ACR Meeting Abstracts

ACR Meeting Abstracts

  • Meetings
    • ACR Convergence 2024
    • ACR Convergence 2023
    • 2023 ACR/ARP PRSYM
    • ACR Convergence 2022
    • ACR Convergence 2021
    • ACR Convergence 2020
    • 2020 ACR/ARP PRSYM
    • 2019 ACR/ARP Annual Meeting
    • 2018-2009 Meetings
    • Download Abstracts
  • Keyword Index
  • Advanced Search
  • Your Favorites
    • Favorites
    • Login
    • View and print all favorites
    • Clear all your favorites
  • ACR Meetings

Abstract Number: 2893

The Use of Natural Language Processing to Identify, Retrieve, Report, and Correct Observational Data on US Veterans Enrolled in the Veterans Affairs Rheumatoid Arthritis Registry

Grant Cannon1, Jorge Rojas1, Neill Bell2, Andreas Reimold3, Ted R. Mikuls4, Namrata Singh5, Gail S. Kerr6, Pascale Schwab7, Jennifer Barton7, Liron Caplan8, Joshua Baker9, Angelo L. Gaffo10, J. Steuart Richards11, Deana Lazaro12, Vikas Majithia13 and Brian Sauer1, 1Salt Lake City VA Medical Center and University of Utah, Salt Lake City, UT, 2Salt Lake City VA Medical Center and University of Utah, Dallas VA Medical Center, Dallas, TX, 3Dallas VA Medical Center and University of Texas Southwestern, Dallas, TX, 4VA Nebraska-Western Iowa Health Care System and University of Nebraska Medical Center, Omaha, NE, 5VA Nebraska-Western Iowa Health Care System and University of Iowa, Omaha, IA, 6Washington DC VA Medical Center, Georgetown and Howard University Hospitals, Washington, DC, 7Portland VA Medical Center and University of Oregon, Portland, OR, 8Denver VA Medical Center and Univerity of Colorado, Denver, CO, 9Philadelphia VA Medical Center and University of Pennsylvania, Philadelphia, PR, 10Birmingham VA Medical Center and University of Alabama at Birmingham, Birmingham, AL, 11Pittsburgh VA Medical Center and University Of Pittsburgh, Pittsburgh, PA, 12Brooklyn VA Medical Center and SUNY Downstate, Brooklyn, NY, 13Jackson VA Medical Center and University of Mississippi, Jackson, MS

Meeting: 2017 ACR/ARHP Annual Meeting

Date of first publication: September 18, 2017

Keywords: Disease Activity, Epidemiologic methods, observation, outcome measures and rheumatoid arthritis (RA)

  • Tweet
  • Click to email a link to a friend (Opens in new window) Email
  • Click to print (Opens in new window) Print
Session Information

Date: Wednesday, November 8, 2017

Title: Health Services Research II: Methods and Technology in Care and Research

Session Type: ACR Concurrent Abstract Session

Session Time: 9:00AM-10:30AM

Background/Purpose:   The Veterans Affairs (VA) Rheumatoid Arthritis (RA) (VARA) registry is an observational cohort study at 12 VA medical centers that prospectively collects clinical and laboratory outcome measures, which are recorded in the VARA database.  To replace manual entry and reduce missing data, a Natural Language Processing (NLP) system was developed to extract outcome measures recorded in standardized  text templates embedded in the electronic health record (EHR) notes for US Veterans enrolled in the registry.  This study compared pre- and post-implementation experiences with this system. 

Methods:   VARA database entries for follow-up observations between January 1, 2016 and April 30, 2016 (pre-implementation) were compared to similar entries after NLP implementation January 1, 2017 to April 30, 2017 (post-implementation) with the number of notes containing outcome measures reported.  Laboratory measure were automatically collected from EHR reports.  At six VARA sites, missing data reports were provided to VARA investigators and clinic notes were reviewed to determine if outcome measures could be retrieved from documentation outside the standardized template. Note addendums were then entered to provide additional data when available.

