ABSTRACT
Objective@#This study aimed to evaluate the validity and predictive value of different methods for myopia screening, and to provide the most effective method of myopia screening.@*Methods@#A total of 7 441 students of grade 1-12 selected from 36 primary and middle schools in 12 counties in Jiangsu Province were enrolled in this study for ophthalmological examination, and uncorrected visual acuity (UCVA, Plan A), noncycloplegic autorefraction (NCAR, Plan B) and combination of UCVA and NCAR (Plan C) were included in the examination. Receiver operating characteristics (ROC) curve was performed to determine the cut-off values (The gold standard is cycloplegic refraction: SE≤0.5 D), and positive predictive value, sensitivities, specificities and other indicators of three screening tests were compared and calculated.@*Results@#The AUC of the combination test for uncorrected visual acuity(UCVA) and noncycloplegic autorefraction (NCAR) was nearly 1(0.91, 95%CI=0.90-0.92), and the sensitivity, specificity, Yoden index, diagnostic coincidence rate, positive predictive value and negative predictive value were 92.4%，89.5%，0.82，91.1%，91.3%，90.8%, respectively, but the index of the younger pupils in primary schools was relatively low. The result of screening plan A for students of middle and high school students was the closest to that of myopia diagnosis.@*Conclusion@#Combination test, which was recommended by China National Health Commission, can be a practical solution for school students for myopic screening. Under the condition of no mydriatic refraction, Plan A could be considered for myopia screening especially for middle and high school students. For the younger pupils, it is suggested to use the internationally recommended myopia screening strategy, that is, SE ≤ 0.50D after mydriasis.