Using a Computerised Staircase and Incremental Optotype Sizes to Improve Visual Acuity Assessment Accuracy.

Abstract

Background: Given the impact of visual acuity results on diagnosis and management, it is essential that the test is accurate, determined by factors such as test-retest variability. Standardisation improves accuracy, which can be performed via a computerised staircase methodology. Standard clinical tests with scoring of 0.02 per optotype implies an incremental score per optotype despite optotype size remaining constant on each line. The aim of this study is to establish if near continuous incremental optotype display and scoring improves test-retest variability compared to current testing methods.

Methods: A computerised three up, one down adaptive staircase was used to display Kay Picture optotypes on an LCD monitor. Three methods of visual acuity assessment were undertaken: ETDRS, Kay Pictures and computerised Kay Pictures. Tests were performed twice under standard clinical conditions.

Results: One hundred nineteen adults were tested. Test-retest variability for computerised Kay pictures was 0.01 logMAR (±0.04, p = 0.001). Good levels of agreement were observed for computerised Kay pictures in terms of test-retest variability, where the test had the smallest mean bias (0.01 logMAR compared to 0.03 and 0.08 logMAR for Kay Pictures and ETDRS respectively) and narrowest limits of agreement. Participants performed better in computerised Kay pictures than Kay Pictures by 0.03 logMAR, and better in ETDRS than computerised Kay pictures by 0.1 logMAR.

Conclusion: Computerised Kay pictures exhibited a low test-retest variability, demonstrating it is reliable and repeatable. This repeatability measure is lower than the test-retest variability of the ETDRS and Kay Pictures tests.