eDIS-ICU - An international validation study to assess accuracy of a new screening tool for delirium in the intensive care unit

Once consent is given, the presence / absence of delirium will be assessed by consecutively using three different test methods: • electronic Delirium Screening in the ICU (eDIS-ICU) • Confusion Assessment Method for the ICU (CAM-ICU) • Expert-administered delirium assessment as per the DSM-V diagnostic criteria CAM-ICU is a delirium assessment tool comprising standardised non-verbal assessments for mechanically ventilated and non-ventilated ICU patients. It assesses four features: (1) the acute onset of mental status changes, or a fluctuating course, (2) inattention, (3) altered levels of consciousness, and (4) disorganised thinking. It takes approximately 1-2 minutes to complete, and will be a paper-based assessment. eDIS-ICU is an electronic delirium screening tool designed specifically for minimally responsive patients in the ICU setting. It is an app containing a bank of 60 questions that test attention, divided attention, and organisation of thinking. Questions are randomised, with up to six questions asked in each assessment with true or false answers. The time required to complete the test is approximately 60 seconds per assessment. eDIS-ICU will be accessed via a weblink which will be sent to the investigators at each recruiting site. The three tests will be administered by three different people at any given time point. The DSM-V assessment will be performed or confirmed (after assessment performed by an appropriately trained and certified psychiatric nurse) by a designated psychiatrist. Patients will be assessed twice within 24 hours (morning and afternoon) immediately following recruitment. The order in which tests are administered (e-DIS-ICU, CAM-ICU and DSM-V assessment) and the clinician administering the CAM-ICU and eDIS-ICU tests will be randomised. A randomisation schedule will be created using statistical software (e.g. the "ralloc" package in Stata), which will then be implemented using the randomisation module in REDCap (the electronic database used for data management – see below). The three tests will all be completed within 1 hour of commencement of the first test to increase the reliability of the comparable data. Assessors will be blinded to previous test results. The time of test commencement and completion will be recorded. Administration of the CAM-ICU and eDIS-ICU tests will be timed (in seconds) and a record will be kept of tests not able to be completed at all assessment timepoints (and reasons for inability to complete).