Delirium is one of the most vexing of problems we face in hospitalized patients. It is hard to treat and just as hard to diagnose. In the next issue of the Journal of Hospital Medicine an original paper “Preliminary Development of an Ultra-Brief 2-Item Bedside Test for Delirium” is being published from Donna Fick, PhD, who led a team of researchers, including Sharon Inouye, MD, MPH and Ed Marcantonio, MD.
As the paper points out only 12-35% of delirium cases are detected in routine care! This paper seeks to find a better and easier to use screening tool for delirium.
The research team recently developed and published on a three minute test called the 3D-CAM. It has a 95% sensitivity and 94% specificity. A great result, but they recognized that even three minutes may be hard to accomplish in the busy atmosphere of a hospital. They then set out to analyze the 20 question database used in the 3D-CAM and see if there were two questions in combination that could give at least a 90% sensitivity. They were successful and the two questions, when used together, with the best sensitivity were: 1.) “What is the day of the week?” and 2.) “Months of the year backwards”. The combination of those two simple questions had a sensitivity of 93%!
I recently asked Dr. Fick, the lead author, about the work.
Delirium is often not diagnosed and as a result we often end up playing catch-up once it becomes obvious. Would you envision that this be a daily screen upon admission for those at risk or more targeted if we suspect something?
Dr. Fick: An order that makes most sense to me would be 1) our new ultra-brief screener, 2) if positive, then follow-up with the 3D-CAM, 3) if the 3D-CAM is positive, then a full clinical assessment. Our new 2-item bedside test should make the process faster and more efficient by only doing the longer assessments on those who need it and are positive on the ultra-brief 2-item screen.
Anything to streamline our work and help us do our job is better! How does the two item screen stack up against our current de facto screen of “oriented X 3”? I saw some of the questions listed are similar to 1. Date, 2. Location, and 3. Reason for hospitalization. Since everyone already asks these questions, were these looked at and compared?
I agree that orientation is frequently asked, and in our prior studies this was the most frequently documented mental status construct in the patient chart by nursing staff. Orientation questions were examined in this study, and were not found to be as informative toward a diagnosis of delirium as the items included in our screener. Inattention is the key deficit in delirium. Our screener includes the single best test of attention (MOYB-Month of the Year Backwards) and the single best test of orientation (“what is the day of the week”).
The 3D-CAM and its twenty questions were used to ferret out the best screening questions. Is it possible that the results might be different if only the two screening questions were performed vs. pulling out of a twenty question sample? In other words, if we were to only ask the two questions, how do you think the results might vary? Perhaps the simple two-question, 36 second screen might have better or worse results? Will this be looked at as part of the future prospective validation you wrote about in the paper?
This is correct since they were pulled from a secondary analysis, and is why our paper is entitled “Preliminary development…” We will need to test the screener, both in other datasets, prospectively when administered in isolation, and with real clinicians rather than research assistants. We are currently proposing this in our future studies and did a small pilot (n=23) where we piloted the approach needed in the larger study.
As you look to study younger populations, based on your review of other studies, how do you think your results might translate to that somewhat lower risk population?
Generally in younger patients, these assessments will perform even better, because there is a lower prevalence of dementia or other forms of cognitive impairment that might cause false positives. On the other hand, many more patients will need to be screened to find 1 case of delirium, so the overall cost-effectiveness of systematic case identification for delirium may be less clear in younger populations. Given limited resources, it might sense to focus case identification efforts on high risk patients, including older patients (70 and up) on general medical/surgical units, and all patients in the ICU. These are again; some questions we hope to answer in future work!
Anything else you would like to say about this research or yourself?
I love working with older adults and have been doing so since very early in my career. Caring for an older adult who was post hip surgery and became very delirious on the evening shift when I was working as a hospital nurse led me to study delirium. After my last NIH funded study I found that nurses and physicians found it difficult to quickly and reliably assess delirium at the bedside–even when they were motivated to do so. When you meet older adults for the first time in the hospital setting (which is sometimes the case) assessing for delirium can be difficult and time consuming, we hope to make it easier.
As you can see from this work, it is very interdisciplinary team work. Geriatrics, by nature is complex and challenging and needs to be interdisciplinary. This is all team work, not my work. Many times we have different perspectives, but my experience in geriatrics is that they are all valued. We all want the same thing (improved care for older adults) and working together I think we are moving closer to that goal. Thank you for being interested in this work and spreading the word about delirium.
Dr. Fick and her team have hopefully given us a new tool to detect delirium and thus improve the clinical response to delirium. So often we miss this subtle diagnosis and our patients pay the price. Having an easy and sensitive screen that we hospitalists can perform in our all-too-busy days would have far reaching impact, from lower morbidity, shorter hospital stays, lower costs, and better experiences for patients.
I look forward to further studies validating and building upon this important work!
Donna Fick, PhD, RN, FAAN is a distinguished professor in the college of nursing at Penn State, Co-Director of the Penn State Hartford Center for Geriatric Nursing Excellence and a board certified gerontological clinical nurse specialist. She received her PhD from the University of California San Francisco, MSN from University of Cincinnati and her BSN from Berea College in Berea, KY. Her area of research is delirium superimposed on dementia and inappropriate medication use in older adults. She is co-chair of the American Geriatrics Society (AGS) Beers Criteria, a member of the AGS board of director, and recently completed two National Institute of Health funded intervention trials for delirium superimposed on dementia. She is a fellow of the Gerontological Society of America, and a Fellow of the American Academy of Nursing. Outside of work, she is married and has three teenagers, Will, Carl and Mary. Donna lives on the edge of a forest and enjoys nature and trail running.