{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,29]],"date-time":"2026-09-29T19:15:21Z","timestamp":1790709321647,"version":"4.1.0"},"reference-count":40,"publisher":"JMIR Publications Inc.","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["JMIR Ment Health"],"abstract":"Abstract<\/jats:title>\n \n Background<\/jats:title>\n Patients\u2019 digital access to their personal health data is becoming increasingly common worldwide. However, medical documentation often contains technical language and sensitive information, which can lead to potential misunderstandings and distress among patients. These issues may be particularly impactful in mental health contexts. Large language models (LLMs) offer a promising approach by transforming clinician-generated health notes into language that is more patient-centered, nonmedicalized, and empathetic. However, risks related to accuracy and clinical safety have not been adequately investigated in psychiatry.<\/jats:p>\n <\/jats:sec>\n \n Objective<\/jats:title>\n This study aimed to qualitatively analyze the errors introduced by LLMs when transforming notes written by psychiatrists into patient-facing formats. It also highlights the implications for clinical communication and patient safety.<\/jats:p>\n <\/jats:sec>\n \n Methods<\/jats:title>\n Clinical notes (n=63) written by 19 psychiatrists in an outpatient treatment setting were collected, anonymized, and translated from German to English by humans. OpenAI GPT-3.5 Turbo was used to develop a preprompt that transformed these notes into a patient-centered, lay-readable form through an iterative process. Three psychiatrists qualitatively analyzed the LLM-revised documentation using Kuckartz content analysis. They compared the preconversion and postconversion notes to systematically identify and categorize LLM-induced errors.<\/jats:p>\n <\/jats:sec>\n \n Results<\/jats:title>\n Five categories of clinically relevant errors were identified: (1) clinical misinterpretations, particularly in critical assessments such as suicidality, where nuanced terminology was oversimplified or inaccurately represented; (2) attribution errors, where behaviors or roles within family dynamics or interactions were incorrectly attributed to different individuals; (3) content distortion errors, which were characterized by speculative additions, emotional exaggerations, and inappropriate contextual assumptions; (4) abbreviation and terminology errors, which resulted from inaccurate expansions of medical abbreviations and terms; and (5) structural and syntax errors, which resulted in ambiguity, particularly when the original notes were brief or bulleted. Despite significant improvements in the readability and overall linguistic fluency of the converted notes, these errors occurred.<\/jats:p>\n <\/jats:sec>\n \n Conclusions<\/jats:title>\n LLMs have the potential to transform psychiatric notes into patient-friendly formats. However, critical errors remain prevalent and can impair clinical judgment, understanding of patient circumstances, clarity of medication regimens, and interpretation of clinical observations. To safely integrate artificial intelligence\u2013generated documentation into psychiatric care, clinician oversight and targeted model refinement are essential. Future research should explore strategies to mitigate these errors, assess their comprehensive clinical impact, and incorporate patient and provider perspectives to ensure robust implementation.