Sixteen Organizations Implement ACG Rx-PM Predictive Modeling from Johns Hopkins University, DST Health...( ACG Rx-PM Software Helps Organizati...),Health

ACG Rx-PM Software Helps Organizations Rapidly Identify High-Risk Cases,

Provide Effective Care Management Services and Improve Underwriting

BIRMINGHAM, Ala., Nov. 6 /PRNewswire/ -- The Johns Hopkins Bloomberg School of Public Health and DST Health Solutions announced that 16 organizations have adopted ACG Rx-PM, pharmacy-based predictive modeling software that helps organizations rapidly identify at-risk patients for care management. These new clients represent the managed care, Medicaid agency, pharmacy benefit administration and consulting sectors. Terms were not disclosed.

Johns Hopkins University developed the ACG (Adjusted Clinical Groups) System's Rx-PM predictive models-based on retail pharmacy claims information -- during 2006. DST Health Solutions has been the exclusive distributor of the Johns Hopkins ACG System since 1992.

"Rx-PM represents a wonderful complement to the ACG System," said Mike Salzillo, chief statistician for Blue Cross & Blue Shield of Rhode Island. "It affords us the opportunity to integrate new data sources into our predictive models for greater statistical performance and more clinical texture. We think that the combination of pharmacy-based predictive models with diagnosis-based predictive models will lead to greater insights into patient quality and population health."

The basis of the system is a new approach for classifying NDC (national drug codes) codes called Rx-MGs (Rx morbidity groups). Rx-MGs group NDC codes by characteristics of underlying patient morbidity. Pharmacy-based predictive models have advantages because pharmacy data is timelier, more complete and has more clinical texture. Rx-PM complements the Johns Hopkins ACG-PM models, which are based on medical claims information. Rx-PM and ACG-PM provide a powerful capability to identify high-risk cases for care management and underwriting.

"Pharmacy is a critical component in understanding predictive modeling," said Jonathan Weine
'/>"/>