Comparison of Ishihara Booklet with Color Vision Smartphone Applications

PURPOSE: To compare the standard Ishihara booklet with color-vision-testing smartphone applications. METHODS: A prospective observational diagnostic study on 42 normal trichromats and 38 color-deficient subjects. Patients were presented with three color vision tests in random order: an Ishihara test booklet and two color-vision-testing smartphone applications: Eye2Phone and the Color Vision Test application (CVT app). Sensitivity and specificity of the electronic tests was compared with Ishihara results, and in each one of these applications every plate was individually analyzed for success/failure rate. RESULTS: Average age was 42.7 ± 12.9 years. There were 57 males (71.2%). Sensitivity and specificity of each test was 100% (38/38) and 95.2% (40/42) for the Eye2Phone, and 100% (38/38) and 54.8% (23/42) for the CVT app. There was no significant difference between the Ishihara booklet and the Eye2Phone (p = 0.500), with a high kappa measure of agreement (0.950, p < 0.001). The CVT app was significantly different than both other tests (p < 0.001) with a low kappa measure of agreement (0.535 with the Ishihara and 0.575 with the Eye2Phone). Of the 21 tested plates, color-deficient subjects failed 11.8 ± 3.1 plates in the Ishihara booklet and 14.1 ± 2.1 plates in the Eye2Phone (p < 0.001). Significant plate-specific differences for the color-deficient group were found in plate numbers 3, 6, 7, 8, 9, 15, and 16. Both tests were poorly able to give an indication of the specific dichromatism type. The Ishihara booklet was rated more comfortable and clearer than the Eye2Phone in color-deficient subjects. The CVT app was rated lowest for comfort and clarity in both groups. CONCLUSIONS: Smartphone applications testing for color vision deficiency are readily available; however, users of these apps should be aware that some may have different sensitivity for detection of color vision deficiency compared to Ishihara booklet, limiting their usefulness for clinical use. Therefore, further validation of these applications is required.