Effect of an AI Agent on Caregiver Communication Experience in the Pediatric Emergency Department

NCT07867782 | Completed

• 2 other identifiers: 2026-05-085-001RS-2023-KH135171
• Study Type: interventional
• Target: 34
• Locations: 1 country
• Sites: 1

Brief Summary

This single-center, single-blind, randomized controlled trial evaluates whether the use of a conversational large language model (LLM)-based AI agent, used by caregivers during the waiting period before physician consultation in a pediatric emergency department (ED), affects the caregiver-perceived effectiveness of clinical information delivery compared with standard care. The AI agent collects and summarizes the child's symptoms via voice before the consultation and does not participate in any diagnostic, prescribing, or treatment decisions. A total of 34 caregivers were enrolled at Samsung Medical Center.

Conditions

Communication

Keywords

pediatric emergency department; AI agent

Outcome Measures

Primary Outcomes (1)

• Caregiver-perceived effectiveness of clinical information delivery (PART 2-1)

  Caregiver-perceived effectiveness of clinical information delivery, measured with an investigator-developed exploratory instrument (PART 2-1). Five items on a 5-point Likert scale (total score range 5 to 25); higher scores indicate better perceived information delivery.

  Once, after completion of the index emergency department visit (assessed within approximately 1 hour after ED discharge)

Secondary Outcomes (2)

• ED communication and service experience (PART 1 and PART 2-2 to 2-9)

  Within 1 hour after emergency department discharge

• Emergency department length of stay

  From emergency department arrival to discharge, assessed up to 24 hours

Study Arms (2)

AI Agent (Intervention)

EXPERIMENTAL

Caregivers used the LLM-based AI agent on a study tablet for approximately 5 to 10 minutes during the pre-consultation waiting period, before the physician consultation.

Other: AI Agent (conversational LLM pre-consultation history-taking tool)

Standard Care (Control)

NO INTERVENTION

Caregivers received usual pediatric emergency department care without using the AI agent.

Interventions

AI Agent (conversational LLM pre-consultation history-taking tool) OTHER

A conversational large language model (LLM)-based AI agent used by caregivers on a study tablet before the physician consultation to collect and summarize the child's symptoms. It functions as a pre-consultation history-taking support tool and does not make diagnostic, prescribing, or treatment decisions.

AI Agent (Intervention)

Eligibility Criteria

• Age: 19 Years+
• Sex: all
• Healthy Volunteers: Yes
• Age Groups: Adult (18-64), Older Adult (65+)

You may qualify if:

• Caregiver aged 19 years or older of a child younger than 18 years presenting to the Samsung Medical Center pediatric emergency department
• Able to self-report and complete the questionnaire
• Provides written informed consent

You may not qualify if:

• Critical or emergent status (e.g., Korean Triage and Acuity Scale level 1)
• Caregiver unable to self-report
• Visual or hearing impairment precluding participation
• Refusal to participate
• Duplicate participation
• Otherwise unsuitable for the study objectives per investigator judgment

Sponsors & Collaborators

• Samsung Medical Center: lead
• Haheho Co., Ltd.: collaborator

Study Sites (1)

Samsung Medical Center

Seoul, 06351, South Korea

Location

MeSH Terms

Conditions

Communication

Condition Hierarchy (Ancestors)

Behavior

Study Design

• Study Type: interventional
• Phase: not applicable
• Allocation: RANDOMIZED
• Masking: SINGLE
• Who Masked: OUTCOMES ASSESSOR
• Purpose: HEALTH SERVICES RESEARCH
• Intervention Model: PARALLEL
• Sponsor Type: OTHER
• Responsible Party: PRINCIPAL INVESTIGATOR
• PI Title: Principal Investigator

Study Record Dates

• First Submitted: September 27, 2026
• First Posted: October 9, 2026
• Study Start: June 18, 2026
• Primary Completion: August 12, 2026
• Study Completion: August 12, 2026
• Last Updated: October 9, 2026

Record last verified: 2026-10

Locations

KR (1)