Do the Scores from ETDRS and SLVA Charts Agree? An Analysis Based on the Acuity Psychometric Function

Abstract Objectives. To determine whether scores from the two most widely-used Early Treatment of Diabetic Retinopathy Study (ETDRS) and Standard for Logarithmic Visual Acuity (SLVA) acuity charts are comparable. Methods. Ten normal adults (24.9±1.9 years; 5 females) with normal (0.0 logMAR) or corrected-to-normal visual acuity (VA) participated in this study. The acuity psychometric function, the probability correct response in identifying the orientation of a tumbling E as a function of letter size, was measured using the Psi method. Each participant finished the test under 12 viewing conditions: monocularly (with the left or right eye) and binocularly with refractive errors corrected or uncorrected, and the dominant eye with 6 levels of Bangerter Occlusion Foils (i.e., 0.1, 0.2, 0.3, 0.4, 0.6, 0.8). Simulated tests with the ETDRS and SLVA charts were repeated 1000 times based on the fitted acuity psychometric functions.Results. The VA scores and probability correct from the ETDRS and SLVA charts were slightly but significantly different (mean difference of 0.018 logMAR and 5%, both P < 0.001). The agreement of the VA scores was excellent (0.997 [95% CI, 0.982-0.999]), while the probability correct was fair (0.453 [95% CI, 0.183-0.636]) between these two charts. The precision of the ETDRS chart (0.04 ± 0.00 logMAR) was significantly better (P < 0.001) than that of the SLVA chart (0.07 ± 0.00 logMAR).Conclusions. In clinical, even with different design principles and scoring rules, the ETDRS and SLVA charts produce comparable VA scores for normal adults with no pathological eye disease.