Truace
Subscribe

Performance of large language models in providing patient- oriented information on gingival recession: a comparative study

Patients are increasingly turning to large language models (LLMs) for health information, but the quality of patient-oriented information remains uncertain. This article evaluated the performance of three major LLMs in answering clinically relevant patient's questions related to gingival recession. A descriptive comparative study was carried out utilizing twenty standardized patient-centered questions related to gingival recession. Each question was submitted once to ChatGPT 5.5, Gemini 3.5, and Claude Sonnet 5,…

By the Truvace desk2 min read1 sourcePeer-reviewedTRV-2026-2140· cite & verify
Performance of large language models in providing patient- oriented information on gingival recession: a comparative study
Dental radiology by Satterlee, Francis Le Roy, jr. Public domain

In brief

Researchers put twenty standardized patient-style questions about gingival recession to three chatbots, producing 60 answers. Five board-certified periodontists scored each answer for accuracy and completeness and ranked their preferences, while word count and response time were also logged.

The results suggest patients seeking dental information from these tools may receive broadly similar quality regardless of which one they use. The authors still frame the tools as complements to professional periodontal consultation, and the study does not show how patients actually use or act on the answers.

Main points

  1. Twenty patient-centered questions were each submitted once to three LLMs, yielding 60 responses.
  2. Accuracy (p = 0.993) and completeness (p = 0.972) did not differ significantly among the models.
  3. ChatGPT was fastest (median 5.82 s) and most concise (median 232 words).
  4. Gemini and Claude were each ranked first in 38% of evaluations versus 24% for ChatGPT.
  5. Post-treatment Care questions received the highest accuracy for all three models.

The gain

Periodontist ratings found ChatGPT, Gemini and Claude similarly accurate and complete on gingival recession questions, suggesting they could supplement patient education.

The rundown

The evaluation used a descriptive comparative design. Five board-certified periodontists independently applied the validated Accuracy of Information Index and a completeness scale, and analysis used non-parametric tests and generalized estimating equations with significance set at p ≤ 0.05.

Speed and length differed markedly. ChatGPT responded in a median of 5.82 seconds, against 8.46 for Gemini and 14.61 for Claude. Median word counts were 232 for ChatGPT, 438 for Gemini and 333.5 for Claude.

Expert preference rankings differed significantly (p = 0.002), with Gemini and Claude favored more often than ChatGPT. The authors conclude the tools may help with patient education but should not replace professional periodontal consultation.

Sources

  1. Peer-reviewedThe Saudi Dental Journal2026-10-19

The debate