Researchers at the University of California at San Francisco have developed an AI tool embedded in the EHR that can help clinicians identify which patients don't need to be woken up during the night to check on vitals.
An EHR tool that uses AI to predict vital signs could be used in hospitals to reduce nighttime check-ups and give patients a better chance to geta good night’s sleep.
That’s the take-away from the study conducted at the University of California at San Francisco and published in JAMA Internal Medicine. It points out that the clinical decision support tool could be used to identify low-risk or stable patients and reduce overnight disruptions in sleep, thus improving the patient’s health and well-being and preventing sleep issues like insomnia.
Researchers analyzed data from more than 1,900 patient encounters involving about 1,700 patients at UCSF between March and November of 2019, and created an algorithm that measures sleep promotion vitals (SPV), or normal nighttime vital signs. When embedded in the EHR, the algorithm could alert care team members that the patient has a 90% chance of maintaining normal vital signs overnight.
“No difference was found between groups in delirium incidence, but physicians usually agreed with the assessment of the clinical decision support tool and therefore discontinued overnight vital sign checks,” the researchers reported. “The intervention group experienced 31% fewer vital sign checks per night with no change in the rates of intensive care unit transfer or code blue alarms.”
The idea is to use technology to better manage data coming in to clinicians and refine workloads to cut out unnecessary tasks or identify important ones. With this particular CDS tool, clinicians could reduce unnecessary disruptions for some patients and concentrate overnight care teams on patients needing more strenuous monitoring.
This could also improve patient engagement and clinical outcomes, due in large part to the positive effects of a good night’s sleep.
The research was led by Nader Nafjani, MD, of the UCSF Department of Medicine, and supported by colleagues Mark J. Pletcher, MD, MPH, and Sajan Patel, MD, as well as Andrew Robinson, BS, from the University of California at San Francisco Medical Center.
Eric Wicklund is the associate content manager and senior editor for Innovation, Technology, and Pharma for HealthLeaders.