Back

Speaker "Eric Fiolkoski" Details Back

 

Topic

Predictive Models for Integrated Care Engagement and Management

Abstract

Value-based reimbursement through integrated models of care requires robust data management and the sophistication of predictive analytics to best determine strategic management of high-risk, high-cost populations. We will discuss machine learning predictive models that identify population members most likely to benefit from integrated models of care and provide methods for engaging them. We will then discuss using claims data to develop predictive machine learning models that strategically direct clinical interventions by quantifying population member risk of hospital admissions and readmissions as well as high cost during future time intervals. Optimization of clinician focus will be emphasized.

Profile

Eric Fiolkoski is the Director of Data Science for DaVita, Inc. where he leads the development and deployment of computationally intensive predictive models for DaVita’s fee-for-outcome, integrated patient care management (VillageHealth and DaVita Health Solutions). Eric is passionate about data science as a renaissance of interdisciplinary scientific exploration that promises unlimited humanitarian, professional, entrepreneurial, and intellectual rewards. Prior to DaVita, Eric developed predictive models for Xcel Energy, led public policy research for the State of Colorado, and was a research fellow at Vanderbilt University. Eric is a native Coloradan who has a B.A. in Philosophy from the University of Kansas, an M.S. in Statistics from South Dakota State University, and is currently completing a PhD in Computational Science and Statistics at South Dakota State University.