{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,1]],"date-time":"2026-10-01T05:12:38Z","timestamp":1790831558745,"version":"4.1.0"},"reference-count":58,"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 Trauma exposure is highly prevalent and associated with various health issues. However, health care professionals can exhibit trauma-related diagnostic overshadowing bias, leading to misdiagnosis and inadequate treatment of trauma-exposed populations. Generative artificial intelligence (GAI) models are increasingly used in health care contexts. No research has examined whether GAI demonstrates this bias in decision-making and how rates of this bias may compare to mental health professionals (MHPs).<\/jats:p>\n <\/jats:sec>\n \n Objective<\/jats:title>\n This study aimed to assess trauma-related diagnostic overshadowing among frontier GAI models and compare evidence of trauma-related diagnostic overshadowing between frontier GAI models and MHPs.<\/jats:p>\n <\/jats:sec>\n \n Methods<\/jats:title>\n \n MHPs (N=232; mean [SD] age 43.7 [15.95] years) completed an experimental paradigm consisting of 2 vignettes describing adults presenting with obsessive-compulsive symptoms or substance abuse symptoms. One vignette included a trauma exposure history (ie, sexual trauma or physical trauma), and one vignette did not include a trauma exposure history. Participants answered questions about their preferences for diagnosis and treatment options for clients within the vignettes. GAI models (eg, Gemini 1.5 Flash, ChatGPT-4o mini, Claude Sonnet, and Meta Llama 3) completed the same experimental paradigm, with each block being reviewed by each GAI model 20 times. Mann-Whitney\n U<\/jats:italic>\n tests and chi-square analyses were used to assess diagnostic and treatment decision-making across vignette factors and respondents.\n <\/jats:p>\n <\/jats:sec>\n \n Results<\/jats:title>\n \n GAI models, similar to MHPs, demonstrated some evidence of trauma-related diagnostic overshadowing bias, particularly in Likert-based ratings of posttraumatic stress disorder diagnosis and treatment when sexual trauma was present (\n \n \n P\n <\/jats:italic>\n <.001). However, GAI models generally exhibited significantly less bias than MHPs across both Likert and forced-choice clinical decision tasks. Compared to MHPs, GAI models assigned higher ratings for the target diagnosis and treatment in obsessive-compulsive disorder vignettes (\n \n r\n b<\/jats:sub>\n <\/jats:italic>\n =0.43\u20100.63;\n P<\/jats:italic>\n <.001) and for the target treatment in substance use disorder vignettes (\n \n r\n b<\/jats:sub>\n <\/jats:italic>\n =0.57;\n P<\/jats:italic>\n <.001) when trauma was present. In forced-choice tasks, GAI models were significantly more accurate than MHPs in selecting the correct diagnosis and treatment for obsessive-compulsive disorder vignettes (\u03c7\u00b2\n 1<\/jats:sub>\n =48.84\u201061.07;\n P<\/jats:italic>\n <.001) and for substance use disorder vignettes involving sexual trauma (\u03c7\u00b2\n 1<\/jats:sub>\n =15.17\u2010101.61;\n P<\/jats:italic>\n <.001).\n <\/jats:p>\n <\/jats:sec>\n \n Conclusions<\/jats:title>\n GAI models demonstrate some evidence of trauma-related diagnostic overshadowing bias, yet the degree of bias varied by task and model. Moreover, GAI models generally demonstrated less bias than MHPs in this experimental paradigm. These findings highlight the importance of understanding GAI biases in mental health care. More research into bias reduction strategies and responsible implementation of GAI models in mental health care is needed.<\/jats:p>\n <\/jats:sec>","DOI":"10.2196\/80801","type":"journal-article","created":{"date-parts":[[2025,9,10]],"date-time":"2025-09-10T12:55:07Z","timestamp":1757508907000},"page":"e80801-e80801","source":"Crossref","is-referenced-by-count":6,"title":["Comparing Generative Artificial Intelligence and Mental Health Professionals for Clinical Decision-Making With Trauma-Exposed Populations: Vignette-Based Experimental Study"],"prefix":"10.2196","volume":"12","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-0821-1622","authenticated-orcid":false,"given":"Katherine E","family":"Wislocki","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0000-1687-3496","authenticated-orcid":false,"given":"Sabahat","family":"Sami","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0000-6339-5706","authenticated-orcid":false,"given":"Gahl","family":"Liberzon","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5159-8431","authenticated-orcid":false,"given":"Alyson K","family":"Zalta","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1010","published-online":{"date-parts":[[2025,10,14]]},"reference":[{"issue":"6","key":"R1","doi-asserted-by":"publisher","first-page":"378","DOI":"10.1177\/1078390311425187","article-title":"Physical health problems after single trauma exposure: when stress takes root in the body","volume":"17","author":"D\u2019Andrea","journal-title":"J Am Psychiatr Nurses Assoc"},{"issue":"1","key":"R2","doi-asserted-by":"publisher","DOI":"10.1186\/s12888-023-04716-w","article-title":"Lifetime posttraumatic stress disorder as a predictor of mortality: a systematic review and meta-analysis","volume":"23","author":"Nilaweera","journal-title":"BMC Psychiatry"},{"issue":"5","key":"R3","doi-asserted-by":"publisher","first-page":"537","DOI":"10.1002\/jts.21848","article-title":"National estimates of exposure to traumatic events and PTSD prevalence using DSM-IV and DSM-5 criteria","volume":"26","author":"Kilpatrick","journal-title":"J Trauma Stress"},{"issue":"1","key":"R4","doi-asserted-by":"publisher","first-page":"46","DOI":"10.1034\/j.1600-0447.2000.101001046.x","article-title":"Traumatic events and post-traumatic stress disorder in the community: prevalence, risk factors and comorbidity","volume":"101","author":"Perkonigg","journal-title":"Acta Psychiatr Scand"},{"key":"R5","doi-asserted-by":"crossref","unstructured":"American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders: DSM-5. 5th ed ed. American Psychiatric Association; 2013. [doi: 10.1176\/appi.books.9780890425596]","DOI":"10.1176\/appi.books.9780890425596"},{"key":"R6","unstructured":"Substance Abuse and Mental Health Services Administration (SAMHSA). Trauma-informed care in behavioral health services. In: Substance Abuse and Mental Health Services Administration (US). 2014. URL: http:\/\/www.ncbi.nlm.nih.gov\/books\/NBK207201"},{"issue":"2","key":"R7","doi-asserted-by":"publisher","first-page":"184","DOI":"10.1080\/15374416.2021.1923020","article-title":"Youth trauma histories are associated with under-diagnosis and under-treatment of co-occurring youth psychiatric symptoms","volume":"52","author":"Becker-Haimes","journal-title":"J Clin Child Adolesc Psychol"},{"issue":"8","key":"R8","doi-asserted-by":"publisher","first-page":"1367","DOI":"10.1037\/tra0001484","article-title":"Assessing the existence of trauma-related diagnostic overshadowing in adult populations","volume":"16","author":"Wislocki","journal-title":"Psychol Trauma"},{"key":"R9","doi-asserted-by":"publisher","DOI":"10.3389\/fpsyt.2023.1285431","article-title":"It\u2019s not in my head: a qualitative analysis of experiences of discrimination in people with mental health and substance use conditions seeking physical healthcare","volume":"14","author":"Cunningham","journal-title":"Front Psychiatry"},{"issue":"11","key":"R10","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0111682","article-title":"Diagnostic overshadowing and other challenges involved in the diagnostic process of patients with mental illness who present in emergency departments with physical symptoms--a qualitative study","volume":"9","author":"Shefer","journal-title":"PLoS ONE"},{"issue":"7","key":"R11","doi-asserted-by":"publisher","DOI":"10.2196\/27370","article-title":"Diagnostic accuracy of artificial intelligence and computer-aided diagnosis for the detection and characterization of colorectal polyps: systematic review and meta-analysis","volume":"23","author":"Nazarian","journal-title":"J Med Internet Res"},{"key":"R12","doi-asserted-by":"publisher","DOI":"10.3389\/fonc.2022.871871","article-title":"Artificial intelligence-based