{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,14]],"date-time":"2026-10-14T05:16:53Z","timestamp":1791955013479,"version":"4.3.4"},"reference-count":54,"publisher":"JMIR Publications Inc.","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["J Med Internet Res"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec sec-type=\"background\">\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Large language models (LLMs) are increasingly applied in health care. However, concerns remain that their nursing care recommendations may reflect patients\u2019 sociodemographic attributes rather than clinical needs. While this risk is acknowledged, there is a lack of empirical evidence evaluating sociodemographic bias in LLM-generated nursing care plans.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec sec-type=\"objective\">\n                    <jats:title>Objective<\/jats:title>\n                    <jats:p>To investigate potential biases in nursing care plans generated by LLMs, we focused on whether outputs differ systematically based on patients\u2019 sociodemographic characteristics and assessed the implications for equitable nursing care.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec sec-type=\"methods\">\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>We used a mixed methods simulation study. A standardized clinical vignette experiment was used to prompt GPT-4 to generate 9600 nursing care plans for 96 patient profiles with varying sociodemographic characteristics (eg, sex, age, income, education, and residence). We first conducted a quantitative analysis of all plans, assessing variations in thematic content. Subsequently, a panel of senior nursing experts evaluated the clinical quality (eg, safety, applicability, and completeness) of a stratified subsample of 500 plans.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec sec-type=\"results\">\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>\n                      We analyzed 9600 LLM-generated nursing care plans and identified 8 consistent themes. Communication and Education (99.98%) and Emotional Support (99.97%) were nearly universal, while Nurse Training and Event Analysis were least frequent (39.3%). Multivariable analyses revealed systematic sociodemographic disparities. Care plans generated for low-income patient profiles were less likely to include the theme Environmental Adjustment (adjusted relative risk [aRR] 0.90). Profiles with lower education were associated with an increased likelihood of including Family Support (aRR 1.10). Similarly, plans generated for older patient profiles were more likely to contain recommendations for Pain Management (aRR 1.33) and Family Support (aRR 1.62) but were less likely to mention Nurse Training (aRR 0.78). Sex and regional differences were also significant. Expert review of 500 plans showed high overall quality (mean 4.47), with strong interrater reliability (\u03ba=0.76\u20100.81). However, urban profiles had higher completeness (\n                      <jats:italic>\u03b2<\/jats:italic>\n                      =.22) and applicability (\n                      <jats:italic>\u03b2<\/jats:italic>\n                      =.14) but lower safety scores (\u03b2=\u20130.09). These findings demonstrate that LLM-generated care plans exhibit systematic sociodemographic bias, raising important implications for fairness and safe deployment in nursing practice.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec sec-type=\"conclusions\">\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>This study identified that LLMs systematically reproduce sociodemographic biases in the generation of nursing care plans. These biases appear in two forms: they shape the thematic content and influence expert-rated clinical quality. These findings reveal a substantial risk that such models may reinforce existing health inequities. To our knowledge, this is the first empirical evidence documenting these nuanced biases in nursing. The study also contributes a replicable framework for evaluating LLM-generated care plans. Finally, it underscores the critical need for robust human oversight to ensure that artificial intelligence serves as a tool for advancing equity rather than perpetuating disparities.