clockIt’s 7 am on a Monday. Your first patient is an 82 year-old woman who was brought in by EMS from an assisted living facility. All EMS can tell you is that she was not acting herself. You enter her room and introduce yourself. “Hello Mrs. Jones. How are you today?” The woman startles, “Well, you see, I went to put my dog out, and then I was just walking, and couldn’t remember. So it’s all coming full circle, and then I ate a sandwich.” Just then EMS rolls in with another patient, a 75 year-old male coming from home, who was found by his wife in his recliner minimally responsive, with a GCS of 6.  He is followed by a 76 year-old female who had a fall from standing three days ago, and has been increasingly confused today, and is currently oriented only to person.

All of these patients are registered with chief complaints of altered mental status (AMS), but their underlying pathology and management are very different. “AMS”, along with the ever-popular “weak and dizzy” and “pain all over” is a chief complaint that creates challenges because of its vague, non-specific nature. Developing a differential diagnosis for an elderly patient with AMS using the traditional ‘worst-first’ frameworks can be difficult, because of the breadth of presentations that are lumped under AMS, as well as its myriad causes. This post outlines a time-based approach and suggests 10 tips to avoid “rookie mistakes” in dealing with elderly patients presenting with AMS.

What is AMS anyway?

“AMS” itself can be a misleading and over-general term. While the full assessment of AMS is beyond the scope of this post,1 a few points are worth mentioning.

It is important to try to identify in what way the mental status is “altered”. The patient may be suffering acute brain dysfunction,which involves a changed level of alertness and attention (coma, stupor, delirium). Coma and stupor are usually obvious, but we frequently fail to explicitly identify the presence of delirium in the ED, particularly the hypoactive form. The patient with delirium may also have a change in cognition that can manifest in different ways, including disorientation, disorganized thinking, and alterations in perception. Alternatively, what is perceived as “AMS” may not actually have a change in mentation, and instead may be an inability to carry out their usual activities due to a loss of higher executive function, decreased coordination, vision loss, depression, or stroke to name a few. Sometimes all the history we can obtain is “Grandma just hasn’t been acting herself,” and it is left up to us to figure out why she is altered, and what should be done next. It is critical to identify delirium in the ED. It can be considered a vital sign in the elderly, alerting us to look for serious underlying problems just as we would with hypotension, tachycardia, or fever. Delirium is an independent predictor of higher mortality.2

Lost and Confused Signpost

The approach to AMS

As EPs we are trained to think of a ‘worst first’ differential diagnosis. However, this approach becomes challenging when the chief complaint is as broad as AMS, which could be applied to someone experiencing a devastating intracranial hemorrhage or –alternatively- to someone who was labeled “unresponsive” at their nursing facility because they had fallen asleep. In an altered patient, it is difficult to weigh which of the many potential maladies could be worst (urosepsis? MI? stroke? ICH?). So it is not always clear where to start. Obtaining more history and performing a physical will usually be helpful to narrow your differential so that you can focus your efforts. However, a time-based approach can also be helpful, especially when you are unable to obtain a history due to AMS, dementia, or a lack of available care-givers who can supply the history.

The time-based approach

The time-based approach looks at several time horizons and asks: What could kill or harm my patient in the next few seconds, minutes, hours, or days? At each point, consider the differential, intervene or test as needed before continuing down the list. Throughout the process, keep in mind the rapidly reversible causes of AMS:

  • Hypoxia
  • Hypercarbia
  • Hypoglycemia
  • Hypovolemia
  • Opioid overdose

What could kill or harm my patient in the next few SECONDS?

This focuses on your ABCs:

  • Cardiac arrest
    • Do they have a strong pulse?
    • Is their HR too fast or too slow to allow adequate perfusion?
  • Airway compromise
    • Is their airway obstructed?
    • Are they making snoring sounds?
    • Are they protecting their airway?
    • Do they need dentures or foreign body removed, jaw thrust, suctioning, intubation?
  • Hypoventilation
    • What is their oxygen saturation and end tidal CO2?
    • Are they hypoxic or hypercarbic?
    • Do they need supplemental oxygen or BiPAP?

 What could kill or harm my patient in the next few MINUTES?

  1. Hypoglycemia

While this condition would rarely be immediately fatal, it is rapidly reversible, so a finger-stick glucose should be checked early in the patient’s evaluation.

