Robust Information and Data Management Support

Robust Information and Data Management Support

Data PointsOrderset Picture 2Orderset Picture 3

As Vanderbilt University Medical Center’s (VUMC) hub for capturing, organizing, and preserving knowledge integral to institutional operations, the Center for Knowledge Management (CKM) creates, connects, and maintains documentation of enterprise-wide clinical decision support (CDS), ensuring the continuance and validity of local practice through the creation of format-agnostic metadata and application of the best current evidence. As VUMC moves towards implementation of eStar as the new clinical system in the winter of 2017, the Center for Knowledge Management (CKM) team is working with HealthIT to understand, extract, and organize CDS content from the existing clinical systems for reuse within eStar and also for future/additional clinical systems. 

The CKM team has longstanding experience in the assessment, critique, and provision of evidence from the biomedical literature to support clinical practice; development of relational databases; and identification and extraction of linkages between disparate sources of information.  Leveraging this unique expertise, CKM began by extracting the Vanderbilt Generalizable Rules (VGRs) residing in Horizon Expert Orders, the computerized physician order entry (CPOE) system, and Indicators from StarPanel,  VUMC’s legacy electronic health record (EHR).  This CDS data is represented as rich metadata within the CKM-created Clinical Support Knowledge Acquisition and Archival Tool (CS-KAAT).  Externalizing and translating CDS content into human-readable form, organized within a structured schema, enables the discovery of linkages among related CDS content and facilitates reuse external to the CDS native environment.

CKM created this schema to represent critical concepts in logic and data flow, content lifecycle, and relationship to underlying evidence.  The knowledge within these two types of clinical decision support (CDS) is represented in a format-agnostic, descriptive format to simplify analysis by many parties including content developers and clinician-users. People responsible for creation and maintenance of CDS rules, version history, and necessary data sources are preserved. Leveraging CKM’s expertise and experience in evidence provision, CKM information scientists use CS-KAAT to document the relationship of each concept within the CDS rules (medication dosing, drug-drug/drug-genome interaction checking, homeostatic ranges, etc.) to the evidence. The evidence-based concepts that unite this embedded CDS content allow CKM to discover and relate key ideas independent of the executable format it currently takes.  Within CS-KAAT, these evidence-based concepts consist of a synthesis of relevant published literature, select filtered references and expertly-formulated search statements enabling dynamic updates on the topic.

Building upon the existing schema created for the VGRs and Indicators and the expertise developed through years of providing evidence to support VUMC order set creation, CKM has next prioritized another part of  the institution’s CDS portfolio - inpatient and newly documented outpatient order sets (called SmartSets in eStar) - for capture and preservation within CS-KAAT.   Likewise, these externalized records are presented in a human-readable format and explicitly linked to evidence-based concepts. Their residence alongside other extracted content enables sharing of evidence and other linkages among various CDS types and allows for a broader visualization of the practices surrounding specific orderables, workflows, and patient types throughout the clinical enterprise.

Publications

  • DesAutels SJ. A Knowledge Management Effort to Evidentiate Laboratory Testing, Document Local Practice and Build a Shared Evidence Infrastructure. Department of Biomedical Informatics Seminar Series. October 4, 2017.

  • Fox, ZE, Frakes ET, Su J, Blasingame MN, Epelbaum M, DesAutels J, Kou Q, Williams AM, Giuse NB. Applying Knowledge Management Best-Practices to Capture, Support, and Archive Systems-Embedded Clinical Decision Support Evidence. Medical Library Association Annual Meeting, Seattle, WA, May 2017.

    Related Projects

    Resources and Services

    Collaborators

    • Health IT Leadership
    • Dario Giuse, Dr. Ing., Associate Professor, Biomedical Informatics
    • William W. Stead, MD, Chief Strategy Information Officer; Professor, Biomedical Informatics
    • Health IT Knowledge Engineering
    • VUMC Order Set and SmartSet development teams