<\/jats:p>\n <\/jats:sec>","DOI":"10.2196\/78351","type":"journal-article","created":{"date-parts":[[2026,2,27]],"date-time":"2026-02-27T12:20:07Z","timestamp":1772194807000},"page":"e78351-e78351","source":"Crossref","is-referenced-by-count":6,"title":["Errors in AI-Transformed Patient-Centered Mental Health Documentation Written by Psychiatrists: Qualitative Pre-Post Study"],"prefix":"10.2196","volume":"13","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-9954-3544","authenticated-orcid":false,"given":"Pelin","family":"Ozkara Menekseoglu","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0002-8999-2431","authenticated-orcid":false,"given":"Mareike","family":"Weibezahl","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0001-0123-6299","authenticated-orcid":false,"given":"Mats","family":"Ellingsen","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0002-7698-9284","authenticated-orcid":false,"given":"Jarl","family":"Sterkenburg","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0908-6173","authenticated-orcid":false,"given":"Anna","family":"Kharko","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2652-135X","authenticated-orcid":false,"given":"Stefan","family":"Hochwarter","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7306-7909","authenticated-orcid":false,"given":"Julian","family":"Schwarz","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1010","published-online":{"date-parts":[[2026,4,29]]},"reference":[{"issue":"2","key":"R1","doi-asserted-by":"publisher","first-page":"121","DOI":"10.7326\/0003-4819-153-2-201007200-00008","article-title":"Open notes: doctors and patients signing on","volume":"153","author":"Delbanco","journal-title":"Ann Intern Med"},{"issue":"8","key":"R2","doi-asserted-by":"publisher","first-page":"572","DOI":"10.1136\/medethics-2021-107746","article-title":"Open notes in patient care: confining deceptive placebos to the past?","volume":"48","author":"Blease","journal-title":"J Med Ethics"},{"issue":"11","key":"R3","doi-asserted-by":"publisher","DOI":"10.2196\/jmir.3363","article-title":"Patients who share transparent visit notes with others: characteristics, risks, and benefits","volume":"16","author":"Jackson","journal-title":"J Med Internet Res"},{"issue":"e1","key":"R4","doi-asserted-by":"publisher","first-page":"e166","DOI":"10.1093\/jamia\/ocw108","article-title":"Inviting patients and care partners to read doctors\u2019 notes: OpenNotes and shared access to electronic medical records","volume":"24","author":"Wolff","journal-title":"J Am Med Inform Assoc"},{"key":"R5","doi-asserted-by":"publisher","DOI":"10.2196\/66690","article-title":"Experiences of older mental health patients and their care partners using a proxy account to access open notes: qualitative interview study","volume":"8","author":"Meier-Diedrich","journal-title":"JMIR Aging"},{"issue":"11","key":"R6","doi-asserted-by":"publisher","DOI":"10.1001\/jamanetworkopen.2022.42972","article-title":"Accuracy in patient understanding of common medical phrases","volume":"5","author":"Gotlieb","journal-title":"JAMA Netw Open"},{"key":"R7","doi-asserted-by":"publisher","DOI":"10.2196\/63713","article-title":"Patient-health care professional communication via a secure web-based portal in severe mental health conditions: qualitative analysis of secure messages","volume":"9","author":"Meier-Diedrich","journal-title":"JMIR Form Res"},{"issue":"2","key":"R8","doi-asserted-by":"publisher","first-page":"129","DOI":"10.1197\/jamia.m1147","article-title":"The effects of promoting patient access to medical records: a review","volume":"10","author":"Ross","journal-title":"J Am Med Inform Assoc"},{"issue":"12","key":"R9","doi-asserted-by":"publisher","DOI":"10.2196\/34170","article-title":"Sharing clinical notes and electronic health records with people affected by mental health conditions: scoping review","volume":"8","author":"Schwarz","journal-title":"JMIR Ment Health"},{"key":"R10","doi-asserted-by":"publisher","DOI":"10.3389\/fpsyt.2019.00917","article-title":"Easy access, difficult consequences? Providing psychiatric patients with access to their health records electronically","volume":"10","author":"Strudwick","journal-title":"Front Psychiatry"},{"key":"R11","doi-asserted-by":"publisher","DOI":"10.1177\/20552076241271813","article-title":"Psychotherapists\u2019 views on open notes: an online survey from Germany","volume":"10","author":"Schwarz","journal-title":"Digit