What’s This About A Rapid 2 Step Delirium Screen?

So the other day, I saw the link to the MedPage article Delirium: What Are the Options? on the American Delirium Society web site. When I saw the reference for a rapid two-step screening method for delirium and looked up the article, I was reminded of a medical student who rotated on the consult service a while ago who actually asked a patient to spell the word “lunch” backward [1].

I remember wondering where he got that. I frequently ask patients to try to spell “world” backward along with using the Mini-Cog to test for inattention and cognitive disorganization along with applying the simple form of the Confusion Assessment Method (CAM) when I’m assessing them for delirium in the general hospital and in the intensive care unit (ICU).

Now I wonder if the medical student knew about the Delirium Triage Screen (DTS), which is checking the patient’s level of consciousness using the Richmond Agitation Sedation Scale (RASS, which is  an arousal scale and the DTS is copyrighted by the way). The other part of the DTS is asking the patient to spell “lunch” backward. I also wonder why use the RASS at all since delirium is ruled out by a normal RASS–which is a normal level of alertness.

I also wonder why bother to copyright the DTS? And remember, the CAM is also copyrighted by the Hospital Elder Life Program, LLC. The reason I mention the CAM is that the rapid two-step delirium screening method described by Han et al include the DTS to rule out delirium, and next, in order to rule in delirium, the Brief Confusion Assessment Method (bCAM), the latter being an adaptation of the CAM-ICU.

Most of us know the stem of the original CAM:

CAM Flow Chart

CAM Flow Chart

However, the bCAM is not an adaptation of the original CAM. It’s an adaptation of the CAM-ICU, developed at Vanderbilt University in Tennessee. See the excellent work being done at Vanderbilt on delirium in the critical care unit here. See the CAM-ICU screen in action in the YouTube video below:

One point to remember about the CAM-ICU is that it’s very well validated for use in critical care units, but there’s been a study showing that it has poor sensitivity on general medical units (18%), casting some doubt on the bCAM performing much better in that area [2].

However, it is very quick, it deserves further study and as long as we remember that delirium fluctuates (which can make doctors miss it) and that there may be a tradeoff in accurate diagnosis of delirium when emphasizing speedy assessment–these guys may be on to something.

Bottom line by the Annals of Emergency Medicine is that it has acceptable sensitivity and specificity in older adults in the emergency room, but it’s not clear yet how it would perform outside of that space or how it might affect health outcomes and resource use.

And watch out for the Fair Use Police.


1. Han, J. H., A. Wilson, et al. (2013). “Diagnosing Delirium in Older Emergency Department Patients: Validity and Reliability of the Delirium Triage Screen and the Brief Confusion Assessment Method.” Annals of emergency medicine 62(5): 457-465.

Delirium is a common form of acute brain dysfunction with prognostic significance. Health care professionals caring for older emergency department (ED) patients miss delirium in approximately 75% of cases. This error results from a lack of available measures that can be performed rapidly enough to be incorporated into clinical practice. Therefore, we developed and evaluated a novel 2-step approach to delirium surveillance for the ED. This prospective observational study was conducted at an academic ED in patients aged 65 years or older. A research assistant and physician performed the Delirium Triage Screen (DTS), designed to be a highly sensitive rule-out test, and the Brief Confusion Assessment Method (bCAM), designed to be a highly specific rule-in test for delirium. The reference standard for delirium was a comprehensive psychiatrist assessment using Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision criteria. All assessments were independently conducted within 3 hours of one another. Sensitivities, specificities, and likelihood ratios with their 95% confidence intervals (95% CIs) were calculated. Of 406 enrolled patients, 50 (12.3%) had delirium diagnosed by the psychiatrist reference standard. The DTS was 98.0% sensitive (95% CI 89.5% to 99.5%), with an expected specificity of approximately 55% for both raters. The DTS’s negative likelihood ratio was 0.04 (95% CI 0.01 to 0.25) for both raters. As the complement, the bCAM had a specificity of 95.8% (95% CI 93.2% to 97.4%) and 96.9% (95% CI 94.6% to 98.3%) and a sensitivity of 84.0% (95% CI 71.5% to 91.7%) and 78.0% (95% CI 64.8% to 87.2%) when performed by the physician and research assistant, respectively. The positive likelihood ratios for the bCAM were 19.9 (95% CI 12.0 to 33.2) and 25.2 (95% CI 13.9 to 46.0), respectively. If the research assistant DTS was followed by the physician bCAM, the sensitivity of this combination was 84.0% (95% CI 71.5% to 91.7%) and specificity was 95.8% (95% CI 93.2% to 97.4%). If the research assistant performed both the DTS and bCAM, this combination was 78.0% sensitive (95% CI 64.8% to 87.2%) and 97.2% specific (95% CI 94.9% to 98.5%). If the physician performed both the DTS and bCAM, this combination was 82.0% sensitive (95% CI 69.2% to 90.2%) and 95.8% specific (95% CI 93.2% to 97.4%). In older ED patients, this 2-step approach (highly sensitive DTS followed by highly specific bCAM) may enable health care professionals, regardless of clinical background, to efficiently screen for delirium. Larger, multicenter trials are needed to confirm these findings and to determine the effect of these assessments on delirium recognition in the ED.

2. Neufeld, K. J., Hayat, M. J., Coughlin, J. M., Huberman, A. L., Leistikow, N. A., Krumm, S. K., & Needham, D. M. (2011). Evaluation of Two Intensive Care Delirium Screening Tools for Non-Critically Ill Hospitalized Patients. Psychosomatics, 52(2), 133-140. doi: 10.1016/j.psym.2010.12.018

%d bloggers like this: