Related Experiment Video
Updated: Mar 2, 2026

Implementation of a Real-Time Psychosis Risk Detection and Alerting System Based on Electronic Health Records using CogStack
Published on: May 15, 2020
Predicting inpatient delirium: The AWOL delirium risk-stratification score in clinical practice
Ethan G Brown1, S Andrew Josephson1, Noriko Anderson2
1Department of Neurology, University of California, San Francisco, USA.
Abstract:
Inpatient delirium improves with multicomponent interventions by hospital staff, though the resources needed are often limited. Risk-stratification to predict delirium is a useful first step to help triage resources, but the performance of risk-stratification as part of a functioning multicomponent pathway has not been assessed. We retrospectively studied the performance of a validated delirium prediction rule, the AWOL score, as a part of a multicomponent delirium care pathway in practice on a university hospital ward. We reviewed the hospitalizations of patients 50 years or older for evidence of delirium and extracted the AWOL score from nursing documentation (n = 347). The area under the receiver operating characteristic curve (AUC) was 0.83 (95% CI 0.77-0.89) for all cases and 0.73 (95% CI 0.60-0.85) when cases of prevalent delirium were removed. Involving minimal additional assessment, this nursing-based risk stratification score performed well as part of a multicomponent delirium care pathway.

