Machine Learning for Predicting Critical Events Among Hospitalized Children - Pediatrics

Q: What problem does the study address in pediatric hospitals?

Unrecognized deterioration among hospitalized children is linked to high mortality and morbidity, while risk stratification is fragmented across hospital units that use different outcome-specific tools.

Q: How did the study define the primary critical events outcome?

Critical events were defined as invasive mechanical ventilation, administration of vasoactive medications, or death within 12 hours of an observation.

Q: Which model performed best for predicting hospitalwide critical events?

The extreme gradient-boosted (XGB) model outperformed existing ward-focused models in discrimination and alerting metrics, while deep learning models did not show improved performance.

Aug 3, 2026

Upload by Sage

Unrecognized deterioration in hospitalized children markedly increases mortality and morbidity, yet pediatric risk stratification remains fragmented because different hospital units rely on different prediction tools for specific outcomes. A retrospective cohort study develops and evaluates a machine learning model intended for early detection of deterioration across emergency, ward, and intensive care units, enabling a unified risk assessment throughout each child’s hospital stay.

LanguageEnglish

CategoryHealthcare

FormatPDF

Pages13

Reading Time33 min

AccessFree to download

Key Points

Question

Findings

Meaning

Introduction

Background on deterioration and risk models

Abstract

Importance, Objective, Design, Main Outcomes, Results, Conclusions

Original Investigation | Pediatrics Machine Learning for Predicting Critical Events Among Hospitalized Children Sierra Strutz, MS; Huan Liang, MS; Kyle Carey, MPH; Fereshteh Bashiri, PhD; Priti Jani, MD, MPH; Emily Gilbert, MD; Julie L. Fitzgerald, MD; Nicholas Kuehnel, MD; Maya Dewan, MD, MPH; L. Nelson Sanchez-Pinto, MD, MBI; Dana Edelson, MD, MS; Majid Afshar, MD, MS; Matthew Churpek, MD, MPH, PhD; Anoop Mayampurath, PhD Abstract IMPORTANCE Unrecognized deterioration among hospitalized children is associated with a high risk of mortality and morbidity. The current approach to pediatric risk stratification is fragmented, as each hospital unit (emergency, ward, or intensive care) uses different tools for predicting specific outcomes. OBJECTIVE To develop a machine learning model for the early detection of deterioration across all units, thereby enabling a unified risk assessment throughout the patient’s hospital stay. DESIGN, SETTING, AND PARTICIPANTS This retrospective cohort study used data from pediatric (age <18 years) admissions to inpatient and intensive care units at 3 tertiary care academic hospitals. Data were analyzed from January 2024 to March 2025 . MAIN OUTCOMESAND MEASURES The primary outcome was critical events, defined as invasive mechanical ventilation, administration of vasoactive medications, or death within 12 hours of an observation. RESULTS The cohort included 135 621 patients (mean [SD] age, 7 [6] years; 60 376 [44.5%] female) . Patient age, hospital unit, vital signs, laboratory results, and prior comorbidities were used to derive a regression-based model, an extreme gradient-boosted machine (XGB) model, and 2 deep learning models. Data from 2 hospitals were used as a derivation cohort, while patients in the third hospital constituted the hold-out external test cohort. The XGB model was the best-performing machine learning model, outperforming 2 existing ward-focused models in terms of discrimination (C statistic: XGB, 0.86; ward-focused models, 0.82 [P < .001] and 0.70 [P < .001]) and the number needed to alert (at an example 80% sensitivity: XGB, 6 ward-focused models: 9 and 11). The deep learning models did not exhibit improved performance. The XGB model performed better or equivalent to models trained for a specific hospital unit. CONCLUSIONSAND RELEVANCE This retrospective cohort study describes the development of a novel hospitalwide model for continuously predicting the risk of critical events through the entirety of a child’s stay. The model facilitated a unified framework for risk assessment in a pediatric hospital. JAMA Network Open. 2025;8(5):e2513149. doi:10.1001/jamanetworkopen.2025.13149 Introduction Key Points Question Can a hospitalwide machine learning model accurately predict critical events among hospitalized children across emergency, ward, and intensive care units? Findings In this cohort study including data from 135 621 pediatric patients, a gradient-boosted machine learning model demonstrated superior performance in predicting hospitalwide critical events, defined as mechanical ventilation, administration of vasoactive drugs, or mortality, compared with clinical standards and other machine learning models. The gradient-boosted machine learning model also showed equivalent or better performance than models trained for a specific hospital unit. Meaning These findings suggest that a gradient-boosted machine learning model can continuously assess risk forchildren as they progress through their hospital stay, potentially improving outcomes for children. + Invited Commentary + Supplemental content Author affiliations and article information are listed at the end of this article. Physiological decompensation in hospitalized children that requires the initiation of mechanical ventilation or administration of vasoactive medications increases the risk for mortality.1-6 Survivors of these events remain at risk for long-term functional or neurodevel

FAQs

What problem does the study address in pediatric hospitals?

Unrecognized deterioration among hospitalized children is linked to high mortality and morbidity, while risk stratification is fragmented across hospital units that use different outcome-specific tools.

How did the study define the primary critical events outcome?

Critical events were defined as invasive mechanical ventilation, administration of vasoactive medications, or death within 12 hours of an observation.

Which model performed best for predicting hospitalwide critical events?

The extreme gradient-boosted (XGB) model outperformed existing ward-focused models in discrimination and alerting metrics, while deep learning models did not show improved performance.

Aug 3, 2026

Upload by Sage