Developing a "Need-Based" Population Management System
February 21, 2017 — 02:30PM - 03:30PM EST
Orange County Convention Center
The search for a population analytics/management application can be harrowing. Many solutions are offered by fledgling companies that may not be viable in the long term. In addition, these products tend to be expensive and hard to implement successfully with extensive and untenable integration requirements. Consequently, Doylestown Health (DH) used existing talent and resources available to develop a population analytics infrastructure that is agile and scalable. This product has been in use for the past year throughout the DH enterprise. The product aggregates multiple data sources against custom risk algorithms to identify at-risk patients across the continuum for intervention. The tool correlates risk score with unwanted utilization an enables the provider to direct actions and interventions toward more efficient and effective care with the objective of reducing readmissions and unwanted utilization. Analysis of risk scoring methodology shows a correlation of increasing risk score with increasing risk of ER visits.
Recognize how to transform existing data systems and quality improvement professionals into a solid population analytics team by transforming skillsets and utilizing current technology
Describe how to use existing integration tools and enterprise infrastructure to build a comprehensive population analytics platform that is agile, scalable, solves specific problems and increases opportunities
Identify how to utilize clinical expertise to "develop the math" and metrics for risk tagging and stratification
Assess data normalization methods that tie in disparate databases for valid interpretation and common meaning