The prevalence of amblyopia among screening defaulters is an important determinant of the efficacy of amblyopia detection by preschool vision screening. A retrospective cohort study was therefore performed to assess an orthoptist-based preschool vision screening programme. The preschool vision screening status of children in the cohort was determined by reviewing their Community Child Health records. The prevalence of amblyopia among screening defaulters was determined by reviewing each child's school entry vision test (performed at 5.5 years of age), with retesting if a Snellen line acuity of 6/6 in each eye had not been documented. For comparison, the prevalence of amblyopia among screening attenders was also determined. The preschool vision screening status was known for 86.0% (772/898) of the cohort. The attendance rate at preschool vision screening was 79.2%. The prevalence of amblyopia among screening defaulters was 1.3% (95% CI 0.2% to 4.5%). The prevalence of amblyopia among screening attenders was 2.5% (95% CI 1.4% to 4.1%). There was no significant difference in the prevalence of amblyopia between screening defaulters and screening attenders (P=0.53). The efficacy of amblyopia detection by preschool vision screening is therefore highly dependent on its attendance rate. Preschool vision screening programmes with a low attendance rate will fail to detect a significant proportion of children with amblyopia.