Validation of a Standardized Home Visual Acuity Test for Teleophthalmology

PURPOSE: The recent exponential growth in teleophthalmology has been limited in part by the lack of a validated method to measure visual acuity (VA) remotely. We investigated the validity of a self-administered Early Treatment Diabetic Retinopathy Study (ETDRS) home VA test. We hypothesized that a home VA test with a printout ETDRS chart is equivalent to a standard technician-administered VA test in clinic. DESIGN: Prospective cohort study. PARTICIPANTS: Two hundred nine eyes from 108 patients who had a scheduled in-person outpatient ophthalmology clinic visit at an academic medical center. METHODS: Enrolled patients were sent a .pdf document consisting of instructions and a printout ETDRS vision chart calibrated for 5 feet. Patients completed the VA test at home before the in-person appointment, where their VA was measured by an ophthalmic technician using a standard ETDRS chart. Survey questions about the ease of testing and barriers to completion were administered. For the bioequivalence test with a 5% nominal level, the 2 1-sided tests procedure was used, and an equivalent 90% confidence interval (CI) was constructed and compared with the prespecified 7-letter equivalence margin. MAIN OUTCOME MEASURES: The primary outcome was the mean adjusted letter score difference between the home and clinic tests. Secondary outcomes included the unadjusted letter difference, absolute letter difference, and survey question responses. RESULTS: The mean adjusted VA letter score difference was 4.1 letters (90% CI, 3.2–4.9 letters), well within the 7-letter equivalence margin. Average unadjusted VA scores in clinic were 3.9 letters (90% CI, 3.1–4.7 letters) more than scores at home. The absolute difference was 5.2 letters (90% CI, 4.6–5.9 letters). Ninety-eight percent of patients agreed that the home test was easy to perform. CONCLUSIONS: An ETDRS VA test self-administered at home following a standardized protocol was equivalent to a standard technician-administered VA test in clinic in the examined population.