Delirium is a common condition in intensive care units (ICUs) that is associated with adverse clinical outcomes. Early identification and intervention are critical for improving the prognosis. This study aimed to develop a predictive model for identifying delirium in critically ill patients at the time of ICU admission.

A multicenter prospective observational cohort study was conducted, enrolling patients aged ≥20 years who were admitted to the ICU for more than 24 h and had no pre-existing cognitive impairment. Two cohorts—development and validation—were recruited during separate periods. Delirium was assessed daily using the Confusion Assessment Method for the ICU and the Diagnostic and Statistical Manual of Mental Disorders, 5th.

Among the 462 patients included in the development cohort, 29.0% developed delirium during their ICU stay. A novel delirium prediction model was developed and presented as a nomogram incorporating only seven easily obtainable variables. This model demonstrated good predictive performance (area under the curve [AUC], 0.736; 95% confidence interval [CI], 0.684–0.787). External validation confirmed its robustness (AUC, 0.810; 95% CI, 0.733–0.887), with superior performance compared to the PRE-DELIRIC model in both cohorts. To facilitate clinical implementation, we developed a user-friendly calculator application (available at shiny_delirium/ ) using shinyapps.io, which is compatible with both smartphones and computers.

We developed and externally validated a nomogram-based model to predict ICU delirium in critically ill patients. The model outperformed existing delirium prediction tools and can be conveniently used at the bedside as part of standard ICU care bundles.