Related Experiment Video
Updated: Jan 11, 2026
Digital Home-Monitoring of Patients after Kidney Transplantation: The MACCS Platform
Published on: April 12, 2021
Performance of AI chatbots in responding to geriatric patient questions on denture issues: A mixed method study of
Indumathi Sivakumar1, Sivakumar Arunachalam2, Praveen Gadde3
1Associate Professor, Faculty of Dentistry, SEGi University, Selangor, Malaysia.
Statement Of Problem:
Artificial intelligence (AI) chatbots have been increasingly used for health information, but their accuracy and ability to convey empathy remain uncertain, raising risks of misinformation and reduced trust among geriatric patients seeking a denture.
Purpose:
The purpose of this study was to evaluate the accuracy and empathy of responses from widely used AI chatbots to questions about complete denture problems in geriatric patients.
Material And Methods:
Five chatbots (ChatGPT GPT-3.5 [CG], DeepSeek R1 [DS], Claude 3.5 Sonnet [CD], Google Gemini [GG], and Microsoft Copilot [MC])) were asked 10 validated denture-related questions. Five prosthodontists independently rated chatbot responses for accuracy and empathy using validated scales. Statistical analysis assessed differences in chatbot accuracy and empathy across platforms and explored their interrelationship (α=.05). Qualitative insights were gathered through open-ended rater comments analyzed using a thematic coding approach.
Results:
Statistically significant differences were observed among platforms in both accuracy and empathy. GG demonstrated the highest overall mean ±standard deviation accuracy (3.3 ±0.50), significantly outperforming MC, which had the lowest (2.5 ±0.58; P<.001). Qualitatively, GG was praised for its comprehensive detail, while MC and CD were often criticized for being excessively concise. For empathy, MC achieved the highest proportion of empathetic responses (52%), with the highest overall mean empathy score (1.52 ±0.50), while CG had the lowest (1.24 ±0.47; P=.003). However, no chatbot consistently demonstrated high empathy. A statistically significant negative correlation was found between accuracy and empathy (r=-0.152, P=.016), indicating that higher accuracy was modestly associated with lower empathy. Qualitative analysis underscored the limitations of text-based AI in conveying genuine empathy.
Conclusions:
Significant variability and inconsistency were found in the accuracy and empathy of current AI chatbot responses in geriatric oral healthcare.
Related Concept Videos
01:15Drug Dosing: Geriatric Patients
01:27Pharmacodynamics in Geriatric Patients: Effects of Age
01:15Techniques of therapeutic communication I: Active Listening, Sharing Observations, Validation, and Using Touch
Therapeutic communication is not the same as social interaction. Social interaction has no goal or purpose and consists of casual information sharing, whereas therapeutic communication has a plan or purpose for the conversation. Therapeutic...
01:27Ethical Issues
Ethical Concerns in Healthcare:
01:13Patient-centered Care
01:21Barriers to Effective Communication II
Cultural barriers:
Differences in values, beliefs, religion, knowledge, and tradition can significantly impact communication. Awareness of nonverbal cues is critical, especially when conversing with a patient from a different culture. What appears appropriate in one culture may be inappropriate in another.
Semantic barriers:
As a result of their tendency to use...