Examine a framework for integrating social determinant of health data into population health analytics, consider application of time-dependent survival modeling in a study to predict survival of patients in hospice, and explore a case study that addresses how much data is enough to build an accurate deep learning model.

This session explores application of risk adjustment and predictive modeling through brief case studies involving key topics; examines the potential of enhanced models to identify patients with rising risk; and considers the impact and implications of analyzing prescription data to determine future patient costs and serve as predicators regarding opioid abuse patients.

Sessions will include: Population Health Management: Innovations in Risk Adjustment and Predictive Modeling; Risk Adjustment and Shared Savings Agreements; and Connecting Predictive Modeling and End-Users: the Last Mile Problem.