Defining delirium

Nosological confusion surrounds the concept of delirium. Historically, acute generalised disturbances of cognition have been referred to by a number of synonyms, each reflecting delirium occurring in particular populations or treatment settings. These range from the clinically intuitive ‘acute confusion’ to the downright bewildering ‘subacute befuddlement’. DSM-III (American Psychiatric Association 1980) provided systematic criteria for diagnosis that have gradually evolved through successive systems to the recently published DSM-5 (American Psychiatric Association 2013). The latter categorises delirium as a neurocognitive disorder based on criteria that reflect the occurrence of generalised disturbance of brain function (evidenced by cognitive and neuropsychiatric symptoms in which inattention and reduced awareness are central) that is relatively recent in onset, fluctuates and occurs in the context of physical morbidity. This concept draws on the relative simplicity that underpinned DSM-IV and has contributed to its popularity among clinicians and researchers (**Reference Morandi, Davis and TaylorMorandi 2013).

Delirium and dementia

Delirium and dementia are the two major neuropsychiatric syndromes affecting elderly people and they have considerable overlap in respect of clinical presentation. It is well recognised that dementia is a major risk factor for developing delirium (and underpins delirium in about 50% of cases), but increasing evidence implicates delirium not only as a marker for emerging dementia, but also as an aggravating factor in dementia that accelerates its course and as an independent risk factor for long-term cognitive impairment, including dementia (**Reference MacLullich, Beaglehole and HallMacLullich 2009). In a recent review, Reference George, Long and VincentGeorge and colleagues (2013) highlight how patients with dementia in acute hospitals have many negative outcomes, such as falls, delirium, loss of function, increased length of stay and increased mortality. Importantly, they identify multiple contributory factors, including inadequate assessment and treatment, inappropriate intervention, discrimination, inadequate staffing levels and poor staff training.

Prevalence and impact

Impact

Delirium has a range of adverse outcomes for the individual and for healthcare services. The cost to the individual includes loss of independence and increased mortality. But most important is the growing recognition that the state of delirium may be more than the neuropsychiatric manifestation or ‘symptom’ of an underlying physiological disturbance: it might in fact be neurotoxic in itself (**Reference MacLullich, Beaglehole and HallMacLullich 2009; **Reference Witlox, Eurelings and De JongheWitlox 2010).

The costs of delirium to healthcare organisations are high, in terms of both economics (Table 1) and service use. The cost of a single admission of a patient with delirium has been estimated at £13 000 in the UK (**Reference Akunne, Murthy and YoungAkunne 2012) and between $16 000 and $64 000 in the USA (**Reference Leslie, Marcantonio and ZhanqLeslie 2008).

TABLE 1

Estimated direct 1-year healthcare costs in the USA

The need for cognitive-friendly hospitals

In view of the significant human, clinical and economic costs of delirium in the acute general hospital, cohesive and multifaceted programmes are required to meet the needs of in-patients with newly emerging and/or long-standing cognitive problems. Reference George, Long and VincentGeorge and colleagues (2013) have proposed a multilevel, multifactorial approach to this problem, aimed at increasing awareness, education and detection and promoting evidence-based management of cognitive problems. Their cognitive-friendly hospital programme focuses on seven levels of care: patient, task, staff, team, environment, organisation and institution.

What constitutes a cognitive-friendly hospital programme?

Cognitive-friendly hospital programmes must be considered in the context of the wider healthcare system, and should be universally sensitive and responsive to the needs of our ageing population. One prominent example is the ‘senior friendly health services’ in Ontario, Canada (**Reference Hart, Frank and HoffmanHart 2006; Senior Friendly Hospitals 2012). Within this clinical construct, cognitive-friendly hospitals are responsive to the needs of a wide range of patients with differing and potentially changing cognitive profiles. For example, a person with pre-existing dementia is ‘flagged’ at entry to hospital as someone who has a high risk of developing delirium and a worsening cognitive profile while in hospital. The individual is closely monitored throughout their hospital stay, with delirium prevention measures and rapid assertive intervention in the event of emerging delirium.

