Professor John Young explains how implementation of the NICE guideline on delirium could help prevent one-third of episodes, resulting in significant cost savings for the NHS
  • Delirium is common across the healthcare system but is a poorly recognised condition. Be aware that people in hospital or long-term care may be at risk of delirium (‘think delirium’)
  • It is especially common in people aged over 65 years, and those who have dementia or hip fracture, or who are acutely ill
  • Reliable diagnosis requires careful clinical assessment to detect the cardinal features of recent onset fluctuating awareness, impairment of memory and attention, and disorganised thinking
  • Diagnosis is important because delirium is associated with increased mortality and risk of functional decline and onset/deterioration of dementia
  • It is estimated that about one-third of delirium episodes could be prevented; this would be hugely cost effective for a healthcare economy
  • Care systems for preventing delirium are currently poorly implemented in the NHS
  • Treatment involves rapid identification of the underlying cause or causes and individualised patient support
  • Treatment with haloperidol* or olanzapine* (at a low dose and for a few days only) may be required if the patient is distressed
  • Delirium is often an unpleasant experience for patients and full or partial recall is common. People should be to encouraged to discuss their experiences in the post-delirium period

*Haloperidol and olanzapine do not have UK marketing authorisation for this indication

Delirium is a common, complex clinical syndrome characterised by disturbed consciousness, cognitive function, or perception, which has an acute onset and fluctuating course.1,2 It is important because it is associated with poor outcomes including an increased risk of death, need for long-term care, and dementia. However, delirium is frequently undetected because patients present non-specifically, and skilful bedside assessment is required for diagnosis. Some case examples are provided in Box 1.

The subtype of hyperactive delirium in which patients have heightened arousal and can be restless and aggressive, is clinically more apparent than the hypoactive form in which the patient becomes withdrawn, quiet, and sleepy.1,2 In recent years it has become apparent that delirium can be prevented in about one-third of patients.3 This is a highly cost-effective strategy for healthcare and social care providers, but current practices and systems to support delirium prevention are poorly developed. All front-line healthcare and social care staff should have knowledge on the recognition, prevention, and treatment of delirium and this article summarises these aspects from the recommendations in the recently published NICE guideline for delirium.1,2

The NICE guideline covers adults in hospital and in residential and nursing homes. It does not cover children, people receiving end-of-life care or those withdrawing from drugs or alcohol. A multidisciplinary Guideline Development Group (GDG) and a technical team from NICE reviewed the clinical and cost-effectiveness evidence and formulated the recommendations. Two supporting algorithms covering the prevention and diagnosis of delirium and its treatment were designed (see Figures 1 and 2, below).4

Box 1: Case examples of people with delirium

Patient 1
Gladys is 75 years old and has been admitted to your ward having been found lying on the floor by home care staff. She appears to be talking to herself, sometimes loudly, but it is hard to understand what she is saying. She seems anxious and repeatedly pulls at her bedclothes. She argues with the nursing staff and has refused, in an angry and snappy fashion, to have a blood sample taken.

Hyperactive delirium indicated by abnormal restless and uncooperative behaviour, incoherent speech, hyper-vigilance, and abnormal hand movements.

Patient 2
Charles is recovering from a heart attack and seems to want to lie in bed. He is polite when approached, but intermittently sleepy and does not seem interested in eating or drinking. Sometimes his speech is incoherent and he does not appear to follow what is said to him.

The clinical clues for delirium in this patient are considerably more attenuated compared with the hyperactive sub-type in Patient 1. This is hypoactive delirium: he has become withdrawn, there is fluctuating drowsiness, incoherent speech, and there is evidence of impaired attention—all cardinal features for delirium. Hypoactive delirium is the more common of the two subtypes, but its more subtle features can be difficult to detect, and careful bedside observation skills are required. This is one important reason why delirium is easily overlooked.

Patient 3
Aariz, is 81 years old, has Alzheimer’s disease, and has been a resident in a care home for 2 years. He is usually quite talkative, although often shifts the conversation to his days as an electrician’s apprentice and frequently seems to think that it is the 1960s. In the last few days, Aariz has had urinary incontinence (unusual for him), and has been shouting out, especially at night. Last night he was found trying to dismantle a plug in his bedroom. One of his visitors tells you that Aariz has said that he believes the staff are trying to poison him.

