What question did this study set out to answer?

The study aims to develop and validate the Pediatric Risk of Mortality Evaluation (PRIME) algorithm using contemporary data.

28: Pediatric Risk of Illness Mortality Evaluation (Prime) Performance at Quaternary Children’s Hospital

View Full Paper

Ask AI

Bookmark

Authors

This retrospective cohort study evaluates mortality prediction in critically ill children, suggesting improvement in risk assessment methods.

The study aims to develop and validate the Pediatric Risk of Mortality Evaluation (PRIME) algorithm using contemporary data.
Conducted a retrospective cohort study of admissions to a quaternary PICU from 2015 to 2021.
Curated EHR data from the first 24 hours post-admission for model training and validation.
Utilized ridge logistic regression, extreme gradient boosting, and random forest models for mortality prediction.
Assessed model performance using AUROC, AUPRC, standardized mortality ratios, and calibration plots.
Out of 15,241 encounters, there were 286 deaths (1.9%) among critically ill children.
PRIME achieved excellent discrimination with an AUROC of 0.93 and AUPRC of 0.46.
PRIME showed good calibration with a Hosmer-Lemeshow statistic of P=0.47 and a Brier score of 0.013.
PRIME significantly outperformed the electronic Global Open Source Severity of Illness Score with a P-value of 0.01.

Nutman et al. (2026) studied this question.

View Full Paper

Ask AI

Bookmark

View Full Paper

Ask AI

Bookmark

Authors

This retrospective cohort study evaluates mortality prediction in critically ill children, suggesting improvement in risk assessment methods.

The study aims to develop and validate the Pediatric Risk of Mortality Evaluation (PRIME) algorithm using contemporary data.
Conducted a retrospective cohort study of admissions to a quaternary PICU from 2015 to 2021.
Curated EHR data from the first 24 hours post-admission for model training and validation.
Utilized ridge logistic regression, extreme gradient boosting, and random forest models for mortality prediction.
Assessed model performance using AUROC, AUPRC, standardized mortality ratios, and calibration plots.
Out of 15,241 encounters, there were 286 deaths (1.9%) among critically ill children.
PRIME achieved excellent discrimination with an AUROC of 0.93 and AUPRC of 0.46.
PRIME showed good calibration with a Hosmer-Lemeshow statistic of P=0.47 and a Brier score of 0.013.
PRIME significantly outperformed the electronic Global Open Source Severity of Illness Score with a P-value of 0.01.

Nutman et al. (2026) studied this question.

View Full Paper

Ask AI

Bookmark