ArticleBMC oral health2026
Accuracy, readability, and content coverage of AI-generated responses to questions on functional appliances.
Article in BMC oral health, 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
9 authors.
Funding
No grant is acknowledged in the PubMed record.
Abstract
backgroundPatients increasingly rely on Large Language Models (LLMs) for health information, yet the accuracy and readability of AI-generated dental advice remain variable across different clinical domains and Artificial Intelligence (AI) models. This study therefore aimed to compare the readability, accuracy, and comprehensiveness of responses generated by four leading AI models (ChatGPT-4o Mini, ChatGPT-5, Google Gemini 2.5 Flash, and DeepSeek V3) to patient questions on functional appliances.
methodsThirty-eight frequently asked questions were identified using a structured Google search and categorized into three domains: "treatment fundamentals and general information," "lifestyle and practical concerns," and "appointments and long-term results". Each question was independently answered by the four AI models. Readability was assessed using the Flesch-Kincaid tools. Accuracy and comprehensiveness were independently rated by two blinded orthodontists.
resultsAI-generated responses were generally accurate and comprehensive but difficult to read, requiring college-level literacy. ChatGPT-5 produced the lowest readability scores (most difficult-to-read responses; P < .001). Although Gemini 2.5 Flash achieved the highest comprehensiveness scores across all three domains, these differences were not statistically significant. Treatment-related questions yielded lower readability scores than lifestyle-related queries across all models (P < .001). No single model demonstrated superior performance across all evaluated domains.
conclusionAI-generated information on functional appliances was generally accurate and comprehensive but often exceeded recommended patient literacy thresholds. Readability must be considered alongside informational quality when deploying AI tools for patient education.
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.