{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,6]],"date-time":"2026-10-06T05:54:21Z","timestamp":1791266061260,"version":"4.1.0"},"reference-count":52,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,10,6]],"date-time":"2026-10-06T00:00:00Z","timestamp":1791244800000},"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. Public Health"],"abstract":"<jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Chinese large language models (LLMs) are increasingly used for HIV\/AIDS health information, yet their multidimensional performance remains unevaluated. This study systematically assessed five mainstream Chinese LLMs across accuracy, caring, completeness, understandability, and actionability.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>A cross-sectional evaluation was conducted using 50 standardized HIV\/AIDS questions. Each model's responses were rated by three independent experts on five dimensions, including adapted PEMAT-based items. Latent profile analysis (LPA) and partial correlation network analysis were employed to identify quality typologies and interdimensional dependencies. Chi-square tests examined associations between profile membership and model identity or question domain.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>\n                      The four general-purpose LLMs showed high accuracy (mean &amp;gt; 4.5\/5) and completeness (86%\u221298% full marks), with no significant differences among them (all P-FDR &amp;gt; 0.05). HuatuoGPT-II, the only medical-domain-specific model, scored significantly lower on both dimensions (all\n                      <jats:italic>P<\/jats:italic>\n                      &amp;lt; 0.001). Its responses were less consistent with current guidelines and included one instance of implicit stigmatizing language. All models showed systemic deficits in caring (median = 3.0\/5) and actionability (max = 1.70\/3). Understandability showed a ceiling effect and no significant profile-level differences, although small model-level differences were observed. LPA identified four profiles: High Accuracy and Completeness (50.0%), Moderate-Balanced (3.6%), Low-Quality (18.8%), and High Completeness and Actionability-Oriented (27.6%). Network analysis identified completeness as the most central node (strength = 0.813) and actionability as a key bridge (betweenness = 0.167). A weak negative partial correlation between accuracy and actionability (\n                      <jats:italic>r<\/jats:italic>\n                      = \u22120.083) suggested a potential trade-off between technical precision and practical guidance at the statistical level. Profile distributions differed by model (\u03c7\n                      <jats:sup>2<\/jats:sup>\n                      = 124.99,\n                      <jats:italic>P<\/jats:italic>\n                      &amp;lt; 0.001) but not domain (\n                      <jats:italic>P<\/jats:italic>\n                      = 0.625), indicating heterogeneity stems from model-specific characteristics. Yuanbao-Hy3 showed the most balanced humanistic-technical profile. DeepSeek-V4 excelled technically but lacked caring.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>In this evaluation, the four general-purpose models outperformed the single medical-domain-specific model examined in factual accuracy and completeness. However, systemic deficits in empathetic communication and actionable guidance persist. Model selection should prioritize holistic profile distributions over single-dimension scores. Developers must integrate de-stigmatizing corpora and action-oriented alignment; clinicians should position AI as supplementary to professional care. Future evaluations should adopt person-centred, multi-turn, and real-user validated frameworks.