Skip to content
View commentaries on this research

This is a plain English summary of an original research article

A new shorter test for delirium appears helpful in assessing older people in hospital who may have the condition. A normal score on the 4 A’s test effectively rules out delirium while an abnormal score is reasonably useful for detecting the condition. People detected by the test would still need a full assessment to confirm the diagnosis.

Delirium is common in older people who have been hospitalised, but it can go undiagnosed. To help combat this, the 4 A’s test was developed as an alternative to a longer screening process. This NIHR-funded study found that the test is easy to use and appears to have been widely adopted in NHS hospitals. The study also reinforced the high costs of patients with delirium to the service.

For high-risk patients and those with sudden-onset confusion, these early results show that the 4 A’s test is a practical tool for initial assessment in time-pressured environments. It will need further testing in other settings.

Why was this study needed?

At least 20% of older people in hospital develop delirium, with the figure being even higher among those with dementia. The condition is treatable but often goes undetected. Symptoms include confusion and mood swings, which are distressing for both the patient and their family.

Delirium can be easily mistaken for dementia. This combined with lengthy assessment tools, have meant that in primary and acute care settings there can be difficulties in diagnosis. To help combat this, a short test was developed which assesses the 4 A’s: Alertness (or Arousal), Attention, Abbreviated Mental Test (4 item version), and Acute change. The test was designed to be more rapid to complete than its counterparts.

This study aimed to assess how far this relatively new tool has been incorporated into practice as well as its diagnostic utility.

What did this study do?

This NIHR-funded study had two main components. The first looked at the use of the 4 A’s test in practice. This was done via surveys and interviews with healthcare professionals. The incorporation of the test into guideline and policy documents was also assessed.

The second determined the test’s diagnostic accuracy. A total of 785 recently admitted patients aged 70 years and over participated. Around one in eight participants had delirium. Of the participants, 45% were male, 99% were white and 9% had a known dementia diagnosis. The participants came from emergency departments or acute general medical wards at three UK sites.

Each patient underwent a reference standard delirium assessment and was also randomised to receive an assessment with either the 4 A’s test or the older Confusion Assessment Method which takes longer to complete. A score of more than 3 on the 4 A’s test was considered abnormal and indicated possible delirium (scores on the test range from 0 to 12).

Only 17% of those eligible consented to take part. This may mean that the sample did not truly reflect the target population.

What did it find?

The reference standard assessment identified 12.1% (95 out of 785) of the participants as having delirium.

  • An abnormal 4 A’s test score had a specificity of 95% (95% confidence interval [CI] 92% to 97%) and a sensitivity of 76% (95% CI 61% to 87%) for reference standard delirium. The Confusion Assessment Method, by contrast, had a specificity of 100% (95% CI 98% to 100%) and a sensitivity of 40% (95% CI 26% to 57%).
  • In this population, the 4 A’s test achieved a positive predictive value of 66% (95% CI 52% to 78%) and a negative predictive value of 96% (95% CI 94% to 98%). This suggested that a normal test is useful for ruling out delirium in this setting.
  • Patients with abnormal 4 A’s test scores had longer lengths of stay (median 5 days, interquartile range 2 to 14 days) than did those with normal scores (median 2 days, interquartile range 1 to 6 days), and they had a higher 12-week mortality rate (16.1% compared with 9.2%, respectively).
  • The estimated costs of the initial inpatient stay for patients with delirium were more than double the costs for patients without delirium, and 12-week costs were also higher. For example, in Scotland, the 12-week costs were £7,559 for a patient with delirium and £4,215 for a patient without the condition. The estimated cost of a false-positive case was £4,653 and of a false-negative case was £8,956. A missed diagnosis increased costs by £2,067.
  • The survey revealed that both staff awareness and the detection of delirium could be improved. Only 20% of respondents reported delirium detection rates of more than 80% in their unit. While 64% of units had guidelines for delirium detection, only 20% reported that these were ‘almost always/always followed’. In terms of the 4 A’s test, 69% of respondents reported that it was routinely used in their clinical area, and 52% used it frequently.

What does current guidance say on this issue?

NICE guidance on delirium (2010) does not currently make any specific recommendations regarding tools to assess delirium at initial presentation. It asks healthcare professionals to determine whether there have been any recent changes in behaviour, either by asking the patient themselves or their relatives or carers.

Changes to look out for include those relating to cognitive function, such as worsened concentration and perception, which could manifest as visual or auditory hallucinations. If behaviour changes are observed, then a full clinical assessment by a trained healthcare professional is required. If there is difficulty distinguishing between dementia and delirium, the guideline recommends treating for delirium first.

What are the implications?

The 4 A’s test, which is shorter than existing full screening tools, appears helpful when used in the diagnostic pathway for people with suspected delirium. People with abnormal 4 A’s test scores stayed in hospital longer and their treatment was more expensive.

However, this study has also highlighted that until staff awareness is improved and there are clearer lines of responsibility for delirium assessment, the opportunity this test provides for better diagnosis could be missed. Additional staff training and promotion of existing internal policies could help with this.

Citation and Funding

MacLullich AMJ, Shenkin SD, Goodacre S et al. The 4 'A's test for detecting delirium in acute medical patients: a diagnostic accuracy study. Health Technol Assess. 2019; 23(40).

This project was funded by the NIHR Health Technology Assessment Programme (project number 11/143/01).



Alzheimer’s Society. Delirium. London: Alzheimer’s Society; 2019.

Healthcare Improvement Scotland. Think delirium. Improving the care for older people: delirium toolkit. Edinburgh: Healthcare Improvement Scotland; 2019.

NICE. Delirium: prevention, diagnosis and management. CG103. London: National Institute for Health and Care Excellence; 2010.

Produced by the University of Southampton and Bazian on behalf of NIHR through the NIHR Dissemination Centre


  • Share via:
  • Print article
Back to top