516 posters, 59 topics, 63 sessions, 1,127 authors, 353 institutions
ePostersLive by SciGen Technologies S.A. All rights reserved.
2026 SOAP Annual Meeting
April 29 - May 3, 2026 | Montreal, Quebec Canada
2341841
Optimizing Obstetric Pre-Admission Unit Efficiency and Enhancing Patient Experience through an AI-Powered Chatbot
Poster Presenter
Part of Topic
Efficiency, Education, Access
Optimizing Obstetric Pre-Admission Unit Efficiency and Enhancing Patient Experience through an AI-Powered Chatbot
Presenting Author: Mostafa Abbasi Dezfouly MD, Bsc. Senior Authors: Elio Belfiore MD, MHSc ; Daniel McIsaac MD MPH ; Shalini Sumathi MPH
Contributing Author: Jayne Acharya
Background
•Patients increasingly seek medical information online prior to clinical encounters, particularly during pregnancy1 (Sayakhot P. BMC Pregnancy Childbirth, 2016.)
•However, access to accurate and personalized obstetric anesthesia information before Pre-Anesthesia Unit visits remains limited.
Purpose
•To determine if a Large Language Model-based digital assistant in the Pre-Anesthesia Unit of Canadian academic hospitals can improve the satisfaction and effectiveness of patient education in obstetrics.
Phase 1 – Chatbot Development
•Obstetric anesthesia literature review to curate knowledge base
•Frequently asked questions (n = 50) identified from obstetric anesthesia resources
•Chatbot responses reviewed by anesthesiologists at The Ottawa Hospital
Phase 2 – Patient Evaluation
•Obstetric patients ≥18 years attending the Pre-Anesthesia Unit
•Patients asked one question (with optional follow-up)
•Post-interaction satisfaction survey
Sample size
•Target n = 20 based on qualitative saturation principles2 (Vasileiou K. et al., BMC Med Res Methodol 2018)
•Clinicians agree the Chatbot gives appropriate, understandable answers.
•Occasional need for clarification
•Patients strongly agree the Chatbot provides appropriate answers.
•Patient comments note that it would be a helpful supplement to the PAU visit
•Recent literature has yielded similar conclusions: that AI chatbots have potential to supplement traditional clinic visits3 (Suri et al. Int J Obstet Anesth, 2025)
•Study limitations: Selection bias, response bias, subjective outcome measures