Health"},{"issue":"4","key":"R12","doi-asserted-by":"publisher","first-page":"e179","DOI":"10.1016\/S2589-7500(23)00048-1","article-title":"Using ChatGPT to write patient clinic letters","volume":"5","author":"Ali","journal-title":"Lancet Digit Health"},{"issue":"1","key":"R13","doi-asserted-by":"publisher","DOI":"10.1186\/s12911-024-02459-6","article-title":"Assessing the research landscape and clinical utility of large language models: a scoping review","volume":"24","author":"Park","journal-title":"BMC Med Inform Decis Mak"},{"issue":"3","key":"R14","doi-asserted-by":"publisher","DOI":"10.1001\/jamanetworkopen.2024.0357","article-title":"Generative artificial intelligence to transform inpatient discharge summaries to patient-friendly language and format","volume":"7","author":"Zaretsky","journal-title":"JAMA Netw Open"},{"key":"R15","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijmedinf.2024.105654","article-title":"Large language models can support generation of standardized discharge summaries - a retrospective study utilizing ChatGPT-4 and electronic health records","volume":"192","author":"Schwieger","journal-title":"Int J Med Inform"},{"issue":"11","key":"R16","doi-asserted-by":"publisher","DOI":"10.1001\/jamanetworkopen.2023.43689","article-title":"Leveraging large language models for decision support in personalized oncology","volume":"6","author":"Benary","journal-title":"JAMA Netw Open"},{"key":"R17","doi-asserted-by":"publisher","DOI":"10.3389\/fmed.2024.1477898","article-title":"Large language models in patient education: a scoping review of applications in medicine","volume":"11","author":"Aydin","journal-title":"Front Med (Lausanne)"},{"issue":"2","key":"R18","doi-asserted-by":"publisher","DOI":"10.1093\/jamiaopen\/ooaf021","article-title":"A proof-of-concept study for patient use of open notes with large language models","volume":"8","author":"Salmi","journal-title":"JAMIA Open"},{"issue":"4","key":"R19","doi-asserted-by":"publisher","first-page":"860","DOI":"10.1093\/cid\/ciad633","article-title":"Black box warning: large language models and the future of infectious diseases consultation","volume":"78","author":"Schwartz","journal-title":"Clin Infect Dis"},{"issue":"Fall","key":"R20","volume":"10","author":"Bowman","journal-title":"Perspect Health Inf Manag"},{"issue":"11","key":"R21","doi-asserted-by":"publisher","DOI":"10.7759\/cureus.73994","article-title":"Evaluating the impact of artificial intelligence (AI) on clinical documentation efficiency and accuracy across clinical settings: a scoping review","volume":"16","author":"Lee","journal-title":"Cureus"},{"issue":"1","key":"R22","doi-asserted-by":"publisher","DOI":"10.1186\/s12888-024-05845-6","article-title":"Psychiatrists\u2019 perceptions of conditions and consequences associated with the implementation of open notes: qualitative investigation","volume":"24","author":"Schwarz","journal-title":"BMC Psychiatry"},{"key":"R23","doi-asserted-by":"publisher","DOI":"10.2196\/51183","article-title":"Generative language models and open notes: exploring the promise and limitations","volume":"10","author":"Blease","journal-title":"JMIR Med Educ"},{"key":"R24","unstructured":"Kuckartz U . Qualitative Inhaltsanalyse Methoden, Praxis, Computerunterst\u00fctzung. 4th ed. Beltz Juventa; 2018. ISBN: 978-3779936824"},{"key":"R25","unstructured":"Vereinbarung zu psychiatrischen institutsambulanzen gem\u00e4\u00df \u00a7 118 abs 2 SGB v [Report in German]. Kassen\u00e4rztliche Bundesvereinigung (KBV); 2010. URL: https:\/\/www.kbv.de\/documents\/infothek\/rechtsquellen\/weitere-vertraege\/praxen\/pia\/Psychiatrische_Institutsambulanzen.pdf [Accessed 07-04-2026]"},{"key":"R26","unstructured":"Neubert O Richter M . Krankenhaus\u2011Report 2016: ambulant im krankenhaus [Report in German]. Schattauer; 2016:63-84. URL: https:\/\/www.wido.de\/fileadmin\/Dateien\/Dokumente\/Publikationen_Produkte\/Buchreihen\/Krankenhausreport\/2016\/Kapitel%20mit%20Deckblatt\/wido_khr2016_kap04.pdf [Accessed 07-04-2026]"},{"key":"R27","unstructured":"Patton MQ . Qualitative Research and Evaluation Methods. Sage Publications; 2005. ISBN: 978-0761919711"},{"key":"R28","unstructured":"LibreTranslate software. URL: https:\/\/libretranslate.com [Accessed 11-04-2026]"},{"key":"R29","unstructured":"Kuckartz U R\u00e4diker S . Qualitative Content Analysis: Methods, Practice and Software. 