This cluster-randomized controlled trial aims to evaluate the effectiveness and cost-effectiveness of an artificial intelligence-based, risk-stratified, tiered intervention model for childhood myopia prevention and control in resource-limited county areas of China, compared with the conventional screening-and-referral approach, with schools in project counties of Guizhou or Yunnan Province randomly assigned at a 1:1 ratio to either the intervention or control group over a 12-month follow-up period.
The study is motivated by the pressing reality that myopia among Chinese children is not only widespread but also occurring at younger ages than ever before, bringing with it a lifelong risk of sight-threatening complications, yet in remote county areas, routine vision screening programs are not yet widely available or well-established, due to a shortage of trained eye care professionals, the practical difficulties of performing cycloplegic refraction in school settings, and the lack of systematic follow-up, all of which make it hard to identify children with myopia or pre-myopia. Given that artificial intelligence may offer a practical way to assess risk without the need for pupil dilation, this study will enroll children in grades 1 through 3 from local primary schools. Over 12 months, an AI model will dynamically categorize children based on screening results and guide management adjustments(outdoor activity and defocus lenses) for the intervention group, while the control group receives routine care.
The findings are expected to provide evidence to inform and optimize AI-based myopia prevention and control strategies in remote areas.