Predicting the Nurse Perceived Adequacy of Staffing Scale (NPASS) Score Using Machine Learning

Ensuring adequate nurse staffing is essential for patient safety, quality of care, and nurse well-being. The Nurse Perceived Adequacy of Staffing Scale (NPASS)1 offers a valid and reliable measure of perceived staffing adequacy, yet it is primarily retrospective and rarely used in real-time decision-making. This project aims to develop and pilot a machine learning (ML)-based system that predicts NPASS scores using routinely collected hospital operational data, including nurse-to-patient ratios, staff and patient characteristics, shift patterns, and ward features. Through a co-design approach involving nurses, hospital administrators, and interdisciplinary experts in AI, nursing science, and healthcare operations, the project aims to produce a predictive tool tailored to clinical realities.
The resulting system will support more balanced and responsive workloads for nursing teams and equip hospital administrators with data-driven insights to optimize shift scheduling, determine Full-Time Equivalent (FTE) needs, and improve workforce allocation. Additionally, the analysis will shed light on work prioritization and task distribution patterns during shifts.