Background: The EQ-5D is a health-related quality of life instrument which provides a simple descriptive health profile and a single index value for health status. The latest version, the EQ-5D-5L, has been translated into more than one hundred languages worldwide - including Thai. This study aims to assess the measurement properties of the Thai version of the EQ-5D-5L (the 5L) compared to the EQ-5D-3L (the 3L).
Methods: A total of 117 diabetes patients treated with insulin completed a questionnaire including the 3L and the 5L. The 3L and 5L were compared in terms of distribution, ceiling, convergent validity, discriminative power, test-retest reliability, feasibility, and patient preference. Convergent validity was tested by assessing the relationship between each dimension of the EQ-5D and SF-36v2 using Spearman's rank-order correlation. Discriminative power was determined by the Shannon index (H ') and Shannon's Evenness index (J '). The test-retest reliability was assessed by examining the intraclass correlation coefficient (ICC) and Cohen's weighted kappa coefficient.
Results: No inconsistent response was found. The 5L trended towards a slightly lower ceiling compared with the 3L (33% versus 29%). Regarding redistribution, 69% to 100% of the patients answering level 2 with the 3L version redistributed their responses to level 2 with the 5L version while about 9% to 22% redistributed their responses to level 3 with the 5L version. The Shannon index (H ') improved with the 5L while the Shannon's Evenness index (J ') reduced slightly. Convergent validity and test-retest reliability was confirmed for both 3L and 5L.
Conclusions: Evidence supported the convergent validity and test-retest reliability of both the 3L and 5L in diabetes patients. However, the 5L is more promising compared to the 3L in terms of a lower ceiling, more discriminatory power, and higher preference by the respondents. Thus, the 5L should be recommended as a preferred health-related quality of life measure in Thailand.