A practical and validated nomogram for predicting delirium in critically ill patients: A multicenter prospective...

Area of Science:

• Critical Care Medicine
• Neuroscience
• Predictive Analytics

Background:

• Delirium is a common, adverse outcome in intensive care units (ICUs).
• Early detection and intervention are crucial for improving patient prognosis.
• Predictive models can aid in identifying high-risk patients upon ICU admission.

Purpose of the Study:

• To develop and validate a predictive model for delirium in critically ill patients.
• To create a user-friendly tool for clinical implementation at the bedside.

Main Methods:

• A multicenter prospective observational cohort study.
• Development and validation cohorts recruited separately.
• Delirium assessed daily using Confusion Assessment Method for the ICU and DSM-5 criteria.

Main Results:

• A novel nomogram-based model incorporating seven variables was developed.
• The model showed good predictive performance (AUC 0.736) in the development cohort.
• External validation confirmed robustness (AUC 0.810), outperforming the PRE-DELIRIC model.

Conclusions:

• A validated nomogram-based model accurately predicts ICU delirium.
• The model offers superior performance compared to existing tools.
• A bedside-applicable calculator app facilitates clinical integration into care bundles.