Original article
Diagnostic accuracy of school vision screening
Authors:
Catherine E. Stewart ,
Department of Optometry & Visual Science, City University, CM466, Tail Building, Northampton Square, London, EC1V 0HB, GB
About Catherine E.
PhD BMedSci (Orthoptics)
James M. Gilchrist,
Department of Optometry, University of Bradford, Bradford, GB
About James M.
PhD BSc
Paul V. McGraw,
School of Psychology, University of Nottingham, Nottingham, GB
About Paul V.
PhD BSc
Barry Winn
Hull York Medical School, University of Hull, Hull, GB
About Barry
PhD BSc
Abstract
Aim: Currently, the Department of Health recommends vision screening of all children between the ages of 4 and 5 years. The aim of this study was to evaluate the diagnostic accuracy of entry-level school- aged vision screening in the UK. The performance of vision screening protocols and referral criterion (Snellen chart) has been compared with an alternative screening tool (Crowded logMAR test). Diagnostic accuracy has been modelled in terms of the screening test and referral criterion.
Methods: A total of 494 children (mean age 5.25 ± 0.28 years) were screened, for the presence of visual anomalies, by school nurses using a Snellen visual acuity chart and Crowded logMAR visual acuity test. Referral was made on the basis of Snellen visual acuity measures using standard NHS protocols (visual acuity 6/9 or worse).
Results: Within the population screened, approximately 14% (n = 68) were referred for further investigation. Of the 68 children referred, over two-thirds were found to be visually normal (false-positive rate of 68%). No false-negatives were identified, indicating that the current screening criterion achieves maximum sensitivity. The Crowded logMAR test at the equivalent referral criteria (0.175 logMAR) also achieved maximum sensitivity but with significantly (p = 0.0008) fewer false-positive referrals (60%).
Conclusion: Current school nurse screening criteria produce a high percentage of false-positives leading to low referral accuracy. However, current protocols provide maximum sensitivity (no false-negatives) for detecting visual anomalies. In order to retain maximum screening sensitivity whilst reducing the rate of false-positive referrals, screening programmes should adopt more precise measures of visual acuity in the form of logMAR-based tests.
How to Cite:
Stewart, C.E., Gilchrist, J.M., McGraw, P.V. and Winn, B., 2008. Diagnostic accuracy of school vision screening. British and Irish Orthoptic Journal, 5, pp.47–53. DOI: http://doi.org/10.22599/bioj.230
Published on
01 Jan 2008.
Peer Reviewed
Downloads