A cognitive-friendly hospital programme needs to consider the patient’s entire ‘journey’ while in hospital in terms of the different clinical areas and services from the point of access to care (e.g. via a medical assessment unit, emergency department or elective admission) through to continuing care (e.g. on medical, psychiatric or surgical wards, or in intensive care or palliative care). This can facilitate a cohesive and multicomponent intervention that is tailored to the patient’s needs at different points of the at-risk/episode/recovery spectrum. Finally, these efforts should link effectively with primary and community care, where possible before, and always during and after, the hospital admission.

Key programme components

A cognitive-friendly programme should focus on the following areas (Fig. 1):

1. • increasing general awareness and recognition of cognitive problems

2. • improving levels of education and training

3. • optimising the use of evidence-based management strategies

4. • discharge planning that reflects the individual’s cognitive profile during the hospital admission.

FIG 1

A cognitive-friendly hospital programme: components relating to delirium.

Increasing awareness

Increasing awareness of delirium and other cognitive problems and of their clinical importance is an essential first step in the development of a cognitive-friendly hospital programme. Although many patients with delirium experience difficulty in communicating their experiences and/or have diminished environmental awareness, studies indicate that many can themselves identify recent confusion and other synonyms for delirium (**Reference Ryan, O'Regan and CaoimhRyan 2013). All front-line clinical staff, such as doctors, nurses, allied healthcare professionals and care attendants, should be involved. Non-clinical staff with high levels of patient contact (e.g. cleaners and catering staff) should also be included, as subtle changes in eating and hygiene/self-care can assist in delirium detection and are often noticed by these staff members. In a UK qualitative study using normalisation process theory (**Reference Godfrey, Smith and GreenGodfrey 2013), staff, volunteers, patients and carers participated in developing a prevention of delirium programme that includes a multicomponent process that aims to introduce and embed delirium prevention into routine clinical practice. This research project led to the development of an integrated model of delirium prevention, known as the Prevention of Delirium (POD) Programme. Significant challenges were cited, including lack of understanding of delirium prevention and difficulties integrating a multifaceted programme into existing hospital systems. The authors have not as yet reported on the implementation of this programme.

Hospital executives should have a clear knowledge of the impact of delirium on factors such as clinical outcomes and length of stay in hospital, especially considering their financial implications (**Reference Leslie and InouyeLeslie 2011). Increasing awareness of delirium among hospital management and budget holders should encourage the development of delirium prevention and treatment programmes with proven cost-effectiveness. These include multicomponent preventive interventions (**Reference Akunne, Murthy and YoungAkunne 2012) such as the Hospital Elder Life Program (HELP). The latter can be tailored to individual hospitals and hospital networks to ensure that the needs of older patients are considered in all relevant aspects of service planning. Delirium diagnosis and associated clinical activity should be routinely captured (**Reference Rubin, Williams and LescisinRubin 2006).

Education and training

Risk factors and delirium prevention

A key objective and quality-defining principle for a cognitive-friendly hospital is its ability to minimise exposure to risks that can result in delirium, and UK guidelines detail a range of evidence-based interventions that can reduce these risks (NICE 2010). Delirium is highly predictable, as a range of predisposing patient, illness and treatment factors interact with acute precipitating insults to produce the acute brain failure of delirium. Many of these factors (including a number with a significant iatrogenic element) are preventable and better service organisation can allow for preventive measures to produce better patient outcomes and reduced healthcare costs.

Reference Inouye and CharpentierInouye & Charpentier’s (1996) model of delirium causation among elderly hospital in-patients predicted a seventeen-fold variation in the relative risk of developing delirium. The model has been validated in elderly patients undergoing hip surgery (**Reference Kalisvaart, De Jonghe and BogaardsKalisvaart 2005). Baseline risk is especially important, as patients with high baseline vulnerability can develop delirium in response to even minor precipitants.

