Views 673
Citations 0
Original Investigation
September 2016

Assessment of the Accuracy of Using ICD-9 Codes to Identify Uveitis, Herpes Zoster Ophthalmicus, Scleritis, and Episcleritis

Author Affiliations
  • 1Francis I. Proctor Foundation for Research in Ophthalmology, University of California, San Francisco
  • 2John A. Burns School of Medicine, University of Hawaii at Manoa, Honolulu
  • 3Department of Ophthalmology, Kaiser Permanente Hawaii, Honolulu
  • 4Pacific Vision Institute of Hawaii, Honolulu
  • 5Center for Health Research, Kaiser Permanente Hawaii, Honolulu
  • 6Department of Epidemiology and Biostatistics, University of California, San Francisco
  • 7Department of Ophthalmology, University of California, San Francisco

Copyright 2016 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.

JAMA Ophthalmol. 2016;134(9):1001-1006. doi:10.1001/jamaophthalmol.2016.2166

Importance  With the increased use of data from electronic medical records for research, it is important to validate International Classification of Diseases, Ninth Revision (ICD-9) codes for their respective diagnoses.

Objective  To assess the accuracy of using ICD-9 codes to identify ocular inflammatory diseases.

Design, Setting, and Participants  Retrospective secondary database analysis. The setting was Kaiser Permanente Hawaii, an integrated managed care consortium that serves approximately 15% of the general Hawaiian population. Participants were patients with ICD-9 diagnosis codes that might be associated with a diagnosis of ocular inflammation seen at Kaiser Permanente Hawaii between January 1, 2006, and December 31, 2007. The data collection and analysis took place from January 2011 to August 2015.

Main Outcomes and Measures  The main outcome was the positive predictive value (PPV) of ICD-9 codes for identifying specific types of ocular inflammatory disease. The PPVs were calculated by determining the ratio of the confirmed cases found by medical record review to the total number of cases identified by ICD-9 code.

Results  Of the 873 patients identified by a comprehensive list of ICD-9 codes for ocular inflammatory diseases, 224 cases were confirmed as uveitis after medical record review. Using a set of uveitis-specific codes and eliminating patients with a history of ocular surgery, the overall PPV for uveitis was 61% (95% CI, 56%-66%). The PPVs for individual uveitis codes ranged from 0% to 100%, and 11 uveitis codes had a PPV exceeding 80%. Herpes zoster ophthalmicus and scleritis/episcleritis ICD-9 codes had PPVs of 91% (95% CI, 86%-95%) and 60% (95% CI, 54%-66%), respectively.

Conclusions and Relevance  Our results suggest that using ICD-9 codes alone to capture uveitis and scleritis/episcleritis diagnoses is not sufficient in the Kaiser Permanente Hawaii healthcare system, although there were specific uveitis codes with high PPVs. However, the electronic medical record can reliably be used to identify herpes zoster ophthalmicus cases. Medical record review, as was done in this study, is recommended to elucidate diagnoses for uveitis and scleritis/episcleritis.