automated treatment planning of postmastectomy volumetric modulated arc radiotherapy","volume":"12","author":"Jiang","journal-title":"Front Oncol"},{"key":"R13","doi-asserted-by":"publisher","DOI":"10.1177\/1533033819873922","article-title":"Artificial intelligence in radiotherapy treatment planning: present and future","volume":"18","author":"Wang","journal-title":"Technol Cancer Res Treat"},{"key":"R14","doi-asserted-by":"publisher","DOI":"10.1016\/j.invent.2022.100495","article-title":"Using AI chatbots to provide self-help depression interventions for university students: a randomized trial of effectiveness","volume":"27","author":"Liu","journal-title":"Internet Interv"},{"key":"R15","doi-asserted-by":"publisher","DOI":"10.2196\/53672","article-title":"Debate and dilemmas regarding generative AI in mental health care: scoping review","volume":"13","author":"Xian","journal-title":"Interact J Med Res"},{"key":"R16","unstructured":"Lai T Shi Y Du Z et al. Psy-LLM: scaling up global mental health psychological services with AI-based large language models. arXiv. Preprint posted online on Sep 1, 2023. [doi: 10.48550\/arXiv.2307.11991]"},{"issue":"10","key":"R17","doi-asserted-by":"publisher","DOI":"10.1001\/jamanetworkopen.2024.40969","article-title":"Large language model influence on diagnostic reasoning: a randomized clinical trial","volume":"7","author":"Goh","journal-title":"JAMA Netw Open"},{"key":"R18","doi-asserted-by":"publisher","DOI":"10.1101\/2023.11.24.23298844","article-title":"ChatGPT influence on medical decision-making, bias, and equity: a randomized study of clinicians evaluating clinical vignettes","author":"Goh","journal-title":"medRxiv"},{"key":"R19","doi-asserted-by":"publisher","DOI":"10.2196\/53043","article-title":"Comparing the perspectives of generative AI, mental health experts, and the general public on schizophrenia recovery: case vignette study","volume":"11","author":"Elyoseph","journal-title":"JMIR Ment Health"},{"issue":"1","key":"R20","doi-asserted-by":"publisher","DOI":"10.1038\/s41746-024-01181-x","article-title":"Large language models outperform mental and medical health care professionals in identifying obsessive-compulsive disorder","volume":"7","author":"Kim","journal-title":"NPJ Digit Med"},{"issue":"2","key":"R21","doi-asserted-by":"publisher","DOI":"10.1002\/pmh.70018","article-title":"Unveiling public stigma for borderline personality disorder: a comparative study of artificial intelligence and mental health care providers","volume":"19","author":"Lauderdale","journal-title":"Personal Ment Health"},{"issue":"4","key":"R22","doi-asserted-by":"publisher","DOI":"10.1136\/fmch-2023-002391","article-title":"Identifying depression and its determinants upon initiating treatment: ChatGPT versus primary care physicians","volume":"11","author":"Levkovich","journal-title":"Fam Med Community Health"},{"issue":"4","key":"R23","doi-asserted-by":"publisher","DOI":"10.4088\/JCP.24m15365","article-title":"Integrating previous suicide attempts, gender, and age into suicide risk assessment using advanced artificial intelligence models","volume":"85","author":"Shinan-Altman","journal-title":"J Clin Psychiatry"},{"issue":"6464","key":"R24","doi-asserted-by":"publisher","first-page":"447","DOI":"10.1126\/science.aax2342","article-title":"Dissecting racial bias in an algorithm used to manage the health of populations","volume":"366","author":"Obermeyer","journal-title":"Science"},{"issue":"5","key":"R25","doi-asserted-by":"publisher","first-page":"1062","DOI":"10.1177\/17456916221134490","article-title":"A call to action on assessing and mitigating bias in artificial intelligence applications for mental health","volume":"18","author":"Timmons","journal-title":"Perspect Psychol Sci"},{"key":"R26","doi-asserted-by":"publisher","DOI":"10.2196\/60589","article-title":"Use of AI in mental health care: community and mental health professionals survey","volume":"11","author":"Cross","journal-title":"JMIR Ment Health"},{"issue":"12","key":"R27","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0240376","article-title":"Artificial