A Bayesian Hierarchical Joint Model to Determine the Bundled Payment Cost for Efficient and Effective Knee Replacement Surgery

• Activity Number: 507
• Type: Contributed
• Date/Time: Wednesday, August 1, 2012 : 10:30 AM to 12:20 PM
• Sponsor: Section on Statistical Consulting
• Author(s): Sherry Lin*+
• Keywords: Bundled Payments ; Surgery ; Joint Bayesian Modeling ; MCMC

Abstract

The prevailing fee-for-service payment method in the U.S. raises health care costs by rewarding providers that perform more treatments, rather than high quality of care. A bundled payment plan has been proposed to compensate providers for the expected treatment costs, plus an extra warranty amount in case an unforeseen complication occurs. With a risk adjusted, fixed bundled payment from each patient, providers are financially motivated to be cost efficient and effective. Current methods to determine the bundled payment amount have relied on large hospital discharge databases and multiple statistical models to define adverse complications, identify inefficient and ineffective hospitals, and separate routine costs from costs incurred during adverse outcomes care. We propose a Bayesian hierarchical joint model to address all these issues and apply it to determine the bundled payment cost for knee replacement surgery.