The Development And Implementation Of A Forecasting Model For Inpatient Nurse Scheduling

Objective: This study was designed to develop a model for predicting nurse scheduling needs in a hospital unit based on historical patient census and nurse staffing requirements. Background: Many hospitals use outdated non-data driven methods for nurse scheduling. Methods: Historical nurse scheduling and staffing datasets for 2015, 2016, and 2017 from a 33-bed surgical unit in an inner-city urban hospital in Portland, Oregon, were used to build a predictive model for nurse scheduling needs. Results: The patient census for 2017 was three patients higher than the two previous years and showed a variation in the day of the week, with a consistent weekly trend of more nurses needed at the beginning of the week and fewer needed during the weekend. Conclusion: Based on model predictions, nurse scheduling in this unit should vary by day of the week, which has not historically been done