Daryl Ramai, MD, MPH, Msc1, Marcello Maida, MD2, Amy Oxentenko, MD3, Jonathan Leighton, MD3, David Hass, MD4, Yuichi Mori, MD5, Mario Dinis-Ribeiro, MD6, Cesare Hassan, MD7, Marianna Arvanitakis, MD8, Monique T.. Barakat, MD9, Raf Bisschops, MD10, Kathryn Byrne, MD11, Evelien Dekker, MD12, John Fang, MD11, Lorenzo Fuccio, MD13, Paraskevas Gkolfakis, MD14, Vicente Lorenzo-Zúñiga, MD15, Andrew Ofosu, MD16, Franco Radaelli, MD17, Jessica Stout, DO11, Luigi Ricciardiello, MD18, Matt Rutter, MD19, Konstantinos Triantafyllou, MD20, Antonio Facciorusso, MD, PhD21 1University of Utah Health, Salt Lake City, UT; 2University of Enna "Kore", Enna, Lombardia, Italy; 3Mayo Clinic, Rochester, MN; 4Yale New Haven Health, Hamden, CT; 5University of Oslo, Blindern, Oslo, Norway; 6Portuguese Institute of Oncology, Porto, Porto, Portugal; 7Humanitas Clinical and Research Hospital, Rozzano, Lombardia, Italy; 8Hôpital Erasme, Brussels, Brussels Hoofdstedelijk Gewest, Belgium; 9Stanford Health Care, Stanford, CA; 10University Hospitals, Leuven, Brussels Hoofdstedelijk Gewest, Belgium; 11University of Utah, Salt Lake City, UT; 12Amsterdam University Medical Center, Amsterdam, Zuid-Holland, Netherlands; 13University of Bologna, Bologna, Marche, Italy; 14Konstantopoulio-Patision General Hospital of Nea Ionia, Athens, Ioannina, Greece; 15La Fe University and Polytechnic Hospital, Valencia, Comunidad Valenciana, Spain; 16University of Cincinnati, Cincinnati, OH; 17Valduce Hospital, Como, Lombardia, Italy; 18University of Bologna, Bologna, Emilia-Romagna, Italy; 19North Tees and Hartlepool NHS Foundation Trust, Stockton-on-Tees, England, United Kingdom; 20Attikon University General Hospital, Athens, Ioannina, Greece; 21University of Foggia, Foggia, Puglia, Italy
Introduction: Despite numerous efforts, adherence to colorectal cancer (CRC) screening remains suboptimal. Artificial intelligence (AI)-assisted chatbots such as ChatGPT (Chat Generative Pretrained Transformer) are emerging as a revolutionary tool that has the potential to provide educational support to patients through an accessible question–answer system. This study aims to evaluate the effectiveness of chat GPT in answering patients' questions about CRC screening, with the ultimate goal of enhancing patients' awareness and adherence to national screening programs worldwide.
Methods: Three experts created a list of 15 questions on CRC screening that were posed to ChatGPT4. Thereafter, the answers were rated by 20 gastroenterology experts and 20 non-experts in three domains (accuracy, completeness, and comprehensibility). The answers were also rated by 100 gastroenterology patients in three dichotomic domains (completeness, comprehensibility and trustability).
Results: According to expert rating, the mean accuracy score on a scale ranging from 1 to 6 was 4.8±1.1, with 10/15 questions receiving a mean rating ³5 points. The average completeness score on a scale ranging from 1 to 3 was 2.1±0.7, with 12/15 questions receiving a mean rating ³2 points. Mean comprehensibility score on a scale from 1 to 3 was 2.8±0.4, with all questions receiving a rating ³2.5 points. Overall, accuracy (4.8±1.1 vs 5.6±0.7, P< 0.001) and completeness (2.1±0.7 vs 2.7±0.4, P< 0.001) scores were significantly lower for expert compared to non-expert, while comprehensibility was comparable among the two groups (2.7±0.4 vs 2.8±0.3, P=0.546). Patients rated all questions as complete, comprehensible and trustable in 97 to 100% of cases.
Discussion: ChatGPT shows good performance with the potential to enhance awareness about CRC and improve screening outcomes. In the future, generative language systems may be further improved after proper training in accordance with scientific evidence and current guidelines.
Figure: Box plot showing the rating of experts in accuracy (a), completeness (b), and comprehensibility (c).
Disclosures:
Daryl Ramai indicated no relevant financial relationships.
Marcello Maida indicated no relevant financial relationships.
Amy Oxentenko indicated no relevant financial relationships.
Jonathan Leighton indicated no relevant financial relationships.
David Hass indicated no relevant financial relationships.
Yuichi Mori indicated no relevant financial relationships.
Mario Dinis-Ribeiro indicated no relevant financial relationships.
Cesare Hassan indicated no relevant financial relationships.
Marianna Arvanitakis indicated no relevant financial relationships.
Monique Barakat indicated no relevant financial relationships.
Raf Bisschops indicated no relevant financial relationships.
Kathryn Byrne indicated no relevant financial relationships.
Evelien Dekker indicated no relevant financial relationships.
John Fang indicated no relevant financial relationships.
Lorenzo Fuccio indicated no relevant financial relationships.
Paraskevas Gkolfakis indicated no relevant financial relationships.
Vicente Lorenzo-Zúñiga indicated no relevant financial relationships.
Andrew Ofosu indicated no relevant financial relationships.
Franco Radaelli indicated no relevant financial relationships.
Jessica Stout indicated no relevant financial relationships.
Luigi Ricciardiello indicated no relevant financial relationships.
Matt Rutter indicated no relevant financial relationships.
Konstantinos Triantafyllou indicated no relevant financial relationships.
Antonio Facciorusso indicated no relevant financial relationships.
Daryl Ramai, MD, MPH, Msc1, Marcello Maida, MD2, Amy Oxentenko, MD3, Jonathan Leighton, MD3, David Hass, MD4, Yuichi Mori, MD5, Mario Dinis-Ribeiro, MD6, Cesare Hassan, MD7, Marianna Arvanitakis, MD8, Monique T.. Barakat, MD9, Raf Bisschops, MD10, Kathryn Byrne, MD11, Evelien Dekker, MD12, John Fang, MD11, Lorenzo Fuccio, MD13, Paraskevas Gkolfakis, MD14, Vicente Lorenzo-Zúñiga, MD15, Andrew Ofosu, MD16, Franco Radaelli, MD17, Jessica Stout, DO11, Luigi Ricciardiello, MD18, Matt Rutter, MD19, Konstantinos Triantafyllou, MD20, Antonio Facciorusso, MD, PhD21. P2125 - The Role of Generative Language Systems in Increasing Patient Awareness of Colon Cancer Screening, ACG 2024 Annual Scientific Meeting Abstracts. Philadelphia, PA: American College of Gastroenterology.