Adding two culture-specific 'bolt-on' dimensions on the Thai version of EQ-5D-5L: an exploratory study in patients with diabetes.

Background: This study explored whether addition of culture-specific bolt-on dimensions, 'interpersonal relationships (IR)' and 'activities related to bending knees (AK)' improves the relevance and validity of the EQ-5D among Thai patients with diabetes.

Methods: A cross-sectional study was conducted among 200 diabetic patients. Ceiling/floor effects, known-groups validity and test-retest reliability were evaluated. Hierarchical multiple regressions was performed to evaluate the incremental value of the EQ-5D+AK+IR over the EQ-5D-5L in predicting utility measured by visual analog scale (VAS).

Results: The EQ-5D+AK+IR reduced ceiling effects by 34%. For known-groups validity, women and lower educated patients reported more AK problems (p-value< 0.05); however, none of the IR dimension hypotheses were supported. Both IR and AK had the weighted kappa coefficients of 0.49, indicating good reliability. Hierarchical multiple regression found that adding AK (adjusted r2: 0.182 vs 0.156), but not IR (adjusted r2: 0.157 vs 0.156) significantly increased the explained variance in VAS score.

Conclusions: Preliminary findings suggest that AK, but not IR, may be a useful addition to the EQ-5D questionnaire for use with Thai patients. Further studies in diverse populations are needed to examine the impacts of the proposed new dimensions and determine if these cultural adaptations are justified.