A Risk Based Random Utility Model for Population Based Discrete Choice Analysis in Health Economics with Applications in Modeling Patients Choices Among Medical Treatment Plans

This paper extends the health economics literature with a risk based Random Utility Model for discrete medical treatments choice analysis using Population level observational data. The model is based on a subjective interpretation of the choice probabilities, and the economic theory of decision making under risk. It lends itself to a more general specification of the utility function, allowing the analyst to link economic theory with a broader class of linear statistical models that subsumes most currently available linear functional form representations as special cases. The model is shown to be in line with generalized linear mixed models (GLMMs), and estimated using Bayesian Markov Chain Monte Carlo (MCMC) methods