Proceedings of the 2011 Winter Simulation Conference S. Jain, R. R. Creasey, J. Himmelspach, K. P. White, and M. Fu, eds. A FRAMEWORK FOR EVIDENCE-BASED HEALTH CARE INCENTIVES SIMULATION Joseph P. Bigus Ching-Hua Chen-Ritzo IBM T.J. Watson Research Center 1101 Kitchawan Road Yorktown Heights, NY 10598, USA Robert Sorrentino IBM T.J. Watson Research Center 19 Skyline Drive Hawthorne, NY 10532, USA ABSTRACT We present a general simulation framework designed for modeling incentives in a health care delivery system. This first version of the framework focuses on representing provider incentives. Key framework components are described in detail, and we provide an overview of how data-driven analytic methods can be integrated with this framework to enable evidence-based simulation. The software implementation of a simple simulation model based on this framework is also presented. 1 INTRODUCTION The health care industry is a complex system involving payers, providers, and patients. This system is currently in a state of flux as government and employers try to reduce costs while simultaneously maintaining or improving health outcomes. The proposed changes impact both payers and providers through use of alternative payment models, new provider business organization models, and use of evidence-based treatment plans. Alternatives to the current fee-for-service payment model are now being evaluated, including bundled episode-based approaches, outcome-based approaches based on improved health outcomes, and gain-sharing approaches, where physician driven cost savings are shared between the payer and the provider organization. Due to the complexity of health care delivery, it is not easy to immediately determine the impact of using those different models on the financial and health outcomes of the healthcare system. The primary approach to exploring changes in payment fee structures have been in-field experiments or pilots focused on certain patient segments or specific provider groups. These pilots usually take much time and effort and may not be generalizable across geographical regions, patient groups and providers. In this paper, we describe a general framework for simulating the impact of alternative healthcare incentives. In addition, we discuss opportunities for using client data to enable evidence-based customization of the patient, provider, disease and intervention models used in a simulation. Figure 1 illustrates the major components of our framework. The three primary groups of models are the sub-systems, the decision-makers, and the evidence-based analytics. The sub-system models describe various aspects of the clinical and socio-economic environment, and includes disease models, provider reimbursement mechanisms, and regulations or guidelines for care. The decision-makers include key players such as the patient, provider and payer. The evidence-based analytics provide the statistical, optimization, machine-learning and visualization models for aiding in model building, model calibration and result interpretation. The evidence-based analytics provide the means for customizing a simulation model for a particular population of patients and providers based on observational data available in the form of medical claims and patient health records. One of the challenges associated with integrating several models into a single simulation has to do with integrating models that may have been developed independently using different modeling paradigms, software languages and applications. While it is outside the scope of our work to address this challenge, we note that Cefkin et al. 2010 have developed the Smarter Planet Platform for Analysis and Simulation of 1103 978-1-4577-2109-0/11/$26.00 ©2011 IEEE