A chained approach to estimating the value of a QALY across Europe

Abstract

A major issue in health economic evaluation is that of the value to place on a quality adjusted life year (QALY), commonly used as a measure of health care effectiveness across Europe. This has come to the fore in several European countries, resulting from the creation of national health technology and pharmaceutical assessment agencies. Methods of estimating cost per QALY, based on rigorous decision analytic models, are now very sophisticated. However, ‘threshold’ values adopted (such as £20-40,000 per QALY above or below which a new therapy will be rejected or recommended for adoption in England) are essentially arbitrary, with little or no economic foundation. This critical policy issue is reflected in the growing interest across Europe in development of more sound methods to elicit such a value.

EuroVaQ has been funded by the European Commission from 2007-2010. It is a collaboration of 28 researchers from 10 European countries, the main aim being to develop more robust methods to determine the monetary value of a QALY based on surveys of the general public. Two methods of deriving a societal willingness-to-pay (WTP) based monetary value of a QALY have been tested. The ‘direct’ approach sets out to ask the question ‘what is the monetary value of a QALY?’ as directly as possible and makes use of diagrammatic representations of a QALY.  The ‘indirect’ approach- which will be the focus of this talk -uses a ‘chained’ approach using the following basic procedure. First, utility values are elicited for health states described using the EQ5D descriptive system.  Second, a monetary value to avoid some risk/duration of that health state is then elicited and the implied WTP per QALY estimated.

Eight versions of the indirect questionnaire have been developed and administered via Internet panels in each participating country. Data collection is almost complete and we are on target to achieve over 21,000 completed indirect interviews by Feb 2010.  An overview of the preliminary results will be presented and plans for modelling the data discussed.