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.2196\/78132","type":"journal-article","created":{"date-parts":[[2025,12,8]],"date-time":"2025-12-08T10:23:13Z","timestamp":1765189393000},"page":"e78132-e78132","source":"Crossref","is-referenced-by-count":8,"title":["Detecting Sociodemographic Biases in the Content and Quality of Large Language Model\u2013Generated Nursing Care: Cross-Sectional Simulation Study"],"prefix":"10.2196","volume":"27","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-8655-5888","authenticated-orcid":false,"given":"Nan","family":"Bai","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0002-0951-6823","authenticated-orcid":false,"given":"Yijing","family":"Yu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0001-8283-9015","authenticated-orcid":false,"given":"Chunyan","family":"Luo","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7793-6876","authenticated-orcid":false,"given":"Si Chen","family":"Zhou","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0008-8091-3809","authenticated-orcid":false,"given":"Qing","family":"Wang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5656-5364","authenticated-orcid":false,"given":"Huijing","family":"Zou","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6833-6131","authenticated-orcid":false,"given":"Qian","family":"Liu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6391-5983","authenticated-orcid":false,"given":"Guanghui","family":"Fu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0009-1762-9570","authenticated-orcid":false,"given":"Wei","family":"Zhai","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9570-9546","authenticated-orcid":false,"given":"Qing","family":"Zhao","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1995-9249","authenticated-orcid":false,"given":"Jianqiang","family":"Li","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2986-012X","authenticated-orcid":false,"given":"Xinni","family":"Wei","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0227-4342","authenticated-orcid":false,"given":"Bing Xiang","family":"Yang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1010","published-online":{"date-parts":[[2025,12,5]]},"reference":[{"issue":"1","key":"R1","doi-asserted-by":"publisher","DOI":"10.1186\/s13012-024-01357-9","article-title":"Generative AI in healthcare: an implementation science informed translational path on application, integration and governance","volume":"19","author":"Reddy","journal-title":"Implement Sci"},{"key":"R2","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijnurstu.2021.104153","article-title":"Artificial Intelligence-based technologies in nursing: a scoping literature review of the evidence","volume":"127","author":"von Gerich","journal-title":"Int J Nurs Stud"},{"key":"R3","unstructured":"Vaswani A Shazeer N Parmar N et al. Attention is all you need. In: Adv Neural Inf Process Syst. 2017. URL: https:\/\/proceedings.neurips.cc\/paper\/2017\/hash\/3f5ee243547dee91fbd053c1c4a845aa-Abstract.html [Accessed 01-10-2025]"},{"key":"R4","doi-asserted-by":"publisher","DOI":"10.1016\/j.ajem.2024.02.008","article-title":"Human intelligence versus Chat-GPT: who performs better in correctly classifying patients in triage?","volume":"79","author":"Zaboli","journal-title":"Am J Emerg Med"},{"issue":"6","key":"R5","doi-asserted-by":"publisher","first-page":"e504","DOI":"10.1097\/BPO.0000000000002682","article-title":"Are generative pretrained transformer 4 responses to developmental dysplasia of the hip clinical scenarios universal? An international review","volume":"44","author":"Luo","journal-title":"J Pediatr Orthop"},{"issue":"3","key":"R6","doi-asserted-by":"publisher","first-page":"206","DOI":"10.1097\/NCQ.0000000000000766","article-title":"Evaluating the quality of postpartum hemorrhage nursing care plans generated by artificial intelligence models","volume":"39","author":"Karacan","journal-title":"J Nurs Care Qual"},{"key":"R7","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijnsa.2024.100181","article-title":"Comparing ChatGPT and clinical nurses\u2019 performances on tracheostomy care: a cross-sectional study","volume":"6","author":"Wang","journal-title":"Int J Nurs Stud Adv"},{"key":"R8","doi-asserted-by":"publisher","DOI":"10.2196\/63188","article-title":"Comparing the accuracy of two generated large language models in identifying health-related rumors or misconceptions and the applicability in health science popularization: proof-of-concept study","volume":"8","author":"Luo","journal-title":"JMIR Form Res"},{"issue":"10","key":"R9","doi-asserted-by":"publisher","DOI":"10.7759\/cureus.71447","article-title":"Assessing the role of artificial intelligence