Utilizing and Optimizing Forecasting Models for Nursing Demand: A Narrative Review.

Area of Science:

• Healthcare Management
• Nursing Workforce Analytics
• Predictive Modeling

Background:

• Accurate nursing demand forecasting is vital for effective healthcare workforce planning.
• Increasing patient volumes and healthcare complexity challenge traditional methods.
• Evolving clinical demands, chronic diseases, and technology impact nursing needs.

Purpose of the Study:

• Map existing nursing demand prediction models.
• Identify key methodologies, data inputs, and evaluation metrics.
• Highlight gaps and implications for healthcare policy and workforce management.

Main Methods:

• Conducted a comprehensive narrative literature synthesis.
• Examined time-series models (ARIMA), machine learning (Random Forest, LSTM), and hybrid frameworks.
• Analyzed studies comparing methodologies, performance, and applicability.

Main Results:

• Advanced models (ML, hybrid) enhance prediction accuracy for proactive workforce planning.
• Limitations include inconsistent data quality, non-standardized validation, and poor contextual adaptation.
• Models often neglect dynamic factors like patient acuity and unit specifics.

Conclusions:

• Forecasting models can significantly improve nursing workforce planning.
• Effectiveness hinges on methodological rigor, data quality, and organizational readiness.
• Future work must integrate dynamic variables and enhance model validation for reliable, context-sensitive predictions.