Abstract Aim: To study the effectiveness and cost of a new school vision screening program involving all class teachers (ACTs) compared with the standard program involving a limited number of selected teachers (STs) in Southern India. Methods: A total of 12 schools were selected for intervention and 12 schools were matched as controls, from in and around Pondicherry, India. Teachers in both the intervention arm (ACTs) and the standard arm (STs) were trained to identify students between the ages of 6 and 17 years with visual acuity 20/30 in either eye or obvious ocular abnormalities and refer them to an ophthalmic team. The ophthalmic team, including an ophthalmologist, visited the schools to examine all children referred by teachers, provided medical treatment or a prescription for glasses, or referred them to the base hospital if required. Results: ACTs (761 teachers) screened 39,357 children (97.7%) and STs (156 teachers) screened 38,469 children (95.7%). ACTs found significantly fewer screen-positive children (n = 3806, 9.7%) than the STs (n = 6387, 16.6%; p < 0.001), but had a significantly larger number of children with actual vision loss and other ocular pathology (2231, 5.7% and 1554, 4.0%, respectively, p < 0.001). More children from ACTs than STs reached the base hospital for further investigation within 3 months (p < 0.001). The cost of screening per child with actual ocular pathology was estimated to be US$1.91 for ACTs and US$4.83 for STs. Conclusions: A school vision screening program involving ACTs resulted in more efficient screening than a program involving STs at about a third of the cost and also improved compliance with hospital referral.