Accuracy of four commonly used color vision tests in the identification of cone disorders

Alberta A H J Thiadens, Carel B Hoyng, Jan Roelof Polling, Riet Bernaerts-Biskop, L Ingeborgh van den Born, Caroline C W Klaver

Research output: Contribution to journalArticleResearchpeer-review

Abstract

PURPOSE: To determine which color vision test is most appropriate for the identification of cone disorders.

METHODS: In a clinic-based study, four commonly used color vision tests were compared between patients with cone dystrophy (n = 37), controls with normal visual acuity (n = 35), and controls with low vision (n = 39) and legal blindness (n = 11). Mean outcome measures were specificity, sensitivity, positive predictive value and discriminative accuracy of the Ishihara test, Hardy-Rand-Rittler (HRR) test, and the Lanthony and Farnsworth Panel D-15 tests.

RESULTS: In the comparison between cone dystrophy and all controls, sensitivity, specificity and predictive value were highest for the HRR and Ishihara tests. When patients were compared to controls with normal vision, discriminative accuracy was highest for the HRR test (c-statistic for PD-axes 1, for T-axis 0.851). When compared to controls with poor vision, discriminative accuracy was again highest for the HRR test (c-statistic for PD-axes 0.900, for T-axis 0.766), followed by the Lanthony Panel D-15 test (c-statistic for PD-axes 0.880, for T-axis 0.500) and Ishihara test (c-statistic 0.886). Discriminative accuracies of all tests did not further decrease when patients were compared to controls who were legally blind.

CONCLUSIONS: The HRR, Lanthony Panel D-15 and Ishihara all have a high discriminative accuracy to identify cone disorders, but the highest scores were for the HRR test. Poor visual acuity slightly decreased the accuracy of all tests. Our advice is to use the HRR test since this test also allows for evaluation of all three color axes and quantification of color defects.

Original languageEnglish
Pages (from-to)114-21
Number of pages8
JournalOphthalmic Epidemiology
Volume20
Issue number2
DOIs
Publication statusPublished - Apr 2013

Keywords

  • Adult
  • Color Perception Tests/instrumentation
  • Color Vision Defects/diagnosis
  • Electroretinography
  • False Positive Reactions
  • Female
  • Humans
  • Male
  • Predictive Value of Tests
  • ROC Curve
  • Reproducibility of Results
  • Retinal Cone Photoreceptor Cells/pathology
  • Sensitivity and Specificity
  • Visual Acuity

Fingerprint Dive into the research topics of 'Accuracy of four commonly used color vision tests in the identification of cone disorders'. Together they form a unique fingerprint.

Cite this