Delirium Prevention Programs…Our Pride and Joy

Today is the day our Epic Delirium Order Set screensavers go live at the University of Iowa Hospitals and Clinics and they’ll run on a lot of computer terminals until May 16, 2014. A couple of medical students and a resident worked pretty hard on them and they should get credit. Thanks Dr. Stephanie Houseton, DO, and senior medical students Paul Meirick and Pablo Kollmar!

This slideshow requires JavaScript.


The population at highest risk for delirium is the elderly, especially those with cognitive impairment. While the group of centenarians (don’t call them “feisty”!) in the video above, filmed in Madison, Wisconsin by St. Mary’s Hospital staff last year look like they’d be at lower risk, advanced age by itself is a risk factor for delirium.

And while nurses are often assumed to carry the heaviest burden of responsibility for screening for delirium–it’s everybody’s job. The recently published study of nurses screening for delirium in older patients in the emergency department (ED) found a relatively low incidence of delirium (around 7%) and acknowledged that “Delirium detection in the ED is almost universally poor.” I thought it was interesting that the risk factors associated with delirium in that study were cognitive impairment, history of depression, and an abnormal heart rate/rhythm [1].

I’m not sure why depression is sometimes highlighted as a risk factor in some studies; possibly it has something to do with depression being associated with inflammation. Alternatively, some older patients could have been taking older antidepressants which tend to be anticholinergic, and which can be associated with delirium.

A critically important safety issue for delirium prevention and management is knowing what to do about the other risk factor, the abnormal heart rate/rhythm, for which a couple of talented trainees who are exceptional doctors already have given us the soul of the story.

The authors indicate a limitation of the study was the inability to use their chosen detection tool, the Confusion Assessment Method (CAM), as a severity rating instrument. However, the developers of the CAM have come up with a new modification that allows researchers to assess delirium severity, the CAM-S [2].

The most important goal is to protect our patients from delirium by preventing it, to which many orgnizations are dedicated including the American Delirium Society. Don’t forget their meeting in June in Baltimore! Once delirium gets started, it’s harder to manage. The disorganized and sometimes dangerous agitation caused by delirium can lead to the agonizing decision to use antipsychotics to calm the patient by reducing or eliminating terrifying fragmented delusions and hallucinations. This is complicated by the new finding that antipsychotics as a class may be associated with a lower Number Needed to Harm (NNH) metric than has been seen in older studies.

I know many of us occasionally feel like starting and sustaining effective, team-powered delirium prevention projects is a Sisyphean task.

On the other hand, The boulder doesn’t always flatten us. When a dedicated group of clinicians work together keeping the end in mind on keeping delirium out of our hospitals–it’s our pride and joy.

References and Web Resources:

1. Hare, M., G. Arendts, et al. (2014). “Nurse screening for delirium in older patients attending the emergency department.” Psychosomatics 55(3): 235-242.
BACKGROUND: Delirium in older emergency department (ED) patients is common, associated with many adverse outcomes, and costly to manage. Delirium detection in the ED is almost universally poor. OBJECTIVES: The authors aimed to develop a simple clinical risk screening tool that could be used by ED nurses as part of their initial assessment to identify patients at risk of delirium. METHODS: A prospective cross-sectional study of patients 65 years and older attending a single ED. RESULTS: Of 320 enrolled patients, 23 (7.2%) had delirium. Logistic regression analysis revealed 3 risk factors strongly associated with delirium risk: cognitive impairment, depression, and an abnormal heart rate/rhythm. Weighting these variables based on the strength of their association with delirium yielded a risk score from 0-4 inclusive. A cutoff of 2 or more in that score would have given a sensitivity of 87%, specificity of 70%, and NPV of 99%, while avoiding further diagnostic workup for delirium in approximately two-thirds of all patients, when used as an initial screen. CONCLUSIONS: A simple risk screening tool using factors evident on initial nurse assessment can be used to identify patients at risk of delirium. Further trials are needed to test whether the tool improves patient outcomes.


