{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,11]],"date-time":"2026-03-11T22:41:37Z","timestamp":1773268897607,"version":"3.50.1"},"reference-count":39,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,3,11]],"date-time":"2026-03-11T00:00:00Z","timestamp":1773187200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Med."],"abstract":"<jats:p>Artificial intelligence (AI) and machine learning have emerged as transformative analytic methods capable of identifying patterns and proposing treatment strategies from complex, heterogeneous data generated in patient care settings. However, translating reports of AI tools into clinical decision-making requires careful interpretation and contextualization. Here, we present a structured framework (ABCDEFG framework) for clinicians to critically appraise studies reporting AI tools using a high-profile reinforcement learning analysis of vasopressin initiation in septic shock as a case example. We highlight key methodological strengths including adherence to reporting standards, alignment with clinical outcomes, and external validation. We also identify limitations related to causal inference, reward structure design, assumptions about state representation, and implementation readiness. Our framework emphasizes the importance of rigorous, clinically grounded critique before integrating AI tools into practice, and provides a roadmap for clinicians to assess the reliability and relevance of emerging AI-based decision support.<\/jats:p>","DOI":"10.3389\/fmed.2026.1787075","type":"journal-article","created":{"date-parts":[[2026,3,11]],"date-time":"2026-03-11T06:53:30Z","timestamp":1773212010000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["ABC\u2019S of AI: proposing a clinician\u2019s framework for the interpretation of artificial intelligence studies"],"prefix":"10.3389","volume":"13","author":[{"given":"Brian","family":"Murray","sequence":"first","affiliation":[{"name":"University of Colorado Skaggs School of Pharmacy","place":["Aurora, CO, United States"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Seth R.","family":"Bauer","sequence":"additional","affiliation":[{"name":"Department of Pharmacy, Cleveland Clinic","place":["Cleveland, OH, United States"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yanjun","family":"Gao","sequence":"additional","affiliation":[{"name":"University of Colorado School of Medicine","place":["Aurora, CO, United States"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Erin F.","family":"Barreto","sequence":"additional","affiliation":[{"name":"Mayo Clinic","place":["Rochester, MN, United States"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Kelli","family":"Henry","sequence":"additional","affiliation":[{"name":"University of Colorado School of Medicine","place":["Aurora, CO, United States"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Kaitlin","family":"Blotske","sequence":"additional","affiliation":[{"name":"University of Colorado School of Medicine","place":["Aurora, CO, United States"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Andrea","family":"Sikora","sequence":"additional","affiliation":[{"name":"University of Colorado School of Medicine","place":["Aurora, CO, United States"]}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1965","published-online":{"date-parts":[[2026,3,11]]},"reference":[{"key":"ref1","doi-asserted-by":"publisher","first-page":"78","DOI":"10.1001\/jama.2023.8288","article-title":"Accuracy of a generative artificial intelligence model in a complex diagnostic challenge","volume":"330","author":"Kanjee","year":"2023","journal-title":"JAMA"},{"key":"ref2","doi-asserted-by":"publisher","first-page":"1951","DOI":"10.1056\/NEJMsa2001090","article-title":"Automated identification of Adults at risk for in-hospital clinical deterioration","volume":"383","author":"Escobar","year":"2020","journal-title":"N Engl J Med"},{"key":"ref3","doi-asserted-by":"publisher","first-page":"1688","DOI":"10.1001\/jama.2025.3046","article-title":"Optimal vasopressin initiation in septic shock: the OVISS reinforcement learning study","volume":"333","author":"Kalimouttou","year":"2025","journal-title":"JAMA"},{"key":"ref4","doi-asserted-by":"publisher","first-page":"e0000651","DOI":"10.1371\/journal.pdig.0000651","article-title":"Bias in medical AI: implications for clinical decision-making","volume":"3","author":"Cross","year":"2024","journal-title":"PLOS Digit Health"},{"key":"ref5","doi-asserted-by":"publisher","first-page":"9072","DOI":"10.48550\/arXiv.2206.03383","article-title":"On the role of discount factor in offline reinforcement learning","author":"Hu","year":"2022","journal-title":"International Conference on Machine Learning"},{"key":"ref6","doi-asserted-by":"publisher","first-page":"7266","DOI":"10.3390\/s22197266","article-title":"Adaptive discount factor for deep reinforcement learning in continuing tasks with