Related Experiment Video
Updated: Jul 29, 2026
Assessment and Evaluation of the High Risk Neonate: The NICU Network Neurobehavioral Scale
Published on: August 25, 2014
Machine Learning-Based Pediatric Early Warning Score: Patient Outcomes in a Pre- Versus Post-Implementation Study,
Anoop Mayampurath1,2, Kyle Carey3, Brett Palama4
1Department of Biostatistics & Medical Informatics, University of Wisconsin-Madison, Madison, WI.
Insights
The pediatric Calculated Assessment of Risk and Triage (pCART) tool significantly reduced critical events in high-risk hospitalized children. This machine learning model improved patient outcomes by enabling timely interventions and preventing direct ward to ICU transfers.
Area of Science:
- Pediatric critical care medicine
- Machine learning in healthcare
- Clinical decision support systems
Background:
- Direct ward to ICU transfers in children pose significant risks.
- Accurate and timely risk stratification is crucial for pediatric patient management.
- Existing methods may not adequately predict the need for intensive care.
Purpose of the Study:
- To describe the implementation of the pediatric Calculated Assessment of Risk and Triage (pCART) machine learning model.
- To evaluate the impact of pCART on predicting direct ward to ICU transfers within 12 hours.
- To assess the associated improvements in outcomes for hospitalized children.
Main Methods:
- A pre- vs. post-implementation study design was employed.
- The study included pediatric admissions (<18 years) at an urban, tertiary-care academic hospital.
- Data were collected from May 2019 to April 2023, divided into baseline and pCART implementation cohorts.
Main Results:
- pCART implementation was associated with a significant decrease in critical events (from 1.4% to 0.4%, p < 0.001).
- High-risk patients identified by pCART had over two-thirds lower adjusted odds of critical events (OR, 0.22; p < 0.001).
- No significant association was found with overall hospital/ICU length-of-stay, but a difference was noted in LOS for ICU-transferred patients.
Conclusions:
- Deployment of the pCART machine learning tool improved clinical decision support for pediatric ward patients.
- pCART implementation was linked to reduced odds of critical events in high-risk pediatric patients.
- The study highlights the potential of AI-driven tools in enhancing pediatric critical care outcomes.
Objectives:
To describe the deployment of pediatric Calculated Assessment of Risk and Triage (pCART), a machine learning (ML) model to predict the risk of the direct ward to the ICU transfer within 12 hours, and the associated improved outcomes among hospitalized children.
Design:
Pre- vs. post-implementation study.
Setting:
An urban, tertiary-care, academic hospital.
Patients:
Pediatric (age < 18 yr) admissions from May 1, 2019, to April 30, 2023.
Interventions:
None.
Measurements And Main Results:
Patients were divided into baseline, pre-pCART implementation (May 1, 2019, to April 30 2021), and post-pCART implementation (May 1, 2021, to April 30, 2023) cohorts. First-ward admissions with a high-risk score (pCART score ≥ 92) were considered as the main cohort. The primary outcome was the occurrence of critical events, defined as invasive mechanical ventilation, vasoactive drug administration, or death within 12 hours of the first high-risk pCART score. There were 2763 and 3943 patients in the baseline and implementation cohorts, respectively. pCART implementation was associated with a decrease in the percentage of the primary outcome from baseline 1.4% to 0.4% (p < 0.001), which converted to more than two-thirds lower adjusted odds of the primary outcome (odds ratio, 0.22 [95% CI, 0.11-0.40]; p < 0.001). pCART implementation was also associated with a decreased prevalence of critical events at 24 and 48 hours after a first high-risk score. We failed to identify any association between cohort period and overall hospital and ICU length-of-stay, number of ICU transfers, and time to ICU transfer. However, there was a difference in hospital length-of-stay among a subpopulation of patients transferred to the ICU (median 6 vs. 7 d; p < 0.001). Analysis of compliance metrics indicates sustained compliance achievements over time.
Conclusions:
The deployment of pCART, a ML-based pediatric risk stratification tool, for clinical decision support for pediatric ward patients, was associated with lower odds of critical events among high-risk patients.
