Accepted for/Published in: Journal of Medical Internet Research
Date Submitted: Apr 7, 2025
Date Accepted: Aug 28, 2025
Warning: This is an author submission that is not peer-reviewed or edited. Preprints - unless they show as "accepted" - should not be relied on to guide clinical practice or health-related behavior and should not be reported in news media as established information.
Better Together: Exploring Patient Perspectives, Engagement, and Output Quality in Doctor-supervised Use of AI During Informed Consent Consultation with ChatGPT and Retrieval Augmented Generation (RAG)
ABSTRACT
Background:
Informed consent is essential for preoperative patient education, ensuring understanding of surgical procedures, risks, and postoperative care. AI chatbots, such as ChatGPT, offer potential to enhance patient education yet caution is needed.
Objective:
To evaluate patient satisfaction and the quality of ChatGPT's responses in informed consent consultations for total hip arthroplasty (THA) and assess the impact of integrating subject- and hospital-specific data via Retrieval-Augmented Generation (RAG).
Methods:
Patients scheduled for elective THA were assigned to three groups (n=12/group): traditional physician-led consultations (control), physician-assisted consultations using native ChatGPT, and physician-assisted consultations using ChatGPT+RAG. Patient satisfaction, preferences, health literacy, anxiety, and general attitude towards artificial intelligence were assessed using Likert scales and questionnaires. ChatGPT responses were evaluated for relevance, accuracy, clarity, completeness, and evidence-based content. Hallucinations were assessed and categorized for severity. Statistical analyses compared group differences.
Results:
Patients using ChatGPT+RAG reported higher satisfaction with information delivery, informedness, and the amount of information received compared to native ChatGPT. AI-assisted groups asked more questions than the control group, with ChatGPT+RAG prompting the most engagement. Despite ChatGPT’s benefits, patients predominantly preferred physician-led consultations, with state anxiety and patients’ general attitude towards AI weakly influencing this preference. ChatGPT+RAG significantly reduced hallucinations and achieved superior scores across all quality metrics compared to native ChatGPT.
Conclusions:
ChatGPT, especially with RAG integration, can enhance preoperative patient education by improving response quality and fostering engagement. However, patients continue to value human interaction, highlighting the complementary role of AI chatbots in informed consent.
Citation
Request queued. Please wait while the file is being generated. It may take some time.
Copyright
© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.