Intelligence in mental health and the biases of language based models","volume":"15","author":"Straw","journal-title":"PLoS ONE"},{"key":"R28","doi-asserted-by":"publisher","DOI":"10.2196\/50696","article-title":"Ethical challenges in AI approaches to eating disorders","volume":"25","author":"Sharp","journal-title":"J Med Internet Res"},{"issue":"1","key":"R29","doi-asserted-by":"publisher","DOI":"10.2196\/70610","article-title":"The application and ethical implication of generative AI in mental health: systematic review","volume":"12","author":"Wang","journal-title":"JMIR Ment Health"},{"issue":"1","key":"R30","doi-asserted-by":"publisher","first-page":"149","DOI":"10.1038\/s43856-022-00214-4","article-title":"Mitigating the impact of biased artificial intelligence in emergency decision-making","volume":"2","author":"Adam","journal-title":"Commun Med (Lond)"},{"key":"R31","unstructured":"Grattafiori A Dubey A Jauhri A et al. The llama 3 herd of models. arXiv. Preprint posted online on Nov 23, 2024. [doi: 10.48550\/arXiv.2407.21783]"},{"key":"R32","unstructured":"OpenAI AJ Adler S et al. GPT-4 technical report. arXiv. Preprint posted online on Mar 4, 2024. [doi: 10.48550\/arXiv.2303.08774]"},{"key":"R33","unstructured":"Gemini AR Borgeaud S et al. Gemini: a family of highly capable multimodal models. arXiv. Preprint posted online on Jun 17, 2024. [doi: 10.48550\/arXiv.2312.11805]"},{"key":"R34","unstructured":"The claude 3 model family: opus, sonnet, haiku. Anthropic. 2024. URL: https:\/\/www-cdn.anthropic.com\/de8ba9b01c9ab7cbabf5c33b80b7bbc618857627\/Model_Card_Claude_3.pdf [Accessed 02-10-2025]"},{"issue":"3","key":"R35","doi-asserted-by":"publisher","first-page":"57","DOI":"10.3390\/informatics11030057","article-title":"Large Language Models in Healthcare and Medical Domain: A Review","volume":"11","author":"Nazi","journal-title":"Informatics (MDPI)"},{"key":"R36","unstructured":"White C Dooley S Roberts M et al. LiveBench: a challenging, contamination-limited LLM benchmark. arXiv. Preprint posted online on Apr 18, 2025. [doi: 10.48550\/arXiv.2406.19314]"},{"issue":"8","key":"R37","doi-asserted-by":"publisher","first-page":"e555","DOI":"10.1016\/S2589-7500(24)00097-9","article-title":"The diagnostic and triage accuracy of the GPT-3 artificial intelligence model: an observational study","volume":"6","author":"Levine","journal-title":"Lancet Digit Health"},{"issue":"1","key":"R38","doi-asserted-by":"publisher","DOI":"10.1038\/s41746-024-01283-6","article-title":"Evaluation and mitigation of cognitive biases in medical language models","volume":"7","author":"Schmidgall","journal-title":"NPJ Digit Med"},{"issue":"10","key":"R39","doi-asserted-by":"publisher","first-page":"6509","DOI":"10.1097\/JS9.0000000000001850","article-title":"The performance of large language model-powered chatbots compared to oncology physicians on colorectal cancer queries","volume":"110","author":"Zhou","journal-title":"Int J Surg"},{"key":"R40","unstructured":"Bommasani R Hudson DA Adeli E et al. On the opportunities and risks of foundation models. arXiv. Preprint posted online on Jul 12, 2022. [doi: 10.48550\/arXiv.2108.07258]"},{"issue":"2","key":"R41","doi-asserted-by":"publisher","first-page":"297","DOI":"10.1177\/001316447603600207","article-title":"Critical values of the rank-biserial correlation coefficient","volume":"36","author":"Willson","journal-title":"Educ Psychol Meas"},{"key":"R42","doi-asserted-by":"crossref","unstructured":"Cohen J . Statistical Power Analysis for the Behavioral Sciences. 2nd ed. Routledge; 2013. [doi: 10.4324\/9780203771587]","DOI":"10.4324\/9780203771587"},{"key":"R43","unstructured":"R Core Team. R: a language and environment for statistical computing (4.3.1). R Foundation for Statistical Computing. 