4.20 webinar LP

Scheduling the Right Staff Every Time with Machine Learning

May 17, 2023
May 17, 2023

This blog is taken from a recent HealthStream® webinar entitled “Scheduling the Right Staff Every Time with Machine Learning,” The webinar was moderated by HealthStream’s Sarah Enders and featured HealthStream Solution Consultant Manager, Terri Epler, MSN, RN and Gia Milo-Slagle, Senior Director of Product Management, HealthStream.

Scheduling and the last-minute scramble to find adequate staff to meet demand are stressful for both leaders and staff. Fortunately, there is a solution that can help your organization to accurately forecast long-term volume to help determine proper shift coverage and create a data-driven schedule up to 120 days in advance, along with a short-term Predictive Census guide that can inform staffing decisions within 7 days of a shift.


ShiftWizard – Transforming Healthcare Scheduling

Milo-Slagle began by introducing HealthStream’s ShiftWizard-a SaaS-based scheduling solution. ShiftWizard is an intuitive, icon-based scheduling experience that is coupled with an interactive communication center. It also integrates seamlessly with electronic medical record (EMR) systems as well as time and attendance systems and includes an easy-to-use mobile app for staff and leaders. In addition, ShiftWizard™ includes Predictive Census components that help ensure smart staffing decisions.


Artificial Intelligence and Machine Learning in Healthcare

Milo-Slagle distinguished artificial intelligence (AI) from machine learning (ML). Machine learning is a subset of AI, and MIT professor, Tommi S. Jaakkola, Ph.D. shared this definition of machine learning “The discipline tries to design, understand, and use computer programs that learn from experience (i.e., data) for the purpose of modeling, prediction or control.” Machine learning trains machines to improve at tasks without explicit programming, while artificial intelligence enables machines to think and make decisions just as a human would. “Machine learning algorithms are able to detect patterns in data and learn from them in order to make their own predictions,” said Milo-Slagle.

Machine learning takes the massive amounts of data collected in healthcare for analytics purposes and uses historical and current data to make predictions. Ultimately, the result is prescriptive information which is understanding the root cause of that prediction and prescribing actions which can change outcomes. The process is about removing the unknown from scheduling, resulting in an optimal experience for patients, their families, and staff and leaders.


Removing the Unknown from Scheduling

Scheduling in healthcare is a constant stressor for healthcare leaders. Predicting census for the organization, units, and departments to make good decisions about staffing is a significant contributor to that stress. ShiftWizard’s Predictive Census component can help forecast census, taking the guesswork out of scheduling.

While it can significantly reduce stress levels for managers, it has another benefit too. Better scheduling means higher levels of staff satisfaction as well, as they will be less likely to be on the receiving end of last-minute requests for additional shifts, floats, and overtime.

Using historical and live feed data, predictive analytics can help you match resources precisely to patient demand to have the resources that patients and families need. It also means that resources are not wasted during any dips in patient demand, which will have been predicted by this feature in ShiftWizard.


ShiftWizard and Decision Support for Leaders

Epler shared some of ShiftWizard’s reporting features that help leaders see at a glance whether or not they are staffed to meet the projected demand. She also pointed out that acuity information can be included to further refine predictions about needed resources.

ShiftWizard breaks down reporting and predictions into 4-hour segments to further ensure that schedules are appropriate for the day and specific shifts. In addition, users can create staffing worksheets that allow the user to compare their staffing matrix to projections even as the census is updated in real-time.

Additionally, Epler shared ShiftWizard’s productivity widget that allows users to monitor productivity metrics based on factors such as hours per patient day (HPPD), average daily census, and predicted daily census, all in four-hour increments to calculate a daily productivity metric.