{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,6]],"date-time":"2026-10-06T01:22:38Z","timestamp":1791249758803,"version":"4.1.0"},"reference-count":39,"publisher":"JMIR Publications Inc.","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["J Med Internet Res"],"abstract":"Abstract<\/jats:title>\n \n Background<\/jats:title>\n Although widespread antiretroviral therapy has extended the life expectancy of people living with HIV, cardiovascular disease (CVD) has emerged as a primary comorbidity. Persistent cross-specialty knowledge gaps in routine clinical practice lead to suboptimal adherence to guidelines. Integrated, evidence-based tools are urgently needed to overcome these interdisciplinary barriers. While large language models (LLMs) have demonstrated significant capabilities in medicine, no systematic evaluation has assessed their ability to facilitate multidisciplinary CVD management for people living with HIV.<\/jats:p>\n <\/jats:sec>\n \n Objective<\/jats:title>\n This study compared the performance of 4 mainstream AI models (DeepSeek-V3, DeepSeek-R1, ChatGPT-4o, and ChatGPT-o4-mini) against 12 human clinicians (8 infectious disease specialists and 4 cardiologists) in addressing guideline-based CVD management tasks for people living with HIV.<\/jats:p>\n <\/jats:sec>\n \n Methods<\/jats:title>\n Based on 4 authoritative domestic and international HIV\/CVD guidelines, a structured 25-question assessment was developed via 2 rounds of Delphi consultation. Standard reference answers and an evaluation framework were finalized through expert consensus. LLM responses were generated using standardized prompts. Clinicians answered identical questions via one-on-one structured interviews, transcribed verbatim. Six multidisciplinary experts independently rated all responses across 4 dimensions\u2014accuracy, completeness, readability, and reliability\u2014using a 4-point ordinal scale (1=poor to 4=excellent). Cumulative link mixed models analyzed intergroup differences.<\/jats:p>\n <\/jats:sec>\n \n Results<\/jats:title>\n \n All AI models achieved significantly higher scores than clinicians across all dimensions (\n P<\/jats:italic>\n <.001). The AI group\u2019s mean scores ranged from 3.44 to 3.68 (median 4, IQR 3.0-4.0; coefficient of variation=0.145-0.178). Conversely, clinicians\u2019 scores were lower (mean 1.78-2.05; median 2, IQR 1.0-3.0; coefficient of variation=0.428-0.473) with marked dispersion. DeepSeek-R1 delivered the optimal performance, significantly outperforming the other 3 models (all\n P<\/jats:italic>\n <.001). Specialty-stratified analysis revealed no significant overall score difference between cardiologists and infectious disease specialists (odds ratio 0.92, 95% CI 0.84-1.01;\n P<\/jats:italic>\n =.09). However, dimension-specific analysis indicated that cardiologists scored higher in accuracy (odds ratio 0.81, 95% CI 0.67-0.97;\n P<\/jats:italic>\n =.03). Domain-specific divergence was evident: cardiologists outperformed infectious disease specialists in CVD risk assessment (2.26 vs 1.83), whereas infectious disease specialists led in drug adverse effect evaluation (2.23 vs 1.65).\n <\/jats:p>\n <\/jats:sec>\n \n Conclusions<\/jats:title>\n In this structured question-and-answer study, LLMs outperformed human clinicians across all metrics for HIV-associated CVD management, with DeepSeek-R1 achieving superior composite scores. These findings validate DeepSeek-R1\u2019s potential as a cross-disciplinary decision-support tool capable of integrating complex clinical knowledge, mitigating specialty gaps, and enhancing information precision. Integrating AI systems into multidisciplinary workflows, complemented by targeted clinical training, may optimize the management of complex comorbidities in people living with HIV.