Validation of an iPad Test of Letter Contrast Sensitivity

Abstract

An iPad-based letter contrast sensitivity test was developed (ridgevue.com) consisting of two letters on each page of an iBook. The contrast decreases from 80% (logCS = 0.1) to 0.5% (logCS = 2.3) by 0.1 log units per page. The test was compared to the Pelli-Robson Test and the Freiburg Acuity and Contrast Test. Twenty normally sighted subjects and 20 low-vision subjects were tested monocularly at 1 m using each test wearing their habitual correction. After a 5-minute break, subjects were retested with each test in reverse order. Two different letter charts were used for both the Pelli-Robson and iPad tests, and the order of testing was varied systematically. For the Freiburg test, the target was a variable contrast Landolt C presented at eight possible orientations and used a 30-trial Best PEST procedure. Repeatability and agreement were assessed by determining the 95% limits of agreement (LoA) ± 1.96 SD of the differences between administrations or tests. All three tests showed good repeatability in terms of the 95% LoA: iPad = ± 0.19, Pelli-Robson = ± 0.19, and Freiburg = ± 0.15. The iPad test showed good agreement with the Freiburg test with similar mean (± SD) logCS (iPad = 1.98 ± 0.11, Freiburg = 1.96 ± 0.06) and with narrow 95% LoA (± 0.24), but the Pelli-Robson test gave significantly lower values (1.65 ± 0.04). Low-vision subjects had slightly poorer repeatability (iPad = ± 0.24, Pelli-Robson = ± 0.23, Freiburg = ± 0.21). Agreement between the iPad and Freiburg tests was good (iPad = 1.45 ± 0.40, Freiburg = 1.54 ± 0.37), but the Pelli-Robson test gave significantly lower values (1.30 ± 0.30). The iPad test showed similar repeatability and may be a rapid and convenient alternative to some existing measures. The Pelli-Robson test gave lower values than the other tests.