[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"health-studies-list:{\"conditionNormalized\":\"diagnostic-errors\",\"overallStatus\":[\"RECRUITING\",\"AVAILABLE\",\"NOT_YET_RECRUITING\"],\"orderBy\":\"LastUpdateSubmitDate:desc\",\"size\":25,\"offset\":0}":3,"health-study-condition:diagnostic-errors":26},{"pageToken":4,"total":5,"offset":6,"count":5,"results":7},null,2,0,[8,43],{"id":9,"slug":10,"hasResults":11,"nctId":12,"briefTitle":13,"officialTitle":14,"acronym":15,"eligibilityCriteria":16,"healthyVolunteers":11,"sex":17,"minAge":18,"maxAge":4,"enrollmentInfo":19,"targetDuration":4,"studyType":22,"phases":4,"briefSummary":23,"conditions":24,"keywords":28,"overallStatus":30,"whyStopped":4,"lastUpdateSubmitDate":31,"lastUpdatePostDateStruct":32,"startDateStruct":35,"completionDateStruct":37,"leadSponsor":39,"locationsCount":42},"100642942","diagnostic-accuracy-of-gpt-4o-and-claude-46-sonnet-in-turkish-ed-anamnesis-notes-100642942",false,"NCT07632859","Diagnostic Accuracy of GPT-4o and Claude 4.6 Sonnet in Turkish ED Anamnesis Notes","Diagnostic Accuracy of Large Language Models From Emergency Department Anamnesis Notes: A Comparison of GPT-4o and Claude 4.6 Sonnet With Emergency Medicine Specialists","LLM-ED-DX-TR","INCLUSION CRITERIA:\n\n* Adult patients (aged 18 years and older) presenting to the emergency department.\n* Complete electronic health record available in the hospital information system (HBYS) containing a detailed anamnesis note with chief complaint, symptom duration, associated symptoms, and relevant medical history.\n* A definitive primary diagnosis recorded by the treating emergency physician using ICD-10 codes at the time of patient file closure.\n\nEXCLUSION CRITERIA:\n\n* Emergency department anamnesis notes containing fewer than 50 words or completely lacking substantive clinical content\\[cite: 1\\].\n* Pediatric cases (age under 18 years)\\[cite: 1\\].\n* Patients critically ill and triaged to high-acuity resuscitation areas (Emergency Severity Index \\[ESI\\] level 1)\\[cite: 1\\].\n* Clinical notes containing residual identifying information that cannot be fully de-identified, preventing compliance with data privacy regulations\\[cite: 1\\].\n* Non-independent clinical notes consisting solely of a brief cross-reference to a prior hospital visit without a new history entry\\[cite: 1\\].","ALL","18 Years",{"count":20,"type":21},600,"ESTIMATED","OBSERVATIONAL","This retrospective diagnostic accuracy study evaluates the ability of two large language models (LLMs) - GPT-4o (gpt-4o-2024-11-20; OpenAI) and Claude 4.6 Sonnet (claude-sonnet-4-6; Anthropic) - to generate correct diagnoses from anonymized Turkish-language emergency department (ED) anamnesis notes, and compares their performance with the diagnosis entered by the treating emergency physician. A consensus gold standard is established by three independent board-certified emergency medicine specialists who blindly review each note and vote on the primary diagnosis using ICD-10 three-character codes; the majority vote (at least 2 of 3 specialists agreeing) constitutes the reference standard. Both LLMs are evaluated using a standardized zero-shot direct prompting strategy (temperature=0, stateless API sessions). The primary outcome is diagnostic accuracy (proportion of ICD-10 chapter-level matches) and Cohen's kappa for each LLM against the gold standard. Secondary outcomes include top-3 accuracy, treating physician accuracy, inter-model agreement, and subgroup analyses by ESI triage level and ICD-10 chapter. Inter-rater reliability among the three specialists is quantified using Fleiss' kappa. Analyses are performed in Jamovi. This study represents the first evaluation of LLM diagnostic accuracy using Turkish-language clinical notes and the first to benchmark LLM performance against an independent three-specialist majority-vote gold standard rather than against the treating physician's own diagnosis.",[25,26,27],"Emergency Medicine","Diagnostic Errors","Artificial Intelligence (AI) in Diagnosis",[29],"Large Language Model; GPT-4o; Claude 4.6 Sonnet; ICD-10; Clinical Coding; Turkish; Emergency Department; Diagnostic Accuracy; STARD; STARD-AI","RECRUITING","2026-06-22",{"date":33,"type":34},"2026-06-25","ACTUAL",{"date":36,"type":21},"2026-06",{"date":38,"type":21},"2026-10",{"name":40,"class":41},"Marmara University Pendik Training and Research Hospital","OTHER",1,{"id":44,"slug":45,"hasResults":11,"nctId":46,"briefTitle":47,"officialTitle":48,"acronym":49,"eligibilityCriteria":50,"healthyVolunteers":11,"sex":17,"minAge":18,"maxAge":4,"enrollmentInfo":51,"targetDuration":4,"studyType":53,"phases":54,"briefSummary":56,"conditions":57,"keywords":4,"overallStatus":58,"whyStopped":4,"lastUpdateSubmitDate":59,"lastUpdatePostDateStruct":60,"startDateStruct":62,"completionDateStruct":64,"leadSponsor":66,"locationsCount":5},"100496708","achieving-diagnostic-excellence-through-prevention-and-teamwork-100496708","NCT05747755","Achieving Diagnostic Excellence Through Prevention and Teamwork","Achieving Diagnostic Excellence Through Prevention and Teamwork (ADEPT)","ADEPT","Inclusion Criteria:\n\n* Adult patients admitted to general medicine services at one of the participating hospitals and who either died during the hospitalization, were transferred to the ICU \\>= 48 hours after admission, or had a rapid response.\n\nExclusion Criteria:\n\n* Admitted for a non-medical reason\n* Patients coded in the field who are moribund on arrival to the hospital",{"count":52,"type":21},7200,"INTERVENTIONAL",[55],"NA","This study seeks to link a group of hospitals to measure and share the rates of diagnostic errors, to understand underlying causes of diagnostic errors, and develop ways that hospitals, clinicians, and patients can work together to avoid diagnostic errors and harms due to those errors. The investigators will test how data sharing and collaboration improve diagnostic processes and develop approaches which can be sustained into the future. The approach represents a novel application of rigorous outcome adjudication to the problem of inpatient diagnostic errors using a learning health system model.",[26],"NOT_YET_RECRUITING","2023-03-03",{"date":61,"type":34},"2023-03-07",{"date":63,"type":21},"2023-03",{"date":65,"type":21},"2026-09",{"name":67,"class":41},"Brigham and Women's Hospital"]