Hartford (Conn.) HealthCare has been able to drop hospital stays by 7% due to an algorithm, the Hartford Courant reported Feb. 27.
The healthcare organization has created a "length of stay algorithm," which helps predict when a patient is ready to be discharged.
"Since it's been implemented, we've dropped seven percent in hospital stays. It's enormous savings. Our goals are typically to drop 1 to 2 percent per year," Hartford Hospital Associate Vice President of Medical Affairs Daniel Kombert, MD, told the publication.
Dr. Kombert called the tool a "discharge consultation," stating that it's still up to the physicians, nurses and case manager to decide when to discharge a patient.
"We use this AI in our daily communication and on all of our medical reports across the system and we've seen improvement across the board," he said.
Additionally, the tool has been able to identify patients with a higher likelihood of being discharged within a day by marking them green, according to Dr. Kombert.
"That information is discussed on the rounds and they can be reassessed whether they can go home safely. We always have cases where the doctor says it's not time to go home yet. There are others where they get the patients home in a timely manner," he said.