Sensitivity and Specificity of Visual Screening Tests in Dezful Schoolchildren, 2004

Early detection can improve the outcome of visual impairment in children, and one method for early detection could be screening of pre-school children with visual acuity tests. The aim of this study was to determine the validity of these tests when they are used on children entering grade school. For this cross-sectional study we drew 39 random clusters, comprising a total of 5721 school children. We then used 2158 student files to extract data on tests of vision performed by school health officers and compared these data with data generated by optometrists. Measurement of uncorrected visual acuity was done with the E Chart by both teachers and optometrists. The sensitivity and specificity of teacher-administered tests were 25.0% and 96.6%, yielding positive and negative predictive values of 13.4% and 98.4%, respectively. Sensitivity and specificity rates did not show any significant difference between male and female populations (P=0.356, P=0.258), but the difference between specificity in urban and rural areas was significant (P<0.001). Screening tests for visual impairment did not attain the desired level of sensitivity or specificity for case detection in school children. More accurate procedures are required to minimize the number of false negative results.