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fpubh.2026.1966522","type":"journal-article","created":{"date-parts":[[2026,10,6]],"date-time":"2026-10-06T05:21:25Z","timestamp":1791264085000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Evaluating Chinese large language models for HIV\/AIDS health information: a multidimensional comparative study of response quality"],"prefix":"10.3389","volume":"14","author":[{"given":"Jinhao","family":"Gu","sequence":"first","affiliation":[{"name":"Department of Biomedical Informatics, School of Life Sciences, Central South University","place":["Changsha, Hunan, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jianwei","family":"Liu","sequence":"additional","affiliation":[{"name":"Department of Biomedical Informatics, School of Life Sciences, Central South University","place":["Changsha, Hunan, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yunrong","family":"Liu","sequence":"additional","affiliation":[{"name":"Department of Biomedical Informatics, School of Life Sciences, Central South University","place":["Changsha, Hunan, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Pingping","family":"Wang","sequence":"additional","affiliation":[{"name":"Department of Biomedical Informatics, School of Life Sciences, Central South University","place":["Changsha, Hunan, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Dehua","family":"Hu","sequence":"additional","affiliation":[{"name":"Department of Biomedical Informatics, School of Life Sciences, Central South University","place":["Changsha, Hunan, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Qizhen","family":"Zhu","sequence":"additional","affiliation":[{"name":"Department of Biomedical Informatics, School of Life Sciences, Central South University","place":["Changsha, Hunan, China"]},{"name":"School of Medical Information Engineering, Jining Medical University","place":["Rizhao, Shandong, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,10,6]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"284","DOI":"10.1089\/apc.2023.0028","article-title":"HIV tissue reservoirs: current advances in research","volume":"37","author":"Li","year":"2023","journal-title":"AIDS Patient Care STDS."},{"key":"B2","unstructured":"Global AIDS update 2026: United to End AIDS\n          \n          2026"},{"key":"B3","doi-asserted-by":"publisher","first-page":"18","DOI":"10.1097\/IM9.0000000000000201","article-title":"Prevalence and spread of HIV in China in the past two decades","volume":"8","author":"Hu","year":"2026","journal-title":"Infect Microb Dis."},{"key":"B4","doi-asserted-by":"publisher","first-page":"671","DOI":"10.1097\/JNC.0000000000000557","article-title":"\u201cI'm So Passive, Every Visit Is a Challenge\u201d: stigma experienced by older people living with HIV during follow-up health care visits in Wuxi, China\u2014a qualitative study","volume":"36","author":"Hao","year":"2026","journal-title":"J Assoc Nurs AIDS Care."},{"key":"B5","doi-asserted-by":"publisher","first-page":"e48586","DOI":"10.1097\/MD.0000000000048586","article-title":"An analysis of HIV-related behaviors and attitudes among college students in eastern China: a cross-sectional study","volume":"105","author":"Zhou","year":"2026","journal-title":"Medicine."},{"key":"B6","doi-asserted-by":"publisher","first-page":"3639","DOI":"10.1038\/s41598-025-33749-2","article-title":"Association of demographics and psychological barriers with late presentation of HIV among men who have sex with men in Chengdu, China","volume":"16","author":"Zhang","year":"2026","journal-title":"Sci Rep."},{"key":"B7","doi-asserted-by":"publisher","first-page":"561","DOI":"10.1080\/09540121.2023.2280460","article-title":"Influence of HIV\/AIDS knowledge on HIV testing behavior among young students (14-27 years) with a history of sexual activity: chain mediation analysis of social discrimination, self-efficacy, and sexual behavior characteristics","volume":"36","author":"Yang","year":"2024","journal-title":"AIDS Care."},{"key":"B8","doi-asserted-by":"publisher","first-page":"72","DOI":"10.1007\/s40471-021-00265-6","article-title":"HIV and aging: double stigma","volume":"8","author":"Brown","year":"2021","journal-title":"Curr Epidemiol Rep."},{"key":"B9","doi-asserted-by":"publisher","first-page":"668","DOI":"10.1016\/j.socscimed.2011.06.051","article-title":"The problem of \u201csignificant risk\u201d: exploring the public health impact of criminalizing HIV non-disclosure","volume":"73","author":"Mykhalovskiy","year":"2011","journal-title":"Soc Sci Med."