Related Experiment Video
Updated: Jun 23, 2026

A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment
Published on: January 11, 2020
Targeted Delirium Screening in At-risk Older Adults in the Emergency Department Improves Screening Rates
Saket A Saxena1, Emily H Weaver2, Evelyn C Brady2
1Center for Geriatric Medicine, Desk X10, Cleveland Clinic, Cleveland, Ohio, USA.
Objectives:
Delirium is prevalent yet underrecognized in the emergency department (ED). Timely identification of delirium is a priority, but results from systematic implementation of targeted delirium screening in the ED are limited.
Methods:
This study analyzes delirium screening rates, positivity rates, and distribution of screening scores before and after implementation of an electronic medical record (EMR)-based risk stratification at ED triage. In the baseline period, delirium screening was based on age (65+ years) and clinical concern for altered mental status (AMS). In the intervention period, eligibility for delirium screening was based on an automated EMR-based risk stratification. The 4AT was used as a delirium screening tool in both periods.
Results:
In the baseline period, only 3267 (3.0%) of patient encounters were noted to have AMS and were eligible for delirium screening; 2241 patients received the screening; 1901 screenings resulted in a positive 4AT. In the intervention period using EMR-based targeted screening, 39,065 encounters were eligible for delirium screening; 14,125 patients received the screening, resulting in 2016 positive 4AT.
Conclusion:
This multicenter study demonstrates the feasibility of EMR-based risk stratification to target encounters for delirium screening at ED triage. The clinical judgment approach resulted in lower screening rates and a greater percentage of high-positive 4AT scores. The risk-stratified approach resulted in a 6-fold increase in number of screenings and identification of low-positive 4AT scores, detecting more patients with subtle delirium. An approach that leverages risk-stratified screening, complemented by screening based on clinical judgment, could capture a wider range of patients with delirium.
