J Kuchenbecker1. 1. Augenklinik des Helios Klinikum Berlin Buch, Schwanebecker Chaussee 50, 13125, Berlin, Deutschland, joern.kuchenbecker@helios-kliniken.de.
Abstract
BACKGROUND: Color vision tests with pseudoisochromatic plates currently represent the most common procedure for the screening of congenital color vision deficiencies. By means of a web-based color vision test, new and old color plates can be tested for diagnostic quality without major effort. METHODS: A total of 16 digitized Stilling's color plates of the 11th edition from 1907 were included in a web-based color vision test (http://www.farbsehtest.de). The χ(2)-test was used to check whether the Stilling color plates showed similar results to the nine previously evaluated Ishihara color plates. RESULTS: A total of 518 subjects including101 (19.5 %) female subjects with a mean age of 34.6 ± 17 years, took the web-based test with the 25 plates. For all participants the range for the correctly recognized plates was between 5.2 % (n = 27) and 97.7 % (n = 506) for the Stilling color plates and between 64.9 % (n = 336) and 100 % (n = 518) for the Ishihara color plates. For participants with more than 5 errors (n = 247), the range for correctly recognized plates was between 2.0 % (n = 5) and 98.0 % (n = 242) for the Stilling plates and between 42.5 % (n = 105) and 100 % (n = 247) for the Ishihara plates. Taking all color plates and all participants into account there was a significantly higher incidence of erroneous recognition of the Stilling color plates (3038 false and 5250 true answers) compared to the Ishihara color plates (1511 false and 3151 true answers) (p < 0.001, χ(2)-test). CONCLUSION: The diagnostic quality of the tested Stilling color pates was very variable. Some of the plates could be used for the test edition of the Velhagen/Broschmann/Kuchenbecker color plates from 2014. Overall, the Stilling color plates were recognized with a higher incidence of error by all participants in the web-based test compared to the utilized Ishihara color plates, which in most cases was attributable to ambiguity of some symbols.
BACKGROUND: Color vision tests with pseudoisochromatic plates currently represent the most common procedure for the screening of congenital color vision deficiencies. By means of a web-based color vision test, new and old color plates can be tested for diagnostic quality without major effort. METHODS: A total of 16 digitized Stilling's color plates of the 11th edition from 1907 were included in a web-based color vision test (http://www.farbsehtest.de). The χ(2)-test was used to check whether the Stilling color plates showed similar results to the nine previously evaluated Ishihara color plates. RESULTS: A total of 518 subjects including101 (19.5 %) female subjects with a mean age of 34.6 ± 17 years, took the web-based test with the 25 plates. For all participants the range for the correctly recognized plates was between 5.2 % (n = 27) and 97.7 % (n = 506) for the Stilling color plates and between 64.9 % (n = 336) and 100 % (n = 518) for the Ishihara color plates. For participants with more than 5 errors (n = 247), the range for correctly recognized plates was between 2.0 % (n = 5) and 98.0 % (n = 242) for the Stilling plates and between 42.5 % (n = 105) and 100 % (n = 247) for the Ishihara plates. Taking all color plates and all participants into account there was a significantly higher incidence of erroneous recognition of the Stilling color plates (3038 false and 5250 true answers) compared to the Ishihara color plates (1511 false and 3151 true answers) (p < 0.001, χ(2)-test). CONCLUSION: The diagnostic quality of the tested Stilling color pates was very variable. Some of the plates could be used for the test edition of the Velhagen/Broschmann/Kuchenbecker color plates from 2014. Overall, the Stilling color plates were recognized with a higher incidence of error by all participants in the web-based test compared to the utilized Ishihara color plates, which in most cases was attributable to ambiguity of some symbols.