in the creation of patient educational videos for corneal refractive surgery","volume":"16","author":"Han","journal-title":"Cureus"},{"issue":"11","key":"R10","doi-asserted-by":"publisher","DOI":"10.1212\/WNL.0000000000209497","article-title":"Implications of large language models for quality and efficiency of neurologic care","volume":"102","author":"Moura","journal-title":"Neurology (ECronicon)"},{"issue":"6","key":"R11","doi-asserted-by":"publisher","DOI":"10.7759\/cureus.40542","article-title":"Harnessing large language models in nursing care planning: opportunities, challenges, and ethical considerations","volume":"15","author":"Nashwan","journal-title":"Cureus"},{"issue":"6","key":"R12","doi-asserted-by":"publisher","first-page":"633","DOI":"10.1016\/j.pmn.2019.03.004","article-title":"Is there gender discrimination in acute renal colic pain management? A retrospective analysis in an emergency department setting","volume":"20","author":"Naamany","journal-title":"Pain Manag Nurs"},{"issue":"6","key":"R13","doi-asserted-by":"publisher","DOI":"10.1111\/ijn.12779","article-title":"Do surgical patients\u2019 characteristics and behaviours affect nurses\u2019 pain management decisions? A qualitative inquiry","volume":"25","author":"Shoqirat","journal-title":"Int J Nurs Pract"},{"key":"R14","doi-asserted-by":"publisher","DOI":"10.1016\/j.nedt.2023.105739","article-title":"Ageism and nursing students, past or reality?: a systematic review","volume":"122","author":"Allu\u00e9-Sierra","journal-title":"Nurse Educ Today"},{"key":"R15","doi-asserted-by":"publisher","DOI":"10.1016\/j.arr.2024.102212","article-title":"Breaking the links between ageism and health: an integrated perspective","volume":"95","author":"Henry","journal-title":"Ageing Res Rev"},{"issue":"10077","key":"R16","doi-asserted-by":"publisher","first-page":"1431","DOI":"10.1016\/S0140-6736(17)30398-7","article-title":"Inequality and the health-care system in the USA","volume":"389","author":"Dickman","journal-title":"The Lancet"},{"key":"R17","doi-asserted-by":"publisher","DOI":"10.1016\/j.ajo.2024.07.039","article-title":"Artificial intelligence and ophthalmic clinical registries","volume":"268","author":"Tran","journal-title":"Am J Ophthalmol"},{"issue":"1","key":"R18","doi-asserted-by":"publisher","first-page":"79","DOI":"10.1111\/jpm.12965","article-title":"Could artificial intelligence write mental health nursing care plans?","volume":"31","author":"Woodnutt","journal-title":"J Psychiatr Ment Health Nurs"},{"key":"R19","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijnurstu.2024.104753","article-title":"A rapid review on current and potential uses of large language models in nursing","volume":"154","author":"Hobensack","journal-title":"Int J Nurs Stud"},{"issue":"9","key":"R20","doi-asserted-by":"publisher","first-page":"46","DOI":"10.1097\/01.NURSE.0000577716.57052.8d","article-title":"Artificial intelligence: essentials for nursing","volume":"49","author":"McGrow","journal-title":"Nursing (Lond)"},{"key":"R21","doi-asserted-by":"publisher","DOI":"10.1016\/j.ebiom.2023.104467","article-title":"Algorithmic encoding of protected characteristics in chest X-ray disease detection models","volume":"89","author":"Glocker","journal-title":"EBioMedicine"},{"key":"R22","unstructured":"Zhang H Dullerud N Roth K Oakden-Rayner L Pfohl SR Ghassemi M . Improving the fairness of chest x-ray classifiers. arXiv. Preprint posted online on Mar 23, 2022. [doi: 10.48550\/arXiv.2203.12609]"},{"key":"R23","doi-asserted-by":"crossref","unstructured":"Raizman E Peng Y Deng A . Towards equitable diagnosis: bias evaluation and mitigation in skin cancer classification. Presented at: 2024 IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology (CIBCB). Aug 27-29, 2024:IEEE. 1-9; Natal, Brazil. [doi: 10.1109\/CIBCB58642.2024.10702173]","DOI":"10.1109\/CIBCB58642.2024.10702173"},{"issue":"2","key":"R24","doi-asserted-by":"publisher","first-page":"239","DOI":"10.1177\/08901171221121639","article-title":"Racial fairness in precision medicine: pediatric asthma prediction algorithms","volume":"37","author":"Pennington","journal-title":"Am J Health Promot"},{"issue":"2","key":"R25","doi-asserted-by":"publisher","first-page":"e117","DOI":"10.1097\/CCM.0000000000005261","article-title":"Association of socioeconomic status with postdischarge pediatric resource use and quality of life","volume":"50","author":"Kachmar","journal-title":"Crit Care