AI glaucoma screening cost-effective in the Netherlands?
Reviewed by Arrane Selvamogan

Population-wide screening for glaucoma has traditionally been deemed not cost-effective due to high costs and low disease prevalence. This Dutch study re-evaluates this position by introducing artificial intelligence (AI) to the screening model. The authors developed a health-economic model comparing repeated AI-based fundus photography screening (every 5 years for ages 50-75) against the current standard of opportunistic case finding. The model, which assumed an AI sensitivity of 85% and specificity of 95%, found that AI screening detected 1.6 times more glaucoma cases and reduced visual impairment by 0.8 months per invited individual. From a societal perspective, the intervention resulted in an incremental cost-effectiveness ratio of €19,311 per quality-adjusted life year. This places it just on the edge of the common €20,000 willingness-to-pay threshold, with the authors calculating a 51.2% probability of the programme being cost-effective. Interestingly, the model showed that AI specificity was more critical than sensitivity for maintaining cost-effectiveness. This study provides the first major evidence that an AI-based screening programme could be a viable and sustainable solution for early glaucoma detection in a Western population, though the authors note outcomes are highly sensitive to disease progression rates.

The cost-effectiveness of an artificial intelligence-based population-wide screening program for primary open-angle glaucoma in the Netherlands. 
Boverhof BJ, Ramos IC, Vermeer KA, et al. 
VALUE IN HEALTH 
2025;28(9):1317–26.