Understanding, Diagnosing, and Improving Delirium Outcomes

The medical community is taking steps to clarify delirium pathology and diagnosis.

Delirium, best described as acute brain failure or loss of brain resilience, is very common, affecting up to 50% of hospital patients aged older than 65 years.1

Delirium has also been linked to a significantly increased risk for dementia and death within the two years after diagnosis, and researchers are just starting to understand how dangerous the condition really is.2 Yet delirium remains frequently unrecognized in most clinical settings.3

“Delirium is strictly a clinical diagnosis. There are no blood tests or scans. Some patients with delirium may just appear sleepy,” said Tamara G. Fong, MD, PhD, assistant professor of neurology at Harvard Medical School, and assistant scientist at the Institute for Aging Research in Boston. “If you don’t wake them up to do an assessment, you can easily miss the diagnosis. We need to start thinking of delirium as the medical emergency that it is.”4

Mortality and dementia risk in the two years following a diagnosis of delirium is supported by data from 2,197 patients included in a 2014 population-based study in BMC Geriatrics. Death occurred in 61% of patients diagnosed with delirium compared with 34% of patients without delirium. Dementia was subsequently diagnosed in 45% of patients diagnosed with delirium compared with 9% without delirium.2

“We are learning that delirium is less reversible than we thought. There is good evidence that delirium makes dementia worse and growing evidence that some of the changes seen in older patients with delirium persist,” Fong said.

How DSM-5 Defines Delirium

The Diagnostic and Statistical Manual of Mental Disorders, 4th edition (DSM-IV) defined delirium as a disturbance in consciousness. The DSM-5 no longer uses the term “consciousness” and instead uses “disturbance in attention and awareness.”

Both definitions explain that the condition should develop over a short period of time, include changes in cognition, and are caused by a physiologic disturbance. But this change in wording may be more than semantics.

“This is an active area of discussion. Taking the emphasis away from level of consciousness allows delirium to be distinct from stupor or coma. This may be appropriate because although level of consciousness is important in delirium in the ICU or following anesthesia, it is less common in general hospital patients,” said Edward R. Marcantonio, MD, professor of medicine at Harvard Medical School and director of the Aging Research Program at Beth Israel Deaconess Medical Center in Boston.

“Also, most of the assessment tools we use to diagnose delirium are based on the DSM-IV definition. It is not yet established if the new definition will significantly change the way we diagnose delirium,” Fong said.

CAM and 3D-CAM

“The best way to diagnose dementia is to have an assessment done by a clinical geriatric specialist or a geriatric psychiatrist. That would be the gold standard,” said Marcantonio.

In the real world of bedside clinical practice, the best diagnostic tool is the Confusion Assessment Method (CAM). Used by a skilled clinician, CAM has a sensitivity of 94% and a specificity of 86%.3

“CAM’s brilliance and its limitation is that it is very flexible. It is a very accurate tool for experts but only as good as the person doing the assessment,” said Marcantonio.

He and his team of researchers have developed a three-minute, user-friendly derivation of CAM called 3D-CAM. This very specific assessment is based on the DSM-5 definition. It lets the examiner know what questions to ask and the diagnostic threshold for the answers.3

“3D-CAM requires less clinical training. For example, in evaluation of inattention, there are just four questions to ask, and the examiner is told how many wrong answers are allowed before flagging the result as positive,” said Marcantonio.

He and colleagues tested 3D-CAM in a validation study with 201 patients. The participants’ delirium diagnosis was confirmed or dismissed by an expert panel, and 3D-CAM achieved 95% sensitivity and 94% specificity. 3D-CAM even did well with patients who had coexisting dementia.3

“We went out of our way to enroll patients with dementia. They made up 30% of the subjects. Even in these patients, who may be unable to answer questions accurately, we achieved a specificity of 86%,” Marcantonio said.

Prevention Also Important — and Possible

Studies show that delirium may be preventable in 30% to 40% of cases. Fong and her colleagues have developed the Hospital Elder Life Program (HELP) to prevent delirium in hospital patients.

HELP is a multicomponent intervention strategy. Interventions include reorientation, therapeutic activities, removal of possible offending drugs, sleep hygiene, mobilization, nutrition, hydration, and hearing and vision support.1

“Simplified, user-friendly assessment tools like 3D-CAM will certainly help. Being aware of the dangers, knowing the risk factors, implementing interventions like HELP, and early treatment should all help to improve outcomes in delirium,” Fong said.

Chris Iliades, MD, is a full-time freelance writer based in Cape Cod, Massachusetts. This article was medically reviewed by Pat F. Bass III, MD, MS, MPH.


  1. Inouye SK, Westendorp RDJ, Saczynski JS. Delirium in elderly people. The Lancet. 2013; doi:10.1016/S0140-6736(13)60688-1.
  2. Davis DH, Barnes LE, Stephan BC, et al. The descriptive epidemiology of delirium symptoms in a large population-based cohort study: results from the Medical Research Council Cognitive Function and Ageing Study (MRC CFAS). BMC Geriatr. 2014;14:87.
  3. Marcantonio ER, Ngo LH, O’connor M, et al. 3D-CAM: derivation and validation of a 3-minute diagnostic interview for CAM-defined delirium: a cross-sectional diagnostic test study. Ann Intern Med. 2014;161(8):554-61.
  4. To learn more about delirium diagnosis, see the 3D-CAM assessment, and find other CAM resources, go to the HELP website at http://www.hospitalelderlifeprogram.org/.