Photo Credit: Jarun011
Deployment of a deep learning model for early prediction of sepsis is associated with significant improvements in outcomes, including mortality, according to a study published in npj Digital Medicine. Aaron Boussina, PhD-candidate, and colleagues assessed how a deep learning model (COMPOSER) for early prediction of sepsis impacted patient outcomes. The analysis included a before-and-after quasi-experimental study design at two EDs that saw 6,217 adult patients who were septic from January 1, 2021, through April 30, 2023. Deployment of COMPOSER was significantly associated with a 1.9% absolute reduction (17% relative decrease) in in-hospital sepsis mortality. Additionally, there was a 5.0% absolute increase (10% relative increase) in sepsis bundle compliance and a 4% reduction in 72-hour sequential organ failure assessment score change after sepsis onset in a causal inference analysis. “Our COMPOSER model uses real-time data to predict sepsis before obvious clinical manifestations,” a coauthor said in a statement.
