Predictive Modeling & Risk Stratification
The MedHOK Predictive Modeling and Risk Stratification module includes over 600 out-of-the-box algorithms to identify and stratify members. The solution allows business users to self-configure various algorithms combining Johns Hopkins risk stratification tool (ACG), HEDIS care gaps (if purchased), and claim and other data resident in the MedHOK system. Included in the solution is the ACG predictive modeling tool, which has financial, risk, utilization, and care gap algorithm components. Identified/stratified members can be triggered for interventions to the Utilization, Pharmacy, Medication Therapy Management, or Care Management modules. The software automatically routes members to a care management queue based on identified risk. This activity supports opening of active case and disease management cases for moderate to high risk members and fully automated health education and other preventive outreach for low risk members.
MedHOK’s predictive modeling, identification, and stratification solution provides hundreds of financial and utilization statistics as well as predicted values from which to stratify for case and disease management and care gaps. Based on the model output, MedHOK’s system also assigns risk scores and creates a comprehensive care profile for each member, providing a detailed assessment of the member’s morbidity, expected annual costs in different medical service areas, and potential for hospital readmission and other high-cost events. Each time additional encounter data is loaded, the predictive model is refreshed, and the new and underlying data is re-analyzed.