Canonical text (the exact bytes fingerprinted)

TRUVACE RECORD VERSION
record: TRV-2026-1386
version: 1
kind: certified
reason: Certified into the record
timestamp: 2026-10-15T01:09:07.015358Z
status: published
lens: g_space
sector: health
headline: Cost-effectiveness of community-based artificial intelligence screening for retinopathy of prematurity
dek: Background Retinopathy of prematurity (ROP) is one of the leading causes of childhood blindness globally, but particularly in middle-income countries, where neonatal survival has improved faster than access to specialist ophthalmic care. Although telemedicine and artificial intelligence (AI)-assisted screening show promise for improving access to ROP screening, their economic value within decentralised health systems remains uncertain. To explore this issue, we evaluated the cost-effectiveness of community-based…
gain_title: In a decision-tree model of 100,000 preterm infants, community-based AI-assisted ROP screening had lower lifetime societal costs and more QALYs than telemedicine or bedside screening in both urban and rural China.
problem_title: (none)
trace_subject: (none)
gain_reading: In a decision-tree model of 100,000 preterm infants, community-based AI-assisted ROP screening had lower lifetime societal costs and more QALYs than telemedicine or bedside screening in both urban and rural China.
gain_evidence: Community-based AI-assisted screening was the most cost-effective strategy in both urban and rural settings. | AI-assisted screening incurred the lowest lifetime costs (USD 1,507 per infant in urban and USD 2,581 in rural areas)
problem_reading: (none)
problem_evidence: (none)
quick_read: Researchers built a decision tree model for a hypothetical cohort of 100,000 preterm infants eligible for retinopathy of prematurity screening in China. They compared community-based AI-assisted screening, community-based telemedicine screening, and bedside screening at specialist facilities, using separate urban and rural parameters and a societal cost perspective.

The model suggests AI-assisted screening could widen access to early eye care while lowering costs, which matters where neonatal survival has outpaced specialist ophthalmic capacity. These are modelled estimates for a hypothetical cohort rather than observed outcomes from a deployed programme, so real-world results may differ.
limitation: The findings come from a decision tree model of a hypothetical cohort, so they are projected estimates rather than outcomes measured in a real screening programme.
tag: Evidence-backed gain
key_points: Telemedicine screening dominated bedside screening, meaning it was less costly and more effective in the base case. | AI-assisted screening further dominated telemedicine, with lifetime costs of USD 1,507 per infant in urban and USD 2,581 in rural areas. | Modelled health gains were 29.7138 QALYs in urban and 29.5835 QALYs in rural settings under AI-assisted screening. | Incremental cost-effectiveness ratios against bedside screening were USD -7,307 per QALY gained in urban and USD -3,688 in rural settings. | Results were robust across scenario and sensitivity analyses.
rundown: The background notes that retinopathy of prematurity is a leading cause of childhood blindness globally, especially in middle-income countries where neonatal survival has improved faster than access to specialist ophthalmic care.

Although telemedicine and AI-assisted screening were seen as promising for access, their economic value within decentralised health systems was uncertain, which motivated this evaluation.

The authors report that AI-assisted screening stopped being cost-effective only at extremely low treatment-requiring ROP incidence, and conclude it is likely the most cost-effective strategy in China, potentially a scalable model for similar resource-constrained regions.
sources:
- peer_reviewed | Journal of Global Health | https://doi.org/10.7189/jogh.16.04357 | 2026-10-02
prev: 0000000000000000000000000000000000000000000000000000000000000000

sha256

f68dca4e9a67b8d86a1f9bdf20f4b0226046ec3fee404141bb3d6390306b44da

previous

0000000000000000000000000000000000000000000000000000000000000000