{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,9]],"date-time":"2026-10-09T19:41:08Z","timestamp":1791574868867,"version":"4.3.4"},"reference-count":49,"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 Outpatients presenting with low back pain (LBP) often require efficient preconsultation triage and early differential diagnostic support. Large language models may assist these text-based tasks, but their performance under different clinical information conditions remains unclear.<\/jats:p>\n <\/jats:sec>\n \n Objective<\/jats:title>\n This study aimed to compare the performance of ChatGPT (5.1; OpenAI) and DeepSeek (V3.2; DeepSeek AI) in musculoskeletal disorders (MSDs) triage and the differential diagnosis of outpatients with LBP using real-world outpatient records under 2 simulated information conditions.<\/jats:p>\n <\/jats:sec>\n \n Methods<\/jats:title>\n This retrospective comparative study was conducted at a tertiary academic teaching hospital in Beijing. A total of 160 cases were included using a balanced design across 8 diagnostic categories (20 per category); 6 MSDs and 2 non-MSDs. Evaluation was performed in 2 phases: Phase 1 (chief complaint) and Phase 2 (structured questionnaire with 7 domains or 33 items), both executed in a zero-shot setting using standardized prompts. Outcomes included (1) triage accuracy, (2) preliminary diagnosis accuracy, and (3) differential diagnosis agreement. In Phase 2, 3 senior orthopedic evaluators additionally rated model rationales across 5 domains using a 5-point Likert scale.<\/jats:p>\n <\/jats:sec>\n \n Results<\/jats:title>\n For triage accuracy across all 160 cases, DeepSeek V3.2 improved from 84.4% to 90.6% (risk difference [RD] 6.2%, 95% CI \u20130.7% to 13.3%), and ChatGPT 5.1 improved from 75.6% to 93.1% (RD 17.5%, 95% CI 10.2%-24.9%). For preliminary diagnosis accuracy across the 120 musculoskeletal cases, DeepSeek V3.2 improved from 48.3% to 76.7% (RD 28.3%, 95% CI 16.8%-38.8%), whereas ChatGPT 5.1 improved from 35.0% to 87.5% (RD 52.5%, 95% CI 42.8%-60.6%). The mean number of correct differential diagnoses increased from 1.27 (SD 0.71) to 2.02 (SD 0.74) for DeepSeek V3.2 and from 1.34 (SD 0.70) to 2.03 (SD 0.77) for ChatGPT 5.1. In Phase 2, rationale ratings were generally good for both models, with ChatGPT 5.1 scoring higher in understanding and reasoning. Recognition of multiple myeloma (MM) remained limited, improving only from 45% to 55% (DeepSeek V3.2) and 55% to 60% (ChatGPT 5.1). Structured input reduced safety-risk errors in both models, but residual errors remained, especially for MM and metastatic spinal tumor.<\/jats:p>\n <\/jats:sec>\n \n Conclusions<\/jats:title>\n Both ChatGPT 5.1 and DeepSeek V3.2 demonstrated potential in text-based triage and differential diagnosis of MSDs for LBP, with structured clinical information generally improving performance, particularly for preliminary diagnosis accuracy and differential diagnosis agreement. However, their suboptimal sensitivity for red-flag conditions such as MM highlights significant safety concerns, indicating that they should not be used as stand-alone triage tools without clinician oversight. ChatGPT 5.1 showed stronger reasoning with structured inputs based on rationale ratings, whereas DeepSeek V3.2 showed better performance under chief-complaint-only input, with significantly higher Phase 1 preliminary diagnostic accuracy and numerically higher Phase 1 triage accuracy. These findings underscore the need for further model refinement, rigorous prospective validation, and integration with clinician oversight before clinical implementation.