Category Archives: Thinking about cases

Delirium: the forgotten medical emergency.

We have all seen a little old lady crumpled onto a hospital trolley. Referred in eye-rollingly from their family who visited the nursing home and meekly say ‘she is more confused than normal’.  Talking to her you think she’s good value, she thinks she’s on a cruise ship, that you’re a terribly nice young man and that you’d make an excellent match for her daughter.

You smile, send off a random panel of blood tests and refer her to care of the elderly. Instruct an exasperated nurse to collect a urine sample (how?!?), maybe you catheterise and cannulate her. No one screams and shouts at you.  The family have got what they want (an admission) and bugger off home.   The poor CofE SHO is used to taking veterinary histories like this and expects nothing better.

This patient has delirium.

Delirium is a medical emergency.

We are emergency physicians.  Patients with delirium stay in hospital longer, are twice as likely to die, and less likely to leave hospital independent than age-matched patients who are with it.  We need to be better at this.

In the last RCEM audit (2014-2015) only 11% of patients were being screened for delirium or dementia, so we have no idea what the ED incidence is.  The incidence of delirium in the community is relatively low, as it will generally lead to someone being admitted quickly.  The incidence in hospital is pretty high.  10-20% on average with a further 10-30% developing delirium during their stay.  Different wards have different incidences; 15-53% of post-op patients get it, and 70-87% of ICU patients.

Now replace ‘delirium’ with ‘sepsis’, or ‘AKI’ and there’d be national outrage that we are not checking people for a condition that increases their risk of death at the front door.  Delirium is a sign of end organ dysfunction, if a patients’ urine output drops to less than 30mls/hr all manner of screaming and shouting occurs, but we don’t do this when people’s brains have stopped working properly.

Delirium is an acute, fluctuating, disturbance in attention, arousal and other aspects of mental status.

If you have delirium you slide up and down a scale with hyperactive symptoms at one end and hypoactive symptoms at the other.  Hyperactive delirium is when people are convinced the nurses are going to kill them, or are restless and agitated.  Hypoactive delirium is more common, and patients with it are withdrawn, quiet and drowsy (we like patients like this, they are no trouble).

The key point is the fluctuations and changes in mental state.  Over time you might initially talk to someone and they seem to make sense, and go back and they might not remember you, or remember why they are here. Hypoactive delirium has a higher mortality and is more common than hyperactive delirium. 


Delirium is an acute illness (it comes on over a period of hours to days), but it has a variable and potentially long course. Recovery time is variable, sometimes as long as weeks and months. It can also lead to long term cognitive impairment.

Patients also recall events while they were delirious and can get flashbacks and PTSD like symptoms.  This can be quite upsetting and affect their feelings about subsequent medical treatment and hospitalisation.


It takes less of an insult to cause delirium in a frail brain.  You or I with our giant young brains require severe sepsis, hypoxia, or lots of beer to make us delirious.  Older people might only need to be slightly dehydrated, or a little constipated. Common causes:-

  1. Drugs (this is a big one, and i’ll try and cover it in more detail later).
  2. Infection
  3. Metabolic
  4. Pain
  5. Infarction
  6. Sleep disturbance
  7. Constipation

(This can be abbreviated to the mnemonic DIMPISS if that’s how you roll).

You are often going to find a story that is like this: An older lady who has been started on a new medication, which has made her less mobile, so she’s become dehydrated, which has led to constipation, and urinary retention and overflow.

In a complete departure from most of your previous medical training, I encourage you to find more than one cause (Occam’s razor be damned), and try and sort out as many simple things as possible.  Don’t be frightened of not knowing the ultimate cause, and having a myriad of options to choose from.  It’s likely to be multifactorial.

What should I do in the ED?

  1. If you think someone might have delirium, ask them to tell you the months of the year backwards, and what day of the week it is. This has a sensitivity of 93% (CI 91-99%) and a specificity of 64% (56-70%).
  2. If they can’t do that try and do an AMT. If they have a lower than normal score, and you’ve got evidence of things fluctuating.  You’ve diagnosed delirium.
  3. Write : Delirium ?Cause in the notes then go onto your differential (You should be able to find 2 or 3 possibilities to fit into DIMPISS).
  4. Investigate and check for each one. That means most patients are going to need a PR, bladder scan, a panel of blood tests, and an ECG.

Please note I have not suggested you do a urine dip.  This is not an omission.  Please come back soon for reasons why…


The biggest problem in the ED is we don’t screen for delirium.  If we start trying to ascertain how big the problem is, we might be able to do something about it.  Now I don’t think that it’s something the nurses need to do at triage, but every time we see an elderly person in majors we need to consider it.

In the ED there is a tendency to stop the diagnostic train once we’ve decided people can’t go home and let other specialities pick up the slack.  Admitting delirious patients to hospital is a bad option.  Often it’s the only one we have, but it’s going to make their delirium last longer, and likely speed their functional decline.  So families need to be aware of this.  It’s important to note that sometimes these delirious patients have been gently starved, and waited many many hours for the paramedics to bring them to you.  People you pick up first thing in the morning will be sleep deprived, dehydrated and hungry.

We also like to find a ‘cause’ for the confusion in the ED.  This probably leads to us diagnosing UTI way more than it actually happens.  This is bad for our patients because it leads admitting specialities to an anchoring bias from our own sloppy work.  Leaving the diagnosis open, encourages further original thought.


What would be ideal for these patients?

Patients with delirium need to be nursed in a quiet calm environment, with consistent staff, and consistent cues for orientation they need to have the bare minimum of procedures done on them (catheters, cannulas etc).  They should not be restrained, and should be allowed to gently wander.  Therefore the ED is almost designed to be a bad place to have delirium.  It is noisy, busy, lacks continuity, and patients are moved around a lot. We need to build ED’s or use processes to limit the bad things, or move patients with delirium out to wards quickly, possibly with investigations en route.


So remember:

Screen all your elderly patients for delirium.  TWO questions.

  1. What day is it?
  2. Months of the year backwards.

Admitting your delirious patients to hospital may cause more harm than good.

Delirium usually has more than one cause and it is better to leave the diagnosis open, rather than prematurely close it.


List of terms used in medical documentation that should make you think of delirium

  • Pleasantly confused
  • Aggressive
  • Muddled
  • Sleepy
  • Drowsy
  • Restless
  • Withdrawn
  • Guarded
  • More confused than normal
  • Not his/her normal self
  • Knocked off
  • Off legs
  • Not quite right
  • Flat
  • Not eating/drinking


Siddiqi, Najma, Allan O. House, and John D. Holmes. “Occurrence and outcome of delirium in medical in-patients: a systematic literature review.” Age and ageing 35.4 (2006): 350-364.

RCEM Clinical Audits.  Assessing for Cognitivie Impairment in Older People. 2014-2015.

Beales L, Mercuri M BET 1: Screening for delirium within the emergency department Emerg Med J 2016;33:741-743.

Fick, D. M., Inouye, S. K., Guess, J., Ngo, L. H., Jones, R. N., Saczynski, J. S. and Marcantonio, E. R. (2015), Preliminary development of an ultrabrief two-item bedside test for delirium. J. Hosp. Med., 10: 645–650. doi:10.1002/jhm.2418