Detecting ocular and visual anomalies in a vision screening setting using the Lang stereotest.

Effectiveness of random dot stereotests (RDS) as vision-screening procedures vary. In this investigation, 183 school-aged children (6-10 years) had their vision screened using the Modified Clinical Technique (20.7 percent referred) and the Lang stereotest (12.0 percent referred)--a type of RDS test not requiring dissociating glasses. For the populations as a whole, the Lang test demonstrated low sensitivity, 31.6 percent; yet specificity was high, 93.1 percent. The phi coefficient (phi) was +0.31 for use as a sole procedure that rose to +0.43 when combined with Snellen acuity and +0.60 when combined with refractive error. Test/retest reliability conducted on a subgroup of 27 children showed a correlation (r) of +0.69. In another subgroup of 17 multiply handicapped children, 41.2 percent were referred by the Lang stereotest; the sensitivity was 100 percent, specificity 90.9 percent and effectivity (phi) +0.88. While not as effective as reports of the Random Dot E stereotest in screening 6-10 year old children, results obtained with the Lang stereotest in the group of exceptional children are remarkable.