Measurement of the validity of a preschool vision screening program.

Abstract

The validity (sensitivity and specificity) of a preschool vision screening program was measured over a 3-year period to determine how well strabismus and significant refractive errors could be detected. Public health nurses were trained to administer tests of visual acuity, stereoacuity, and ocular alignment. Failure on any test, visual acuity of 6/9 or less, stereoacuity of less than 100 seconds of arc, or an apparent misalignment of the eyes resulted in referral to an eye care practitioner. An age-matched control was also referred. Analysis of practitioner reports used predefined study-based criteria for ocular abnormalities. More than 1100 children were screened each year. The annually calculated prevalence of vision problems ranged between 10.5% and 13.8%. The estimated sensitivity varied from 60.4% to 70.9% (specificity, 69.6% to 79.9%). The yield indicated that a very high percentage of children with vision problems were identified for the first time. The validity of this screening is comparable to that of other school screenings. The limitations are predictable. Consideration should be given to replacing visual acuity tests with a rapid, objective measure of refractive error and ocular alignment.