External Validation of Delirium Prediction Models for Intensive Care Patients

Overview

Delirium is frequently observed among critically ill patients and is associated with detrimental outcomes. Currently, no evidence-based prevention or treatment exists for delirium, and especially, the inability to identify effective preventive measures of delirium has increased the focus on identifying patients with a high risk of delirium through prediction models. Two prediction models have been developed to estimate the risk of delirium in ICU patients: the prediction model for delirium (PRE-DELIRIC) and the early prediction model for delirium (E-PRE-DELIRIC). These robust and well-calibrated prediction models have the potential of assisting in identifying patients with the highest risk for delirium and thereby to focusing preventive strategies on the most vulnerable group. However further validation is needed in a Danish population.

Adult ICU patients admitted to the participating ICUs in the Region Zealand are included according to eligibility. Eligible patients will be assessed with E-PREDELIRIC at admission to the ICU and PREDELIRIC 24h after admission to the ICU. Patients will be screened twice daily with CAM-ICU until ICU discharge, death, transferal to non-participating ICU or for a maximum observation period of 14 days. Central screening via EPIC PDM and registration in central database (REDCap, ZUH) by the principal investigator. The following are registered: Age, sex, history of cognitive impairment, history of alcohol abuse, admission category, urgent admission, mean arterial blood pressure at the time of ICU admission, use of corticosteroids, respiratory failure, Blood Urea Nitrogen/Carbamide, APACHE-II score, administered morphine does, use of sedatives, metabolic acidosis, coma, presence of infection, delirium status by CAM-ICU, time of ventilation, length of ICU and hospital stay, mortality.

Conditions

Interventions