<\/jats:p>\n <\/jats:sec>","DOI":"10.2196\/92315","type":"journal-article","created":{"date-parts":[[2026,6,10]],"date-time":"2026-06-10T14:35:05Z","timestamp":1781102105000},"page":"e92315-e92315","source":"Crossref","is-referenced-by-count":3,"title":["Performance of DeepSeek V3.2 and ChatGPT 5.1 in Musculoskeletal Triage and Differential Diagnosis of Outpatients With Low Back Pain: Multidimensional Comparative Study"],"prefix":"10.2196","volume":"28","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-1245-378X","authenticated-orcid":false,"given":"Ziqian","family":"Ma","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0003-0745-4427","authenticated-orcid":false,"given":"Ruiyuan","family":"Chen","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3271-1953","authenticated-orcid":false,"given":"Aobo","family":"Wang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0005-3022-9281","authenticated-orcid":false,"given":"Yu","family":"Xi","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0006-4010-0243","authenticated-orcid":false,"given":"Minghui","family":"Liang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5668-9527","authenticated-orcid":false,"given":"Shuo","family":"Yuan","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0095-9476","authenticated-orcid":false,"given":"Ning","family":"Fan","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0001-2164-1494","authenticated-orcid":false,"given":"Jianwei","family":"Zang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-5016-858X","authenticated-orcid":false,"given":"Tianyi","family":"Wang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1403-4159","authenticated-orcid":false,"given":"Lei","family":"Zang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1010","published-online":{"date-parts":[[2026,7,3]]},"reference":[{"issue":"1","key":"R1","doi-asserted-by":"publisher","DOI":"10.1186\/s12916-018-1151-2","article-title":"Musculoskeletal conditions may increase the risk of chronic disease: a systematic review and meta-analysis of cohort studies","volume":"16","author":"Williams","journal-title":"BMC Med"},{"issue":"10258","key":"R2","doi-asserted-by":"publisher","first-page":"1285","DOI":"10.1016\/S0140-6736(20)30677-2","article-title":"Fertility, mortality, migration, and population scenarios for 195 countries and territories from 2017 to 2100: a forecasting analysis for the Global Burden of Disease Study","volume":"396","author":"Vollset","journal-title":"Lancet"},{"issue":"5","key":"R3","doi-asserted-by":"publisher","first-page":"100707","DOI":"10.1016\/j.lanhl.2025.100707","article-title":"Addressing the growing burden of musculoskeletal diseases in the ageing US population: challenges and innovations","volume":"6","author":"Nguyen","journal-title":"Lancet Healthy Longev"},{"key":"R4","doi-asserted-by":"publisher","DOI":"10.1016\/j.lana.2023.100661","article-title":"Musculoskeletal health: an ecological study assessing disease burden and research funding","volume":"29","author":"Nguyen","journal-title":"Lancet Reg Health Am"},{"issue":"2","key":"R5","doi-asserted-by":"publisher","first-page":"79","DOI":"10.1136\/bjsports-2018-099878","article-title":"What does best practice care for musculoskeletal pain look like? Eleven consistent recommendations from high-quality clinical practice guidelines: systematic review","volume":"54","author":"Lin","journal-title":"Br J Sports Med"},{"key":"R6","doi-asserted-by":"publisher","DOI":"10.2196\/67269","article-title":"Improving safety, efficiency, cost, and satisfaction across a musculoskeletal pathway using the digital assessment routing tool for triage: quality improvement study","volume":"27","author":"Lowe","journal-title":"J Med Internet Res"},{"issue":"4","key":"R7","doi-asserted-by":"publisher","first-page":"277","DOI":"10.1016\/j.physio.2014.03.007","article-title":"Musculoskeletal triage: a mixed methods study, integrating systematic review with expert and patient perspectives","volume":"100","author":"Joseph","journal-title":"Physiotherapy"},{"key":"R8","doi-asserted-by":"publisher","DOI":"10.2196\/51711","article-title":"Evolution