Related Experiment Video
Updated: Sep 17, 2025
Implementation of a Real-Time Psychosis Risk Detection and Alerting System Based on Electronic Health Records using CogStack
Published on: May 15, 2020
Development and validation of a risk prediction model for unplanned 7-day readmission to PICU
Min Ding1, Chunfeng Yang1, Yumei Li2
1Department of Pediatric Intensive Care Unit, Children's Medical Center, The First Hospital of Jilin University, 1 Xinmin Street, Changchun, 130012, China.
Insights
A new risk prediction model identifies children likely to be readmitted to the pediatric intensive care unit (PICU) within 7 days. This tool aids in early detection and intervention for high-risk pediatric patients.
Area of Science:
- Pediatric Intensive Care
- Clinical Risk Prediction
- Healthcare Outcomes
Background:
- Unplanned readmissions to the pediatric intensive care unit (PICU) pose significant challenges.
- Developing accurate risk prediction models is crucial for timely intervention.
Purpose of the Study:
- To develop and validate a risk prediction model for 7-day unplanned readmission to the PICU.
- To identify key risk factors associated with PICU readmission in children.
Main Methods:
- A retrospective case-control study was conducted from November 2011 to September 2022.
- A risk prediction model was developed using a developing set and validated on a separate set.
- Model performance was assessed using receiver operating characteristic curves, calibration curves, and decision curve analysis.
Main Results:
- The final model incorporated five risk factors: complex chronic conditions, Pediatric Logistic Organ Dysfunction 2 scores, sedation, and Functional Status Scale score.
- The model demonstrated strong predictive performance with areas under the curve of 0.851 and 0.811 in the developing and validation sets, respectively.
- The model showed good calibration, decision curve analysis performance, and reproducibility.
Conclusions:
- The developed risk prediction model effectively assesses the probability of 7-day PICU readmission in children.
- This model can support early detection and targeted interventions for at-risk pediatric patients.
- Identifying children at high risk for readmission can improve patient outcomes and resource allocation.
Abstract:
We studied risk factors and predicted the probability of a child being readmitted to the pediatric intensive care unit (PICU) within 7 days of being discharged home. From November 2011 and September 2022, a retrospective case-control study was conducted to develop a risk prediction model in the PICU. The case group included children aged 1 month to 18 years discharged home who required unplanned 7-day readmission to the PICU. Non-readmitted children were chosen as controls. Characteristics were collected on the first admission and divided into a developing set and a validation set. In the developing set, a nomogram was established to predict the risk of readmission, and its performance was assessed by receiver operating characteristic curves, calibration curves, and decision curve analysis (DCA). Internal validation was eventually performed on the model. 5266 children were involved in the study, with 173 eligible children in the case group and 184 in the control group. The model included five risk characteristics: complex chronic conditions, higher Pediatric Logistic Organ Dysfunction 2 scores on admission and discharge, sedation, and the Functional Status Scale score. In the two datasets, the area under the curves was 0.851 and 0.811, respectively. The calibration curve and DCA both performed well. And the model showed great reproducibility. The model demonstrated good capability for assessing the risk of 7-day readmission to the PICU, which could support early detection and intervention.
