Revisiting the axiom of completeness in health care.

Experiments concerned with modelling individual preferences are based on the assumption of completeness i.e. it is assumed that individuals have well-defined preferences for any choice they are presented with. However, this may not be the case for goods such as health care, where individuals are not used to making choices. If this assumption is violated, the large body of experimental economic literature eliciting patient preferences in health care may be challenged. This paper reports the results of a discrete choice experiment carried out to examine the assumption of complete preferences within health care. The tests carried out are based on the comparison of preferences for three different goods for which different levels of formed preferences are expected: a supermarket; dentist consultation and bowel cancer screening. The results do not provide sufficient evidence to support this hypothesis. However, further research is required before these results are generalised. Copyright 2002 John Wiley & Sons, Ltd.