OBJECTIVES: The EQ-5D-5L questionnaire (5L) has demonstrated better measurement properties than the EQ-5D-3L questionnaire (3L) in many cross-sectional studies. However, there is scarcity of evidence on responsiveness (sensitivity to change) of 5L index relative to 3L index in longitudinal studies. In this study, we aimed to (i) compare the responsiveness of 3L and 5L indices to the treatment benefit of cataract surgery, and (ii) compare them with the SF-6D and HUI3 indices in patients receiving cataract surgery.

METHODS: Patients who planned for the first or second cataract surgery in Singapore National Eye Centre were interviewed face-to-face before and after the surgery using 3L, 5L, SF-6D, HUI3, and a 12-item visual function (VF-12). Responsiveness of all indices was assessed using the standardized effect size (SES), standardized response mean (SRM), and F-statistic. The 95% confidence intervals (CI) of the responsiveness measures were calculated using 1000 bootstrap samples.

RESULTS: The analysis included 148 patients who completed both pre- and post-surgery assessments. Mean change in 3L index was 0.016 (Singapore value set) and 0.013 (UK value set), and in 5L index 0.028 (Singapore value set) and 0.016 (England value set). Using the Singapore value sets, the SES (95% CI) for EQ-5D-3L and EQ-5D-5L indices was 0.097 (-0.074, 0.259) and 0.199 (0.045, 0.350); the SRM (95% CI) was 0.091 (-0.063, 0.274) and 0.196 (0.034, 0.354); and the F-statistic (95% CI) was 1.2 (0.0, 10.8) and 5.7 (0.3, 18.6), respectively. Similar trends were observed using the UK/England EQ-5D value sets, although the magnitudes were slightly smaller. The SES, SRM and F-statistic (95% CI) for SF-6D and HUI3 were slightly larger than EQ-5D, but smaller than VF-12.

CONCLUSIONS: It appears that the 5L index tends to be more responsive than the 3L index to treatment benefit of cataract surgery.