A fully bayesian cost-effectiveness analysis using conditionally specified prior distributions

Abstract

The Bayesian approach to statistics has been growing rapidly in popularity as an alternative to the classical approach in the economic evaluation of health technologies. Bayesian approach has significant benefits over the classical approach. One of the most important advantages of the Bayesian methods is the incorporation of prior information. Thus, Bayesian methods can make use of more available information, and so produce stronger results than frequentist methods. In a cost ¿effectiveness analysis we relate the costs and efficacies of the two technologies compared. The parameters of interest are the mean efficacy and mean cost of each technology. The most common prior structure for both parameters is the bivariate normal structure. In this paper we study the use of a more general family of prior distributions for the parameters. In particular we assume that conditional densities of the parameters are all normal. This structure allows incorporating a large range of prior information. The bivariate normal distribution is included as a particular case of the conditional prior structure.