  • A “silent” MI
    • Elderly patients are less likely to have chest pain with MIs. Check an ECG.
  • Metabolic derangements such as kyperkalemia
    • Does the patient have any reason to be hyperkalemic (check the med list if you have it, look for a dialysis fistula)?
    • Order an ECG and an i-stat (where available), if you are concerned.
  • Aortic dissection or AAA rupture
    • Consider a bedside ultrasound, bilateral blood pressures, and portable CXR.
  • Overdose on prescribed medications or illicit drugs
    • Do they have signs of QRS widening suspicious for TCA overdose?
    • Are they bradycardic and hypotensive concerning for one of the “brady bunch” (beta-blockers, calcium-channel blockers, clonidine, digoxin)?
    • Are they exhibiting signs of a toxidrome? Check their skin, pupils, heart, muscle tone, temperature, reflexes, and an ECG.
    • Look for transdermal medication patches such as fentanyl, clonidine, or nicotine. A tip from personal experience: look in the patient’s mouth and on the perineum for patches as well.
  • Intracranial hemorrhage
    • The textbooks say the patient should complain of sudden onset, severe headache, but their altered mental status may make it impossible for them to voice their complaints. Alternatively, their mental status initially may not be severely altered, but can rapidly progress to coma as the hematoma expands. You may be able to intervene to help slow the hematoma expansion through blood pressure control and reversal of anticoagulation.3

All of the above can be performed within the first 5-10 minutes. The next question to ask is:

 What could kill or harm my patient in the next few hours?

  • Hypovolemia or hypotension
    • Do they need fluids?
    • Have they had blood loss from a GI bleed or trauma?
  • Alcohol withdrawal
    • We under-recognize alcohol abuse and dependence in the emergency department. These behaviors are important to keep in mind with elderly as well as younger patients.
  • Non-convulsive status epilepticus
    • This condition is less common, but is still an important consideration, particularly when other causes are not evident.
  • Ischemic CVA that, if missed, will be outside the tPA window.
    • The use of tPA for stroke is a hotly debated topic.4 Whether it is even useful at all is beyond the scope of this post (see this SMART EM podcast for a great review). However, if your patient is within the tPA window on arrival, but stroke symptoms are not recognized until several hours after arrival, they may no longer tPA candidates.
  • Sepsis
    • Are there signs or symptoms of infection? Elderly patients are more susceptible to infections than younger people, have higher infection-related morbidity and mortality, and often are more subtle in their presentations.5 They may not be able to mount a fever, tachycardia, or a white count as robustly as younger people.
    • Where could the infection be? I will borrow a mnemonic from Rob Orman of ERCAST of “LUCAS”:
      • Lung
      • Urine
      • CNS
      • Abdomen
      • Skin
    • Also consider osteomyelitis and endocartiditis, if none of the above fit.
  • Necrotizing fasciitis or other serious skin infections
    • This deserves special attention because of the risk of decubitus ulcers in elderly patients. Look at the patient’s skin, all their skin.
    • During an overnight shift, I was assessing an elderly, wheelchair-bound male who had come in with general malaise. As I was walking away from the bed I realized we had not rolled him. With the help of two nurses, we rolled the patient onto his left. On the right buttock I noted a 5 cm stage 1 pressure sore, and was feeling good that I had made sure to note and document it. Then as we rolled him a little bit more, we found that his entire left buttock was eroded down to the bone with necrotic ulceration and surrounding cellulitis. It turned out he was septic from this infection, which could have easily gone unrecognized if the patient’s skin had not been thoroughly examined.
  • Intra-abdominal catastrophe
    • Does the patient have an acute abdomen? It might be hard to tell. Elderly patients may not have the usual rigidity and rebound associated with peritonitis that we expect in younger people.6
    • Regarding elderly patients with abdominal pain:

Physical examination findings cannot reliably predict or exclude significant disease. These patients should be strongly considered for hospital admission, particularly when fever, hypotension, leukocytosis, or abnormal bowel sounds are present. — Marco et al7

  • Other metabolic derangements
    • Consider hypo/hypernatremia, hypokalemia, dehydration, hypo/hypercalcemia, DKA, HHNK, and adrenal disorders.
  • Medication interactions or side effects.
    • Are there any new medications, or changes in doses?

What could kill or harm my patient in the next few HOURS TO DAYS?

At this stage, you have more time to think. You can go through your usual, full differential diagnosis for AMS. The AEIOU TIPS mnemonic is one popular memory aid.

Hopefully the outline above will help guide your approach to an altered, elderly patient when little additional information can be obtained.


10 Tips to help avoid Rookie Mistakes in the Evaluation of Elderly Patients with AMS

For the interns who are just getting their feet wet, embrace the challenge that vague complaints such as “AMS” or “weak and dizzy” present. Accept it as an opportunity to hone your clinical skills and uncover subtle presentations of potentially serious illness!

