DTRI paper recognized as one of the top five papers at AMIA Summit

May 28, 2014

June 12, 2014

A paper by DTRI and Duke faculty and staff was recognized at the American Medical Informatics Association's (AMIA) Summit on Clinical Research Informatics last month in San Francisco.

The paper, “Use of RxNorm and NDF-RT to normalize and characterize participant-reported medications in an i2b2-based research repository,” was recognized as one of the top five papers at the conference. Colette Blach; Guilherme Del Fiol, MD; Chandel Dundee, RN; Julie Frund; Rachel Richesson, PhD; Michelle M. Smerek; Anita Walden; and Jessica Tenenbaum, PhD, (pictured, right) authored the paper.

Standardized drug terminologies are used for clinical drugs and drug delivery devices in the United States to enable interoperability and clear communication between electronic systems, regardless of software and hardware compatibility. However, medication information in clinical care and research is often stored only using non-standardized identifiers with drug names stored as text.

The paper's authors theorized that two widely used standardized terminologies, RxNorm and the VA National Drug File Reference Terminology (NDF-RT), together with freely available web-based tools, could be used to code patient-provided drug data. If possible, this would maximize the utility of the data while minimizing the costs associated with collecting and analyzing them. In this case, the drug data came from the MURDOCK Study, a longitudinal, large scale epidemiological study in which drug information was reported by study participants as free text.

The researchers obtained 130,273 individual entries. Of these, they were able to accurately map 122,523 (94 percent) to RxNorm concepts, and 106,135 (85 percent) of those concepts to nodes in NDF-RT. This approach, they concluded, could enable the use of drug data in combination with other information for cohort identification. The method could also be generalized to other projects requiring coding of medication data from free text.

“The question we wanted to answer was, could we use publicly available tools to take this free text data and do something useful with it,” said Tenenbaum. “This paper demonstrated that, in fact, we could.”

A list of other Duke faculty and staff whose work was featured at the Joint Summits, along with photographs, is available on the Duke Center for Health Informatics (DCHI) website.