Speech recognition technology in prehospital documentation: A scoping review

Abstract

Objectives

The nature of paramedic workflows, where paramedics are responsible to provide care and chart concurrently, can result in incomplete or non-existent patient care reports on patient handover to the emergency department (ED). Charting delays and retrospective recollection of care may lead to patient information gaps, which can increase ED workload, cause care delays, and increase the risk of adverse events. Speech recognition documentation technology has the potential to produce complete patient care reports quicker and improve paramedic-to-ED handover. We performed a scoping review to determine paramedic perceptions and user requirements for speech recognition documentation technology.

Methods

MEDLINE, Google Scholar, IEEE Explore, ProQuest, and CINAHL were searched from 2014 to March 2024. Criteria included studies focused on paramedics’ use or perceptions of speech recognition documentation technology. This review included studies conducted in the prehospital environment and adjacent agencies (i.e., ED, fire, police, military).

Results

The review identified eight articles that met inclusion criteria. All eight articles were small focus group-based studies in laboratory settings published on or after 2020. Five studies were conducted in the United States, two in Switzerland, and one in Japan. Of the eight studies, five recommended further live environment testing of the technology examined, and three underscored the importance of a user-centred design. The top user requirements for speech recognition adoption was hands-free use, noise reduction technology, battery life, and word accuracy. All eight studies recommended further research and development of speech recognition documentation technology in the prehospital workflow.

Conclusion

This scoping review has highlighted that while there is a growing interest in speech recognition documentation technology in the paramedicine workflow, more research is needed, especially with larger samples in a live environment. The user requirements and perceptions of speech recognition documentation technology in paramedicine must be better understood to design systems with high adoption rates.