Why the study?
Unrecognized clinical deterioration during pediatric hospitalization is linked to high mortality and morbidity, motivating development of machine learning algorithms to identify ICU transfers within 12 hours.
Does a machine learning model improve prediction of direct ward to ICU transfer within 12 hours in pediatric inpatients compared to the Bedside Pediatric Early Warning Score?
Population
Pediatric inpatients (age <18 yr) across 50,830 admissions at site 1 and 88,970 admissions at site 2
Comparison
Machine learning algorithms vs modified Bedside Pediatric Early Warning Score
Design
Observational cohort study across two urban, tertiary-care, academic hospitals
Authors
Loading...
ML models may enable earlier detection of deterioration in hospitalized children; hypothesis-generating and requires prospective validation before clinical adoption.
Does a machine learning model improve prediction of direct ward to ICU transfer within 12 hours in pediatric inpatients compared to the Bedside Pediatric Early Warning Score?
A machine learning model using age, vital signs, and lab results accurately predicts ICU transfer in hospitalized children, outperforming standard early warning scores.
Mayampurath et al. (2022) studied this question.
Synapse has enriched 4 closely related papers on similar clinical questions. Consider them for comparative context: