Verification of a nationwide population-based myopia growth chart in a large longitudinal cohort of 1155 Korean children.

OBJECTIVE:To evaluate predictive value of the myopia growth chart based on population-based health survey data using longitudinal cohorts. METHODS:Patients aged from 5 to 18 years and underwent two or more cycloplegic refraction (CR) exams with at least one year of interval were included. Percentile deviation was calculated by subtracting percentile at final exam from the percentile at initial exam based on the chart. Spherical equivalent (SE) deviation was calculated by SE at final CR subtracted from predicted SE based on initial CR using the chart. RESULTS:2310 eyes from 1155 subjects were included. There were 1344 eyes (58.2%) categorized as inliers, where both initial and final CR were within the 2nd to 99th percentile. Mean percentile and SE deviations were + 11.0 ± 22.9 percentiles and -0.60 ± 1.33 diopters, each. Outliers, those except the inliers, were 966 eyes (41.8%). Most outliers (709 eyes, 73.4%) were outside the chart for both initial and final exam. The rest of the outliers (257 eyes, 26.2%) were within the 2 to 99 percentile range on the chart at least once, either at initial or final exams; most of those (202 eyes, 78.6%) progressed toward myopia more than predicted. CONCLUSIONS:In our large cohorts, both inliers and outliers tended to progress toward more myopia than predicted from the chart. This suggests the chart predicts childhood myopia rather conservatively. The myopia growth chart may be useful as a screening tool in detecting children at high risk of developing high myopia.