Capturing Metadata for the BioSense Surveillance System

Background:

 The Federal Enterprise Architecture (FEA) is a set of reference models to classify and organize federal IT assets as well as augment more detailed agency enterprise architectures. Through its published Data Reference Model (DRM), the FEA program has provided the framework and standards to help enable data discovery and sharing. Using a DRM schema, CDC captures metadata for the BioSense system associated with the processing of HL7 messages from hospitals in the U.S.

 Methods:

 The metadata is organized according to logical, business-related design elements which include data concepts, information class, entities, etc. The captured metadata also include more detailed structural elements such as the physical databases, tables, views and control structures in which data elements are stored. Analysis and management of these metadata will support improvements in data architecture, governance, reuse, and enablement of service-oriented architectures (SOAs).

 Results:

 Within BioSense, a primary near-term goal of this process is the collection and synthesis of end-to-end documentation for data process flow within the separate business components. A longer-term goal is the creation of a metadata repository based on identified requirements for a searchable system to facilitate analysis and sharing of the data’s business context, purpose, quality, and potential suitability for other purposes. This repository can also be used to identify the impact of change (e.g., in transformation rules), identify data and structures no longer used, facilitate architectural improvements, and, ultimately, enable SOA.

 Discussion:

The existing BioSense system holds a large amount of critical data and business logic. CDC can leverage a bottom-up approach to “meet -in-the-middle” with the more typical top-down analysis for enablement of architectural improvements.  This approach allows near-term and long-term opportunities for system improvements including SOA compliance.