The Dutch tariff: results and arguments for an effective design for national EQ-5D valuation studies.

The objective of this study was to estimate a Dutch EQ-5D tariff and to determine in a simulation study using the dataset of the original UK valuation study, the number of health states and respondents needed to estimate a reliable tariff. In all, 300 Dutch respondents directly valued 17 states compared to 3000 respondents and 42 states in the original MVH protocol. The results reaffirmed differences in health-related preferences between countries, justifying the estimation of national tariffs. The mean absolute error was 0.030. The design of this study is recommended for national EQ-5D valuation studies.