OBJECTIVE: The purpose of this research is to develop a decision model for evaluating the impact of information regarding alternative treatments, toxicity, efficacy, delivery method, and cost on demand for unique compounds. METHODS: This paper characterizes the relationship between multi-attribute utility theory and health-state preferences. Health outcomes arising from pharmaceutical interventions are viewed as multiattribute commodities. Presenters describe a method for reliable, utility-theoretic quantification of health-state preferences. This procedure requires estimating utility weights from stated-preference (SP) data. Including health cost as an SP attribute facilitates conversion of marginal utilities to marginal dollar values to explicitly account for cost in determining market share. Additionally, various utility specifications and simplifying assumptions are described. Finally, a rule for simulating aggregate choice behavior over time is presented. The prediction rule employed draws upon random utility maximization with adaptive expectations. This method addresses the probabilistic nature of the choice process by inclusion of a residual term representing the effect of unobserved factors on perceived utility. Expectation updating deals with imperfect information about product attributes such as efficacy and side effects by allowing learning to take place over successive drug administrations. RESULTS: A numerical example considering migraine medications demonstrates the capability of the decision model. In this example, a general, preference-based form for health-related utility facilitates direct estimation of health attribute utility weights arising from pharmaceutical consumption. Manipulating pharmaceutical attributes in a random utility framework simulates choice probabilities. Repeatedly simulating choice probabilities with attribute expectation updating provides market penetration estimates for unique compounds over time. CONCLUSIONS: Combining stated preference survey techniques with random utility and adaptive expectations provides a unique and realistic method for predicting demand for novel compounds.