of the \u201cInternet Plus Health Care\u201d mode enabled by artificial intelligence: development and application of an outpatient triage system","volume":"26","author":"Yang","journal-title":"J Med Internet Res"},{"issue":"1","key":"R9","doi-asserted-by":"publisher","DOI":"10.1038\/s41746-025-01684-1","article-title":"Evaluating large language model workflows in clinical decision support for triage and referral and diagnosis","volume":"8","author":"Gaber","journal-title":"NPJ Digit Med"},{"issue":"11","key":"R10","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"},{"issue":"11","key":"R11","doi-asserted-by":"publisher","DOI":"10.1001\/jamanetworkopen.2024.43925","article-title":"Natural language processing of clinical documentation to assess functional status in patients with heart failure","volume":"7","author":"Adejumo","journal-title":"JAMA Netw Open"},{"key":"R12","doi-asserted-by":"publisher","DOI":"10.2196\/59435","article-title":"Application of large language models in medical training evaluation-using ChatGPT as a standardized patient: multimetric assessment","volume":"27","author":"Wang","journal-title":"J Med Internet Res"},{"issue":"3","key":"R13","doi-asserted-by":"publisher","DOI":"10.1016\/S2589-7500(23)00023-7","article-title":"ChatGPT: friend or foe?","volume":"5","journal-title":"Lancet Digit Health"},{"issue":"1","key":"R14","doi-asserted-by":"publisher","DOI":"10.1038\/s41746-024-01258-7","article-title":"A framework for human evaluation of large language models in healthcare derived from literature review","volume":"7","author":"Tam","journal-title":"NPJ Digit Med"},{"key":"R15","doi-asserted-by":"publisher","DOI":"10.2196\/63010","article-title":"Comparative study to evaluate the accuracy of differential diagnosis lists generated by Gemini Advanced, Gemini, and Bard for a case report series analysis: cross-sectional study","volume":"12","author":"Hirosawa","journal-title":"JMIR Med Inform"},{"key":"R16","doi-asserted-by":"publisher","DOI":"10.2196\/45312","article-title":"How does ChatGPT Perform on the United States Medical Licensing Examination (USMLE)? The implications of large language models for medical education and knowledge assessment","volume":"9","author":"Gilson","journal-title":"JMIR Med Educ"},{"issue":"4","key":"R17","doi-asserted-by":"publisher","first-page":"586","DOI":"10.1007\/s11604-024-01712-2","article-title":"Structured clinical reasoning prompt enhances LLM\u2019s diagnostic capabilities in diagnosis please quiz cases","volume":"43","author":"Sonoda","journal-title":"Jpn J Radiol"},{"issue":"1","key":"R18","doi-asserted-by":"publisher","first-page":"143","DOI":"10.1136\/ard-2024-226202","article-title":"Assessing the performance of AI chatbots in answering patients\u2019 common questions about low back pain","volume":"84","author":"Scaff","journal-title":"Ann Rheum Dis"},{"key":"R19","doi-asserted-by":"publisher","DOI":"10.2196\/75567","article-title":"Evaluating the performance of state-of-the-art artificial intelligence chatbots based on the WHO global guidelines for the prevention of surgical site infection: cross-sectional study","volume":"27","author":"Wang","journal-title":"J Med Internet Res"},{"issue":"5","key":"R20","doi-asserted-by":"publisher","first-page":"1043","DOI":"10.1093\/mr\/road115","article-title":"Large language model may assist diagnosis of SAPHO syndrome by bone scintigraphy","volume":"34","author":"Mori","journal-title":"Mod Rheumatol"},{"issue":"7","key":"R21","doi-asserted-by":"publisher","first-page":"2058","DOI":"10.1016\/j.arthro.2023.12.009","article-title":"ChatGPT responses to common questions about anterior cruciate ligament reconstruction are frequently satisfactory","volume":"40","author":"Li","journal-title":"Arthroscopy"},{"issue":"8","key":"R22","doi-asserted-by":"publisher","first-page":"ITC113","DOI":"10.7326\/AITC202108170","article-title":"Low