Med"},{"issue":"4","key":"R26","doi-asserted-by":"publisher","first-page":"285","DOI":"10.1016\/j.clml.2024.11.011","article-title":"Effects of socioeconomic status and healthcare resource availability on survival in older (\u226566 years) non-Hispanic Black patients versus non-Hispanic White patients with multiple myeloma","volume":"25","author":"Li","journal-title":"Clin Lymphoma Myeloma Leuk"},{"key":"R27","doi-asserted-by":"publisher","DOI":"10.1016\/j.socscimed.2024.117163","article-title":"Impact of medical insurance integration on reducing urban-rural health disparity: evidence from China","volume":"357","author":"Tang","journal-title":"Soc Sci Med"},{"issue":"6654","key":"R28","doi-asserted-by":"publisher","first-page":"150","DOI":"10.1126\/science.adh2713","article-title":"Mitigating bias in AI at the point of care","volume":"381","author":"DeCamp","journal-title":"Science"},{"issue":"5","key":"R29","doi-asserted-by":"publisher","first-page":"32","DOI":"10.16088\/j.issn.1001-6597.2021.05.003","article-title":"Does \u201cdiscrimination\u201d exist in the doctor-patient relationship? An experimental investigation on \u201cbehavior\u201d","volume":"57","author":"Guo","journal-title":"J Guangxi Norm Univ"},{"issue":"1","key":"R30","doi-asserted-by":"publisher","DOI":"10.1186\/s12910-017-0179-8","article-title":"Implicit bias in healthcare professionals: a systematic review","volume":"18","author":"FitzGerald","journal-title":"BMC Med Ethics"},{"issue":"9650","key":"R31","doi-asserted-by":"publisher","first-page":"1661","DOI":"10.1016\/S0140-6736(08)61690-6","article-title":"Closing the gap in a generation: health equity through action on the social determinants of health","volume":"372","author":"Marmot","journal-title":"The Lancet"},{"issue":"7","key":"R32","doi-asserted-by":"publisher","first-page":"653","DOI":"10.1089\/152460902760360595","article-title":"Gender bias in physicians\u2019 management of neck pain: a study of the answers in a swedish national examination","volume":"11","author":"Hamberg","journal-title":"J Womens Health Gend Based Med"},{"key":"R33","unstructured":"Tyson A Pasquini G Spencer A Funk C . 60% of americans would be uncomfortable with provider relying on AI in their own health care. Pew Research Center. Jan 2, 2023. URL: https:\/\/www.pewresearch.org\/science\/2023\/02\/22\/60-of-americans-would-be-uncomfortable-with-provider-relying-on-ai-in-their-own-health-care\/ [Accessed 05-11-2025]"},{"issue":"2","key":"R34","doi-asserted-by":"publisher","DOI":"10.1111\/opn.12226","article-title":"A systematic review of patient participation during bedside handovers on wards with older patients indicates evidence is urgently needed","volume":"14","author":"Malfait","journal-title":"Int J Older People Nurs"},{"issue":"1","key":"R35","doi-asserted-by":"publisher","DOI":"10.1186\/s12939-024-02261-w","article-title":"Rural-urban differences in health service utilization in upper-middle and high-income countries: a scoping review","volume":"23","author":"Liepins","journal-title":"Int J Equity Health"},{"issue":"1","key":"R36","doi-asserted-by":"publisher","first-page":"47","DOI":"10.1007\/s44155-024-00108-x","article-title":"Healthcare access and deprivation in low-income urban households","volume":"4","author":"Mazenda","journal-title":"Discov Soc Sci Health"},{"issue":"1","key":"R37","doi-asserted-by":"publisher","DOI":"10.2196\/46888","article-title":"Digital health literacy and its association with sociodemographic characteristics, health resource use, and health outcomes: rapid review","volume":"13","author":"Yuen","journal-title":"Interact J Med Res"},{"issue":"1","key":"R38","doi-asserted-by":"publisher","first-page":"e12","DOI":"10.1016\/S2589-7500(23)00225-X","article-title":"Assessing the potential of GPT-4 to perpetuate racial and gender biases in health care: a model evaluation study","volume":"6","author":"Zack","journal-title":"Lancet Digit Health"},{"issue":"6","key":"R39","doi-asserted-by":"publisher","first-page":"1873","DOI":"10.1038\/s41591-025-03626-6","article-title":"Sociodemographic biases in medical decision making by large language models","volume":"31","author":"Omar","journal-title":"Nat Med"},{"issue":"4","key":"R40","doi-asserted-by":"publisher","first-page":"695","DOI":"10.1080\/17437199.2022.2161594","article-title":"Is thematic analysis used well in health psychology? A critical review of published research, with recommendations for quality practice and reporting","volume":"17","author":"Braun","journal-title":"Health