Although these clinical features may result from progression of the dementia, the change in behaviour and paranoid ideas suggest that delirium superimposed on dementia, with a urinary infection (new urinary incontinence) as a precipitant, should be considered.

‘Think delirium’

The full guideline includes a review of the prospective cohort and cross sectional studies investigating delirium occurrence.2 This research evidence demonstrates that delirium is common and occurs throughout the healthcare system. The estimated prevalence of delirium in different settings is shown below:2

  • 20%–30% in medical wards
  • 10%–15% in surgical wards
  • 30%–80% in intensive care units
  • 10% in accident and emergency departments
  • possibly >15% in long-term care facilities.

However, the GDG found 70–700 fold disparity between the expected and observed delirium rates as reported and routinely collected using NHS coding data.2 Delirium appears to be invisible within the healthcare service. A recent well-conducted study reported that the delirium detection rate was only 25% for older people presenting to a medical assessment unit.5 Improved recognition of this serious syndrome seems an essential first step. Too many cases of delirium remain unrecognised: hence, the strap line for the NICE guideline is ‘Think delirium’.1,2

Risk factor assessment
Some groups of patients are at a higher risk of delirium than others. It therefore makes sense to target delirium prevention at these groups. The GDG identified four easy-to-define clinical groups that have a greater than five-fold increased risk of delirium. As most people in long-term care are aged over 65 years, and many have cognitive impairment, it follows that most residents will be at high risk for delirium and therefore the prevention programme should be widely deployed.

The NICE guideline recommends that when patients first present to hospital or long-term care, they should be assessed for a number of risk factors (listed below); the person is at risk of delirium if any of these are present:1,2

  • Age 65 years or older
  • Cognitive impairment (past or present) and/or dementia. If cognitive impairment is suspected, confirm it using a standardised and validated cognitive impairment measure
  • Current hip fracture
  • Severe illness (a clinical condition that is deteriorating or is at risk of deterioration).

People should be observed at every opportunity for any changes in the risk factors for delirium.1,2

Indicators of delirium
It is recommended that people who are at risk of delirium should be assessed at presentation for recent changes or fluctuations in behaviour (i.e. within hours or days). These behavioural changes may be reported by the at-risk person, or a carer/relative. The healthcare professional should be particularly vigilant for behaviour indicating hypoactive delirium (marked in bold below), as this subtype is problematic to detect. These behaviour changes may affect:1,2

  • Cognitive function (e.g. worsened concentration, slow responses, confusion)
  • Perception (e.g. visual or auditory hallucinations)
  • Physical function (e.g. reduced mobility, reduced movement, restlessness, agitation, changes in appetite, sleep disturbance)
  • Social behaviour (e.g. lack of cooperation with reasonable requests, withdrawal, or alterations in communication, mood and/or attitude).

If any of these behaviour changes are present, a clinical assessment should be undertaken to confirm the diagnosis; this should be carried out by a healthcare professional who is trained and competent in diagnosing delirium.1,2

‘At presentation’ is not simply about admission to hospital but includes a sick person being seen by a GP, and particularly urgent visits to people in care homes (where there is a high incidence of delirium). The GDG considered that an emphasis on these common delirium symptoms would be more likely to improve detection rather than the routine use of standardised diagnostic instruments that require considerable training. Moreover, this less technical approach might encourage patients’ families and healthcare assistants based in long-term care facilities to help identify features of delirium. These people are likely to know the person well and therefore are able to appreciate the small, subtle changes occurring in an individual who is developing early delirium.

Prevention of delirium

Delirium is an unpleasant illness. It is distressing for both patients and families, outcomes are poor (increased mortality, onset/deterioration of dementia) and it is expensive (increased length of hospital stay, increased use of long-term care). Prevention is therefore a highly attractive proposition.

