PURPOSE: Previous research has documented the limited use of risk-sharing agreements (RSAs) in the U.S. commercial sector for both medicines and devices.  Research has also identified numerous practical barriers, including the costs of reaching an agreement and the lack of adequate data infrastructure.  From a theoretical perspective, these agreements attempt to address the key uncertainties by (1) collecting real-world data post-launch, and (2) sharing risk by adjusting post-launch reimbursement.  This problem has been characterized from a variety of theoretical perspectives--from value of information theory to real option theory to Bayes theorem.   This workshop explores the utility of these approaches to help parties to the agreement construct an arrangement appropriate to the source of the uncertainty and their risk preferences.   DESCRIPTION: Discussion leaders will describe practical challenges and how theory can help to structure appropriate incentives for addressing specific sources of uncertainty.  Five different, but complementary theoretical frameworks will be described and their utility discussed:  value-of-information theory, money-back guarantees, real option theory, portfolio theory, and Bayes theorem.  The first four have received more attention, so the discussion will focus on a Bayesian approach that, recognizing payer/manufacturer asymmetry of beliefs on outcomes, optimizes the trade-off between predictive likelihood priors (i.e., sensitivity and specificity in identifying the target population for the risk-sharing agreement) to maximize a relevant, posterior objective (gain/loss) function.  Alternative loss functions—payer budget impact (neutrality or maximum increment), population net health benefit, and target product market share—will be explored interactively, using hypothetical data, with workshop participants. Included will be a discussion of how modern item response theory (IRT) can inform population selection for RSAs.  This workshop will be valuable to health outcomes researchers interested in crafting RSAs on behalf of their manufacturer or managed care plan institutions, and in critically reviewing potential agreements in terms of optimizing patient population selection, outcomes choice, and administrative considerations.