Effectiveness of preschool vision screening: Four studies reported on the effectiveness of preschool visual screening. No RCTs were found that compared screening against no screening. One large (3,490 participants) fair quality nested RCT found that repeated orthoptist screening from ages eight to 37 months was associated with a statistically significant reduction in likelihood of amblyopia at age 7.5 years compared with one-time orthoptist screening at age 37 months on one of two definitions of amblyopia. One large, prospective cohort study from the same population found that one-time orthoptist screening at 37 months was associated with no significant difference in risk for amblyopia at age 7.5 years compared with no screening or school entry screening. Three retrospective cohort studies found that preschool screening was associated with improved school-age vision outcomes compared with no screening but each study had methodological limitations and none evaluated school performance or other functional outcomes.
Impact of age of assessment on effectiveness of preschool vision screening: No RCTs compared outcomes of preschool vision screening in different age groups. Evidence from other studies was poor quality, subject to confounding or did not address relevant clinical outcomes.
Accuracy of screening tests for vision impairment in preschool children: Thirty-one studies evaluated the diagnostic accuracy of preschool visual screening tests. None was consistently associated with both high (>90%) sensitivity and specificity. In the largest comparative study, differences in likelihood ratio estimates and diagnostic odds ratios for 10 different screening tests were generally small except for the Random Dot E stereoacuity test, which was associated with a lower diagnostic odds ratio. Combinations of clinical tests generally showed stronger likelihood ratios than individual tests.
Impact of age of assessment on accuracy of screening tests: Diagnostic accuracy of preschool vision tests did not clearly differ in children stratified by age. Testability was generally slightly lower in children aged one to three years with the possible exception of the MTI Photoscreener, for which one large study reported testability as being 94% at age one year.
Harms of preschool vision screening: One large cohort study reported a 50% reduction in the odds of being bullied at age 7.5 years among children offered screening compared with those not offered screening. Six of seven studies reported false-positive rates greater than 70% but none evaluated the effects of unnecessary corrective lenses or treatment for amblyopia on long-term vision or functional outcomes.
Effectiveness of treatment for vision impairment in children aged one to five years: Three fair or good quality trials reported that treatment resulted in small (<1 line on the Snellen eye chart) improvements in visual acuity in the amblyopic or worse eye compared with no treatment after five weeks to one year of follow-up in older children (four to five years). One trial found larger benefits in the subgroup of children with worse baseline visual impairment. No trials reported school performance or other measures of function. Evidence on the impact of age on effectiveness of treatment was inconsistent.
Harms of treatment for vision impairment in children aged one to five years: Evidence from five good-quality trials suggested that amblyopia treatments were associated with reversible visual acuity loss in the non-amblyogenic eye. Evidence on adverse psychosocial effects and effects of suboptimal compliance with amblyopia treatments was limited.