Canonical text (the exact bytes fingerprinted)

TRUVACE RECORD VERSION
record: TRV-2026-1441
version: 1
kind: certified
reason: Certified into the record
timestamp: 2026-10-15T10:43:05.604540Z
status: published
lens: g_space
sector: health
headline: Cost-Effectiveness of AI in Breast Cancer Screening and Lung Cancer Diagnostics: A Case Study From Costa Rica
dek: Objectives This study evaluated the cost-effectiveness of integrating artificial intelligence (AI) into breast and lung cancer diagnostic workflows in Costa Rica, addressing the scarcity of health economic evidence for AI adoption in low- and middle-income countries and focusing on context-specific systemic barriers. Methods Two state-transition Markov models compared AI-assisted workflows (AI as a first reader) with conventional radiologist-only strategies over a 10-year horizon. AI-supported mammography screen…
gain_title: In Markov models for Costa Rica, using AI as a first reader in breast screening and lung cancer diagnosis was projected to lower costs while increasing quality-adjusted life-years over 10 years.
problem_title: (none)
trace_subject: (none)
gain_reading: In Markov models for Costa Rica, using AI as a first reader in breast screening and lung cancer diagnosis was projected to lower costs while increasing quality-adjusted life-years over 10 years.
gain_evidence: AI in breast cancer screening demonstrated economic dominance
problem_reading: (none)
problem_evidence: (none)
quick_read: Researchers built two state-transition Markov models to compare AI-assisted diagnostic workflows, with AI acting as first reader, against radiologist-only strategies in Costa Rica over a 10-year horizon. One model covered mammography screening for asymptomatic women aged 50 to 70, and the other covered symptom-driven lung cancer pathways.

In both models the AI strategy came out cheaper and produced more QALYs, which the authors describe as economic dominance. The result matters for low- and middle-income countries, where health economic evidence on AI is scarce. These are modelled projections rather than observed outcomes from deployed systems, so real-world results will depend on implementation and system conditions.
limitation: The findings come from simulation models over a 10-year horizon, with the authors noting AI's impact is context-dependent and tied to system barriers and implementation realities, so results are not measured outcomes from actual deployment.
tag: Evidence-backed gain
key_points: Breast screening model: QALYs rose 49% (10.46 vs 7.04) while costs fell 49% (USD 93 102 vs USD 183 278). | Net monetary benefit for breast screening was USD 212 119 at Costa Rica's willingness-to-pay threshold. | Lung cancer model: total costs fell to USD 1 218 507 from USD 1 337 640 while QALYs rose to 6.88 from 4.15. | Savings in lung cancer were attributed to earlier detection that averts expensive late-stage palliative care. | Authors call for early detection investment and equity-focused implementation.
rundown: The breast cancer model evaluated AI-supported mammography in asymptomatic women aged 50 to 70. Compared with radiologist-only reading, the AI strategy improved QALYs and reduced costs, making it dominant in the model.

The lung cancer model used symptom-driven pathways based on GLOBOCAN 2022 incidence data. Earlier detection under the AI strategy reduced the need for costly late-stage palliative care.

Model inputs used region-specific costs in USD, QALYs, and progression probabilities calibrated to Costa Rica's cancer registry. Reporting followed the Consolidated Health Economic Evaluation Reporting Standards 2022 guidelines.
sources:
- peer_reviewed | Value in Health Regional Issues | https://doi.org/10.1016/j.vhri.2026.102123 | 2026-10-01
prev: 0000000000000000000000000000000000000000000000000000000000000000

sha256

534249dea214fd2bf8d3b5e85fcfe28a846b0b7668d185c4aa907474245285d6

previous

0000000000000000000000000000000000000000000000000000000000000000