Feld, Andrew D.

INTRODUCTION: The American College of Gastroenterology assembled a multidisciplinary task force to evaluate the current state and future direction of artificial intelligence (AI) in gastroenterology, hepatology, and endoscopy leading to the development of consensus-based recommendations for responsible AI integration in clinical practice. METHODS: A total of 32 subject-matter experts and 12 industry partners, representing diverse practice settings and expertise, conducted subgroup literature reviews across 5 key areas (endoscopy, practice management clinical applications, training and education, inflammatory bowel disease and liver disease, ethics and equity). Draft statements were developed and rated on a 5-point Likert scale using a modified Delphi process. A consensus was set at ≥70% combined agreement. Nonconsensus items were revised and revoted electronically. RESULTS: A total of 43 statements, 40 (93%) reached consensus in round 1 and the remaining 3 achieved consensus after round 2. Evidence supports computer-aided detection improving adenoma detection rate and miss rate in controlled studies, with mixed real-world impact and insufficient long-term outcomes (e.g., interval colon cancer rate). Recommendations emphasize thorough validation and reduction of bias by heterogeneous Data sets. Outside endoscopy, ambient AI scribes, natural language processing (NLP)-enabled coding, workflow optimization, and previous authorization support show potential. Training recommendations endorse a structured AI curriculum while preserving independent procedural competence to avoid deskilling. In inflammatory bowel disease and hepatology, AI could help improve diagnostic accuracy, help predict risk of disease progression, and help guide therapy. Equity, governance, and reimbursement statements call for chain-of-custody data protections, specialty-society oversight, and payment models that reward quality and cost reduction. DISCUSSION: This consensus outlines how AI can augment rather than replace clinical expertise while promoting safety, transparency, interoperability, and equity. Priorities include pragmatic and prospective trials, multi-institutional data-sharing consortia, bias mitigation, and workforce training to enable trustworthy and clinically impactful AI adoption in gastroenterology, liver, and endoscopy care.

INTRODUCTION: Modeling supporting recommendations for colonoscopy and stool-based colorectal cancer (CRC) screening tests assumes 100% sequential participant adherence. The impact of observed adherence on the long-term effectiveness of screening is unknown. We evaluated the effectiveness of a program of screening colonoscopy every 10 years vs annual high-sensitivity guaiac-based fecal occult blood testing (HSgFOBT) using observed sequential adherence data. METHODS: The MIcrosimulation SCreening ANalysis (MISCAN) model used observed sequential screening adherence, HSgFOBT positivity, and diagnostic colonoscopy adherence in HSgFOBT-positive individuals from the National Colonoscopy Study (single-screening colonoscopy vs ≥4 HSgFOBT sequential rounds). We compared CRC incidence and mortality over 15 years with no screening or 10 yearly screening colonoscopy vs annual HSgFOBT with 100% and differential observed adherence from the trial. RESULTS: Without screening, simulated incidence and mortality over 15 years were 20.9 (95% probability interval 15.8–26.9) and 6.9 (5.0–9.2) per 1,000 participants, respectively. In the case of 100% adherence, only screening colonoscopy was predicted to result in lower incidence; however, both tests lowered simulated mortality to a similar level (2.1 [1.6–2.9] for screening colonoscopy and 2.5 [1.8–3.4] for HSgFOBT). Observed adherence for screening colonoscopy (83.6%) was higher than observed sequential HSgFOBT adherence (73.1% first round; 49.1% by round 4), resulting in lower simulated incidence and mortality for screening colonoscopy (14.4 [10.8–18.5] and 2.9 [2.1–3.9], respectively) than HSgFOBT (20.8 [15.8–28.1] and 3.9 [2.9–5.4], respectively), despite a 91% adherence to diagnostic colonoscopy with FOBT positivity. The relative risk of CRC mortality for screening colonoscopy vs HSgFOBT was 0.75 (95% probability interval 0.68–0.80). Findings were similar in sensitivity analyses with alternative assumptions for repeat colonoscopy, test performance, risk, age, and projection horizon. DISCUSSION: Where sequential adherence to stool-based screening is suboptimal and colonoscopy is accessible and acceptable—as observed in the National Colonoscopy Study—offering screening colonoscopy can increase screening effectiveness.