},{"key":"B10","doi-asserted-by":"publisher","first-page":"3603","DOI":"10.1109\/TKDE.2024.3392335","article-title":"A survey of large language models","volume":"36","author":"Zhao","year":"2024","journal-title":"IEEE Trans Knowl Data Eng."},{"key":"B11","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3672393","article-title":"Large language models: a survey","volume":"56","author":"Minaee","year":"2024","journal-title":"ACM Computing Surveys."},{"key":"B12","doi-asserted-by":"publisher","first-page":"1","DOI":"10.48550\/arXiv.1910.10683","article-title":"Exploring the limits of transfer learning with a unified text-to-text transformer","volume":"21","author":"Raffel","year":"2020","journal-title":"J Mach Learn Res."},{"key":"B13","doi-asserted-by":"publisher","first-page":"1591","DOI":"10.1186\/s12879-025-12022-x","article-title":"Comparison of the accuracy and reliability of ChatGPT-4o and Gemini in answering HIV-related questions","volume":"25","author":"Tarhan","year":"2025","journal-title":"BMC Infect Dis."},{"key":"B14","doi-asserted-by":"publisher","first-page":"2746","DOI":"10.1007\/s10461-024-04391-2","article-title":"Assessing ChatGPT's potential in HIV prevention communication: a comprehensive evaluation of accuracy, completeness, and inclusivity","volume":"28","author":"De Vito","year":"2024","journal-title":"AIDS Behav."},{"key":"B15","doi-asserted-by":"publisher","first-page":"1624","DOI":"10.1186\/s12879-025-11621-y","article-title":"Readability and information quality of LLM-generated HIV content: a methodological content evaluation","volume":"25","author":"Chen","year":"2025","journal-title":"BMC Infect Dis."},{"key":"B16","doi-asserted-by":"publisher","first-page":"112","DOI":"10.1177\/09564624251372369","article-title":"Artificial intelligence meets HIV education: comparing three large language models on accuracy, readability, and reliability","volume":"37","author":"Eren Korkmaz","year":"2026","journal-title":"Int J STD AIDS"},{"key":"B17","doi-asserted-by":"publisher","first-page":"1828810","DOI":"10.3389\/fcvm.2026.1828810","article-title":"Cross-platform evaluation of LLM-generated educational texts on cardiac myxoma: quality, readability, and actionability using network analysis and latent profile analysis","volume":"13","author":"Deng","year":"2026","journal-title":"Front Cardiovasc Med."},{"key":"B18","volume-title":"The 55th statistical report on internet development in China","year":"2025"},{"key":"B19","unstructured":"HuatuoGPT-II: one-stage adaptation of LLMs for medical question answering\n          \n          \n            \n              Zhang\n              Y\n            \n            \n              Chen\n              H\n            \n            \n              Zhu\n              Z\n            \n            \n              Wang\n              J\n            \n            \n              Liu\n              S\n            \n            \n              Li\n              X\n            \n            \n              Wang\n              B\n            \n          \n          arXiv\n          arXiv:2311.09774\n          \n          2024"},{"key":"B20","unstructured":"Questions and Answers about HIV\/AIDS\n          \n          2023"},{"key":"B21","doi-asserted-by":"publisher","first-page":"719","DOI":"10.1002\/jso.27966","article-title":"Assessment of artificial intelligence chatbot responses to common patient questions on Bone Sarcoma","volume":"131","author":"Khabaz","year":"2025","journal-title":"J Surg Oncol."},{"key":"B22","unstructured":"Geneva\n          World Health Organization\n          Consolidated guidelines on differentiated HIV testing services.\n          \n          2024"},{"key":"B23","volume-title":"National Technical Specification for HIV Testing (2025 Revision).","year":"2025"},{"key":"B24","doi-asserted-by":"publisher","first-page":"4","DOI":"10.1097\/ID9.0000000000000152","volume":"5","year":"2025","journal-title":"Chinese guidelines for the diagnosis and treatment of human immunodeficiency virus infection\/acquired immunodeficiency syndrome (2024 edition). Infect Dis Immun."