Purchase this article with an account.
E.M. Harvey, V. Dobson, J.M. Miller, C.E. Clifford; Effectiveness of Recommended Visual Acuity Screening Referral Criteria in a Population With a High Prevalence of Astigmatism . Invest. Ophthalmol. Vis. Sci. 2006;47(13):691.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
To determine optimal visual acuity screening referral criteria for kindergarten and grade–school children who are members of a population with a high prevalence of astigmatism.
Monocular visual acuity (VA) screening was conducted using ETDRS logMAR letter charts on 1003 K–2 and 4–6 grade children attending school on the Tohono O’odham Reservations. Following acuity screening, each child received a cycloplegic eye examination, and a best estimate of refractive error was determined by cycloplegic autorefraction (Nikon Retinomax K+), confirmed by retinoscopy.
High refractive error (hyperopia > 2.50 D or myopia > 0.75 D in either meridian, astigmatism > 1.00 D, or anisometropia > 1.50 D spherical equivalent) or ocular abnormalities (exam "positive") were present in 45% percent of children, with 89% of positive exam findings meeting the criteria due to high astigmatism (> 1.00 D RE or LE). Area under the ROC curve for visual acuity screening was 0.85, with little variability across grade (ranging from 0.83 to 0.90). AAO, AAPOS, and AAO recommended referral criteria (> 2 line difference or 20/50 or worse for K, 20/40 or worse for 1st–6th grade) resulted in sensitivity of 0.67 and specificity of 0.80 for kindergarteners and sensitivities of ranging from 0.73 to 0.80 and specificities ranging from 0.83–0.95 for 1st–6th graders. Lowering referral criteria to > 2 line difference between eyes or 20/40 or worse for kindergarten (sens 0.91, spec 0.70) and 20/32 or worse for children in grades 1st –6th (sens 0.79 to 0.93, spec 0.67 to 0.85), resulted in costs to specificity, but introduced significant gains in sensitivity.
Lowering referral criteria, particularly for kindergarten children, should be considered for this population with a high prevalence of astigmatism. Further studies of populations with different distributions of refractive errors and abnormalities will help determine optimal referral criteria and will determine if data obtained in this study apply to for kindergarten and grade–school children in general.
This PDF is available to Subscribers Only