Measuring the loss of consumer choice in mandatory health programs using Discrete Choice Experiments. CHERE Working Paper 2012/04

Economic evaluation of mandatory health programs generally do not consider the utility impact of a loss of consumer choice upon implementation, despite evidence suggesting that consumers do value having the ability to choose. The primary aim of this study was to explore whether the utility impact of a loss of consumer choice from implementing mandatory health programs can be measured using discrete choice experiments (DCEs). Three case studies were used to test the methodology: fortification of bread-making flour with folate, mandatory influenza vaccination of children, and the banning of trans-fats. Attributes and levels were developed from a review of the literature. An orthogonal, fractional factorial design was used to select the profiles presented to respondents to allow estimation of main effects. Overall each DCE consisted of 64 profiles which were allocated to 4 versions of 16 profiles. Each choice task compared two profiles, one being voluntary and the other being mandatory, plus a ?no policy? option, thus each respondent was presented with 8 choice tasks. For each choice task respondents were asked which health policy they most preferred and least preferred. Data was analysed using a mixed logit model with correlated coefficients (200 Halton draws). The compensating variation required for introducing a program on a mandatory basis (versus achieving the same health impacts with a voluntary program) that holds utility constant was estimated. Responses were provided by 535 participants (a response rate of 83%). For the influenza vaccination and folate fortification programs, the results suggested that some level of compensation may be required for introducing the program on a mandatory basis. Introducing a mandatory influenza vaccination program required the highest compensation ($113, 95%CI: -$61, $286) compared to folate fortification ($18, 95%CI: -$4, $40). No compensation was required for introducing the trans-fats program ($0, 95%CI: -A$6 to A$6). In addition to the type of MHP, the compensation required was also found to be dependent on a number of other factors. In particular, the study found an association between the compensation required and stronger libertarian preferences. We conclude that DCEs can be used to measure the utility impact of a loss of consumer choice. Excluding the utility impact of a loss of consumer choice from an economic evaluation taking a societal perspective may result in a sub-optimal, or incorrect, funding decision.