[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-85104-en":3,"doc-seo-85104-105":29,"detail-sidebar-cat-0-en-105":90},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":13,"seo_description":14,"update_tm":27,"read_time":28},85104,687197207057,"Sage","https://ap-avatar.wpscdn.com/davatar_29158cc5080c5b710cf443261637dec0",7,"Healthcare","The Complexities of Patient-Centred Conversational Artificial Intelligence","Consumer-facing health chatbots powered by large language models are increasingly used for symptom assessment, yet development and evaluation often depend on cooperative simulated patients. Analysis of 2,053 real patient-chatbot conversations shows wide variation in communication patterns and emotional expression. A patient simulator was built to model clinical content, emotion, strategy, and communication style. In a Turing-inspired realism test, simulated conversations were nearly indistinguishable. Using five patient personae, communication style substantially altered triage outcomes, underscoring the need to accommodate patient diversity.","The complexities of patient-centred conversational artificial intelligence  \nJoão Matos1,2*, Olivia Buege1, Donny Cheung1, Gary S. Collins3,4, Paula Dhiman2, Nan Li1, Bingyu Mao1, Benjamin W. Nelson1,5, Michail Ouroutzoglou1, Paul Varghese1, Jonathan Amar1  \n1 Verily Health, Dallas, TX, USA  \n2 Centre for Statistics in Medicine, University of Oxford, Oxford, UK  \n3 Department of Applied Health Sciences, School of Health Sciences, University of Birmingham, Birmingham, UK  \n4 National Institute for Health and Care Research (NIHR) Biomedical Research Centre: Birmingham, Birmingham, University Hospitals Birmingham NHS Foundation Trust and University of Birmingham, Birmingham, UK  \n5 Division of Digital Psychiatry, Department of Psychiatry, Harvard Medical School and Beth Israel Deaconess Medical Center, Boston, MA, US  \n* Corresponding Author, [joao.matos@ndorms.ox.ac.uk](joao.matos@ndorms.ox.ac.uk)  \nAbstract  \nConsumer-facing health chatbots powered by large language models (LLMs) are increasingly used for symptom assessment. However, chatbot development and evaluation often rely on cooperative, articulate, simulated patients. We analysed 2,053 real patient-chatbot conversations and found that communication patterns and expression of emotions vary widely across users. We developed a patient simulator that separately models clinical content, emotional state, conversational strategy, and communication style. In a Turing-inspired evaluation of realism with 15 human graders, simulated conversations were nearly indistinguishable from real ones, with human graders achieving an accuracy of 55% . We used five distinct patient personae, across 1, 164 clinician-graded cases, to evaluate the performance of four LLMs in urgency assessment. We found that communication style can significantly alter triage outcomes. Patient-centred conversational artificial intelligence must accommodate communication diversity: systems designed for idealised, rather than realistic, interactions risk underperforming and amplifying health disparities when deployed in the real world.  \nConversational artificial intelligence (AI) is becoming an important interface between patients and healthcare 1. Consumer-facing applications with health assistants powered by large language models (LLMs), including ChatGPT Health 2, Verily Me 3, Microsoft Copilot Health 4, and, more recently, Google Health 5 allow individuals to describe symptoms, receive preliminary assessments, and decide when to seek care.  \nWhat these human-chatbot conversations gain in convenience and cost reduction is arguably lost in the wealth of contextual patient information. A physician conducting an in-person consultation draws on visual cues, body language, vocal tone, and physical examination findings that text-based exchange cannot fully capture 6. Multimodal inputs such as images 7,8 , sensor data from wearable devices 9, or synchronised electronic health records 10 can only partially compensate, hardly substituting for the contextual richness of direct clinical encounter.  \nConsumer-facing medical AI differs fundamentally from clinician-facing AI. In clinician-facing systems, such as imaging decision-support tools, a trained professional remains responsible for interpreting the output and integrating it into their clinical workflow 11. In consumer-facing applications, such as health chatbots, the system interacts directly with patients. The performance of the system will also depend on what patients understand and disclose, and how they communicate. The onus now falls more heavily on the patient: often without clinician supervision, users themselves must judge whether they are receiving safe, useful, and fair guidance.  \nThis makes conversational medical AI uniquely complex. Emotional, linguistic, and literacy nuances mediate every exchange. For instance, stigma or embarrassment may make a patient omit a relevant symptom unless the system elicits it sensitively. An anxious user may inadvertently ","cbCaij1hyg3UM69P","https://ap.wps.com/l/cbCaij1hyg3UM69P","pdf",2780916,1,36,"English","en",105,"# Abstract\n## Conversational AI in healthcare\n## Challenges unique to consumer-facing systems\n## Patient diversity, emotion, and missing information\n## Limits of current benchmarks","[{\"question\":\"How did the study assess realism between simulated and real patient–chatbot conversations?\",\"answer\":\"It used a Turing-inspired evaluation with 15 human graders to compare simulated conversations against real ones, finding simulated exchanges were nearly indistinguishable.\"},{\"question\":\"What did the analysis of 2,053 real conversations reveal?\",\"answer\":\"Communication patterns and emotional expression varied widely across users, indicating that patient interactions are not uniform.\"},{\"question\":\"How can communication style affect clinical triage outcomes?\",\"answer\":\"The study found that communication style can significantly change urgency assessment and triage dispositions, so models must handle interaction variability rather than idealized communication.\"}]",1784201131,91,{"code":4,"msg":30,"data":31},"ok",{"site_id":24,"language":23,"slug":32,"title":13,"keywords":33,"description":14,"schema_data":34,"social_meta":85,"head_meta":87,"extra_data":89,"updated_unix":27},"the-complexities-of-patient-centred-conversational-artificial-intelligence","",{"@graph":35,"@context":84},[36,53,67],{"@type":37,"itemListElement":38},"BreadcrumbList",[39,43,47,50],{"item":40,"name":41,"@type":42,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":44,"name":45,"@type":42,"position":46},"https://docshare.wps.com/document/","Document",2,{"item":48,"name":12,"@type":42,"position":49},"https://docshare.wps.com/document/healthcare/",3,{"item":51,"name":13,"@type":42,"position":52},"https://docshare.wps.com/document/the-complexities-of-patient-centred-conversational-artificial-intelligence/85104/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":23,"description":14,"dateModified":61,"datePublished":61,"encodingFormat":60,"isAccessibleForFree":62,"interactionStatistic":63},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":40,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-16",true,{"@type":64,"interactionType":65,"userInteractionCount":4},"InteractionCounter",{"@type":66},"ViewAction",{"@type":68,"mainEntity":69},"FAQPage",[70,76,80],{"name":71,"@type":72,"acceptedAnswer":73},"How did the study assess realism between simulated and real patient–chatbot conversations?","Question",{"text":74,"@type":75},"It used a Turing-inspired evaluation with 15 human graders to compare simulated conversations against real ones, finding simulated exchanges were nearly indistinguishable.","Answer",{"name":77,"@type":72,"acceptedAnswer":78},"What did the analysis of 2,053 real conversations reveal?",{"text":79,"@type":75},"Communication patterns and emotional expression varied widely across users, indicating that patient interactions are not uniform.",{"name":81,"@type":72,"acceptedAnswer":82},"How can communication style affect clinical triage outcomes?",{"text":83,"@type":75},"The study found that communication style can significantly change urgency assessment and triage dispositions, so models must handle interaction variability rather than idealized communication.","https://schema.org",{"og:url":51,"og:type":86,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":88,"canonical":51},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":91},[92,96,100,104,109,114,117,122,127,130,134],{"id":20,"doc_module":4,"doc_module_name":45,"category_name":93,"show_sort_weight":94,"slug":95},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":97,"show_sort_weight":98,"slug":99},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":45,"category_name":101,"show_sort_weight":102,"slug":103},"Exam",70,"exam",{"id":105,"doc_module":4,"doc_module_name":45,"category_name":106,"show_sort_weight":107,"slug":108},5,"Comic",60,"comic",{"id":110,"doc_module":4,"doc_module_name":45,"category_name":111,"show_sort_weight":112,"slug":113},6,"Technology",50,"technology",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":115,"slug":116},40,"healthcare",{"id":118,"doc_module":4,"doc_module_name":45,"category_name":119,"show_sort_weight":120,"slug":121},8,"Research & Report",30,"research-report",{"id":123,"doc_module":4,"doc_module_name":45,"category_name":124,"show_sort_weight":125,"slug":126},9,"Religion & Spirituality",20,"religion-spirituality",{"id":125,"doc_module":4,"doc_module_name":45,"category_name":128,"show_sort_weight":125,"slug":129},"World Cup","world-cup",{"id":131,"doc_module":4,"doc_module_name":45,"category_name":132,"show_sort_weight":131,"slug":133},10,"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":45,"category_name":136,"show_sort_weight":105,"slug":137},19,"General","general"]