Implementing a prediction driven framework for emergency department nurse staffing to optimize real time decisions

Area of Science:

• Healthcare Management
• Nursing Informatics
• Emergency Medicine

Background:

• Emergency departments face challenges in optimizing nurse staffing due to unpredictable patient volumes.
• Inefficient staffing can lead to increased costs and compromised patient throughput.

Purpose of the Study:

• To implement and evaluate a prediction-driven nurse staffing framework in a large adult emergency department.
• To assess the impact of the framework on patient throughput and staffing costs.

Main Methods:

• A pre-post study design was used to compare outcomes before and after framework implementation.
• A two-stage prediction model forecasted patient volume to guide staffing decisions.
• Key performance indicators included door-to-evaluation time, length of stay, left-without-being-seen rate, and hourly nurse staffing costs.

Main Results:

• The prediction model demonstrated accuracy with RMSE values of 11.261 (base stage) and 9.973 (surge stage).
• Hourly nurse staffing costs were reduced by $162.04 post-implementation.
• No negative impact on patient throughput metrics was observed.
• Reducing nurse staffing by one hour increased wait times by two minutes.

Conclusions:

• Prediction-driven nurse staffing frameworks can effectively reduce operational costs in emergency departments.
• This approach maintains or improves patient throughput while optimizing resource allocation.
• Further research can explore optimal staffing thresholds to balance cost savings and patient wait times.