Multiple moves within an acute hospital are now common. Many patients will move sequentially from accident and emergency to assessment units, to acute wards, and sometimes to post-acute wards. This is an example of how the whole hospital environment fails to promote a patient-centred approach. This could make it difficult for a sick person on the brink of a delirium episode to maintain their orientation and contact with reality. Another example is the excessive noise in hospital wards, which disrupts sleep—an important risk factor for delirium. By systematically attending to issues such as these, the occurrence and impact of delirium can be reduced. This is the concept of risk factor modification to reduce delirium incidence.

The NICE guideline recommends ensuring that people at risk of delirium are cared for by a team of healthcare professionals who are familiar to the person at risk. Moving people within and between wards or rooms should be avoided unless absolutely necessary.1,2

Preventative interventions
The GDG examined possible preventative strategies that included pharmacological, single component, and multicomponent interventions. The research evidence base is only sufficiently robust for complex multicomponent interventions and this approach is therefore recommended as suitable for uptake into routine care, both in hospitals (reasonable evidence) and long-term care (indirect evidence only).2 These interventions target and modify risk factors associated with delirium. The research literature suggests that about one-third of incidental delirium in hospitals (and perhaps care homes) could be prevented by this approach.3

At-risk patients should be given a tailored multicomponent intervention package:1,2

  • Within 24 hours of admission; people should be assessed for clinical factors contributing to delirium
  • Based on the results of this assessment, provide a multicomponent intervention tailored to the person’s individual needs and care setting.

The key components of the multicomponent intervention package are shown in the quick reference guide ( Although these interventions do not seem challenging (and may even be considered as ‘basic care’), the challenge for delirium prevention is one of high fidelity. We do some of these things to some of the patients some of the time, but prevention of delirium requires that we do all of these things, all the time to all patients at risk. It is quite a challenge and one that requires more than ‘motivated staff’ (although this is clearly important!). It requires a healthcare system or systems that support comprehensive and reliable delivery of these tasks.


The process for diagnosing delirium is summarised in the algorithm in Figure 1.1,2

There are no biomarkers that can reliably diagnose delirium. Rather, the diagnosis rests on bedside assessment. This is based on the Diagnostic and statistical manual of mental disorders (DSM-IV) criteria that comprise the international standard for the condition.6 The short Confusion Assessment Method is a simple algorithm that operationalises the somewhat interpretative components of the DSM-IV criteria.7 Both assessment methods can be readily applied in community settings but require considerable training and experience for accurate use.

Distinguishing delirium from dementia is a common clinical dilemma (See patient 3 in Box 1); the safest clinical approach is to manage all older people presenting with confusion as if they also have delirium until proven otherwise, and this is reflected in the guideline.1,2

Figure 1: Preventing and diagnosing delirium4


*If cognitive impairment is suspected, confirm using a standardised and validated cognitive impairment measure. If dementia is suspected, refer to ‘Dementia: supporting people with dementia and their carers in health and social care’ (NICE Clinical Guideline 42).
For further information on recognising and responding to acute illness in adults in hospital see ‘Acutely ill patients in hospital’ (NICE Clinical Guideline 50).
A healthcare professional trained and competent in the diagnosis of delirium should carry out this assessment.
DSM=Diagnostic and statistical manual; CAM=confusion assessment method; ICU=intensive care unit
National Institute for Health and Care Excellence (NICE) (2010) CG103. Delirium: diagnosis, prevention and management. London: NICE. Available from Reproduced with permission.

Figure 2: Treating delirium4

figure 2

*See ‘Violence’ (NICE Clinical Guideline 25).
Haloperidol and olanzapine do not have UK marketing authorisation for this indication.
For more information on the use of antipsychotics for these conditions, see ‘Parkinson’s disease’ (NICE Clinical Guideline 35) and ‘Dementia’ (NICE Clinical Guideline 42).
§For more information on dementia see ‘Dementia’ (NICE Clinical Guideline 42).
National Institute for Health and Care Excellence (NICE) (2010) CG103. Delirium: diagnosis, prevention and management. London: NICE. Available from Reproduced with permission.

