangeably. The EQ-5D-5 L was superior to the SF-6D in Chinese low back pain patients in this research, with stronger correlation to ODI and better known-groups validity. Further study needs to evaluate other factors, such as responsiveness and reliability.