Endoscopy
Signal score, how each part rates
ActionabilityPrevalenceStakesEvidence strengthNovelty
See the full rubric →Cost-effectiveness and environmental impact of artificial intelligence-assisted colonoscopy with the resect-and-discard strategy: a microsimulation study.
Clinical takeawayThis is a modeling study comparing strategy costs and outcomes, not a clinical trial of patient outcomes. It suggests that computer-aided detection improved adenoma detection without improving health outcomes compared to non-AI resect-and-discard; adding diagnostic AI (CADx) increased costs without clinical benefit. The findings suggest improved adenoma detection alone may not justify added costs and environmental impact if baseline health outcomes remain unchanged. However, these are model-based conclusions dependent on input assumptions about adenoma progression, detection accuracy, and mortality rates. Real-world adoption decisions about AI tools should consider actual performance, costs, and environmental impact in your setting. Individual gastroenterologists should not change current resect-and-discard practice based on this model.
What it foundIn a Markov model simulating 6 million individuals aged ≥45 years undergoing screening colonoscopy, non-AI resect-and-discard achieved the highest quality-adjusted life years (51.37 million) at lowest cost ($36.6B) and carbon emissions (0.90 million kg CO₂) compared to CADe-assisted RD ($36.9B) or CADe+CADx-assisted RD ($37.4B); differences in QALYs across strategies were <0.01%, and CADe modestly increased adenoma detection without improving health outcomes.
ContextCurrent practice widely promotes computer-aided detection (CADe) and diagnostic (CADx) tools to increase adenoma detection rates in screening colonoscopy. This model challenges the assumption that detection improvements alone drive health benefit in populations where screening is already reasonably effective, suggesting diminishing returns beyond baseline screening performance. The finding illustrates a general principle: increasing test performance does not always improve patient outcomes if the underlying disease is already well-managed with existing strategies.
Emergingsuggested applicable standard· U.S. Multi-Society Task Force on Colorectal Cancer (Rex DK, Boland CR, Dominitz JA, et al.), "Colorectal Cancer Screening: Recommendations for Physicians and Patients From the U.S. Multi-Society Task Force on Colorectal Cancer," Gastroenterology, 2017
Decision at stakeHow to manage diminutive polyps identified during screening colonoscopy
U.S. Multi-Society Task Force on Colorectal Cancer (Rex DK, Boland CR, Dominitz JA, et al.), "Colorectal Cancer Screening: Recommendations for Physicians and Patients From the U.S. Multi-Society Task Force on Colorectal Cancer," Gastroenterology, 2017 · reviewed 2026-07-23 ↗Begin average-risk colorectal cancer screening at age 45 using a patient-centered shared-decision modality choice, colonoscopy every 10 years (preferred) or annual FIT as Tier 1 options, with multi-target stool DNA every 3 years, CT colonography every 5 years, or flexible sigmoidoscopy every 5 to 10 years as Tier 2 alternatives. A positive stool-based test requires diagnostic colonoscopy, and stool tests should not be ordered for patients who would decline follow-up colonoscopy. Generally stop at age 75 with individualized decisions for ages 76-85 and no screening beyond 85.
Hernandez LV … Pohl H · Endoscopy · IF 11.8 · PubMed ↗