Improving Data Access and Analysis through Implementation of SAP HANA (Carolina Data Warehouse for Health 3.0)

The CDW-H 3.0 project will introduce a virtual access layer into the data integration environment. ISD is currently collaborating with key partners and business stakeholders to complete this project. The CDWH3.0 project team selected the SAP HANA platform to support UNCHCS’s move to a virtual access layer, and is on track for an initial go-live targeted for October 2019.

UNC’s Carolina Data Warehouse for Health (CDW-H) is home to over 6 TB of data that can be leveraged to answer important questions about how to provide better patient care, reduce costs, and improve the overall quality of the healthcare we deliver.  As the ways we gather, store, and use that data have changed, the structure of our CDW-H needs to evolve to allow us to quickly access needed information and turn it into actionable insights that provide real-time business guidance.

The CDW-H 3.0 project will introduce a virtual access layer into the data integration environment. ISD is currently collaborating with key partners and business stakeholders to complete this project.  The CDWH3.0 project team selected the SAP HANA platform to support UNCHCS’s move to a virtual access layer, and is on track for an initial go-live targeted for October 2019 (Editors Note: This date was initially incorrectly listed as September 2019). The CDWH3.0 Steering committee is represented by ISD Analytics, Data Governance, NC TraCS (Translational and Clinical Sciences), Hospital Quality & Innovation, and EADS.

Implementation of a virtual access layer has other added benefits.  It aligns with Epic’s reporting strategy, and allows our ISD team to focus efforts on development needs that aren’t captured elsewhere.  It also supports our data governance strategy, ensuring that we treat our data as an asset and provide trust and reliability to all users of our data.  End users will be able to abstract data from multiple sources and bring it together in an integrated, friendly view. With quicker, simpler access to critical data, users will be better able to use advanced business intelligence and analytic tools to answer important healthcare delivery questions. 

Implementation of the virtual access layer will reduce the requirement to Extract, Transform, and Load (ETL) new sources or different data elements into our data warehouse, since HANA can virtually connect to source system for input into data views.  It will also serve as a tool that applies appropriate security and access levels and controls runtimes protecting our hardware resources.  As a result, advanced users with demonstrated skills/abilities may be able to directly access the database.  The CDW-H Universe currently available in Business Objects will still be available, but the background data will be sourced from HANA after completion of the project.  The goal is to implement this platform without requiring major updates to user’s existing content pointed to the CDW-H Universe.

Furthermore, this process will help reduce confusion about the correct “source” of data and where the data lives.  Reduced redundancy / data replication across multiple systems will free up system time, reduce development efforts, and decrease data errors.  Data becomes more governed, so that the same data used in Epic reporting (SlicerDicer, for example) is available in the data warehouse. 

Achievement of the ROI of this investment will require governance that ensures we develop a reusable, well understood “data lake”, rather than a “data swamp.”  Greater access at the end user level necessitates a higher level of compliance with processes established to ensure analysts have access to trusted sources of data.

All these improvements combine to ensure UNCHCS has a solid data foundation that provides access to reliable, traceable and authentic data assets that support our need for real-time data analysis and reporting. 

Filed under: ,