Impact of Large Language Model-Based AI Tools on Physician-Patient Communication: Systematic Review and Meta-Analysis

Area of Science:

• Artificial Intelligence in Medicine
• Clinical Communication Technology
• Natural Language Processing Applications

Background:

• Large language models (LLMs) like GPT-3/4 are driving AI chatbot development for medical applications.
• These AI tools aim to assist physician-patient communication, potentially improving empathy and clarity.
• The precise impact of LLM-based interventions on communication outcomes remains under investigation.

Purpose of the Study:

• To systematically review and meta-analyze studies on LLM interventions in physician-patient communication.
• To evaluate the effects of LLMs on empathy, clarity, trust, and patient understanding.
• To synthesize evidence from peer-reviewed literature published between 2020 and 2025.

Main Methods:

• Conducted a systematic review and meta-analysis following PRISMA 2020 guidelines.
• Searched major databases (PubMed, Embase, Scopus, Web of Science) for relevant studies (2020-2025).
• Included diverse study designs (randomized, observational, cross-sectional, qualitative) and performed qualitative synthesis and random-effects meta-analysis.

Main Results:

• Ten quantitative studies were included, predominantly cross-sectional.
• LLM-generated responses were rated significantly higher in empathy across multiple studies (SMD 1.02, 95% CI 0.44-1.60).
• LLMs improved perceived clarity and understanding, with one study showing increased comprehension scores and reduced consultation time.

Conclusions:

• LLM-based chatbots show potential to enhance physician-patient communication through more empathetic and understandable responses.
• Potential drawbacks include overly lengthy or inaccurate information, necessitating physician oversight.
• Further robust randomized controlled trials and longitudinal studies are required to confirm benefits and guide clinical integration.