USING BIG DATA TO IMPROVE POPULATION HEALTH
The U.S. health care system is focused on treating expensive clinical events after they occur as oppose to negating these events via population health management. With the recent explosion in the volume and types of data collected in medicine, popularly called “Big Data,” there is the opportunity for a paradigm shift to a health care system that is proactive in targeting disease and anticipates adverse clinical events before they occur. Among patients with chronic conditions, approximately 1 in 5 is readmitted to the hospital within 30 days after discharge. Underserved and minority populations often have even higher rates of readmission. Many of these readmissions could be prevented if higher risk patients were identified and effective interventions then targeted towards these individuals. This session includes three papers using big data to improve population health, especially for chronically ill patients. The first paper focuses on the role of risk adjustment in maximizing welfare. The second paper uses big data and machine learning methodologies to improve on readmission predictions. The third paper evaluates the performance of an intervention triggered by on an insurance-based big-data algorithm.