},{"key":"B25","doi-asserted-by":"publisher","first-page":"395","DOI":"10.1016\/j.pec.2014.05.027","article-title":"Development of the Patient Education Materials Assessment Tool (PEMAT) for patient education materials","volume":"96","author":"Shoemaker","year":"2014","journal-title":"Patient Educ Couns."},{"key":"B26","doi-asserted-by":"publisher","first-page":"892","DOI":"10.1037\/abn0000446","article-title":"What do centrality measures measure in psychological networks?","volume":"128","author":"Bringmann","year":"2019","journal-title":"J Abnorm Psychol"},{"key":"B27","doi-asserted-by":"crossref","first-page":"953","DOI":"10.1002\/9781118489772.ch30","article-title":"Network psychometrics","volume-title":"The Wiley Handbook of Psychometric Testing: A Multidisciplinary Reference on Survey, Scale and Test Development","author":"Epskamp","year":"2018"},{"key":"B28","doi-asserted-by":"publisher","first-page":"5","DOI":"10.1002\/wps.20375","article-title":"A network theory of mental disorders","volume":"16","author":"Borsboom","year":"2017","journal-title":"World Psychiatry"},{"key":"B29","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1007\/s00127-016-1319-z","article-title":"Mental disorders as networks of problems: a review of recent insights","volume":"52","author":"Fried","year":"2017","journal-title":"Soc Psychiatry Psychiatr Epidemiol."},{"key":"B30","doi-asserted-by":"publisher","first-page":"539","DOI":"10.1080\/00273171.2019.1614898","article-title":"Bridge centrality: a network approach to understanding comorbidity","volume":"52","author":"Jones","year":"2017","journal-title":"Multivar Behav Res."},{"key":"B31","unstructured":"Patient Education Materials Assessment Tool (PEMAT)\n          \n          2026"},{"key":"B32","doi-asserted-by":"publisher","first-page":"105","DOI":"10.1037\/0022-3514.72.1.105","article-title":"Empathy and attitudes: can feeling for a member of a stigmatized group improve feelings toward the group?","volume":"72","author":"Batson","year":"1997","journal-title":"J Pers Soc Psychol."},{"key":"B33","doi-asserted-by":"publisher","first-page":"103275","DOI":"10.1016\/j.msksp.2025.103275","article-title":"ChatGPT is a comprehensive education tool for patients with patellar tendinopathy, but it currently lacks accuracy and readability","volume":"76","author":"Deng","year":"2025","journal-title":"Musculoskelet Sci Pract."},{"key":"B34","doi-asserted-by":"publisher","first-page":"385","DOI":"10.20344\/amp.22805","article-title":"Use of ChatGPT in HIV infection counselling and literacy","volume":"38","author":"Ganicho","year":"2025","journal-title":"Acta M\u00e9d Portug."},{"key":"B35","doi-asserted-by":"publisher","first-page":"104317","DOI":"10.1016\/j.ijoa.2024.104317","article-title":"Readability, quality and accuracy of generative artificial intelligence chatbots for commonly asked questions about labor epidurals: a comparison of ChatGPT and Bard","volume":"61","author":"Lee","year":"2025","journal-title":"Int J Obstet Anesth"},{"key":"B36","doi-asserted-by":"publisher","first-page":"934","DOI":"10.1016\/j.scib.2025.01.034","article-title":"Large language models for diabetes training: a prospective study","volume":"70","author":"Li","year":"2025","journal-title":"Sci Bull"},{"key":"B37","unstructured":"HIV\/AIDS Health Education Resource Database\n          \n          2026"},{"key":"B38","doi-asserted-by":"publisher","first-page":"e49373","DOI":"10.7759\/cureus.49373","article-title":"Pilot testing of a tool to standardize the assessment of the quality of health information generated by artificial intelligence-based models","volume":"15","author":"Sallam","year":"2023","journal-title":"Cureus."},{"key":"B39","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1080\/09540121.2026.2660134.","article-title":"Cultural formation and determinants of HIV stigma among a minority group in Southwest China","author":"Ting","year":"2026","journal-title":"AIDS Care."},{"key":"B40","doi-asserted-by":"publisher","first-page":"1080","DOI":"10.1093\/jamia\/ocy052","article-title":"Good intentions are not enough: how informatics interventions can worsen inequality","volume":"25","author":"Veinot","year":"2018","journal-title":"J Am Med Inf Assoc."