June 2021
Volume 62, Issue 8
Open Access
ARVO Annual Meeting Abstract  |   June 2021
Inconsistencies in Visual Acuity Data in Electronic Health Records
Author Affiliations & Notes
  • Judith E Goldstein
    Ophthalmology, Johns Hopkins Medicine, Baltimore, Maryland, United States
  • Xinxing Guo
    Ophthalmology, Johns Hopkins Medicine, Baltimore, Maryland, United States
  • Michael V Boland
    Massachusetts Eye and Ear Infirmary Department of Ophthalmology, Boston, Massachusetts, United States
  • Footnotes
    Commercial Relationships   Judith Goldstein, None; Xinxing Guo, None; Michael Boland, None
  • Footnotes
    Support  Readers Digest Partners For Sight Foundation
Investigative Ophthalmology & Visual Science June 2021, Vol.62, 3496. doi:
  • Views
  • Share
  • Tools
    • Alerts
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Judith E Goldstein, Xinxing Guo, Michael V Boland; Inconsistencies in Visual Acuity Data in Electronic Health Records. Invest. Ophthalmol. Vis. Sci. 2021;62(8):3496.

      Download citation file:

      © ARVO (1962-2015); The Authors (2016-present)

  • Supplements

Purpose : Challenges exist in harmonizing electronic health record (EHR)-derived data for secondary analysis, yet the problem is rarely studied. We aim to characterize data quality and quantify inconsistencies in EHR-derived visual acuity (VA) data during ophthalmology encounters.

Methods : Retrospective analysis of all VA entries from all eye care encounters across nine clinical locations and 9 subspecialties of the Wilmer Eye Institute between August 1, 2013 and December 31, 2015. Using the pre-defined VA menu options as the standard, we compared the VA entries to the menu options to determine their agreement on an overall and subspecialty level. VA entries were classified info 3 categories: (1) exact match of the VA entry to one of the 24 menu options; (2) partial discordance between the VA entry and the menu options but the entry could be converted to an accepted VA value; (3) total discordance between the VA entry and the menu options and not able to convert it to an accepted VA value.

Results : All VA entries from 513,036 encounters representing 166,212 patients were included. Of the 1,573,643 VA entries, 1,142,738 (72.6%) were an exact match to one of the menu options, 295,943 (18.8%) were partially discordant, and the remaining 134,962 (8.6%) were totally discordant, and classified as missing data. Documented VA entries with providers from comprehensive eye care (86.5%), oculoplastics (81.2%), and pediatrics/strabismus (77.1%) yielded the highest proportions of exact match with the menu options. VA entries during visits with providers from retina (17.5%), glaucoma (14.0%), neuro-ophthalmology (8.8%), and low vision (8.8%) had the highest rates of total discordance / missing VA data.

Conclusions : Inconsistencies exist in VA entries by ophthalmology subspecialty, affecting data quality measures of completeness, correctness and concordance. Transparency regarding data quality may reveal potential biases in VA documentation and should be considered before analyzing data between providers or institutions.

This is a 2021 ARVO Annual Meeting abstract.


This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.