At What Level Should Pass/ Fail Criteria Be Set?
Independent of the screening method chosen, the pass/fail criteria established are essential for the sensitivity and specificity of the test. A low (i.e., 0.5 instead of 0.8) pass/fail threshold will lower the number of over-referrals, but it will also increase the number of under-referrals and vice versa. Both under- and over-referrals are considered disadvantages in a screening system. Under-referrals give a false impression of visual and ocular health, while over-referrals lead to an unnecessarily large number of subjects referred for more specialized examinations.
In a recent policy statement by the American Academy of Pediatrics, visual acuity of less than 20/40 in either eye, or a two-line interocular difference irrespective of visual acuity, at age 3-5 years is suggested for referral criteria . For children 6 years and older, visual acuity less than 20/30 in either eye, or a two-line interocular difference irrespective of visual acuity, is suggested. For both age groups, any abnormalities of ocular alignment or ocular media also constitute reasons for referral.
In Scandinavia, the visual acuity criteria have traditionally been stricter. In Sweden the referral criteria for 4-year-olds has been less than 0.8 (20/25) in either eye, or two lines of interocular difference. For 5.5-year-olds, the criteria has been less than 1.0 (20/20) in either eye, or two lines interocular difference . Due to large numbers of over-referrals, a project was carried out in the Goteborg region with less strict referral criteria . Children with visual acuity (0.65 in each eye or 0.65 in one eye and 0.8 in the other) where re-tested at age 5.5 years and then referred if visual acuity is less than 0.8 in either eye. The project showed that few children with slightly reduced visual acuity at age 4 years had conditions needing specialized ophthalmologic care. For those requiring treatment, outcome was good.
In a study on randomized treatment of unilateral visual impairment detected at preschool vision screening in 3- to 5-year-old children, Clarke et al.  found no difference in outcome for children with initial visual acuity 0.5-0.67, when comparing subjects who received treatment to those who did not receive any treatment. They argue that "... children with 6/9 (approximately 0.65) in one eye no longer constitute screen failures and do not justify treatment, even with glasses."
Summary for the Clinician
• The ability of a test to correctly identify affected subjects is termed sensitivity.
• The ability of a test to correctly identify healthy subjects is termed specificity.
• The cost and accessibility of vision screening depends on the profession conducting the test. By using tests that can be administered by personnel with only a minimum of ophthalmologic training, costs for screening can be kept low.
• Choosing appropriate pass/fail criteria is crucial for a screening system to be efficient. Referral criteria for visual acuity at preschool screening differ between countries. Recent studies suggest that 3- to 5-year-old children with moderately reduced visual acuity (0.65) should not be considered screening failures due to lack of effect on outcome compared to controls.
Was this article helpful?