Utilizing standardized nursing terminologies in implementing an AI-powered fall-prevention tool to improve patient...

Area of Science:

• Nursing Informatics
• Clinical Decision Support Systems
• Health Information Technology

Background:

• Standardized nursing terminologies (SNTs) are crucial for consistent knowledge representation and comparing nursing practice effectiveness.
• Implementing AI-powered tools requires semantic interoperability across diverse electronic medical record (EMR) systems.

Purpose of the Study:

• To investigate if SNTs can support semantic interoperability and outcome tracking.
• To evaluate the implementation of an AI-powered clinical decision support (CDS) tool for fall prevention across multiple EMR systems.

Main Methods:

• An AI-powered CDS tool was implemented in 4 hospitals with different EMR systems.
• The tool included a fall risk prediction model and CDS functions utilizing SNTs mapped to an ICNP-based catalog.
• Patient outcomes and nursing activities were tracked longitudinally.

Main Results:

• The AI-powered CDS tool was successfully disseminated across hospitals with consistent prediction model performance (AUROCs 0.8051-0.9581).
• Nursing process data significantly contributed to fall risk predictions.
• While fall rates did not significantly decrease, the rate of nursing interventions notably increased.

Conclusions:

• SNTs facilitated semantic interoperability, enabling the dissemination of AI-powered CDS tools.
• SNTs supported automatic tracking of nursing and patient outcomes across multiple EMR systems.