Screening efficacy of a simplified logMAR chart

Abstract

Background: Snellen acuity charts are the most commonly used method for visual acuity (VA) testing in screening programmes despite comparative studies verifying that the logarithm of minimum angle of resolution (logMAR) acuity measurement charts are more accurate than the Snellen chart acuity measurements. LogMAR acuity measurement charts however are not well implemented in routine clinical practice because of the increased testing time and the complexity of scoring. To implement the logMAR method in a screening programme, there has to be some simplification of it.Aim: This study evaluates the efficacy of a simplified logMAR chart, designed for VA testing over the conventional Snellen chart, in a school-based vision-screening programme.Methods: We designed a simplified logMAR chart by employing the principles of the Early Treatment Diabetic Retinopathy Study (ETDRS) chart in terms of logarithmic letter size progression, inter-letter spacing, and inter-line spacing. Once the simplified logMAR chart was validated by students in the Elite school vision-screening programme, we set out to test the chart in 88 primary and middle schools in the Tiruporur block of Kancheepuram district in Tamil Nadu. One school teacher in each school was trained to screen a cross-sectional population of 10 354 primary and secondary school children (girls: 5488; boys: 4866) for VA deficits using a new, simplified logMAR algorithm. An experienced paediatric optometrist was recruited to validate the screening methods and technique used by the teachers to collect the data. Results: The optometrist screened a subset of 1300 school children from the total sample. The optometrist provided the professional insights needed to validate the clinical efficacy of the simplified logMAR algorithm and verified the reliability of the data collected by the teachers. The mean age of children sampled for validation was 8.6 years (range: 9–14 years). The sensitivity and the specificity of the simplified logMAR chart when compared to the standard logMAR chart were found to be 95% and 98%, respectively. Kappa value was 0.97. Sensitivity of the teachers’ screening was 66.63% (95% confidence interval [CI]: 52.73–77.02) and the specificity was 98.33% (95% CI: 97.49–98.95). Testing of VA was done under substandard illumination levels in 87% of the population. A total of 10 354 children were screened, 425 of whom were found to have some form of visual and/or ocular defect that was identified by the teacher or optometrist.Conclusion: The simplified logMAR testing algorithm proved to be less time consuming than the standard logMAR test. This suggests that the simplified logMAR chart is effective in vision-screening programmes and would be a reliable alternative to the standard logMAR chart and therefore replace the use of Snellen chart acuity tests in vision-screening programmes. The study also showed that non-healthcare providers, such as teachers, can reliably administer the simplified logMAR test.Keywords: vision screening; school children; vision screening standards; screening reliability