Psychol Rev"},{"key":"R41","doi-asserted-by":"crossref","unstructured":"Institute of Medicine (US) Committee on Quality of Health Care in America. Crossing the Quality Chasm: A New Health System for the 21st Century. National Academies Press; URL: http:\/\/www.ncbi.nlm.nih.gov\/books\/NBK222274\/ [Accessed 28-09-2025] [doi: 10.17226\/10027]","DOI":"10.17226\/10027"},{"issue":"2","key":"R42","doi-asserted-by":"publisher","first-page":"420","DOI":"10.1037\/\/0033-2909.86.2.420","article-title":"Intraclass correlations: uses in assessing rater reliability","volume":"86","author":"Shrout","journal-title":"Psychol Bull"},{"issue":"1","key":"R43","doi-asserted-by":"publisher","first-page":"159","DOI":"10.2307\/2529310","article-title":"The measurement of observer agreement for categorical data","volume":"33","author":"Landis","journal-title":"Biometrics"},{"issue":"6","key":"R44","doi-asserted-by":"publisher","first-page":"e406","DOI":"10.1016\/S2589-7500(22)00063-2","article-title":"AI recognition of patient race in medical imaging: a modelling study","volume":"4","author":"Gichoya","journal-title":"Lancet Digit Health"},{"issue":"23","key":"R45","doi-asserted-by":"publisher","first-page":"12592","DOI":"10.1073\/pnas.1919012117","article-title":"Gender imbalance in medical imaging datasets produces biased classifiers for computer-aided diagnosis","volume":"117","author":"Larrazabal","journal-title":"Proc Natl Acad Sci U S A"},{"key":"R46","unstructured":"Sogancioglu G Mijsters F van Uden A Peperzak J . Sex bias in (non)-contextual clinical word embeddings for stereotypical medical categories. arXiv. Preprint posted online on 2022. URL: http:\/\/arxiv.org\/abs\/2208.01341 [Accessed 01-10-2025]"},{"issue":"6464","key":"R47","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"},{"key":"R48","unstructured":"Jeoung S Ge Y Wang H Diesner J . Examining alignment of large language models through representative heuristics: the case of political stereotypes. arXiv. Preprint posted online on 2025. URL: http:\/\/arxiv.org\/abs\/2501.14294 [Accessed 30-09-2025]"},{"issue":"4","key":"R49","doi-asserted-by":"publisher","first-page":"1066","DOI":"10.1037\/xge0001547","article-title":"Do large language models show decision heuristics similar to humans? A case study using GPT-3.5","volume":"153","author":"Suri","journal-title":"J Exp Psychol Gen"},{"key":"R50","doi-asserted-by":"publisher","DOI":"10.1016\/j.chbah.2025.100138","article-title":"Artificial intelligence and human decision making: exploring similarities in cognitive bias","volume":"4","author":"Campbell","journal-title":"Comput Human Behav Artif Humans"},{"issue":"1","key":"R51","doi-asserted-by":"publisher","DOI":"10.1186\/s12913-025-13282-8","article-title":"Culture of paternalism in the emergency department: a critical ethnographic study","volume":"25","author":"Davoudi","journal-title":"BMC Health Serv Res"},{"key":"R52","unstructured":"Zea J . Patients and clinicians: llms achieve high QA accuracy but require human evaluation for clinical safety. ArkangelAI. URL: https:\/\/arkangel.ai\/research\/patients-and-clinicians-llms-achieve-high-qa-accuracy-but-require-human-evaluation-for-clinical-safety?utm_source=chatgpt.com [Accessed 30-09-2025]"},{"issue":"4","key":"R53","doi-asserted-by":"publisher","first-page":"559","DOI":"10.1016\/j.mcpdig.2024.08.004","article-title":"Challenges and limitations of human oversight in ethical artificial intelligence implementation in health care: balancing digital literacy and professional strain","volume":"2","author":"van Voorst","journal-title":"Mayo Clin Proc Digit Health"},{"issue":"1","key":"R54","doi-asserted-by":"publisher","DOI":"10.1038\/s41746-025-01732-w","article-title":"Identifying and mitigating algorithmic bias in the safety net","volume":"8","author":"Mackin","journal-title":"NPJ Digit Med"}],"container-title":["Journal of Medical Internet Research"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.jmir.org\/2025\/1\/e78132","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,30]],"date-time":"2026-06-30T18:10:37Z","timestamp":1782843037000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.jmir.org\/2025\/1\/e78132"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,12,5]]},"references-count":54,"URL":"https:\/\/doi.org\/10.2196\/78132","relation":{},"ISSN":["1438-8871"],"issn-type":[{"value":"1438-8871","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,12,5]]},"article-number":"v27i1e78132"}}