DeLLiriuM: A large language model for delirium prediction in the ICU using structured EHR

Delirium is an acute syndrome characterized by fluctuating attention, cognitive impairment, and severe disorganization ofbehavior, which has been shown to affect up to 31% of patients in the intensive care unit (ICU). Early detection can enabletimely interventions and improved health outcomes. While artificial intelligence (AI) models have shown great potential for ICUdelirium prediction using structured electronic health records (EHR), most studies have either not leveraged state-of-the-artAI models, been limited to single-center cohorts, or relied on small datasets for development and validation. In this study,we introduce DeLLiriuM, a novel LLM-based delirium prediction model that utilizes EHR data from the first 24 hours of ICUadmission to estimate a patient’s risk of developing delirium for the remainder of their ICU stay. We developed and validatedDeLLiriuM using ICU admissions from 104,303 patients across 195 hospitals in three large databases: the eICU CollaborativeResearch Database, the Medical Information Mart for Intensive Care (MIMIC)-IV, and the University of Florida’s Integrated DataRepository. Our DeLLiriuM model achieved superior performance compared to all baseline models on the external validationset, measured by the area under the receiver operating characteristic curve (AUROC) metric. DeLLiriuM attained 82.5 (95%confidence interval 81.8-83.1) across 77,543 patients spanning 194 hospitals. Our approach of transforming structured EHRdata into an unstructured text format, the primary data modality for LLMs, enables our DeLLiriuM model to capture clinicalcontextual information, resulting in improved predictive performance. To the best of our knowledge, DeLLiriuM is the firstLLM-based delirium prediction tool for the ICU that utilizes structured EHR data with LLMs rather than clinical notes with LLMsor traditional structured feature representations used in AI models.