Certain factors are more relevant in particular settings and patient groups, but age extremes (very young and very old), pre-existing cognitive problems, severe comorbid illness and psychotropic medication are robust predictors of delirium risk across populations. All of these risks factors are readily identifiable and many are modifiable. Others can inform the risk–benefit analysis of surgical and other interventions in deciding on optimal care, especially for frail elderly patients with cognitive impairments (Box 1).

Minimising exposure to modifiable factors has been the focus of intervention studies that demonstrate that delirium is highly preventable (**Reference O'Hanlon, O'Regan and MacLullichO’Hanlon 2014). Many elements simply reflect good medical and nursing care (e.g. avoiding unnecessary polypharmacy, correcting sensory deficits). That these relatively simple practices need to be protocolised within a formal ‘intervention’ model reflects the challenges to the provision of routine care in most hospitals: delirium is often a marker of the dysfunction within our healthcare systems and processes.

Management of delirium

Although most of the literature reporting on the treatment of delirium derives from secondary care, the key messages are equally germane to healthcare professionals working in primary care. Delirium management can be thought of broadly in terms of prevention and treatment. These overlap and include both environmental and pharmacological approaches. Fundamental to these efforts is the identification of patients at high risk of developing delirium as well as the timely recognition of delirium when it occurs. However, these efforts must be aligned to meaningful interventions, as identification of patients with delirium does not always translate into better care. Several studies have identified substantial challenges to the management of delirium once identified, not least the knowledge and skills of healthcare workers (**Reference Godfrey, Smith and GreenGodfrey 2013; **Reference Teodorczuk, Mukaetova-Ladinska and CorbettTeodorczuk 2013).

Despite our increased understanding of delirium as a cohesive clinical entity, its presentation, severity and course can be highly variable, for example involving both syndromal and subsyndromal levels of severity and varying degrees of persistence or residual features over time. These factors highlight the complex clinical nature of delirium and associated difficulties in ensuring timely detection and treatment.

Given the number and complexity of biopsychosocial triggers of delirium, it is unsurprising that simple one-dimensional interventions (e.g. consensus guidelines, education) have limited impact (**Reference Young and GeorgeYoung 2003; **Reference Gagnon, Allard and GagnonGagnon 2010) and that multifaceted interventions that are sensitive to the complexities of individual patients’ needs are required.

Discharge planning and delirium management post-discharge

Despite the high prevalence of delirium in the general hospital and its significant impact on long-term general and cognitive health and subsequent mortality, delirium occurrence is frequently omitted from discharge planning and in communications with primary care and other community services. There is evidence that the use of structured discharge reports can increase the reporting of delirium in discharge summaries (**Reference Van Zyl and DavidsonVan Zyl 2003).

The occurrence of delirium has important implications for post-acute care, where ongoing risk factors can be minimised and any residual functional deficits or psychological sequelae can be addressed. A substantial number of patients experience long-term cognitive and functional difficulties after resolution of the initial delirium event. One study, involving 433 older adults, found that 83% of those who were discharged from hospital with persistent delirium, and 68% of those whose episode of delirium had resolved before discharge home, required long-term care or had died within a year; these figures compared with 42% for patients who were never delirious (**Reference McAvay, van Ness and BogardusMcAvay 2006). Furthermore, an episode of delirium accelerates cognitive decline in patients with a pre-existing dementia (**Reference Fong, Jones and ShiFong 2009), emphasising the need for community care partners to be aware of the potential need for increased support for such individuals.

