Improving the quality and cost-effectiveness of publicly financed health care
Current Applications of Predictive Modeling in Medicaid Managed Care
OVERVIEW
Medicaid managed care plans are at a disadvantage when enrolling new members. Without information about their medical history, the medical needs for these enrollees may not be identified for some time, potentially leading to adverse clinical outcomes.
One tool to help plans prioritize and anticipate the future medical needs of their members is predictive modeling. This is generally defined as, “A process of analyzing currently available data to prospectively identify specific individuals who are at high-risk of having adverse outcomes in the near future (Eisenberg).” It is a statistical representation based on a member's clinical history as compared to population-based clinical outcomes, defining the probability of future adverse events for the individual.
During this April 5, 2005 CHCS webinar, three expert organizations presented national, state, and health plan experiences using predictive modeling. Highlights from the call include an overview given by presenters from Mercer Human Resources Consulting, followed by guest experts from Wisconsin's Bureau of Managed Health Care Programs and Priority Partners Health Plan, who reviewed how predictive modeling is employed by their organizations.
Steve Johnson, PhD, Mercer Government Human Services Consulting - Presentation (PDF)
Michael Fox, Bureau of Managed Care Programs, WI Division of Health Care Financing - Presentation (PDF)
Linda Dunbar, Priority Partners Health Plan (Maryland) - Presentation (PDF)