2nd ed. SAGE Publications Ltd; 2023. ISBN: 9781529609141"},{"key":"R30","doi-asserted-by":"publisher","DOI":"10.1016\/j.mex.2023.102545","article-title":"K-alpha calculator-Krippendorff\u2019s alpha calculator: a user-friendly tool for computing Krippendorff\u2019s alpha inter-rater reliability coefficient","volume":"12","author":"Marzi","journal-title":"MethodsX"},{"key":"R31","doi-asserted-by":"publisher","DOI":"10.2196\/54419","article-title":"Using ChatGPT-4 to create structured medical notes from audio recordings of physician-patient encounters: comparative study","volume":"26","author":"Kernberg","journal-title":"J Med Internet Res"},{"issue":"1","key":"R32","doi-asserted-by":"publisher","DOI":"10.1038\/s41746-025-01670-7","article-title":"A framework to assess clinical safety and hallucination rates of LLMs for medical text summarisation","volume":"8","author":"Asgari","journal-title":"NPJ Digit Med"},{"issue":"2","key":"R33","doi-asserted-by":"publisher","DOI":"10.7759\/cureus.35179","article-title":"Artificial hallucinations in ChatGPT: implications in scientific writing","volume":"15","author":"Alkaissi","journal-title":"Cureus"},{"key":"R34","doi-asserted-by":"crossref","unstructured":"Williams CYK Bains J Tang T et al. Evaluating large language models for drafting emergency department discharge summaries. medRxiv. Preprint posted online on Apr 4, 2024. [doi: 10.1101\/2024.04.03.24305088]","DOI":"10.1101\/2024.04.03.24305088"},{"issue":"1","key":"R35","doi-asserted-by":"publisher","DOI":"10.1093\/bib\/bbad493","article-title":"Opportunities and challenges for ChatGPT and large language models in biomedicine and health","volume":"25","author":"Tian","journal-title":"Brief Bioinform"},{"issue":"1","key":"R36","doi-asserted-by":"publisher","first-page":"9","DOI":"10.1177\/10398562241280348","article-title":"The uses and misuses of artificial intelligence in psychiatry: promises and challenges","volume":"33","author":"Reutens","journal-title":"Australas Psychiatry"},{"key":"R37","doi-asserted-by":"crossref","unstructured":"Dignum V . Responsible Artificial Intelligence: How to Develop and Use AI in a Responsible Way. Springer; 2019. [doi: 10.1007\/978-3-030-30371-6]","DOI":"10.1007\/978-3-030-30371-6"},{"issue":"37","key":"R38","doi-asserted-by":"publisher","DOI":"10.1126\/scirobotics.aay7120","article-title":"XAI-explainable artificial intelligence","volume":"4","author":"Gunning","journal-title":"Sci Robot"},{"issue":"7","key":"R39","doi-asserted-by":"publisher","first-page":"1740","DOI":"10.1111\/jerd.13447","article-title":"Evaluating the accuracy, reliability, consistency, and readability of different large language models in restorative dentistry","volume":"37","author":"Ozdemir","journal-title":"J Esthet Restor Dent"},{"key":"R40","doi-asserted-by":"publisher","DOI":"10.2196\/53164","article-title":"Hallucination rates and reference accuracy of ChatGPT and bard for systematic reviews: comparative analysis","volume":"26","author":"Chelli","journal-title":"J Med Internet Res"}],"container-title":["JMIR Mental Health"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/mental.jmir.org\/2026\/1\/e78351","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,30]],"date-time":"2026-06-30T18:10:57Z","timestamp":1782843057000},"score":1,"resource":{"primary":{"URL":"https:\/\/mental.jmir.org\/2026\/1\/e78351"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,4,29]]},"references-count":40,"URL":"https:\/\/doi.org\/10.2196\/78351","relation":{},"ISSN":["2368-7959"],"issn-type":[{"value":"2368-7959","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,4,29]]},"article-number":"v13i2e78351"}}