Sherif B, Graham CN, Neidhardt K, Gilloteau I, O'Neill CB, McBride D, Augustin M. EQ-5D-3L utilities tariffs: differences in German and UK utilities and QALYS in patients with moderate to severe psoriasis. Poster presented at the 2017 ISPOR 22nd Annual International Meeting; May 22, 2017. Boston, MA. [abstract] Value Health. 2017 May; 20(5):A331.

OBJECTIVES: Health utilities needed for quality-adjusted life-years (QALYs) are often measured by instruments such as the EQ-5D-3L, with utility weights calculated by country-specific algorithms. This analysis explored the differences in utilities and QALYs when applying two different national EQ-5D-3L algorithms.

METHODS: EQ-5D-3L data came from CLEAR, a phase 3b, 52-week, head-to-head study comparing the efficacy and safety of secukinumab versus ustekinumab in adults with moderate to severe plaque psoriasis. EQ-5D-3L was assessed at baseline and weeks 4, 8, 12, 16, 28, 48, and 52. Utilities were calculated from observed data, using algorithms from Germany and the United Kingdom (UK). Treatment response was defined as Psoriasis Area Severity Index (PASI) reduction of < 50% (i.e., PASI 50), 50-74, 75- 89, and 90-100. A linear mixed model evaluated the relationship between EQ-5D-3L utility score and PASI, with the baseline utility score used as a covariate. QALYs were calculated in an existing cost-effectiveness Markov model evaluating biologic psoriasis treatments over 10 years.

RESULTS: Utility estimates derived from the German algorithm were higher than those derived from the UK algorithm. Differences decreased with higher PASI response. The mean baseline utility scores (0.8039 and 0.6812 for Germany and UK, respectively) were included in the Markov model to estimate the utility scores by PASI response. Estimated utility weights using German and UK algorithms were 0.888 and 0.801 for PASI 50; 0.924 and 0.850 for PASI 50-74; 0.937 and 0.880 for PASI 75-89; 0.951 and 0.908 for PASI 90-100, respectively. This resulted in lower incremental QALYs with German utilities than with UK utilities (0.068 vs. 0.095, respectively).

CONCLUSIONS: Applying different country-specific utility algorithms to EQ-5D-3L clinical trial data provided substantially different utility weights, affecting incremental QALY estimates in moderate to severe psoriasis. This variance should be considered when interpreting the results of cost-effectiveness analyses.