<\/jats:p>\n <\/jats:sec>","DOI":"10.2196\/89858","type":"journal-article","created":{"date-parts":[[2026,7,15]],"date-time":"2026-07-15T04:20:05Z","timestamp":1784089205000},"page":"e89858-e89858","source":"Crossref","is-referenced-by-count":1,"title":["Comparative Performance of AI Models and Clinicians in Evidence-Based Cardiovascular Disease Management for People Living With HIV: Comparative Study"],"prefix":"10.2196","volume":"28","author":[{"ORCID":"https:\/\/orcid.org\/0009-0002-4529-4850","authenticated-orcid":false,"given":"Tianqi","family":"Kong","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Liqin","family":"Sun","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0007-6557-3503","authenticated-orcid":false,"given":"Yinsong","family":"Luo","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0001-2862-4419","authenticated-orcid":false,"given":"Xi","family":"Xiao","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jin","family":"Li","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2863-7006","authenticated-orcid":false,"given":"Jiaye","family":"Liu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1010","published-online":{"date-parts":[[2026,8,10]]},"reference":[{"key":"R1","unstructured":"Data on the size of the HIV epidemic. World Health Organization. 2024. URL: https:\/\/www.who.int\/data\/gho\/data\/themes\/hiv-aids\/data-on-the-size-of-the-hiv-aids-epidemic [Accessed 26-04-2025]"},{"key":"R2","unstructured":"Joint United Nations Programme on HIV\/AIDS. 2017 global AIDS update\u2014ending AIDS: progress towards the 90\u201390\u201390 targets. UNAIDS. 2017. URL: https:\/\/www.unaids.org\/en\/resources\/documents\/2017\/20170720_Global_AIDS_update_2017 [Accessed 21-06-2026]"},{"issue":"1","key":"R3","doi-asserted-by":"publisher","first-page":"4","DOI":"10.1097\/ID9.0000000000000152","article-title":"Chinese guidelines for the diagnosis and treatment of human immunodeficiency virus infection\/acquired immunodeficiency syndrome (2024 edition)","volume":"5","author":"Acquired Immunodeficiency Syndrome Professional Group, Society of Infectious Diseases, Chinese Medical Association; Chinese Center for Disease Control and Prevention","journal-title":"Infect Dis Immun"},{"issue":"6","key":"R4","doi-asserted-by":"publisher","first-page":"1754","DOI":"10.1093\/ije\/dyt198","article-title":"Association of HIV and ART with cardiometabolic traits in sub-Saharan Africa: a systematic review and meta-analysis","volume":"42","author":"Dillon","journal-title":"Int J Epidemiol"},{"issue":"1","key":"R5","doi-asserted-by":"publisher","first-page":"44","DOI":"10.1161\/HYPERTENSIONAHA.118.10893","article-title":"Hypertension in HIV-infected adults: novel pathophysiologic mechanisms","volume":"72","author":"Fahme","journal-title":"Hypertension"},{"issue":"6","key":"R6","doi-asserted-by":"publisher","first-page":"847","DOI":"10.7326\/ANNALS-24-03564","article-title":"Statin therapy as primary prevention for persons with HIV: a synopsis of recommendations from the U.S. Department of Health and Human Services Antiretroviral Treatment Guidelines Panel","volume":"178","author":"Beavers","journal-title":"Ann Intern Med"},{"key":"R7","doi-asserted-by":"publisher","DOI":"10.7189\/jogh.14.04078","article-title":"Higher cardiovascular disease risks in people living with HIV: a systematic review and meta-analysis","volume":"14","author":"Zhu","journal-title":"J Glob Health"},{"issue":"8","key":"R8","doi-asserted-by":"publisher","first-page":"687","DOI":"10.1056\/NEJMoa2304146","article-title":"Pitavastatin to prevent cardiovascular disease in HIV infection","volume":"389","author":"Grinspoon","journal-title":"N Engl J Med"},{"key":"R9","doi-asserted-by":"publisher","DOI":"10.3389\/fnut.2025.1635682","article-title":"Large language models in clinical nutrition: an overview of its applications, capabilities, limitations, and potential future prospects","volume":"12","author":"Belkhouribchia","journal-title":"Front Nutr"},{"key":"R10","doi-asserted-by":"publisher","DOI":"10.2196\/51282","article-title":"Assessing GPT-4\u2019s performance in delivering medical advice: comparative analysis with human experts","volume":"10","author":"Jo","journal-title":"JMIR Med Educ"},{"issue":"5","key":"R11","doi-asserted-by":"publisher","first-page":"937","DOI":"10.1111\/bju.16873","article-title":"Chatbots in urology: accuracy, calibration, and comprehensibility; is DeepSeek taking over the throne?","volume":"136","author":"Asker","journal-title":"BJU Int"},{"issue":"14","key":"R12","doi-asserted-by":"publisher","DOI":"10.3390\/s25144305","article-title":"Synthetic patient-physician conversations simulated by large language models: a multi-dimensional evaluation","volume":"25","author":"Haider","journal-title":"Sensors (Basel)"},{"key":"R13","doi-asserted-by":"publisher","DOI":"10.2196\/69955","article-title":"Enhancing the readability of online patient education materials using large language models: cross-sectional study","volume":"27","author":"Will","journal-title":"J Med Internet Res"},{"key":"R14","doi-asserted-by":"publisher","DOI":"10.1177\/19160216251360651","article-title":"Readability, reliability, and quality analysis of internet-based patient education materials and large language models on Meniere\u2019s disease","volume":"54","author":"Alamleh","journal-title":"J Otolaryngol Head Neck Surg"},{"key":"R15","doi-asserted-by":"publisher","DOI":"10.2147\/PPA.S527922","article-title":"Generative AI\/LLMs for plain language medical information for patients, caregivers and general public: opportunities, risks and ethics","volume":"19","author":"Pal","journal-title":"Patient Prefer Adherence"},{"issue":"2","key":"R16","doi-asserted-by":"publisher","DOI":"10.1001\/jamanetworkopen.2024.57879","article-title":"Large language models for chatbot health advice studies","volume":"8","author":"Huo","journal-title":"JAMA Netw Open"},{"issue":"1","key":"R17","doi-asserted-by":"publisher","DOI":"10.1186\/s12883-025-04280-8","article-title":"Evaluating ChatGPT and DeepSeek in postdural puncture headache management: a comparative study with international consensus guidelines","volume":"25","author":"Deng","journal-title":"BMC Neurol"},{"key":"R18","doi-asserted-by":"publisher","DOI":"10.2196\/65365","article-title":"Assessing the role of large language models between ChatGPT and DeepSeek in asthma education for bilingual individuals: comparative study","volume":"13","author":"Liu","journal-title":"JMIR Med Inform"},{"issue":"8","key":"R19","doi-asserted-by":"publisher","first-page":"3025","DOI":"10.1002\/ksa.12711","article-title":"Evaluating DeepResearch and DeepThink in anterior cruciate ligament surgery patient education: ChatGPT\u20104o excels in comprehensiveness, DeepSeek R1 leads in clarity and readability of orthopaedic information","volume":"33","author":"G\u00fcltekin","journal-title":"Knee Surg Sports Traumatol Arthrosc"},{"issue":"2","key":"R20","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","journal-title":"Int J STD AIDS"},{"issue":"1","key":"R21","doi-asserted-by":"publisher","DOI":"10.1186\/s12911-025-02972-2","article-title":"Effectiveness of various general large language models in clinical consensus and case analysis in dental implantology: a comparative study","volume":"25","author":"Wu","journal-title":"BMC Med Inform Decis Mak"},{"issue":"12","key":"R22","doi-asserted-by":"publisher","first-page":"1000","DOI":"10.3760\/cma.j.cn112148-20201009-00796","article-title":"Chinese guideline on the primary prevention of cardiovascular diseases","volume":"48","author":"Chinese Society of Cardiology of Chinese Medical Association","journal-title":"Zhonghua Xin Xue Guan Bing Za Zhi"},{"issue":"6","key":"R23","doi-asserted-by":"publisher","first-page":"1261","DOI":"10.12290\/xhyxzz.2024-0766","article-title":"Chinese guideline for diagnosis and treatment of human immunodeficiency virus infection\/acquired immunodeficiency syndrome (2024 edition)","volume":"15","author":"Acquired Immunodeficiency Syndrome Professional Group of Society of Infectious Diseases, Chinese Medical Association; Chinese Center for Disease Control and Prevention","journal-title":"Med J Peking Union Med Coll Hosp"},{"key":"R24","unstructured":"EACS Guidelines Version 13.0. European AIDS Clinical Society. 2024. URL: https:\/\/www.eacsociety.org\/guidelines\/eacs-guidelines [Accessed 21-06-2026]"},{"issue":"7","key":"R25","doi-asserted-by":"publisher","first-page":"609","DOI":"10.1001\/jama.2024.24543","article-title":"Antiretroviral drugs for treatment and prevention of HIV in adults: 2024 recommendations of the International Antiviral Society-USA Panel","volume":"333","author":"Gandhi","journal-title":"JAMA"},{"key":"R26","doi-asserted-by":"publisher","DOI":"10.2196\/65785","article-title":"Comparison of an AI chatbot with a nurse hotline in reducing anxiety and depression levels in the general population: pilot randomized controlled trial","volume":"12","author":"Chen","journal-title":"JMIR Hum Factors"},{"issue":"1","key":"R27","doi-asserted-by":"publisher","DOI":"10.1186\/s41747-025-00591-0","article-title":"Five