Improving situation awareness to reduce healthcare-acquired urinary tract infection
Date
2024
Authors
Alqarrain, Yaser
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Reducing healthcare-acquired urinary tract infections (HAUTI) is a common goal among healthcare providers and organizations. Nurses' situation awareness (SA) skills would likely improve patient status recognition and prevent healthcare-acquired urinary tract infections. Healthcare providers, such as nurses, need eHealth systems that support their situation awareness as they provide care.
Integrating Endsley's design principles with machine learning offers a promising approach for developing an SA-oriented dashboard that could help reduce HAUTI. This study takes an initial step toward this goal by exploring context-based variables contributing to HAUTI. I included a comprehensive list of nursing assessments and implemented multiple methodologies to handle the datasets and address missing data. The XGBoost model emerged as the most effective model in predicting HAUTI, isolating factors such as improving skin integrity and mobility and monitoring neurological status as key factors in reducing HAUTI rates. However, these results should be carefully interpreted, given this study's significant missing data. The finding of this study reinforces the necessity of high-quality data to support the interpretation of Machine Learning (ML) models in clinical settings.
Description
Keywords
Nursing, Situation Awareness, Machine Learning, Endsley Situation Awareness model, Urinary Tract Infection