Ambient Artificial Intelligence Scribe Adoption and Documentation Time in the Emergency Department

Area of Science:

• Emergency Medicine
• Health Informatics
• Artificial Intelligence

Background:

• The integration of artificial intelligence (AI) scribes into emergency departments (EDs) is an emerging trend.
• Real-world adoption patterns and efficiency impacts of these tools require thorough investigation.

Purpose of the Study:

• To evaluate the adoption rate and characteristics of ambient AI scribe usage in a real-world ED setting.
• To compare documentation time and electronic health record (EHR) note characteristics between encounters utilizing ambient AI and standard documentation methods.

Main Methods:

• A retrospective observational study was conducted on adult ED encounters at a tertiary academic medical center.
• Data were extracted from EHR audit logs for encounters with and without ambient AI scribe use, excluding those with human scribes.
• Adoption was analyzed by physician, ED zone, and patient acuity, with median times compared between groups.

Main Results:

• Ambient AI scribe use was observed in 11.2% of eligible encounters, with adoption concentrated among a small group of physicians and in specific ED zones (telemedicine, vertical care).
• Physicians preferentially used ambient AI for lower-acuity patients and those not requiring interpreters.
• Encounters using ambient AI demonstrated significantly shorter median on-shift documentation time (-28%) and total EHR time (-16%) compared to standard encounters, with shorter note lengths.

Conclusions:

• Early implementation of ambient AI scribes in the ED reveals low but skewed adoption patterns.
• The use of ambient AI is associated with significant reductions in documentation time and EHR workload.
• Physician preference for specific patient types (lower acuity, no interpreters) highlights areas for targeted implementation and further study.