Success Stories

Research Accelerates at NIH with One-Stop-Shopping for Data

Client:

National Institutes of Health, Clinical Center

Challenge:

Improve usability of research data stored in isolated databases across NIH’s Institutes and Centers.
Automate and integrate data access.

Solution:

A standards-based biomedical data repository and reporting tool set – BTRIS.
Uses Research Entities Dictionary (RED) that standardizes data.
BTRIS queries provide comprehensive data aggregation from disparate sources.

Results:

BTRIS’s initial release (July 2009) has many committed users, who are completing work in minutes that had required hours.
Old data that had not been analyzed in decades are now available electronically.
Future releases will serve the entire NIH research community of about 5,000.

At the National Institutes of Health, medical research moves a bit faster these days because of a new repository for clinical protocol data called BTRIS – the Biomedical Translational Research Information System.

Research data is stored in many databases throughout NIH, but they are not easily accessible and do not use standard terminology. When NIH researchers need information from more than one source, they can face quite a chore because the data must be standardized before the information can be combined. However, there is so much value in larger data sets that researchers are willing to do the tedious, time-consuming work.

Translation is Key

BTRIS translates data from different sources into a standard structure and language, so researchers can query several sources at once and receive a single, comprehensive, relevant data set. Result: less time spent preparing data for analysis and more time actually analyzing it. And because BTRIS eliminates the manual processes involved in working with non-standard data, researchers are confident that data sets are accurate.

BTRIS also has increased the sheer amount of electronic data available to researchers. Though all data now in BRTIS were stored in some electronic form before being loaded, some of the old storage media were not accessible in any practical way. Now researchers can easily organize data going back a quarter-century.

Finally, BTRIS automates routine administrative reports that used to consume hours of researcher time and are now produced in seconds.

Design and Implementation

CSC collaborated with NIH to design, develop and implement BTRIS. CSC also was responsible for testing the system before it went live as scheduled on July 30, 2009. Since then, we have managed BTRIS operations and helped with system deployment and user training. Our 20-member team brings skills in project management, architecture, healthcare data warehousing, Web services development, medical terminology, data analysis, testing, database administration, systems administration, change control and other key areas.

BTRIS is being implemented incrementally. At launch, the repository contained clinical research data from two NIH Institutes in addition to the Clinical Center. In mid-2010 (the time of this writing), data from a third large Institute is being added, and the team is working on improving usability and expanding functionality. It is anticipated that data from all remaining Institutes and Centers will be integrated in the system by 2013. Another BTRIS expansion is expected to follow, enabling the system to collect information from NIH research partners such as academic medical centers across the United States.

Incremental implementation allows design issues to be identified and addressed in early stages, when changes are easier and less costly to make. It also begins to deliver ROI well before the system is complete.

A Promising Future

The BTRIS solution architecture is flexible enough to accommodate all data types that may be encountered, even those that don’t yet exist, and the system is designed to be easily adapted for interoperability with other biomedical data repositories. BTRIS is well-positioned to become a core NIH research system, with an intramural user population of about 5,000 researchers, who will have access to standardized data from thousands of protocols.