Certain AI tools may help alleviate emergency department overcrowding, researchers at Rochester, Minn.-based Mayo Clinic found.
The study authors conducted a prospective study evaluating an AI model that predicts hospital admission risk, alternately showing and hiding its suggestions to clinicians over an 11-month period and covering 52,394 ED visits in 2023. The AI solution reduced median ED length of stay by 12 minutes while not changing the number of ED patients discharged per day, the researchers concluded in the May 12 Nature Communications study. Hospitalists found the tool more useful than ED clinicians did.
“These findings show that integrating a low-burden AI prediction tool into ED workflows can improve operational efficiency,” the authors wrote.
