The Impact of a Revised EQ-5D Population Scoring on Preference-Based Utility Scores in an Inflammatory Arthritis Cohort

Abstract

Background and Objective It is well established that there are problems with the EQ-5D. This is due to the original scoring methods used and how negative time trade-off (TTO) values were treated. A revised scoring method has been published. This article applies this to an inflammatory arthritis cohort. The objective is to examine the impact of a revised scoring system for the EQ-5D (UK) TTO on the utility estimates and in the case of rheumatoid arthritis, to explore the impact of using different utility metrics on the incremental cost-effectiveness ratio (ICER) results of an economic model. Methods A total of 504 patients with inflammatory arthritis were rescored using revised EQ-5D scoring, which uses an episodic random utility model to deal with negative TTO values. Differences in utility scores were compared and the new mapping coefficients were obtained. These were then used in an economic model to examine the impact on the ICER. Results In rheumatoid arthritis, the overall change is less for the revised EQ-5D scoring than with the original EQ-5D (TTO) but greater than the SF-6D: EQ-5D UK ?0.22 (95% confidence interval [CI] ?0.30 to ?0.15), revised EQ-5D UK ?0.16 (95% CI ?0.21 to ?0.10) and SF-6D ?0.08 (95% CI ?0.11 to ?0.05). A similar trend is seen in the psoriatic arthritis group. The economic model produced different ICERs, when different utility measures were used; EQ-5D (TTO) ?42,402, SF-6D ?111,788, and revised EQ-5D (TTO) ?57,747. Conclusion In the context of inflammatory arthritis, this article demonstrates that a revised scoring for EQ-5D may have a significant impact on utility estimates and on the output of the economic model.