Introduction: Delirium, a common complication of an intensive care unit (ICU) admission, is inconsistently diagnosed by clinicians. Current screening tools require specialist expertise and/or training. Some are time-consuming to administer, and reliability in routine clinical practice is questionable. An innovative app designed to enable efficient and sensitive screening for delirium without specialist training (eDIS-ICU) has recently been described. This pilot study compared the eDIS-ICU against the reference standard expert assessment using DSM-V (Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition) criteria and the Confusion Assessment Method for the ICU (CAM-ICU).
Methods: In this prospective, single-centre pilot study, a convenience sample of 29 ICU patients were recruited at a tertiary referral hospital between November 2018 and August 2019. After obtaining written consent, demographic and clinical data were collected, and the patients were screened for delirium using eDIS-ICU and CAM-ICU by two clinician researchers in random order. The patients were also assessed for presence of delirium independently by an expert clinician using a structured interview to diagnose as per DSM-V criteria. The results of screening and diagnosis were tabulated to allow comparison of screening tools against diagnosis; sensitivity and specificity of the tools were calculated.
Results: Seven participants were diagnosed with delirium as per DSM-V criteria. The eDIS-ICU tool correctly identified six of these participants compared with two identified by CAM-ICU. The sensitivity of the eDIS-ICU tool was 86% (95% confidence interval [CI] = 81.5-100.0) compared with 29% (95% CI = 5.1-69.7) for CAM-ICU (p < 0.05), and the specificity was 73% (95% CI = 81.5-100.0) versus 96% (95% CI = 75.1-99.8), respectively.
Conclusion: The simple and novel eDIS-ICU delirium screening tool was noninferior to the CAM-ICU in detecting delirium as per DSM-V criteria. A definitive validation study is warranted.
Keywords: Delirium; ICU; Screening tools; Technology.
Crown Copyright © 2021. Published by Elsevier Ltd. All rights reserved.