Evaluation of a Home-Printable Vision Screening Test for Telemedicine

Importance: Many ophthalmology appointments have been converted to telemedicine assessments. The use of a printed vision chart for ophthalmology telemedicine appointments that can be used by people who are excluded from digital testing has yet to be validated. Objectives: To evaluate the repeatability of visual acuity measured using the Home Acuity Test (HAT) and the agreement between the HAT and the last in-clinic visual acuity. Design, Setting, and Participants: This diagnostic study was conducted from May 11 to 22, 2020, among 50 control participants and 100 adult ophthalmology outpatients who reported subjectively stable vision and were attending routine telemedicine clinics. Bland-Altman analysis of corrected visual acuity measured with the HAT was compared with the last measured in-clinic visual acuity on a conventional Early Treatment Diabetic Retinopathy Study logMAR chart. Main Outcomes and Measures: For control participants, repeatability of the HAT and agreement with standard logMAR visual acuity measurement. For ophthalmology outpatients, agreement with the last recorded in-clinic visual acuity and with the International Classification of Diseases and Related Health Problems, 11th Revision visual impairment category. Results: A total of 50 control participants (33 [66%] women; mean [SD] age, 36.0 [10.8] years) and 100 ophthalmology patients with a wide range of diseases (65 [65%] women; mean [SD] age, 55.3 [22.2] years) were recruited. For control participants, mean (SD) test-retest difference in the HAT line score was -0.012 (0.06) logMAR, with limits of agreement (LOA) between -0.13 and 0.10 logMAR. The mean (SD) difference in visual acuity compared with conventional vision charts was -0.14 (0.14) logMAR (range, -0.4 to 0.18 log MAR) (-7 letters) in controls, with LOA of -0.41 to 0.12 logMAR (-20 to 6 letters). For ophthalmology outpatients, the mean (SD) difference in visual acuity was -0.10 (0.17) logMAR (range, -0.5 to 0.3 logMAR) (1 line on a conventional logMAR sight chart), with the HAT indicating poorer visual acuity than the previous in-clinic test, and LOA of -0.44 to 0.23 logMAR (-22 to 12 letters). There was good agreement in the visual impairment category for ophthalmology outpatients (Cohen κ = 0.77 [95% CI, 0.74-0.81]) and control participants (Cohen κ = 0.88 [95% CI, 0.88-0.88]). Conclusions and Relevance: This study suggests that the HAT can be used to measure visual acuity by telephone for a wide range of ophthalmology outpatients with diverse conditions. Test-retest repeatability is relatively high, and agreement in the visual impairment category is good for this sample, supporting the use of printed charts in this context.