DOCUMENTS

reports

Selection of Medicaid beneficiaries for chronic care management programs: overview and uses of predictive modeling

Published: April 1, 2009
Category: Reports
Authors: Jones WC, Weir S
Country: United States
Language: null
Type: Care Management
Setting: Academic

Shrewsbury, MA: Center for Health Policy and Research.

Center for Health Policy and Research, University of Massachusetts Medical School, Shrewsbury, MA, USA

Effective use of care management techniques may help Medicaid agencies reduce costs and improve health care quality. However, identifying members at highest risk of costly, preventable service utilization remains a challenge. Even choosing from the many software models designed to predict health risk can be difficult.
In order to make it easier for Medicaid programs to select appropriate software, UMass Medical School’s Center for Health Policy and Research evaluated the strengths and weaknesses of three widely-available solutions. Using Medicaid claims data from the State of Vermont, we compared these pre-existing health risk predictive models in terms of their capacity to predict the most costly Medicaid members with chronic conditions. All three models — Chronic Illness and Disability Payment System (CDPS),Diagnostic Cost Groups (DCG), and Adjusted Clinical Groups Predictive Model (ACG-PM) — were designed specifically to analyze Medicaid populations.
For predicting the very highest-cost members (i.e., the 99th percentile), the DCG model is most effective. However, our research showed that the ACG-PM model performed best overall for the Office of Vermont Health Access, the Vermont Medicaid program. Since ACG-PM is free for Medicaid agencies, it presents the most cost-effective solution.

Predictive Risk Modeling,Cost Burden Evaluation,Care Management,High-Impact Chronic Conditions,United States

Please log in/register to access.

Log in/Register

© The Johns Hopkins University, The Johns Hopkins Hospital, and Johns Hopkins Health System.
All rights reserved. Terms of Use Privacy Statement

Back to top