A Pilot Study for Integrating Automatic Speech Recognition Into Nursing Documentation

The purpose of this study is to develop and integrate Automatic Speech Recognition (ASR) technology into electronic health record (EHR) for nursing documentation. This study was conducted at a hospital in Seoul, South Korea, and followed the constructive ehealth evaluation method (CeHEM) framework, encompassing 6 phases: research and planning, design, development, implementation, evaluation, and redesign. A multidisciplinary ASR Project task force team, composed of experts from diverse domains including nursing, ASR technology, and EHR systems, led the project. Version 1 of the Nursing ASR system underwent 2 pilot deployments. It was initially integrated into nursing notes and later expanded to the Kardex, which entails considerable nursing documentation. Following the pilots, feedback was gathered from clinical nurses through interviews and an online survey to identify ASR system issues and areas for improvement. Based on these findings, the task force team proposed targeted solutions that informed the development of Version 2. The updated system was redesigned to support a broader range of nursing tasks, offer more intuitive voice commands, and provide a mobile application version to enhance accessibility. This study highlights the importance of nurse-centered, iterative development and demonstrates how interdisciplinary collaboration can translate user feedback into system improvements. The findings provide practical insights for health care organizations adopting ASR technologies.