Multi-Level Uncinatectomies and Posterior Column Osteotomies to Correct a Cervical Kyphotic Deformity: Case Instruction With Intraoperative Picture and Video: Corrigendum
Year: 2026
Type: erratum
Abstract: Plain Language Summary BACKGROUND AND OBJECTIVES: Large Language Models (LLMs) are increasingly used by healthcare professionals and patients for medical information synthesis. Unfortunately, their tendency to generate fabricated information (“hallucinations”) poses a significant risk. This study quantifies and compares citation accuracy of 5 prominent LLMs for common neurosurgical spine topics. METHODS: Five LLMs (ChatGPT, Gemini, Claude, Microsoft Copilot, and OpenEvidence) were evaluated. Ten clinician-level questions derived from North American Spine Society guidelines were posed using a prompt instructing the LLM to act as an “experienced spine surgeon.” Ten corresponding patient-level questions were posed using a “patient” persona prompt. Both prompts requested a concise answer and 3 to 5 peer-reviewed references. Each citation was manually verified and scored as 2 (accurate), 1 (real but misrepresented), or 0 (fabricated). RESULTS: Performance varied significantly across models and question types. OpenEvidence, which directly indexes PubMed and requires user verification as a healthcare professional, achieved perfect accuracy. Among general purpose LLMs, Claude demonstrated the highest accuracy (average score 1.67, 78.0% accurate citations). Gemini had the lowest accuracy (average score 0.97) and the highest rate of fabrications (26.2%). All general purpose LLMs performed worse on patient-level queries. ChatGPT's accuracy score dropped from 1.83 to 1.08 for patient questions while its fabrication rate increased from 2.1% to 20.0%. Error profiles differed by model. Copilot frequently misrepresented article metadata (55.1% score-1 citations) and cited non–peer-reviewed websites (31.6%), particularly for patient questions. CONCLUSION: General purpose LLMs exhibit substantial and variable citation errors when queried on spine surgery topics. Accuracy is substantially lower for patient-facing prompts, which often yield fabricated references or nonscientific sources. While specialized, access-restricted platforms such as OpenEvidence provide high accuracy, the performance of widely accessible models highlights the need for careful verification of all LLM-generated medical information to ensure patient safety. Plain Language SummaryThis study examined how accurately 5 Large Language Models (LLMs)—ChatGPT, Gemini, Claude, Microsoft Copilot, and OpenEvidence—cite medical literature on common neurosurgical spine topics. Ten clinician-level and ten patient-level questions based on North American Spine Society guidelines were asked, each requesting 3 to 5 peer-reviewed references. OpenEvidence, which directly indexes PubMed and restricts access to healthcare professionals, showed perfect citation accuracy. Among general-purpose LLMs, Claude performed best (average score 1.67, 78.0% accurate citations), while Gemini performed worst (average score 0.97, 26.2% fabricated citations). All general-purpose LLMs were less accurate for patient-style questions, underscoring the need to independently verify LLM-generated medical information. Text is machine generated and may contain inaccuracies. FAQ
Source: Neurosurgery Open
Language: English
Subfield: Surgery
Field: Medicine
Domain: Health Sciences
Open Access status: diamond
OpenAlex ID: W7148628117
Cites:
Cited by:
Related to: