Grainne Scanlon

Significance Clinicians, occupational health personnel, and educationalists need to make an appropriate choice of color vision screening test or tests when screening for color vision deficiency (CVD). Four color vision screening tests were assessed on the same sample of subjects against the anomaloscope as a reference, enabling direct comparison of these tests. Two of the tests are available in revised form, one has received little attention for inherited CVDs. Purpose The objective of this study was to examine three new or revised color vision screening tests, together with the Ishihara, on their (1) sensitivity, (2) specificity, and (3) ability to provide a tentative assessment of severity and of differentiation between protanomaly and deuteranomaly deficiencies. Methods Data from 104 color deficient and 38 color normal subjects were analyzed. The Hardy–Rand–Rittler (4th edition), City University (3rd edition), Ishihara (2005), and Mollon–Reffin tests were evaluated against the Oculus Heidelberg multi‐color anomaloscope. All screening tests were performed before anomaloscopy. Results Sensitivity was comparable for the Ishihara, Hardy–Rand–Rittler, and City University tests (Chi‐square = 3.26, df = 2, p > 0.05), whereas the Mollon–Reffin had best specificity (100% using a threshold value of two). Compared with all other screening tests the Hardy–Rand–Rittler was the best at correctly classifying a deficiency: protanomaly (75%) and deuteranomaly (82%). While the Ishihara was good at detecting deutans (100%), it misclassified 100% of protans as being deutan. Finally, the Hardy–Rand–Rittler was the only screening test to successfully separate mild from medium severity of deficiency. Conclusions Selection of screening test is dependent on the intended outcome of screening. Referral for more definitive CVD assessment is the preferred option.