  1. Have a systematic approach to rapidly identify the time-sensitive pathologies.
  2. Don’t underestimate the severity of injuries that can result from a simple fall from standing, particularly for those on anti-coagulants.8 C-spine fractures and intracranial hemorrhages are common. Also, ask about falls in the last week (not just today). Elderly patients are more at risk for delayed subdural hematomas.9
  3. Look at their skin… all their skin.
  4. Look carefully at their medications, and ask about changes.
  5. Evaluate for delirium, and think of it as a vital sign alerting you to potentially serious underlying pathology in the elderly patient. The Confusion Assessment Method (CAM) is a simple, common sense tool to used for delirium assessment.  In an agitated, combative elderly patient, delirium is obvious. In the hypoactive form, it is easier to miss. Always look for an underlying medical problem in a hypoactive or delirious patient.
  6. In addition to a good neuro exam, do a brief cognitive evaluation. Patients can often compensate for deficits. For example, they may sincerely answer your various questions regarding nausea, vomiting, recent illnesses, medication changes, but on cognitive evaluation, you find they have severe short term memory deficits, and think that Reagan is the president. The minicog is a quick assessment of underlying cognitive dysfunction.10
  7. Think about alcohol or substance abuse and withdrawal. It is easy to assume that elderly patients do not use drugs or alcohol. The truth is that we under-recognize the prevalence of these problems in elderly patients.11,12
  8. Do not stop looking for causes of AMS simply because you found 7 WBC’s in the urine. We likely over-diagnose UTIs in the elderly.
  9. Don’t attribute AMS and vomiting to a GI bug. It could just as easily be increased ICP, meningitis, posterior circulation CVA, or bowel perforation.
  10. Get collateral information! Call a family member or the facility where the patient lives as soon as possible. If there is too much delay, then the patient’s nurse will have gone for the day, leaving no one who can explain why the patient was sent to the ED. It is equally important to establish the patient’s baseline mental status and level of functioning. If they normally drive and play multi-lingual scrabble, then the fact that they cannot remember the year is a big change. It is easy to erroneously attribute a patient’s deficits to advanced age and dementia. Obtaining collateral information is the best way to avoid this mistake.

There are 2 ways to learn – from your own mistakes and from someone else’s. The latter is always preferable.


Han J, Wilber S. Altered mental status in older patients in the emergency department. Clin Geriatr Med. 2013;29(1):101-136. [PubMed]
Han J, Wilson A, Ely E. Delirium in the older emergency department patient: a quiet epidemic. Emerg Med Clin North Am. 2010;28(3):611-631. [PubMed]
Elliott J, Smith M. The acute management of intracerebral hemorrhage: a clinical review. Anesth Analg. 2010;110(5):1419-1427. [PubMed]
Weintraub M. Thrombolysis (tissue plasminogen activator) in stroke: a medicolegal quagmire. Stroke. 2006;37(7):1917-1922. [PubMed]
Norman D. Fever in the elderly. Clin Infect Dis. 2000;31(1):148-151. [PubMed]
Kauvar D. The geriatric acute abdomen. Clin Geriatr Med. 1993;9(3):547-558. [PubMed]
Marco C, Schoenfeld C, Keyl P, Menkes E, Doehring M. Abdominal pain in geriatric emergency patients: variables associated with adverse outcomes. Acad Emerg Med. 1998;5(12):1163-1168. [PubMed]
Nishijima D, Offerman S, Ballard D, et al. Risk of traumatic intracranial hemorrhage in patients with head injury and preinjury warfarin or clopidogrel use. Acad Emerg Med. 2013;20(2):140-145. [PubMed]
Itshayek E, Rosenthal G, Fraifeld S, Perez-Sanchez X, Cohen J, Spektor S. Delayed posttraumatic acute subdural hematoma in elderly patients on anticoagulation. Neurosurgery. 2006;58(5):E851-6; discussion E851-6. [PubMed]
Borson S, Scanlan J, Brush M, Vitaliano P, Dokmak A. The mini-cog: a cognitive “vital signs” measure for dementia screening in multi-lingual elderly. Int J Geriatr Psychiatry. 2000;15(11):1021-1027. [PubMed]
O’Connell H, Chin A, Cunningham C, Lawlor B. Alcohol use disorders in elderly people–redefining an age old problem in old age. BMJ. 2003;327(7416):664-667. [PubMed]
Kalapatapu R, Paris P, Neugroschl J. Alcohol use disorders in geriatrics. Int J Psychiatry Med. 2010;40(3):321-337. [PubMed]
Christina Shenvi, MD PhD
Associate Professor
University of North Carolina
Christina Shenvi, MD PhD


Emergency Medicine and Geriatrics trained, Educator, Professional nerd, mother of 4, excited about #educationaltheory, #MedEd, #EM, #Geriatrics, #FOAMed.
Christina Shenvi, MD PhD

Latest posts by Christina Shenvi, MD PhD (see all)