Management of delirium

The DSM-IV criteria emphasise that delirium is triggered by a physical cause or, more usually, a combination of causes. A common example would be an older person with dementia (high baseline risk of delirium) who suffers a fall and develops delirium because of pain, a urinary infection, dehydration, and multiple medications. The literature and the personal accounts provided by the lay members of the GDG offered ample evidence that delirium is a largely negative and unpleasant experience and therefore management should also prioritise a ‘human’ approach.

Initial management of delirium
The possible cause or combination of causes of delirium should be identified and managed in people who have been diagnosed with the condition.1,2

Healthcare professionals should ensure effective communication and reorientation (e.g. explaining where the patient is, who they are, and the role of the professional) throughout the management process. Patients should be reassured and it may be useful to involve the family, friends, and carers in supporting them. It is important to ensure that patients are cared for in a suitable environment (e.g. avoid moving patient between rooms).1,2

Distressed people
If the patient is distressed or is deemed a risk to themselves or others, and verbal and non-verbal de-escalation techniques are ineffective or inappropriate, it is worth considering short-term (1 week) haloperidol or olanzapine. Treatment should be initiated at the lowest clinically appropriate dose and be titrated cautiously according to symptoms. It should be noted that neither of these drugs have UK marketing authorisation for this indication.1,2

Given the unpleasant experience and distress associated with delirium, there could be a strong argument for sedating pharmacological treatments. Moreover, delirium is in part a disturbance of neurotransmitter function (reduced acetylcholine, increased dopamine) and drugs may have a role in ameliorating the underlying neuropathophysiological processes. However, the evidence base is far from robust and the potential harms of sedating and antipsychotic medications needs to be considered. For example, benzodiazepines are in themselves a risk factor for delirium and are associated with a five-fold increased risk of delirium.8 It is counterintuitive to treat a person using a medication that may cause the condition. As far as antipsychotics such as haloperidol and olanzapine are concerned, recent large scale observational studies have reported neurotoxicity with increased stroke risk, especially for people with dementia (a common group to experience delirium).9 Hence the recommendation on the management of delirium emphasised non-pharmacological approaches with a recommendation for anti-psychotic drug treatment only for selected patients, and only for a few days.1,2

It is increasingly recognised that delirium may not fully resolve and that a proportion of patients will go on to develop dementia.2 Thus it is important to review patients with persisting symptoms. Patients may also benefit from psychological support in the post-delirium period. Partial or complete recall after an episode of delirium is common. The distress of a ‘waking dream’ and ‘illogical uncertainties’ needs to be acknowledged and recognised as it can adversely affect the psychological health of patients. The GDG therefore recommends that patients should be encouraged to discuss their experiences in the post-delirium period.1,2

Health economic model

A health economic model was developed to assess the economic impact of delirium prevention in at-risk patients:2

  • for the surgical prevention model— the incremental net monetary benefit (INMB) for the prevention system of care compared with usual care is £8180
  • for the medical prevention model—the INMB for the prevention system of care compared with usual care is £2200.

These results were robust to various sensitivity analyses that assessed for imprecision in the model parameters. Thus, a widely deployed delirium prevention strategy in hospitals and in care homes would be expected to save money.


The NICE guideline on delirium is accompanied by several implementation tools, but effectively putting these recommendations into practice will require three interrelated actions from healthcare professionals, care staff, and their employing organisations.

Firstly, we need to improve the awareness of delirium as a common and serious illness. The currently poor detection of delirium and the common misdiagnosis of this condition as dementia needs to be addressed by educational programmes for all front-line staff working in care homes and general hospitals. Clinical coding in hospital and primary care records needs to be improved to document the condition for local and national audits. Healthcare and social care monitoring organisations such as the Care Quality Commission could use delirium incidence rates as a ‘window on care quality’.

Secondly, staff in hospitals and care homes should adopt a new culture of delirium prevention. This will require both specific education to change attitudes and improve knowledge and skills, and a re-design of care systems. Clear clinical and managerial leadership and responsibility are therefore required as a whole hospital or a whole care home approach is necessary. The efficiency gains for a local healthcare economy are likely to be substantial if delirium prevention and treatment is improved. A helpful analogy is that of pressure ulcers. Before the 1990s, the NHS was prepared to commit substantial resources on the treatment of pressure ulcers. A realignment of staff skills and care systems to support pressure ulcer prevention has been hugely successful. The same process needs to be repeated for delirium.

