Vanderbilt University Medical Center’s (VUMC) long history of innovation in biomedical informatics has led to the creation of unique and increasingly complex clinical decision support (CDS) over multiple decades. These CDS resources are embedded in the many information systems of the institution at the point of care. As VUMC moves towards implementation of Epic as the new clinical system in the winter of 2017, the CKM team is working with HealthIT to understand, extract, and organize CDS content from the existing clinical systems for reuse within Epic and also for future/additional clinical systems. As a starting point, CKM has extracted and is examining the Vanderbilt Generalizable Rules (VGRs) residing in Horizon Expert Orders, the computerized physician order entry (CPOE) system and the Indicators from StarPanel, the electronic health record (EHR). The evidence-based concepts that unite this embedded CDS content allow CKM to extract and support logic and key ideas independent of the format it currently takes.
The CKM team has used their unique skillset to create the Clinical Support Knowledge Acquisition and Archival Tool (CS-KAAT). CS-KAAT provides organization of and rich metadata for CDS content in a human-readable format, enables the discovery of linkages between related CDS content, and facilitates reuse external to the current systems. Dependencies between files can be visualized to explore directed relationships and build holistic understanding of a concept or functionality. In the hive plot below, files that work together to display a notification that the desired drug is non-formulary are linked - from hidden root files (top), through intermediate files that reroute the user (right) to final display files (left). Additionally, CKM information scientists work within CS-KAAT to provide the latest evidence to support CDS rules.
Hive plot of relationships between files for non-formulary drugs.
DesAutels SJ, Fox ZE, Giuse DA, Williams AM, Kou Q, Weitkamp A, Patel NR, Giuse NB. Using Best Practices to Extract, Organize, and Reuse Embedded Decision Support Content Knowledge Rules from Mature Clinical Systems. AMIA Annual Symposium, Chicago, IL, November 2016.