back pain","volume":"174","author":"Chou","journal-title":"Ann Intern Med"},{"issue":"10294","key":"R23","doi-asserted-by":"publisher","first-page":"78","DOI":"10.1016\/S0140-6736(21)00733-9","article-title":"Low back pain","volume":"398","author":"Knezevic","journal-title":"Lancet"},{"key":"R24","doi-asserted-by":"publisher","DOI":"10.2196\/50638","article-title":"Prompt engineering as an important emerging skill for medical professionals: tutorial","volume":"25","author":"Mesk\u00f3","journal-title":"J Med Internet Res"},{"issue":"2","key":"R25","doi-asserted-by":"publisher","DOI":"10.1002\/imt2.178","article-title":"ChatGPT and generative AI are revolutionizing the scientific community: a janus-faced conundrum","volume":"3","author":"Pu","journal-title":"Imeta"},{"key":"R26","doi-asserted-by":"publisher","DOI":"10.3389\/fmed.2024.1504532","article-title":"A guide to prompt design: foundations and applications for healthcare simulationists","volume":"11","author":"Maaz","journal-title":"Front Med (Lausanne)"},{"issue":"2","key":"R27","doi-asserted-by":"publisher","DOI":"10.1001\/jamanetworkopen.2024.57879","article-title":"Large language models for chatbot health advice studies: a systematic review","volume":"8","author":"Huo","journal-title":"JAMA Netw Open"},{"issue":"10","key":"R28","doi-asserted-by":"publisher","DOI":"10.2106\/JBJS.RVW.24.00122","article-title":"Generative artificial intelligence and prompt engineering: a primer for orthopaedic surgeons","volume":"12","author":"Carroll","journal-title":"JBJS Rev"},{"issue":"4","key":"R29","first-page":"343","volume":"85","author":"Casazza","journal-title":"Am Fam Physician"},{"issue":"9","key":"R30","doi-asserted-by":"publisher","first-page":"2788","DOI":"10.1007\/s00586-016-4684-0","article-title":"Red flags presented in current low back pain guidelines: a review","volume":"25","author":"Verhagen","journal-title":"Eur Spine J"},{"issue":"5","key":"R31","doi-asserted-by":"publisher","first-page":"1438","DOI":"10.1016\/j.arthro.2024.06.021","article-title":"The large language model ChatGPT-4 exhibits excellent triage capabilities and diagnostic performance for patients presenting with various causes of knee pain","volume":"41","author":"Kunze","journal-title":"Arthroscopy"},{"key":"R32","doi-asserted-by":"publisher","DOI":"10.1016\/j.csbj.2024.12.013","article-title":"Evaluating ChatGPT, Gemini and other large language models (LLMs) in orthopaedic diagnostics: a prospective clinical study","volume":"28","author":"Pagano","journal-title":"Comput Struct Biotechnol J"},{"issue":"1","key":"R33","doi-asserted-by":"publisher","DOI":"10.1186\/s10195-023-00740-4","article-title":"Arthrosis diagnosis and treatment recommendations in clinical practice: an exploratory investigation with the generative AI model GPT-4","volume":"24","author":"Pagano","journal-title":"J Orthop Traumatol"},{"issue":"12","key":"R34","doi-asserted-by":"publisher","first-page":"e538","DOI":"10.1016\/S1470-2045(14)70442-5","article-title":"International myeloma working group updated criteria for the diagnosis of multiple myeloma","volume":"15","author":"Rajkumar","journal-title":"Lancet Oncol"},{"issue":"5","key":"R35","doi-asserted-by":"publisher","first-page":"464","DOI":"10.1001\/jama.2022.0003","article-title":"Diagnosis and management of multiple myeloma: a review","volume":"327","author":"Cowan","journal-title":"JAMA"},{"issue":"1","key":"R36","doi-asserted-by":"publisher","DOI":"10.1038\/s41746-024-01074-z","article-title":"Foundation metrics for evaluating effectiveness of healthcare conversations powered by generative AI","volume":"7","author":"Abbasian","journal-title":"NPJ Digit Med"},{"issue":"4","key":"R37","doi-asserted-by":"publisher","DOI":"10.1016\/j.injury.2025.112297","article-title":"Artificial intelligence versus orthopedic surgeons as an orthopedic