More 'con' than 'joint': Problems with the application of conjoint analysis to participatory healthcare decision making.

Conjoint analysis (CA) is a survey method of data collection and analysis that proponents claim is a rigorous technique for eliciting consumer preferences for healthcare provision. Essentially the technique entails respondents making trade-offs between different healthcare attributes. Regression techniques are applied to the data to produce utility values for the different attributes, which can then be used to inform service provision. Earlier criticisms have tended to focus on aspects of data collection and analysis. This paper offers a broader critique of CA based on its underlying assumptions and claims. The technique is briefly described, followed by a series of criticisms, including the tendency to exclude from the analysis respondents who are unwilling or unable to make trade-offs; the extent to which the options tend to be defined within existing bureaucratic constraints; the absence of a logical pathway between the attributes that are to be traded; the way in which the creation of a common currency of utility value contributes to the further marketization of healthcare; the ontological assumptions made about the respondents' capacity to anticipate future states of subjectivity; and the epistemological claims made for the technique as a 'scientifically defensible' account of consumer preferences. It is concluded that there is an inverse relationship between the magnitude of a decision and the value of applying CA to it. And that rather than facilitating 'joint' decision making, CA uses methodological artifice to 'con' respondents into contributing to a discourse of consumer preference that may not reflect their actual beliefs.