uncertainty","volume":"22","author":"Kim","year":"2022","journal-title":"Sensors"},{"key":"ref7","doi-asserted-by":"publisher","first-page":"337","DOI":"10.1038\/s41746-024-01316-0","article-title":"A primer on reinforcement learning in medicine for clinicians","volume":"7","author":"Jayaraman","year":"2024","journal-title":"NPJ Digit Med"},{"key":"ref8","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/s41597-022-01899-x","article-title":"MIMIC-IV, a freely accessible electronic health record dataset","volume":"10","author":"Johnson","year":"2023","journal-title":"Sci Data"},{"key":"ref9","doi-asserted-by":"publisher","first-page":"180178","DOI":"10.1038\/sdata.2018.178","article-title":"The eICU collaborative research database, a freely available multi-center database for critical care research","volume":"5","author":"Pollard","year":"2018","journal-title":"Sci Data"},{"key":"ref10","doi-asserted-by":"publisher","first-page":"1380","DOI":"10.1097\/CCM.0000000000006338","article-title":"The impact of common variations in sequential organ failure assessment score calculation on Sepsis measurement using Sepsis-3 criteria: a retrospective analysis using electronic health record data","volume":"52","author":"Alrawashdeh","year":"2024","journal-title":"Crit Care Med"},{"key":"ref11","doi-asserted-by":"publisher","first-page":"1920","DOI":"10.1038\/s41598-024-51989-6","article-title":"Subtle variation in sepsis-III definitions markedly influences predictive performance within and across methods","volume":"14","author":"Cohen","year":"2024","journal-title":"Sci Rep"},{"key":"ref12","doi-asserted-by":"publisher","first-page":"e1063","DOI":"10.1097\/CCM.0000000000005337","article-title":"Surviving Sepsis campaign: international guidelines for Management of Sepsis and Septic Shock 2021","volume":"49","author":"Evans","year":"2021","journal-title":"Crit Care Med"},{"key":"ref13","doi-asserted-by":"publisher","first-page":"940","DOI":"10.1097\/CCM.0000000000002323","article-title":"The septic shock 3.0 definition and trials: a vasopressin and septic shock trial experience","volume":"45","author":"Russell","year":"2017","journal-title":"Crit Care Med"},{"key":"ref14","doi-asserted-by":"publisher","first-page":"e1284","DOI":"10.1097\/CCE.0000000000001284","article-title":"Vasopressin initiation timing and in-hospital mortality in septic shock: an observational study of large public databases","volume":"7","author":"Sacha","year":"2025","journal-title":"Crit Care Explor"},{"key":"ref15","doi-asserted-by":"publisher","first-page":"326","DOI":"10.1093\/jamiaopen\/ooaa033","article-title":"Evaluating artificial intelligence in medicine: phases of clinical research","volume":"3","author":"Park","year":"2020","journal-title":"JAMIA Open"},{"key":"ref16","doi-asserted-by":"publisher","first-page":"1486","DOI":"10.1007\/s00134-020-06045-y","article-title":"Machine learning in intensive care medicine: ready for take-off?","volume":"46","author":"Fleuren","year":"2020","journal-title":"Intensive Care Med"},{"key":"ref17","doi-asserted-by":"publisher","first-page":"1399","DOI":"10.1001\/jamainternmed.2023.5002","article-title":"Analysis of devices authorized by the FDA for clinical decision support in critical care","volume":"183","author":"Lee","year":"2023","journal-title":"JAMA Intern Med"},{"key":"ref18","volume-title":"EU Guidelines on Ethics in artificial Intelligence: Context and Implementation","author":"Madiega","year":"2019"},{"key":"ref19","volume-title":"Ethics and Governance of artificial Intelligence for Health: Guidance on large multi-Modal Models","year":"2025"},{"key":"ref20","doi-asserted-by":"publisher","first-page":"12","DOI":"10.1007\/s10916-021-01790-z","article-title":"Trustworthy augmented intelligence in health care","volume":"46","author":"Crigger","year":"2022","journal-title":"J Med Syst"},{"key":"ref21","volume-title":"An Artificial Intelligence Code of Conduct for Health and Medicine: Essential Guidance for Aligned Action","author":"Krishnan","year":"2025"},{"key":"ref22","doi-asserted-by":"publisher","first-page":"585","DOI":"10.1093\/jamia\/ocac006","article-title":"Defining AMIA\u2019s artificial intelligence principles","volume":"29","author":"Solomonides","year":"2022","journal-title":"J Am Med Inform Assoc"},{"key":"ref23","doi-asserted-by":"publisher","first-page":"1806","DOI":"10.1001\/jama.2019.16489","article-title":"How to read articles that use machine learning: users' guides to the medical literature","volume":"322","author":"Liu","year":"2019","journal-title":"JAMA"},{"key":"ref24","doi-asserted-by":"publisher","first-page":"e078378","DOI":"10.1136\/bmj-2023-078378","article-title":"TRIPOD+AI statement: updated guidance for reporting clinical prediction models that use regression or machine learning methods","volume":"385","author":"Collins","year":"2024","journal-title":"BMJ"},{"key":"ref25","doi-asserted-by":"publisher","first-page":"1351","DOI":"10.1038\/s41591-020-1037-7","article-title":"Guidelines for clinical trial protocols