},{"key":"B41","unstructured":"Geneva\n          World Health Organization\n          Overview of WHO recommendations on HIV and sexually transmitted infection testing, prevention, treatment, care and service delivery\n          \n          2025"},{"key":"B42","unstructured":"Geneva\n          World Health Organization\n          WHO Updated Recommendations on HIV Clinical Management: Recommendations for a Public Health Approach.\n          \n          2025"},{"key":"B43","unstructured":"Geneva\n          World Health Organization\n          WHO consolidated operational handbook on sexually transmitted infections\n          \n          2026"},{"key":"B44","doi-asserted-by":"publisher","first-page":"1706","DOI":"10.1007\/s00266-025-05458-8","article-title":"Evaluating large language models as medical consultation tools for double eyelid surgery: cross-language study in English and Chinese","volume":"50","author":"Xin","year":"2026","journal-title":"Aesthet Plast Surg."},{"key":"B45","doi-asserted-by":"publisher","first-page":"153","DOI":"10.1097\/ID9.0000000000000126","article-title":"Human immunodeficiency virus\/acquired immunodeficiency syndrome prevention and control in China: statistics, challenges, and the road ahead","volume":"3","author":"Luo","year":"2024","journal-title":"Infect Dis Immun"},{"key":"B46","doi-asserted-by":"publisher","first-page":"688","DOI":"10.1097\/JU.0000000000003615","article-title":"Evaluating the effectiveness of artificial intelligence-powered large language models application in disseminating appropriate and readable health information in urology","volume":"210","author":"Davis","year":"2023","journal-title":"J Urol"},{"key":"B47","doi-asserted-by":"publisher","first-page":"406","DOI":"10.1097\/QAI.0000000000002593","article-title":"Disclosure concerns and viral suppression in people newly initiating HIV care: the role of internalized HIV stigma","volume":"86","author":"Yigit","year":"2021","journal-title":"J Acquir Immune Defic Syndr"},{"key":"B48","doi-asserted-by":"publisher","first-page":"e38502","DOI":"10.2196\/38502","article-title":"Examining Chinese users' feedback comments on HIV self-testing kits from e-commerce platforms: thematic and content analysis","volume":"24","author":"Liu","year":"2022","journal-title":"J Med Internet Res."},{"key":"B49","doi-asserted-by":"publisher","first-page":"107682","DOI":"10.1016\/j.pec.2023.107682","article-title":"Readability of publicly available mental health information: a systematic review","volume":"111","author":"Boutemen","year":"2023","journal-title":"Patient Educ Couns"},{"key":"B50","article-title":"Models that guide health education and communication","author":"Nutbeam","year":"2024","journal-title":"Theory in a Nutshell: A Practical Guide to Health Promotion Theories"},{"key":"B51","doi-asserted-by":"crossref","DOI":"10.1145\/3613905.3651077","article-title":"Rapport matters: Enhancing HIV mHealth communication through linguistic analysis and large language models","volume-title":"Extended Abstracts of the CHI Conference on Human Factors in Computing Systems (CHI EA '24)","author":"Wang","year":"2024"},{"key":"B52","doi-asserted-by":"publisher","first-page":"1804524","DOI":"10.3389\/fpubh.2026.1804524","article-title":"Challenges of using generative artificial intelligence for diabetes patient education: a cross-platform analysis of the quality, readability, and actionability of text generated by large language models","volume":"14","author":"Wang","year":"2026","journal-title":"Front Public Health"}],"container-title":["Frontiers in Public Health"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fpubh.2026.1966522\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,10,6]],"date-time":"2026-10-06T05:21:28Z","timestamp":1791264088000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fpubh.2026.1966522\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,10,6]]},"references-count":52,"alternative-id":["10.3389\/fpubh.2026.1966522"],"URL":"https:\/\/doi.org\/10.3389\/fpubh.2026.1966522","relation":{},"ISSN":["2296-2565"],"issn-type":[{"value":"2296-2565","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,10,6]]},"article-number":"1966522"}}