Results:   In comparison to 640 notes on 540 patients from 8 VARA sites in the pre-implementation period, 798 notes on 671 unique patients were recorded in the VARA database from 11 VARA sites during the post-implementation period.  This represents an increase of 24.7%, 24.3%, and 37.5% in the number of notes, unique patients captured, and sites engaged, respectively.  This increase in data capture was much larger than the 8.3% increase in VARA patient enrollment during the same period.  The successful capture of specific outcome measures is listed in the Table, which shows an absolute increase in all elements between observation periods.   In a pilot effort to do data corrections at six VARA test sites, missing data on 31 notes were evaluated to see if corrections were possible.  Investigators were able to retrieve additional information for 10 (32%) notes from data outside the standard template and use note addendums to report these data.  An automated NLP extraction was then employed to add this additional information to the database.  Permanently missing data were universally due to the failure to document information at the point-of-care, as opposed to failure of the algorithm. 

Conclusion: This study demonstrates that NLP can be leveraged in conjunction with standardized EHR templates to successfully retrieve patient data from across a national health care system, eliminate the need for manual data entry, and substantially increase the rate of data capture.  By facilitating the collection of clinical and laboratory measures in RA patients, these efforts will further enhance the feasibility of conducting epidemiologic and outcomes studies of RA.


Disclosure: G. Cannon, None; J. Rojas, None; N. Bell, None; A. Reimold, None; T. R. Mikuls, None; N. Singh, None; G. S. Kerr, None; P. Schwab, None; J. Barton, None; L. Caplan, None; J. Baker, None; A. L. Gaffo, None; J. S. Richards, None; D. Lazaro, None; V. Majithia, None; B. Sauer, None.

To cite this abstract in AMA style:

Cannon G, Rojas J, Bell N, Reimold A, Mikuls TR, Singh N, Kerr GS, Schwab P, Barton J, Caplan L, Baker J, Gaffo AL, Richards JS, Lazaro D, Majithia V, Sauer B. The Use of Natural Language Processing to Identify, Retrieve, Report, and Correct Observational Data on US Veterans Enrolled in the Veterans Affairs Rheumatoid Arthritis Registry [abstract]. Arthritis Rheumatol. 2017; 69 (suppl 10). https://acrabstracts.org/abstract/the-use-of-natural-language-processing-to-identify-retrieve-report-and-correct-observational-data-on-us-veterans-enrolled-in-the-veterans-affairs-rheumatoid-arthritis-registry/. Accessed .
  • Tweet
  • Click to email a link to a friend (Opens in new window) Email
  • Click to print (Opens in new window) Print

« Back to 2017 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/the-use-of-natural-language-processing-to-identify-retrieve-report-and-correct-observational-data-on-us-veterans-enrolled-in-the-veterans-affairs-rheumatoid-arthritis-registry/

Advanced Search

Your Favorites

You can save and print a list of your favorite abstracts during your browser session by clicking the “Favorite” button at the bottom of any abstract. View your favorites »

All abstracts accepted to ACR Convergence are under media embargo once the ACR has notified presenters of their abstract’s acceptance. They may be presented at other meetings or published as manuscripts after this time but should not be discussed in non-scholarly venues or outlets. The following embargo policies are strictly enforced by the ACR.

Accepted abstracts are made available to the public online in advance of the meeting and are published in a special online supplement of our scientific journal, Arthritis & Rheumatology. Information contained in those abstracts may not be released until the abstracts appear online. In an exception to the media embargo, academic institutions, private organizations, and companies with products whose value may be influenced by information contained in an abstract may issue a press release to coincide with the availability of an ACR abstract on the ACR website. However, the ACR continues to require that information that goes beyond that contained in the abstract (e.g., discussion of the abstract done as part of editorial news coverage) is under media embargo until 10:00 AM ET on November 14, 2024. Journalists with access to embargoed information cannot release articles or editorial news coverage before this time. Editorial news coverage is considered original articles/videos developed by employed journalists to report facts, commentary, and subject matter expert quotes in a narrative form using a variety of sources (e.g., research, announcements, press releases, events, etc.).

Violation of this policy may result in the abstract being withdrawn from the meeting and other measures deemed appropriate. Authors are responsible for notifying colleagues, institutions, communications firms, and all other stakeholders related to the development or promotion of the abstract about this policy. If you have questions about the ACR abstract embargo policy, please contact ACR abstracts staff at [email protected].

Wiley

  • Online Journal
  • Privacy Policy
  • Permissions Policies
  • Cookie Preferences

© Copyright 2025 American College of Rheumatology