EQ-5D-5L valuation project for the Spanish populationa descriptive overview and preliminary results

Resumen

The latest EQ - 5D instrument, EQ - 5D - 5L, needs new country - specific valuation studies to obtain a value set adapted to the characteristics of the updated instrument. Eight countries from Europe, North and South America and Asia have participated on pilot exercises to develop a final protocol which will be commonly used to perform the valuation studies in each country. Spain is the first country where this protocol ha s be en introduced as part of the Valuation Project for the Spanish Population. T his discussion paper report s a descriptive overview of EQ - 5D - 5L valuation results in the Spanish population and its preliminary results. The survey has a two - stage sample plan. The first stage concentrates on the selection of Spanish regions. The 50 Spanish regions were ordered by p opulation size, and the first 20 regions covering 80% of the total Spanish population were selected. In the second stage, a simple random sampling strategy on each of the selected regions was conducted. The sample size on each region was calculated multipl ying the total sample size (1,000) by the percentage of the population on the region respect the total population of the select 20 regions. Data will be collected between 21st May and 15th June 2012 using the final agreed protocol by the EuroQol group. Pri mary data collection will be conducted by a specialist survey company with a second company conducting a strict quality control process to ensure interviews and data collection of highest quality. The final survey has three blocks of questions. The first block includes patient characteristics (age, gender, socioeconomic status and so on), and a respondent valuation of own health using the EQ - 5D - 5L. The second block contains 10 composite time trade - offs (TTO) questions, for states better than death classic TTO is used and the �Lead Time� TTO is used for those health states considered worse than death. In this case the lead time (period in full health) is 10 years and the time in the disease is another 10 years to be comparable with the classical TTO were the re are 10 years in full health and 10 years in the disease. The last block contains 7 discrete choice experiments questions, where the participant has to choose between two states. Finally some questions about the difficulties of the survey are also incl uded. Descriptive statistics of the final sample are reported. A detailed overview of summary statistics for the health state valuations is included. Different models were explored; hybrid TTO and DC model and DC conditional logi s t ic rescaled with TTO valu es were used and compared