Reporting Guideline for Chatbot Health Advice Studies

作者：The CHART Collaborative, Bright Huo, Professor Gary S. Collins, David Chartash, Arun James Thirunavukarasu, Annette Flanagin, Alfonso Iorio, Giovanni Cacciamani, Xi Chen, Nan Liu, Piyush Mathur, An‐Wen Chan, Christine Lainé, Daniela Pacella, Michael Berkwits, Stavros A. Antoniou, Jennifer Camaradou, Carolyn Canfield, Michael Mittelman, Timothy Feeney, Elizabeth Loder, Riaz Agha, Ashirbani Saha, Julio Mayol, Anthony Paulo Sunjaya, Hugh Harvey, Jeremy Y. Ng, Tyler McKechnie, Yung Lee, Nipun Verma, Gregor Štiglic, Melissa D. McCradden, Karim Ramji, Vanessa Boudreau, Monica Ortenzi, Joerg J Meerpohl, Per Olav Vandvik, Thomas Agoritsas, Diana Samuel, Helen Frankish, Michael C. Anderson, Xiaomei Yao, Stacy Loeb, Cynthia Lokker, Xiaoxuan Liu, Eliseo Güallar, Gordon Guyatt · 发表于：JAMA Network Open · 年份：2025 · DOI：10.1001/jamanetworkopen.2025.30220 · 被引用次数：27 · 研究领域：Digital Mental Health Interventions、Artificial Intelligence in Healthcare and Education、AI in Service Interactions

Importance: The rise in chatbot health advice (CHA) studies is accompanied by heterogeneity in reporting standards, impacting their interpretability. Objective: To provide reporting recommendations for studies evaluating the performance of generative artificial intelligence (AI)-driven chatbots when summarizing clinical evidence and providing health advice. Design, Setting, and Participants: CHART was developed in several phases after performing a comprehensive systematic review to identify variation in the conduct, reporting, and methodology in CHA studies. Findings from the review were used to develop a draft checklist that was revised through an international, multidisciplinary modified asynchronous Delphi consensus process of 531 stakeholders, 3 synchronous panel consensus meetings of 48 stakeholders, and subsequent pilot testing of the checklist. Results: CHART includes 12 items and 39 subitems to promote transparent and comprehensive reporting of CHA studies. These include title (subitem 1a), abstract or summary (subitem 1b), background (subitems 2ab), model identifiers (subitem 3ab), model details (subitems 4abc), prompt engineering (subitems 5ab), query strategy (subitems 6abcd), performance evaluation (subitems 7ab), sample size (subitem 8), data analysis (subitem 9a), results (subitems 10abc), discussion (subitems 11abc), disclosures (subitem 12a), funding (subitem 12b), ethics (subitem 12c), protocol (subitem 12d), and data availability (subitem 12e). Conclusions a...