ArticleBMC anesthesiology2026
Evaluation of the accuracy, quality, and readability of large language models on local anesthesia and general anesthesia-sedation in dentistry.
Article in BMC anesthesiology, 2026. The graph could read no effect estimate from its abstract, so it casts no vote on the map. Not yet cited in PubMed.
What it found
Each row is one number read from the abstract, on the scale the paper reported it, with its interval. Left of the dashed line favours the treatment, right favours the comparator. Under each row is the sentence it came from. New to these charts? A ten-minute tutorial.
The abstract states no effect estimate the extractor could read, or names no intervention and outcome on the map, so this paper lights no cell and moves no belief. It is still indexed, cited and linked below.
The trial behind it
Trials whose registry record cites this paper, or whose number appears in the abstract. A trial that started after this paper was published is citing it as background, not reporting it.
Neither the registry nor the abstract names a trial number. If this is a trial report, that itself is worth knowing.
Who cites it
0 citing papers in PubMed.
No citing paper in PubMed yet.
Corrections and comments
PubMed lists nothing against this paper. Absence here is not a guarantee, only a check that was made.
Authors and funding
2 authors.
Funding
No grant is acknowledged in the PubMed record.
Abstract
purposeThis study aims to compare the responses provided by commonly used artificial intelligence-based chatbots such as ChatGPT-3.5, ChatGPT-4o, Gemini 2.0 Flash, and DeepSeek-R1 about dental local anesthesia, sedation, and general anesthesia in terms of accuracy, reliability, and readability.
methodsSixty questions were created from the American Dental Association (ADA) and American Society of Anesthesiologists (ASA) guidelines. Thirty were patient questions, thirty professional questions. Each group contained ten open-ended, ten multiple-choice, and ten true/false questions. The questions were submitted to ChatGPT-3.5, ChatGPT-4o, DeepSeek-R1, and Gemini 2.0 Flash. Four blinded pediatric dentists evaluated the answers with a modified global quality scale. Clinical safety and risk analysis were evaluated using a 3-point Likert scale. Readability was measured by Flesch Reading Ease Score (FRE) and Flesch-Kincaid Grade Level (FKGL). The intraclass correlation coefficient (ICC) tested inter-rater reliability. Significance was set at p < 0.05.
resultsDeepSeek-R1 demonstrated the highest overall accuracy and inter-rater agreement, providing the most accurate and reliability responses across all question types (p < 0.001). It was followed by Gemini 2.0 Flash, ChatGPT-4o, and ChatGPT-3.5. In terms of readability, Gemini 2.0 Flash consistently produced the most accessible responses, while DeepSeek-R1 was significantly less readable (p = 0.012). GPT-3.5 showed variability by question type, with MCQs being easier to read than open-ended ones (p = 0.027). No significant readability differences were observed across question types for ChatGPT-4o, Gemini 2.0 Flash, or DeepSeek-R1 (p > 0.05).
conclusionChatbot performance depends on both question type and evaluation criteria. DeepSeek-R1 excelled in accuracy and quality. Gemini 2.0 Flash produced the clearest, patient-friendly responses. AI chatbots can support communication in dental anesthesia. Choosing the right model may improve education, assist clinical training, and guide professional decisions.
Indexed as
Identifiers
What OpenQuestion holds
Registered trials
Read under generation 80e0d062 · epoch 390. Bibliography from PubMed, PubMed Central and OpenAlex; grants from NIH RePORTER; trial links from ClinicalTrials.gov; estimates, votes and beliefs from the OpenQuestion graph.