for interventions involving artificial intelligence: the SPIRIT-AI extension","volume":"26","author":"Cruz Rivera","year":"2020","journal-title":"Nat Med"},{"key":"ref26","doi-asserted-by":"publisher","first-page":"60","DOI":"10.1038\/s41591-024-03425-5","article-title":"The TRIPOD-LLM reporting guideline for studies using large language models","volume":"31","author":"Gallifant","year":"2025","journal-title":"Nat Med"},{"key":"ref27","doi-asserted-by":"publisher","first-page":"2011","DOI":"10.1093\/jamia\/ocaa088","article-title":"MINIMAR (MINimum information for medical AI reporting): developing reporting standards for artificial intelligence in health care","volume":"27","author":"Hernandez-Boussard","year":"2020","journal-title":"J Am Med Inform Assoc"},{"key":"ref28","doi-asserted-by":"publisher","first-page":"1364","DOI":"10.1038\/s41591-020-1034-x","article-title":"Reporting guidelines for clinical trial reports for interventions involving artificial intelligence: the CONSORT-AI extension","volume":"26","author":"Liu","year":"2020","journal-title":"Nat Med"},{"key":"ref29","doi-asserted-by":"publisher","first-page":"1320","DOI":"10.1038\/s41591-020-1041-y","article-title":"Minimum information about clinical artificial intelligence modeling: the MI-CLAIM checklist","volume":"26","author":"Norgeot","year":"2020","journal-title":"Nat Med"},{"key":"ref30","doi-asserted-by":"publisher","first-page":"e240300","DOI":"10.1148\/ryai.240300","article-title":"Checklist for artificial intelligence in medical imaging (CLAIM): 2024 update","volume":"6","author":"Tejani","year":"2024","journal-title":"Radiol Artif Intell"},{"key":"ref31","doi-asserted-by":"publisher","first-page":"e070904","DOI":"10.1136\/bmj-2022-070904","article-title":"Reporting guideline for the early stage clinical evaluation of decision support systems driven by artificial intelligence: DECIDE-AI","volume":"377","author":"Vasey","year":"2022","journal-title":"BMJ"},{"key":"ref32","doi-asserted-by":"publisher","first-page":"806","DOI":"10.1136\/bmj.39335.541782.AD","article-title":"Strengthening the reporting of observational studies in epidemiology (STROBE) statement: guidelines for reporting observational studies","volume":"335","author":"von Elm","year":"2007","journal-title":"BMJ"},{"key":"ref33","doi-asserted-by":"publisher","first-page":"AIp2401106","DOI":"10.1056\/aip2401106","article-title":"Development, evaluation, and assessment of large language models (DEAL) checklist: a technical report","volume":"2","author":"Tripathi","year":"2025","journal-title":"NEJM AI"},{"key":"ref34","doi-asserted-by":"publisher","first-page":"1760","DOI":"10.1513\/AnnalsATS.201604-259OC","article-title":"Epidemiology of vasopressin use for Adults with septic shock","volume":"13","author":"Vail","year":"2016","journal-title":"Ann Am Thorac Soc"},{"key":"ref35","doi-asserted-by":"publisher","first-page":"668","DOI":"10.1097\/SHK.0000000000002558","article-title":"Adjunctive vasopressors and short-term mortality in ADULTS with septic shock: a systematic review and META-analysis","volume":"63","author":"Bauer","year":"2025","journal-title":"Shock"},{"key":"ref36","doi-asserted-by":"publisher","first-page":"509","DOI":"10.1001\/jama.2016.10485","article-title":"Effect of early vasopressin vs norepinephrine on kidney failure in patients with septic shock: the VANISH randomized clinical trial","volume":"316","author":"Gordon","year":"2016","journal-title":"JAMA"},{"key":"ref37","doi-asserted-by":"publisher","first-page":"1743","DOI":"10.1097\/CCM.0000000000004023","article-title":"Vasopressin versus norepinephrine for the Management of Septic Shock in Cancer patients: the VANCS II randomized clinical trial","volume":"47","author":"Hajjar","year":"2019","journal-title":"Crit Care Med"},{"key":"ref38","doi-asserted-by":"publisher","first-page":"877","DOI":"10.1056\/NEJMoa067373","article-title":"Vasopressin versus norepinephrine infusion in patients with septic shock","volume":"358","author":"Russell","year":"2008","journal-title":"N Engl J Med"},{"key":"ref39","doi-asserted-by":"publisher","first-page":"1216","DOI":"10.1016\/j.chest.2023.07.009","article-title":"Optimizing vasopressin use and initiation timing in septic shock: a narrative review","volume":"164","author":"Sacha","year":"2023","journal-title":"Chest"}],"container-title":["Frontiers in Medicine"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fmed.2026.1787075\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,3,11]],"date-time":"2026-03-11T06:53:32Z","timestamp":1773212012000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fmed.2026.1787075\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,3,11]]},"references-count":39,"alternative-id":["10.3389\/fmed.2026.1787075"],"URL":"https:\/\/doi.org\/10.3389\/fmed.2026.1787075","relation":{},"ISSN":["2296-858X"],"issn-type":[{"value":"2296-858X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,3,11]]},"article-number":"1787075"}}