Delirium is a distressing experience for patients and their caregivers. Patients who have recovered from delirium may be uncomfortable discussing it because they equate delirium with senility or being ‘mad’. Reference Breitbart, Gibson and TremblayBreitbart et al (2002) explored perceptions of the experience of delirium among patients with a resolved delirium episode and their spouses. They found that distress correlated with severity of perceptual disturbances for the patients and functional loss for their spouses. In elderly patients without dementia who experienced delirium, more than half could recall their psychotic symptoms and many were still distressed by their recollections 6 months later (**Reference O'Malley, Leonard and MeagherO’Malley 2008).

Persistent psychological disturbances are a particular target for interventions as they can affect subsequent help-seeking behaviour. Formal followup visits can facilitate post-delirium adjustment by allowing for discussion of the meaning of delirium and planning of how to minimise future risk (e.g. by addressing risk factors such as medication exposure and sensory impairments).

A plan for improved delirium care

A cognitive-friendly hospital environment depends on a holistic approach to the identification, prevention and management of delirium. It requires patient-centred clinical care and system-wide support. The cost-effectiveness of such interventions and programmes should be evaluated and demonstrated. The evidence in support of the development of delirium care pathways already exists and can be embedded into the everyday practice of all hospitals. While the general principles for complete delirium care are well described, these care pathways should be developed with the specific pressure points of each institution in mind.

Particular challenges include:

1. • promoting awareness of delirium by including its detection and management as a key educational component of medical and nursing programmes at undergraduate and postgraduate level

2. • ensuring that delirium risk status monitoring becomes embedded into daily routines across all healthcare settings, including hospitals and community care

3. • systematic application of preventive measures, particularly for high-risk patients

4. • earlier and more consistent detection of delirium through formal systematic screening in everyday practice across healthcare settings, using computer-assisted and other technologies to facilitate more accurate assessment

5. • linking diagnosis of delirium to evidence-based action through protocolised management based on a more coherent understanding of how best to manage specific clinical presentations of delirium, the optimal duration of treatment and associated risk/benefit ratios

6. • implementation of formal strategies to promote the active management of the post-delirium phase, including the risk of subsequent episodes and the prevention of secondary psychological sequelae; this should include liaison between primary and secondary care especially in the post-discharge follow-up of patients diagnosed with delirium while in hospital

7. • monitoring the frequency of delirium and its impact on outcomes as a routine performance indicator within services

8. • conducting studies to clarify how non-pharmacological and pharmacological interventions for delirium management interact across populations.

MCQs

Select the single best option for each question stem

1. 1 Delirium:

   1. a occurs in approximately 1% of acute general hospital admissions

   2. b is routinely recorded within healthcare activity statistics

   3. c is usually caused by MRSA infection

   4. d is missed or misdiagnosed in approximately 50% of cases

   5. e can be readily identified by neuroimaging techniques.

2. 2 ‘Cognitive-friendly’ hospitals:

   1. a are the norm in high-income countries

   2. b can produce better clinical and economic outcomes

   3. c can be facilitated by faster patient throughput

   4. d are linked to longer durations of hospital stay

   5. e can be achieved with simple educational interventions.

3. 3 Dementia:

   1. a is uncommonly present in patients with delirium

   2. b is a more urgent clinical diagnosis than delirium

   3. c is characterised by more profound disturbance of awareness than delirium

   4. d always follows an episode of delirium

   5. e can undergo an accelerated course after an episode of delirium.

4. 4 In the treatment of delirium:

   1. a benzodiazepines are never indicated

   2. b NICE guidelines support the use of olanzapine and haloperidol where delirium is associated with risk or distress

   3. c environmental measures are ineffective

   4. d procholinergic agents are effective in prophylaxis

   5. e polypharmacy is usually warranted.

5. 5 In the aftermath of delirium:

   1. a patients usually cannot recall psychotic symptoms

   2. b recurrence is uncommon

   3. c long-term cognitive problems are rare

   4. d modifiable risk factors can be usually identified

   5. e antipsychotic agents should be continued for 3 months.

MCQ answers

1

d

2

b

3

e

4

b

5

d