2. Inouye, S. K., C. M. Kosar, et al. (2014). “The CAM-S: Development and Validation of a New Scoring System for Delirium Severity in 2 CohortsThe CAM-S Score for Delirium Severity.” Annals of Internal Medicine 160(8): 526-533.
Background: Quantifying the severity of delirium is essential to advancing clinical care by improved understanding of delirium effect, prognosis, pathophysiology, and response to treatment.Objective: To develop and validate a new delirium severity measure (CAM-S) based on the Confusion Assessment Method.Design: Validation analysis in 2 independent cohorts.Setting: Three academic medical centers.Patients: The first cohort included 300 patients aged 70 years or older scheduled for major surgery. The second included 919 medical patients aged 70 years or older.Measurements: A 4-item short form and a 10-item long form were developed. Association of the maximum CAM-S score during hospitalization with hospital and posthospital outcomes related to delirium was evaluated.Results: Representative results included adjusted mean length of stay, which increased across levels of short-form severity from 6.5 days (95% CI, 6.2 to 6.9 days) to 12.7 days (CI, 11.2 to 14.3 days) (P for trend < 0.001) and across levels of long-form severity from 5.6 days (CI, 5.1 to 6.1 days) to 11.9 days (CI, 10.8 to 12.9 days) (P for trend < 0.001). Representative results for the composite outcome of adjusted relative risk of death or nursing home residence at 90 days increased progressively across levels of short-form severity from 1.0 (referent) to 2.5 (CI, 1.9 to 3.3) (P for trend < 0.001) and across levels of long-form severity from 1.0 (referent) to 2.5 (CI, 1.6 to 3.7) (P for trend < 0.001).Limitation: Data on clinical outcomes were measured in an older data set limited to patients aged 70 years or older.Conclusion: The CAM-S provides a new delirium severity measure with strong psychometric properties and strong associations with important clinical outcomes.Primary Funding Source: National Institute on Aging.




  1. At our Delirium Quality Improvement committee meeting this afternoon, a colleague criticized the CAM-S and mentioned a new Delirium Observation Screening Scale (DOSS) severity rating instrument she believes is superior. She said it was published only a couple of months ago and that she’d have her secretary send it to me. Out of curiosity I looked for it myself on PubMed, but the only article I could find was an “old” one from 2011:

    Scheffer, A. C., B. C. van Munster, et al. (2011). “Assessing severity of delirium by the delirium observation screening scale.” International Journal of Geriatric Psychiatry 26(3): 284-291.
    Objective Delirium is the most common acute neuropsychiatric disorder in hospitalized elderly. Assessment of the severity of delirium is important for adjusting medication. The minimal dose of medication is preferable to prevent side effects. Only few nurse based severity measures are available. The aim of this study was to validate a scale developed to assess symptoms of delirium during regular nursing care, the Delirium Observation Screening (DOS) Scale, for monitoring severity of delirium. Method Delirious patients of 65 years and older were included. Delirium was diagnosed according to DSM-IV criteria and the Confusion Assessment Method. The DOS Scale was compared to the Dutch version of the Delirium Rating Scale-Revised-98 (DRS-R-98). Global cognitive functioning was assessed by the Informant Questionnaire Cognitive Decline in the Elderly-Short Form (IQCODE-SF) and the KATZ-ADL Scale was used for functional impairment. Results Ninety seven delirious patients were included: 41 hip fracture patients and 56 medical patients. The correlation between total DRS-R-98 scores and DOS Scale scores was 0.67 (p = 0.01). For the cognitive impaired group (IQCODE-SF ≥3.9) this correlation was 0.61 (p = 0.01); for the group with no global cognitive impairment, this correlation was 0.67 (p = 0.01). Correlations between DRS-R-98 and DOS Scale for hypoactive, hyperactive and mixed delirium subtype were 0.40 (p = 0.32), 0.44 (p = 0.01) and 0.69 (p = 0.05), respectively. Conclusions The DOS Scale is able to measure severity of delirium. In routine daily clinical practice, the DOS Scale is a time-efficient, easy to use and reliable method for measuring and monitoring severity of delirium by nurses. Copyright © 2010 John Wiley & Sons, Ltd.


  2. Of course I had to pin this. Did so on my Health and Mental Health News. But will also post it in time on my Aging Isn’t for Sissies Boards. As always thank you and your residents for all you do.

    Liked by 1 person

  3. Sharon Van Fleet says:

    Congratulations, Dr. Amos! This achievement is something to celebrate!

    Liked by 1 person

%d bloggers like this: