Methodological issues in the application of conjoint analysis in health care.

This paper adds to an increasing literature on methodological questions addressed in the application of conjoint analysis (CA) in health care. Three issues are addressed: ordering effects; internal validity; and internal consistency. The results of an application of CA in a primary care setting provide no evidence that the ordering of scenarios was important. Evidence was found of both internal validity and internal consistency. In addition, individual preferences were found to be determined by experiences, which raise potentially important questions regarding the elicitation and use of such preferences in economic evaluation.