Abstract
Purpose: :
To examine the sensitivity and specificity of visual acuity (VA) measures to screen for significant refractive error in a population-based random cluster sample of 12-year old school children.
Methods: :
The Sydney Myopia Study randomly selected 21 secondary schools, stratified by socio-economic status. Year 7 students were invited to participate. VA was performed monocularly, unaided at 2.44m using a retro-illuminated logMAR chart (CSV1000). Cycloplegic (Cyclopentolate 1%) auto-refraction was conducted. Clinically significant refractive error was defined as spherical equivalent ≤-1.0D for myopia; ≥+2.0D for hyperopia and ≤-1.0D for astigmatism.
Results: :
Of the 2353 Year 7 students (mean age 12.7 years), data for both eyes were pooled for a total of 4670 observations, after excluding 18 children with reduced VA from ocular pathology or amblyopia. The sensitivity and specificity for all clinically significant refractive errors at a logMAR VA cut-off level of 6/7.5 were 69% & 96%, respectively. At 6/9, these were 61% & 98%, respectively. Screening sensitivities and specificities were then examined for individual refractive errors. For myopia, sensitivity and specificity were 99% & 93% for 6/7.5 cut-off and with better specificity at 97% & 96% for 6/9 cut-off. For hyperopia, sensitivity and specificity was 37% & 86% using 6/7.5 cut-off and sensitivity reduced to 25% with 90% specificity using 6/9 cut-off. For astigmatism, sensitivity and specificity was 61% & 88% using 6/7.5 cut-off and sensitivity decreased to 45% with 92% specificity using 6/9 cut-off. However, screening sensitivity slightly improved for astigmatism using a 6/6 VA cut-off (69%), but specificity was reduced (84%). Reliability of VA at the 6/12 cut-off was insensitive for hyperopia (17%) and astigmatism (37%), but sensitive for myopia (92%), specificity was sound for all types 98%, 91%, 93% respectively, at this VA level.
Conclusions: :
In this adolescent group, sensitivity and specificity values for a VA 6/7.5 cut-off detects myopic refractive error quite reliably. Astigmatism is detected moderately well only when the VA cut-off is 6/6. No VA level adequately screens for clinically significant hyperopia, presumably due to the accommodative ability of young adolescents. Therefore while you can appropriately screen for myopia using a VA cut off of 6/7.5, at present cycloplegic refraction seems to be the only way to reliably screen for hyperopia and astigmatism.
Keywords: visual acuity • refraction • clinical (human) or epidemiologic studies: systems/equipment/techniques