Evaluation of a Generative AI-Based Voice Documentation System for Patient-Clinician Communication in the Emergency Department and Its Impact on Clinical Workflow Efficiency and Usability

Evaluation of a Generative AI-Based Communication and Clinical Documentation System and Its Impact on Workflow Efficiency in the Emergency Department

This research study at Samsung Medical Center evaluates the impact of Ambient AI system, a Generative AI-based communication and documentation system developed by NAVER, on the efficiency and quality of medical record-keeping within the Emergency Department (ED). By converting patient-clinician dialogues into automated medical summaries, the study aims to reduce the administrative burden on healthcare providers and address the issue of physician burnout caused by excessive documentation. The trial involves approximately 1,148 patients and 20 medical staff members, comparing a control period of conventional documentation against an experimental phase where AI-generated drafts are utilized to assist in clinical recording. Through quantitative analysis of documentation time, patient length of stay, and speech-to-text accuracy, alongside qualitative assessments of usability and satisfaction, the research seeks to determine if this AI intervention can effectively streamline ED workflows and enhance the overall quality of patient care in overcrowded medical environments.

Inclusion Criteria:

• Healthcare providers who have performed triage and initial examinations in the emergency department for at least 6 months
• Individuals who act as the primary subjects of communication during the triage and initial examination process
• Individuals who have voluntarily signed the informed consent form after receiving a detailed explanation of the research study

Exclusion Criteria:

• Individuals who do not agree to participate in the research
• Individuals who do not participate in the communication during the triage or initial examination process