The Ultra-Brief Confusion Assessment Method (UB-CAM): A New Approach for Rapid Diagnosis of CAM-Defined Delirium

Contributed by Edward R. Marcantonio MD SM, Donna M. Fick, RN PhD, Richard N. Jones ScD, Sharon K Inouye MD MPH

The 3D-CAM. As discussed previously on NIDUS, the Confusion Assessment Method (CAM)1 has a 30-year track record during which it has become the standard for delirium identification in both clinical care and research. Moreover, the CAM diagnostic algorithm defines a clear phenotype for delirium. In applying the CAM over the past 3 decades, a number of questions have arisen:  1) what are the best questions for assessing each of the CAM diagnostic features?, 2) how many “positive” items (e.g. wrong answers) are needed to endorse the presence of each feature?, and 3) can we make the assessment as short as possible?  To address all of these questions, we developed the 3-Minute Diagnostic Assessment for CAM-defined delirium (3D-CAM). Using a database of nearly 5000 CAM assessments with over 120 assessment items each, we used modern measurement methods to identify the best items to assess each CAM diagnostic feature.2 We determined the number of “positives” required for the presence of each feature, which turned out to be one! And, we put the cognitive testing and CAM algorithm together in a short structured assessment that is easy to apply on the wards. We then prospectively validated the 3D-CAM in 201 general medicine patients—a purposeful “challenge” sample with average age over 80, and nearly a third with dementia. In comparison to the “gold standard” clinical evaluation for delirium, the 3D-CAM (performed blinded to the gold standard) had outstanding test characteristics, with sensitivity of 95% and specificity of 94%.3 Moreover, it performed well in challenging groups, such as those with hypoactive delirium, and delirium superimposed on dementia, and took only 3 minutes to perform. The 3D-CAM is now freely available along with a User’s Manual, has been translated into 10 languages, and has been widely adopted in both clinical and research settings. Two methods for measuring delirium severity using the 3D-CAM that require no additional questions/ratings are also available.4,5

The UB-2.  Shortly after publication of the 3D-CAM, several of our colleagues challenged us to make it even shorter. We surmised that using a highly sensitive ultra-brief screener at the start of the assessment could rule out delirium quickly, and reduce the fraction of patients requiring the full 3D-CAM. Using the pool of 3D-CAM items, we identified two items—Months of the Year Backwards, and What is the Day of the Week?—as the most sensitive pair of items for the presence of delirium. Ability to answer both questions correctly is considered a negative screen; anything else (either one or both questions answered incorrectly or not at all) is considered positive. This new Ultra-Brief 2-Item Screen, the UB-2, takes 35-40 seconds to administer, and has 93% sensitivity for delirium, but only 64% specificity.6 Negative screens can quickly rule out delirium, while positive screens require further evaluation to determine if delirium is present. The UB-2 is very easy to complete and requires only a few minutes to train staff.  It has been administered by nursing assistants at the bedside with high sensitivity.  A short free training video on the UB-2 is available at

The UB-CAM.  Since the UB-2 items come from the 3D-CAM, it makes sense to use them together as a two-step protocol to identify CAM-defined delirium.7  Additionally, since the presence of only one positive item triggers presence of a CAM feature in the 3D-CAM, we developed a skip pattern—as soon as one “sign” (an incorrect answer or positive patient symptom report or interview observation) is positive, the remainder of the items in that feature can be skipped—to further shorten the instrument. We call the combination of the UB-2 followed in “positives” by the 3D-CAM with skip the Ultra-Brief CAM (UB-CAM). In preliminary studies, it is highly accurate, with sensitivity of 93% and specificity of 95%, and can be completed in about 1 minute (median 40 seconds, mean 74 seconds).8 We attach a simple paper form that leads the assessor through the UB-CAM. While the UB-CAM has 20 items, only a minority are asked in most encounters—the median number of items administered is 2, and the mean is 6. Moreover, delirium is diagnosed quickly in severely impaired patients, and ruled out quickly in intact patients. So, the most items are administered to those with intermediate levels of impairment, as is appropriate. Given the adaptive testing approach (the questions asked depend on answers to previous questions) we have developed a UB-CAM App, which makes administration even easier, and are working on refining it for release in the near future. The UB-CAM’s speed, accuracy, and ability to identify CAM-defined delirium offers advantages over all other brief delirium identification tools available at this time. For questions about the UB-CAM, please reach us at


  1. Inouye SK, Van Dyck CH, Alessi CA, Balkin S, Siegal AP, Horwitz RI. Clarifying confusion: The Confusion Assessment Method. A new method for detection of delirium. Ann Intern Med. 1990; 113: 941-948.
  2. Yang FM, Jones RN, Inouye SK, Tommet D, Crane PK, Rudolph JL, Ngo LH, Marcantonio ER. Selecting optimal screening items for delirium: an application of item response theory. BMC Medical Research Methodology. 2013 Jan 22;13:8. doi: 10.1186/1471-2288-13-8.
  3. Marcantonio ER, Ngo L, O’Connor MA, Jones RN, Crane PK, Metzger ED, Inouye SK. 3D-CAM: Validation of a 3-Minute Diagnostic Interview for CAM-defined Delirium. Ann Int Med. 2014;161(8):554-61.
  4. Vasunilashorn SM*, Guess J* (*co-first), Ngo L, Fick D, Jones RN, Schmitt E, Kosar CM, Saczynski JS, Travison TG, Inouye SK**, Marcantonio ER** (**co-last). Derivation and Validation of a Severity Scoring Method (3D-CAM-S) for the 3-Minute Diagnostic Interview for CAM-defined Delirium. J Am Geriatr Soc. 2016; 64(8):1684-9.
  5. Vasunilashorn SM*, Devinney MJ* (*co-first), Acker L, Jung Y, Ngo L, Cooter M, Huang R, Marcantonio ER**, Berger M** (co-last). A New Severity Scoring Scale for the 3-Minute Confusion Assessment Method (3D-CAM) J Am Geriatr Soc. 2020 Jun 1. doi: 10.1111/jgs.16538. Online ahead of print.
  6. Fick DM, Inouye SK, Guess J, Long LH, Jones RN, Saczynski JS, Marcantonio ER. Preliminary development of an ultra-brief 2-item bedside test for delirium. J Hosp Med. 2015;10(10):645-50.
  7. Fick, DM, Inouye, SK, McDermott, C, Zhou, W, Ngo, L, Gallagher, J, McDowell, J, Penrod J, Siuta, J, Covaleski, T, Marcantonio, ER. Pilot Study of a Two-Step Delirium Detection Protocol Administered By Certified Nursing Assistants, Physicians and Registered Nurses. J Gerontol Nurs. 2018; 44(5):18-24.
  8. Motyl C, Zhou W, Ngo L, Fick D, Marcantonio E. Comparative Effectiveness and Efficiency of 4 Delirium Screening Protocols. J Am Geriatr Soc. Abstr Suppl, 2019.
Posted in Uncategorized.

Leave a Reply