SHM Converge Scientific Abstract ePoster Gallery-Implementation & Validation of a Delirium Risk Score

Pdf Summary

This study implemented and validated an automatic EHR-based delirium risk score for hospitalized medical patients. Delirium is common in older inpatients and is linked to worse outcomes, including longer hospital stays, higher costs, and increased morbidity and mortality. Because many existing delirium tools are reactive and require manual completion, the authors aimed to create a low-touch predictive tool using routinely collected EHR data to identify patients at risk before delirium develops.<br /><br />The risk score used four variables: age 75 or older, dementia, hearing impairment, and use of psychotropic drugs. Each factor was assigned points, which were summed into a total score. The researchers then performed a retrospective chart review of 112 consecutive patients aged 65 and older on a general hospitalist service. Delirium was identified using both the Confusion Assessment Method (CAM) and documentation review for terms such as “delirium,” “encephalopathy,” or “altered mental status.”<br /><br />Results showed that 31% of patients developed delirium. Each component of the score was associated with increased odds of delirium, and each 1-point increase in the total score raised the odds of delirium by 28%. A cutoff score of 3, previously validated in emergency department patients, produced a sensitivity of 68.6% and specificity of 54.5%. Using this cutoff, the odds of developing delirium were 2.62 times higher.<br /><br />The authors conclude that an automatically calculated delirium risk score based on EHR data may help clinicians target preventive interventions earlier. They recommend further refinement and larger-scale evaluation, including assessment of whether the tool can predict delirium severity and guide prevention strategies.

Asset Subtitle

Amber Balzer

Meta Tag

Author List Amber Balzer, Leigh West, Melissa Bui, Stephen Biederman, Timothy R. Ford

Category Research

Concept Delirium

Concept Delirium Risk Score

Concept Risk Calculator

Concept Electronic Health Record

Concept Risk Factor

Distinguished Non-Finalist

Presenter Organization Virginia Commonwealth University Health System

Presenting Author Amber Balzer

Track Quality Improvement

Keywords

delirium risk score

electronic health records

hospitalized medical patients

predictive tool

older inpatients

dementia

hearing impairment

psychotropic drugs

Confusion Assessment Method

retrospective chart review

Delirium

Delirium Risk Score

Risk Calculator

Electronic Health Record

Risk Factor