Efficacy of Forced-Choice versus ETDRS visual acuity testing

Abstract

Purpose : Electronic displays of randomized Snellen acuity charts are common; however, the accuracy of a forced-choice visual acuity (FCVA) test has not been explored. The purpose of this study is to assess the test-retest reliability (TRR), criterion validity (CV), and discriminant validity (DV) of the Early Treatment Diabetic Retinopathy (ETDRS) and FCVA test in a prospective comparative clinical study. We predict that there will be no difference between the logarithm of the minimal angle of resolution (logMAR) results of both exams.

Methods : Data of the right eye (69 healthy, 59 disease) from subjects 18-85 years, VA better than counting fingers, and no central vision loss were included. VA testing was done at 4 meters and repeated at least 1 hour later. Subjects were separated by primary ophthalmic diagnosis. Results were recorded in logMAR. All subjects ETDRS and FCVA results and TRR were analyzed with paired t-test, DV by independent t-test, and CV by Pearson correlation. All subject ETDRS and FCVA results were further analyzed by Bland-Altman (B-A) analysis.

Results : A paired sample t-test of all subjects between ETDRS and FCVA showed significant differences at the first visit (95% CI -.07 to -.02, p < .01), B-A analysis showed that 43% of subjects had at least 0.1 difference in logMAR between testing modalities. CV analysis demonstrated r=0.836 with p value less than .01. TRR analysis showed insignificant differences at two different time points. DV analysis showed significant differences amongst disease and healthy subjects within each test.

Conclusions : ETDRS and FCVA exams showed high TRR, but a significant difference was detected between ETDRS and FCVA. Though FCVA may solve issues such as crowding and memorization, the significant difference between logMAR results suggests that FCVA cannot replace ETDRS at this time.

This is a 2021 ARVO Annual Meeting abstract.