Lastly, greater involvement of patients and relatives is required to inform timely patient-centred care (the key to delirium prevention), and in the redesign of care systems (e.g. noise reduction, maintaining hydration, nutrition).


Delirium has been a massively neglected condition relative to its frequency and serious consequences. The newly published NICE guideline contains three headline conclusions:

  • delirium is hugely under recognised and under diagnosed in the NHS
  • approximately one-third of all delirium episodes could be prevented
  • prevention would be a hugely cost-effective strategy for the NHS.
Implementation tools

NICE has developed the following tools to support implementation of Clinical Guideline 103 on Delirium: diagnosis, prevention and management. The tools are now available to download from the NICE website:

Audit support
Audit support has been developed to support the implementation of this guideline. The aim is to help NHS organisations with a baseline assessment and to assist with the audit process, thereby helping to ensure that practice is in line with the NICE recommendations. The audit support is based on the key recommendations of the guidance and includes criteria and data collection tools.

Baseline assessment tool
The baseline assessment tool is an Excel spreadsheet that can be used by organisations to identify if they are in line with practice recommended in NICE guidance and to help them plan activity that will help them meet the recommendations.

Costing tools
National cost reports and local cost templates for the guideline have been produced:

  • Costing reports are estimates of the national cost impact arising from implementation based on assumptions about current practice, and predictions of how it might change following implementation of the guideline
  • Costing templates are spreadsheets that allow individual NHS organisations and local health economies to estimate the costs of implementation taking into account local variation from the national estimates, and they quickly assess the impact the guideline may have on local budgets.

Slide set
The slides provide a framework for discussing the NICE guideline with a variety of audiences and can assist in local dissemination. This information does not supersede or replace the guidance itself.

Admission assessment for delirium
This aims to help healthcare professionals identify people at risk of and with indicators of delirium.

Care plan for the prevention of delirium
This aims to help healthcare professionals prevent delirium in people considered at risk of the condition.

Delirium prevention awareness workshop
An interactive workshop can help to raise awareness among healthcare professionals and care staff about how delirium can be prevented.

Implementation advice
This has been produced to support implementers in developing an action plan to put in place the key priorities for implementation for this guideline.



  1. National Institute for Health and Care Excellence. Delirium: Delirium—diagnosis, prevention and management. Clinical Guideline 103. London: NICE, 2010. Available at:
  2. National Clinical Guideline Centre. Delirium: diagnosis, prevention and management. London: NCGC, 2010. Available at:
  3. Young J, Inouye S. Delirium in older people. BMJ 2007; 334: 842–846.
  4. National Institute for Health and Care Excellence. Delirium: diagnosis, prevention and management. Quick reference guide. London: NICE, 2010. Available at:
  5. Collins N, Blanchard M, Tookman A, Sampson E. Detection of delirium in the acute hospital. Age Ageing 2010; 39 (1): 131–135.
  6. American Psychiatric Association. Diagnostic and statistical manual of mental disorders (DSM-IV). 4th ed. Washington: American Psychiatric Association, 2004.
  7. Inouye S, van Dyck C, Alessi C et al. Clarifying confusion: the confusion assessment method. A new method for detection of delirium. Ann Intern Med 1990; 113 (12): 941–948.
  8. Clegg A, Young J. Which medications to stop in people at risk of delirium: A systematic review. Age and Ageing (in press).
  9. Douglas I, Smeeth L. Exposure to antipsychotics and risk of stroke: self controlled case series study. BMJ 2008; 337: a1227.G
  • The NICE guideline is adamant that the prevention of delirium has the potential to generate cost savings
  • A simple risk assessment tool to identify individuals at risk of delirium can be adopted locally
  • The use of such a risk assessment, linked to prevention of delirium, could be written into contracts with local community and acute hospital providers by commissioners
  • In primary care, GPs and nursing teams could carry out a similar risk assessment when people are admitted to residential care (this could be commissioned as a local enhanced service)
  • Commissioners should consider a delirium awareness and prevention educational package for primary care healthcare professionals and carers