BACKGROUND: Vision screenings of a school-based program were conducted in state-mandated grades (pre-kindergarten [pre-K] or kindergarten [K], 1st and 8th grade), and nonmandated grades (2nd to 7th). METHODS: During school years 2016-19, 51,593 pre-K to 8th grade students from 123 Baltimore City Public Schools underwent vision screenings, with 85% of the schools qualifying for Free and Reduced Price Meals. Assessments included distance visual acuity, Spot photoscreening, stereopsis, and cover testing. Screening failures were analyzed by grade using aggregate data. Failure rates for mandated and nonmandated grades were compared using a logistic regression model, and visual acuity distributions were analyzed using individual data. RESULTS: Over the 3-year period, 17,414 (34%) of students failed vision screening. Failure rates by grade ranged from 28% to 38%. Children in kindergarten and 3rd grade and higher were statistically more likely to fail screening than those in 1st grade. Reduced visual acuity was the most common reason for failure (91%). Failure rates were significantly higher in nonmandated grades than in state-mandated testing grades (34.7% vs 32.5% [P < 0.001]). Mean visual acuity of all students who failed vision screening was 20/50 in the worse-seeing eye and was 20/40 in the better-seeing eye. CONCLUSIONS: One-third of students failed vision screening. High screening failure rates across all grades suggest that screening in select grade levels, as currently mandated in Maryland schools, is inadequate for detecting vision problems in the low-income communities served by this program.