The COVID-19 pandemic revealed new threats and opportunities throughout global healthcare infrastructure. As the world grappled with a deadly virus, healthcare professionals faced logistical challenges from the front lines. Now, with lockdowns lifted and the world of healthcare forever changed by the pandemic, what lessons can we learn and use to improve our hospitals, urgent care facilities, and emergency rooms moving forward?

At UAB, our Health Informatics masters program is dedicated to extracting insights from data and using those insights to improve healthcare. In the emerging field of predictive analytics, we may find answers to some of healthcare’s greatest hurdles. Here’s how UAB’s faculty plan to tackle emergency room crowding through predictive analytics.

Emergency Department Overcrowding

Tools and methods learned in a graduate Health Informatics program can be used to solve a multitude of contemporary healthcare problems. One of those problems has to do with overcrowding in Emergency Departments (EDs).

While overcrowding is not unique within the University of Alabama at Birmingham Health System (UABHS), our professors are lending their expertise in machine learning and predictive analytics to mitigate the problem of ED overcrowding. This is a critical problem to solve because it has been correlated to poor healthcare outcomes such as higher mortality rates, ambulance diversions, treatment delays, patients leaving without being seen, etc. (Moore et al., 2017; Arya et al., 2013; Sun et al., 2013; Araz et al., 2019).

What is Predictive Analytics?

Over the past ten years, machine learning algorithms have evolved and been used across various domains of healthcare for their ability to handle multidimensional and complex healthcare datasets. In the case of ED overcrowding, the main problem is that there is a lack of information that can be used to make operational decisions far enough in advance to impact care and flow of the ED.

Consider that a patient needs to be admitted from the ED to the hospital. Due to the limited information at the point of care, the decision to admit the patient is made after the patient is checked by a physician. This information is communicated with inpatient units too late and the patient has to wait until a bed is assigned. Predictive analytics uses very large data sets of different variables to help predict the likelihood of an admission the moment a patient arrives at an ED - sometimes even before the patient arrives. Then, the predicted information can be communicated with the inpatient unit early and a bed can be found by the time the patient is ready for admission.

This is a simple, yet effective method of using machine learning to develop predictive analytics to streamline hospital operations. Other important information that can be predicted using machine learning includes the volume of ED patients, average number of waiting patients, and more. Such information can be predicted early before an ED gets overcrowded and then used to manage daily operations in EDs more efficiently.

Better Care Through Analytics

At UAB, we’re exploring the myriad applications of predictive analytics throughout the world of healthcare and beyond. Our health informatics master’s program is dedicated to understanding how predictive analytics can be used throughout a health system for enhancing health outcomes, accelerating streamlined operations, and supporting point of care decisions.

Through accreditation, intimate class sizes, diverse learning opportunities, and expert faculty, UAB’s health informatics master’s program is training the innovators of tomorrow to predict unique challenges today.