Probabilistic priority assessment of nurse calls

Area of Science:

• Nursing Informatics
• Artificial Intelligence in Healthcare
• Clinical Workflow Optimization

Background:

• Current nurse call systems are static and lack situational adaptability.
• Existing systems do not leverage patient or staff profiles for intelligent call management.
• There is a need for dynamic and intelligent nurse call systems to improve patient care and operational efficiency.

Purpose of the Study:

• To introduce a probabilistic extension of the ontology-based Nurse Call System (oNCS).
• To develop a sophisticated nurse call algorithm that dynamically assigns priorities based on patient risk factors and call type.
• To evaluate the performance of the probabilistic oNCS compared to traditional systems using real-world data.

Main Methods:

• Development of a probabilistic extension to the ontology-based Nurse Call System (oNCS).
• Implementation of a prototype and simulations using data from three nursing departments.
• Comparative analysis of nurse arrival times, workload distribution, and call priority assignment against current systems.

Main Results:

• The probabilistic oNCS significantly improves nurse assignment to calls, leading to quicker response times.
• Workload distribution among nurses is enhanced, promoting a more balanced workflow.
• The system effectively incorporates patient risk factors and call types into priority assignments, with an average algorithm execution time of 50.333 ms.

Conclusions:

• The probabilistic oNCS offers a substantial improvement over traditional nurse call systems.
• Dynamic prioritization and intelligent algorithms enhance patient care by ensuring timely and appropriate responses.
• The developed system demonstrates the potential of ontology-based AI in optimizing clinical workflows and nurse resource allocation.