consultant in the emergency department","volume":"56","author":"Liu","journal-title":"Injury"},{"key":"R38","doi-asserted-by":"publisher","DOI":"10.3389\/fmed.2024.1366514","article-title":"Common differential diagnosis of low back pain in contemporary medical practice: a narrative review","volume":"11","author":"Ferdinandov","journal-title":"Front Med (Lausanne)"},{"issue":"3","key":"R39","doi-asserted-by":"publisher","DOI":"10.1007\/s11916-019-0757-1","article-title":"Low back pain, a comprehensive review: pathophysiology, diagnosis, and treatment","volume":"23","author":"Urits","journal-title":"Curr Pain Headache Rep"},{"key":"R40","doi-asserted-by":"publisher","DOI":"10.1177\/20552076241265215","article-title":"Diagnostic performance of generative artificial intelligences for a series of complex case reports","volume":"10","author":"Hirosawa","journal-title":"Digit Health"},{"key":"R41","doi-asserted-by":"publisher","DOI":"10.2196\/53008","article-title":"Generative AI in medical practice: in-depth exploration of privacy and security challenges","volume":"26","author":"Chen","journal-title":"J Med Internet Res"},{"issue":"5","key":"R42","doi-asserted-by":"publisher","first-page":"451","DOI":"10.1016\/j.clindermatol.2024.06.014","article-title":"Legal implications of artificial intelligence in health care","volume":"42","author":"Rosic","journal-title":"Clin Dermatol"},{"issue":"1","key":"R43","doi-asserted-by":"publisher","DOI":"10.1038\/s41746-025-01900-y","article-title":"A practical framework for appropriate implementation and review of artificial intelligence (FAIR-AI) in healthcare","volume":"8","author":"Wells","journal-title":"NPJ Digit Med"},{"key":"R44","doi-asserted-by":"publisher","DOI":"10.2196\/48633","article-title":"Barriers to and facilitators of artificial intelligence adoption in health care: scoping review","volume":"11","author":"Hassan","journal-title":"JMIR Hum Factors"},{"key":"R45","doi-asserted-by":"publisher","DOI":"10.2196\/56110","article-title":"ChatGPT with GPT-4 outperforms emergency department physicians in diagnostic accuracy: retrospective analysis","volume":"26","author":"Hoppe","journal-title":"J Med Internet Res"},{"issue":"11","key":"R46","doi-asserted-by":"publisher","first-page":"1362","DOI":"10.1001\/jamadermatol.2021.3129","article-title":"Lack of transparency and potential bias in artificial intelligence data sets and algorithms: a scoping review","volume":"157","author":"Daneshjou","journal-title":"JAMA Dermatol"},{"issue":"11","key":"R47","doi-asserted-by":"publisher","first-page":"2929","DOI":"10.1038\/s41591-023-02608-w","article-title":"The value of standards for health datasets in artificial intelligence-based applications","volume":"29","author":"Arora","journal-title":"Nat Med"},{"key":"R48","doi-asserted-by":"publisher","DOI":"10.2196\/53787","article-title":"The role of large language models in transforming emergency medicine: scoping review","volume":"12","author":"Preiksaitis","journal-title":"JMIR Med Inform"},{"key":"R49","doi-asserted-by":"publisher","DOI":"10.3389\/frai.2024.1452469","article-title":"Large language model triaging of simulated nephrology patient inbox messages","volume":"7","author":"Pham","journal-title":"Front Artif Intell"}],"container-title":["Journal of Medical Internet Research"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.jmir.org\/2026\/1\/e92315","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,31]],"date-time":"2026-07-31T22:23:57Z","timestamp":1785536637000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.jmir.org\/2026\/1\/e92315"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,3]]},"references-count":49,"URL":"https:\/\/doi.org\/10.2196\/92315","relation":{},"ISSN":["1438-8871"],"issn-type":[{"value":"1438-8871","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,3]]},"article-number":"v28i1e92315"}}