OBJECTIVES:  The EuroQol EQ-5D is one of the most frequently used generic questionnaires to assess the patients’ health states and their quality of life. A new version (EQ-5D-5L) with five instead of three response categories (EQ-5D-3L) was introduced. Currently, six directly elicitated country value sets are available. For ten countries “crosswalk” value sets derived from EQ-5D-3L are available. We compared health state utility valuations from each of the value sets. METHODS: The value sets for the different countries were obtained from literature and the crosswalk value sets were downloaded from the EuroQol.org website. Differences of the utility indices between the value sets were analyzed by descriptive statistics and scatter plots. RESULTS: There were substantial differences in the utility index between countries in the values attributed to each health state. The median difference between the country with the highest index and the country with the lowest index was 0.417 across all 3125 possible health states for the 10 countries with crosswalk sets. For the six countries with literature data the median discrepancy was 0.260. For 99% of the health states the difference was more than 0.190 in the crosswalk sets. The highest difference in indices of 0.626 was observed for health state 15155 with Japan scoring 0.440 and the UK scoring -0.186. In almost all health states, Germany reported higher indices than France. Large discrepancies were also obtained when analyzing changes from baseline. In many cases not only the magnitude of the change was different between the countries but also the direction of the change. CONCLUSIONS: The values attributed to the different health states of the EQ-5D-5L questionnaire vary substantially by country. When analyzing a multinational clinical trial several relevant value sets should be used to judge the treatment effect and not only one set as it is commonly done.