Should we use willingness to pay to elicit community preferences for health care? New evidence from using a 'marginal' approach.

We describe a willingness to pay (WTP) survey in which values were elicited from the public for three disparate health care programmes. Previous applications of WTP in this context have revealed a high proportion of preference reversals between WTP values and ordinal ranking of the programmes. In view of the doubts these findings raise over the use of WTP in this context, our aim was to develop a method of eliciting WTP values which we considered would improve consistency between respondents' explicit ranking of the programmes and their WTP values. Compared to the standard approach, the structure of the new design (the marginal approach) reduced the number of possible preference reversals, thus encouraging a degree of consistency among respondents. Despite this, the marginal approach did not result in fewer preference reversal being observed in actuality, thus casting doubt on the application of WTP in this context.