Case Study: Steward Health Care achieves $2M annual savings and 95% staffing prediction accuracy with DataRobot

Reducing Costs with DataRobot at Steward Health

Steward Health Care, the largest for‑profit private hospital operator in the U.S. with 38 hospitals, faced a major challenge in staffing: static, average‑based schedules caused understaffing at peaks and overstaffing at valleys, driving up labor and overtime costs. Erin Sullivan, Steward’s Executive Director of Information Systems and Software Development, partnered with DataRobot to apply predictive analytics, AI, and machine learning—using DataRobot’s automated machine learning platform—to turn Steward’s extensive historical data into actionable forecasts.

DataRobot provided both the AutoML tools and hands‑on data science expertise (including support from Sergey Yurgenson) to build and deploy 384 day‑specific and 1,152 shift‑specific models into Steward’s Proactive Labor Management (PLM) dashboard on Azure. The models currently track daily accuracy around 95% and have produced measurable impact: a 1% reduction in registered nurse hours per patient day saved $2 million annually across eight hospitals, a 0.1% reduction in length of stay projects over $10 million in annual savings, and improved proactive staffing and supply planning.

Steward Health Care

Erin Sullivan

Executive Director of Information Systems and Software Development