ASSESSING THE PREDICTIVE ABILITY OF THE TEST-POSITIVE FINDINGS OF AN ELEMENTARY-SCHOOL VISION SCREENING

Purpose. The purpose of this study is to assess the predictive ability of the test-positive findings of an elementary school vision screenin g program, conducted by 2nd-year optometry students, in identifying ch ildren with eye or vision disorders. Methods. A modified version of th e Modified Clinical Technique (MCT) vision screening was administered to elementary children from a below-average socioeconomic neighborhood in Houston, Texas. Comprehensive eye and vision examinations were pro vided to the available children who failed the vision screening. The s creening was administered by groups of 2nd-year optometry students wit h the assistance of a 4th-year optometry student, and supervised by a faculty member licensed to practice optometry. The follow-up examinati ons were provided by supervised 4th-year optometry students in a clini cal setting based at the elementary school. Positive predictive values calculated from the screening and examination findings estimate the p robability that a failure on one car more of the screening tests would identify children with eye or Vision disorders, Results. Sixty-nine p ercent of the test-positive children examined were found to be true po sitives by the criteria developed in a study of vision screening metho ds in Orinda, California from 1954 through 1956. Conclusions. The pred ictive ability of this study's test-positive findings for identifying eye and vision disorders was found to be less than the predictive abil ity of the Orinda Study findings. This reduced predictive ability of t he present study resulted in a larger number of children being overref erred for examinations than had occurred in the Orinda Study. The redu ction in the ability of the test-positive findings of the current stud y's screening program to identify accurately children with eye and vis ion problems is most likely due to the limited experience of the optom etry students conducting the screening program. The inability of this retrospective study to evaluate the accuracy of the test-negatives is a major limitation in assessing the total effectiveness of this vision screening program. Although the present study may indicate some value in optometry students conducting elementary vision screening programs , a prospective study which could assess the predictive ability of bot h test-positive and test-negative findings, as well as determine the s ensitivity and specificity of the screening program, is needed to asse ss more fully the effectiveness of school vision screening programs us ing professional students.