2023. URL: https:\/\/www.r-project.org\/ [Accessed 02-10-2025]"},{"issue":"11","key":"R44","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pdig.0000651","article-title":"Bias in medical AI: implications for clinical decision-making","volume":"3","author":"Cross","journal-title":"PLOS Digit Health"},{"issue":"12","key":"R45","doi-asserted-by":"publisher","DOI":"10.1056\/AIcs2400639","article-title":"Cognitive biases and artificial intelligence","volume":"1","author":"Wang","journal-title":"NEJM AI"},{"key":"R46","doi-asserted-by":"publisher","DOI":"10.1016\/j.cmpb.2024.108013","article-title":"ChatGPT in healthcare: a taxonomy and systematic review","volume":"245","author":"Li","journal-title":"Comput Methods Programs Biomed"},{"key":"R47","unstructured":"Huang Y Bie B Na Z et al. An empirical study of the anchoring effect in llms: existence, mechanism, and potential mitigations. arXiv. Preprint posted online on May 21, 2025. [doi: 10.48550\/arXiv.2505.15392]"},{"issue":"1","key":"R48","doi-asserted-by":"publisher","DOI":"10.1038\/s41746-025-01512-6","article-title":"Assessing and alleviating state anxiety in large language models","volume":"8","author":"Ben-Zion","journal-title":"NPJ Digit Med"},{"issue":"sup5","key":"R49","doi-asserted-by":"publisher","DOI":"10.1080\/20008198.2017.1353383","article-title":"Trauma and PTSD in the WHO world mental health surveys","volume":"8","author":"Kessler","journal-title":"Eur J Psychotraumatol"},{"issue":"1","key":"R50","doi-asserted-by":"publisher","DOI":"10.1038\/s44184-024-00067-w","article-title":"Behavioral health and generative AI: a perspective on future of therapies and patient care","volume":"3","author":"Sezgin","journal-title":"Npj Ment Health Res"},{"issue":"1","key":"R51","doi-asserted-by":"publisher","DOI":"10.1038\/s41746-025-01460-1","article-title":"Transforming diagnosis through artificial intelligence","volume":"8","author":"D\u2019Adderio","journal-title":"NPJ Digit Med"},{"issue":"2","key":"R52","doi-asserted-by":"publisher","first-page":"221","DOI":"10.1038\/s42256-024-00976-7","article-title":"What large language models know and what people think they know","volume":"7","author":"Steyvers","journal-title":"Nat Mach Intell"},{"issue":"1","key":"R53","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-023-42384-8","article-title":"Humans inherit artificial intelligence biases","volume":"13","author":"Vicente","journal-title":"Sci Rep"},{"key":"R54","doi-asserted-by":"crossref","unstructured":"Pan L Saxon M Xu W Nathani D Wang X Wang WY . Automatically correcting large language models: surveying the landscape of diverse automated correction strategies. Trans Assoc Comput Linguist. May 3, 2024;12:484-506. [doi: 10.1162\/tacl_a_00660]","DOI":"10.1162\/tacl_a_00660"},{"issue":"1","key":"R55","doi-asserted-by":"publisher","DOI":"10.1038\/s41746-023-00913-9","article-title":"Considerations for addressing bias in artificial intelligence for health equity","volume":"6","author":"Abr\u00e0moff","journal-title":"NPJ Digit Med"},{"issue":"9","key":"R56","doi-asserted-by":"publisher","first-page":"1324","DOI":"10.1080\/0144929X.2022.2072768","article-title":"Measuring user competence in using artificial intelligence: validity and reliability of artificial intelligence literacy scale","volume":"42","author":"Wang","journal-title":"Behav Inf Technol"},{"issue":"1","key":"R57","doi-asserted-by":"publisher","DOI":"10.1038\/s41746-022-00737-z","article-title":"AI in the hands of imperfect users","volume":"5","author":"Kostick-Quenet","journal-title":"NPJ Digit Med"},{"key":"R58","doi-asserted-by":"publisher","DOI":"10.3389\/fpsyt.2023.1277756","article-title":"ChatGPT is not ready yet for use in providing mental health assessment and interventions","volume":"14","author":"Dergaa","journal-title":"Front Psychiatry"}],"container-title":["JMIR Mental Health"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/mental.jmir.org\/2025\/1\/e80801","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,30]],"date-time":"2026-06-30T18:14:27Z","timestamp":1782843267000},"score":1,"resource":{"primary":{"URL":"https:\/\/mental.jmir.org\/2025\/1\/e80801"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,10,14]]},"references-count":58,"URL":"https:\/\/doi.org\/10.2196\/80801","relation":{},"ISSN":["2368-7959"],"issn-type":[{"value":"2368-7959","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,10,14]]},"article-number":"v12i6e80801"}}