Fall Risk Prediction in Older Adults Using Free-Text Nursing Notes and Medications in Electronic Health Records.

Area of Science:

• Gerontology
• Health Informatics
• Artificial Intelligence in Healthcare

Background:

• Nursing notes within Electronic Health Records (EHR) contain valuable data for predicting patient outcomes.
• Fall risk prediction in older adults is crucial for preventing injuries and improving quality of life.
• Existing research has not fully explored the potential of nursing notes for fall risk prediction.

Purpose of the Study:

• To investigate the efficacy of deep learning models in predicting fall risk among older adults.
• To utilize nursing notes and medication data from EHR for fall risk prediction.
• To explore the application of natural language processing (NLP) in clinical data analysis for healthcare.

Main Methods:

• Recurrent neural network (RNN)-based NLP models were developed and trained.
• Pre-trained BioWordVec embeddings were used to represent clinical notes and medication data.
• Data from 162 older adults at a senior living facility were analyzed.

Main Results:

• The final deep learning model achieved an accuracy of 0.81.
• The model demonstrated a sensitivity of 0.75 and a specificity of 0.83.
• An F1 score of 0.82 indicates robust performance in predicting fall events.

Conclusions:

• Fall risk can be effectively predicted using nursing notes and medication information from EHR.
• Deep learning and NLP offer powerful tools for extracting actionable insights from clinical text.
• Future research should incorporate additional EHR data modalities to enhance prediction accuracy.