Evaluating the accuracy of picture and letter visual acuity tests for detecting interocular acuity difference

PURPOSE: It is commonly reported that children should proceed from picture visual acuity (VA) tests to letter VA tests as soon as possible due to improved accuracy. However, there is a lack of evidence to support this claim in relation to sensitivity and specificity for detecting an interocular acuity difference (IAD). The aim of the study is to assess VA using two commonly used picture tests and compare the results to a gold- standard letter test. METHODS: Uniocular VA was tested using Crowded Kay Pictures, Crowded Lea Symbols and Early Treatment Diabetic Retinopathy Study (ETDRS) chart. Testing order and which eye tested first were randomised. Inclusion criteria were participants aged 5 years and older, VA better than 1.30logMAR in each eye with a mix of patients with and without IAD of ≥0.2 logMAR. RESULTS: = 51) of participants had an IAD of ≥0.2logMAR on ETDRS. The mean IAD for participants with an IAD of ≥0.2 logMAR for Crowded Kay Pictures was 0.37 logMAR (SD ± 0.21), for Lea Symbols 0.34 logMAR (SD ± 0.20) and for ETDRS 0.36 logMAR (SD ± 0.21). The sensitivity of Crowded Kays and Lea Symbols for detecting an IAD of ≥0.2logMAR was 94% for both. Specificity was 93% for Crowded Kays and 98% for Lea symbols. CONCLUSION: Analysis of this data suggests picture VA tests are comparably accurate to letter VA tests in relation to sensitivity and specificity to detect IAD of ≥0.2 logMAR.