advanced chatbots solving European Diploma in Radiology (EDiR) text-based questions: differences in performance and consistency","volume":"9","author":"Pristoupil","journal-title":"Eur Radiol Exp"},{"issue":"10","key":"R28","doi-asserted-by":"publisher","DOI":"10.1001\/jamanetworkopen.2024.37711","article-title":"Performance of multimodal artificial intelligence chatbots evaluated on clinical oncology cases","volume":"7","author":"Chen","journal-title":"JAMA Netw Open"},{"issue":"5","key":"R29","doi-asserted-by":"publisher","DOI":"10.1148\/radiol.230922","article-title":"How AI responds to common lung cancer questions: ChatGPT vs Google Bard","volume":"307","author":"Rahsepar","journal-title":"Radiology"},{"key":"R30","doi-asserted-by":"publisher","DOI":"10.2196\/64486","article-title":"Accuracy of large language models when answering clinical research questions: systematic review and network meta-analysis","volume":"27","author":"Wang","journal-title":"J Med Internet Res"},{"key":"R31","doi-asserted-by":"publisher","DOI":"10.2196\/54571","article-title":"Assessing generative pretrained transformers (GPT) in clinical decision-making: comparative analysis of GPT-3.5 and GPT-4","volume":"26","author":"Lahat","journal-title":"J Med Internet Res"},{"issue":"7","key":"R32","doi-asserted-by":"publisher","first-page":"2776","DOI":"10.1007\/s00266-025-05108-z","article-title":"A comparative study on the use of DeepSeek-R1 and ChatGPT-4.5 in different aspects of plastic surgery","volume":"50","author":"Ali","journal-title":"Aesth Plast Surg"},{"issue":"16","key":"R33","doi-asserted-by":"publisher","DOI":"10.3390\/jcm14165697","article-title":"How accurate is AI? A critical evaluation of commonly used large language models in responding to patient concerns about incidental kidney tumors","volume":"14","author":"Ralla","journal-title":"J Clin Med"},{"issue":"8","key":"R34","doi-asserted-by":"publisher","DOI":"10.3390\/diseases12080185","article-title":"AI-driven patient education in chronic kidney disease: evaluating chatbot responses against clinical guidelines","volume":"12","author":"Acharya","journal-title":"Diseases"},{"issue":"3","key":"R35","doi-asserted-by":"publisher","first-page":"348","DOI":"10.1016\/j.ajodo.2025.04.008","article-title":"Information from digital and human sources: a comparison of chatbot and clinician responses to orthodontic questions","volume":"168","author":"Metin","journal-title":"Am J Orthod Dentofacial Orthop"},{"issue":"1","key":"R36","doi-asserted-by":"publisher","first-page":"18","DOI":"10.1002\/jpn3.70069","article-title":"Parental education in pediatric dysphagia: a comparative analysis of three large language models","volume":"81","author":"Alyanak","journal-title":"J Pediatr Gastroenterol Nutr"},{"issue":"6","key":"R37","doi-asserted-by":"publisher","first-page":"1281","DOI":"10.1007\/s40123-025-01142-x","article-title":"Large language models: pioneering new educational frontiers in childhood myopia","volume":"14","author":"Delsoz","journal-title":"Ophthalmol Ther"},{"issue":"1","key":"R38","doi-asserted-by":"publisher","DOI":"10.1186\/s12911-024-02824-5","article-title":"Assessing the accuracy and quality of artificial intelligence (AI) chatbot-generated responses in making patient-specific drug-therapy and healthcare-related decisions","volume":"24","author":"Shiferaw","journal-title":"BMC Med Inform Decis Mak"},{"issue":"1","key":"R39","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-025-04309-5","article-title":"Evaluating performance of large language models for atrial fibrillation management using different prompting strategies and languages","volume":"15","author":"Li","journal-title":"Sci Rep"}],"container-title":["Journal of Medical Internet Research"],"original-title":[],"language":"en","deposited":{"date-parts":[[2026,8,10]],"date-time":"2026-08-10T19:21:50Z","timestamp":1786389710000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.jmir.org\/2026\/1\/e89858"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,10]]},"references-count":39,"URL":"https:\/\/doi.org\/10.2196\/89858","relation":{},"ISSN":["1438-8871"],"issn-type":[{"value":"1438-8871","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,10]]},"article-number":"v28i1e89858"}}