The essential features of delirium include disturbances of consciousness, attention, cognition, and perception. The disturbance develops over a short period of time (usually hours to days) and tends to fluctuate during the course of the day. Following are the DSM-IV criteria for delirium (1):

• Disturbance of consciousness (i.e., reduced clarity of awareness of the environment) with reduced ability to focus, sustain, or shift attention.
• A change in cognition (such as memory deficit, disorientation, language disturbance) or the development of a perceptual disturbance that is not better accounted for by a preexisting, established, or evolving dementia.
• The disturbance develops over a short period of time (usually hours to days) and tends to fluctuate during the course of the day.

According to DSM-IV, delirium frequently represents a sudden and significant decline from a previous level of functioning and cannot be better accounted for by a preexisting or evolving dementia. There is usually evidence from the history, physical examination, or laboratory tests that the delirium is a direct physiological consequence of a general medical condition, substance intoxication or withdrawal, use of a medication, toxin exposure, or a combination of these factors. The disorders included in the DSM-IV delirium section have a common symptom presentation of a disturbance in consciousness and cognition but are differentiated by etiology:

• Delirium due to a general medical condition.
• Substance-induced delirium.
• Delirium due to multiple etiologies.
• Delirium not otherwise specified.

The disturbance in consciousness or arousal can be manifested by a reduced clarity or awareness of the environment that does not reach the level of stupor or coma. In addition, the ability to focus, sustain, or shift attention is frequently impaired and may result in the patient's being easily distracted.

There is also an accompanying decline in other areas of cognition. Cognitive deficits can include memory and visuoconstructional impairment, disorientation, or language disturbance. Memory impairment is most commonly evident in recent memory. Disorientation is usually manifested as disorientation to time (e.g., thinking it is morning in the middle of the night) or place (e.g., thinking one is at home rather than in the hospital). Disorientation to other persons occurs commonly, but disorientation to self is rare. It may be difficult for the clinician to fully assess cognitive function because the patient is inattentive and incoherent. Obtaining information from the medical chart, medical staff, and other informants, particularly family members, is often helpful in these circumstances.

Dysarthria is a frequent speech and language disturbance, and dysnomia (i.e., impaired ability to name objects), dysgraphia (i.e., impaired ability to write), or even frank aphasia may be observed.

Perceptual disturbances may include misinterpretations, illusions, or hallucinations. For example, the patient may see the nurse mixing intravenous solutions and conclude the nurse is trying to poison him or her (misinterpretation); the folds of the bedclothes may appear to be animate objects (illusion); or the patient may see a group of people around the bed when no one is actually there (hallucination). Although visual misperceptions and hallucinations are most common in delirium, auditory, tactile, gustatory, and olfactory misperceptions or hallucinations can also occur. Misperceptions range from simple and uniform to highly complex. A patient with delirium may have a delusional conviction of the reality of a hallucination and exhibit emotional and behavioral responses consistent with the hallucination's content.

Other commonly associated features of delirium include disturbances of sleep, psychomotor activity, and emotion. Disturbances in the sleep-wake cycle observed in delirium include daytime sleepiness, nighttime agitation, and disturbances in sleep continuity. In some cases, complete reversal of the night-day sleep-wake cycle or fragmentation of the circadian sleep-wake pattern can occur.

Delirium is often accompanied by disturbed psychomotor activity. Lipowski (2, 3) clinically described two subtypes of delirium based on psychomotor activity and arousal levels. These delirium subtypes included the "hyperactive" (or agitated, hyperalert) subtype and the "hypoactive" (lethargic, hypoalert) subtype. Others have included a "mixed" delirium subtype with alternating features of both. Ross et al. (4) suggested that the hyperactive form is more often characterized by hallucinations, delusions, agitation, and disorientation, while the hypoactive form is characterized by confusion and sedation and is less often accompanied by hallucinations, delusions, or illusions. Comparable levels of cognitive impairment have been observed with both motor subtypes.

The delirious individual may also exhibit emotional disturbances, such as anxiety, fear, depression, irritability, anger, euphoria, and apathy. There may be affective lability, with rapid and unpredictable shifts from one emotional state to another.

Depending on the etiology, delirium can be associated with a number of nonspecific neurological abnormalities, such as tremor, myoclonus, asterixis, and reflex or muscle tone changes. For example, nystagmus and ataxia may accompany delirium due to medication intoxications; cerebellar signs, myoclonus, and generalized hyperreflexia may be seen with lithium intoxication; cranial nerve palsies may occur with Wernicke's encephalopathy; and asterixis may be observed with renal or hepatic insufficiency. The background rhythm seen on EEG is typically abnormal, usually showing generalized slowing. However, in alcohol or sedative-hypnotic withdrawal, the EEG usually shows fast activity. In addition, laboratory findings that are characteristic of associated or etiological general medical conditions (or intoxication or withdrawal states) may be seen.

The differential diagnosis of patients with features of delirium is discussed in the delirium section of DSM-IV. The most common issue in differential diagnosis is whether the patient has dementia rather than delirium, has delirium alone, or has a delirium superimposed on a preexisting dementia. Cognitive disturbances, such as memory impairment, are common to both delirium and dementia; however, the patient with dementia usually is alert and does not have the disturbance of consciousness or arousal that is characteristic of delirium. The temporal onset of cognitive deficit symptoms and the temporal course and reversibility of cognitive impairments are helpful in distinguishing between delirium and dementia. The severity of delirium symptoms characteristically fluctuates during a 24-hour period, while dementia symptoms generally do not. Information from medical records, other caregivers, and family members may be helpful in determining whether a dementia was present before the onset of a delirium.

The prevalence of delirium in the hospitalized medically ill ranges from 10% to 30%. In the hospitalized elderly, the delirium prevalence ranges from 10% to 40% (2). As many as 25% of hospitalized cancer patients (5) and 30%–40% of hospitalized AIDS patients (6) develop delirium. As many as 51% of postoperative patients develop delirium (7), and up to 80% of patients with terminal illnesses develop delirium near death (8). Patients who have just had surgery, particularly cardiotomy, hip surgery, or a transplant, and patients with burns, dialysis, or central nervous system lesions are at increased risk for delirium.

Some patients manifest subclinical delirium or prodromal symptoms such as restlessness, anxiety, irritability, distractibility, or sleep disturbance in the days before the onset of overt delirium. Prodromal symptoms may progress to full-blown delirium over 1–3 days. The duration of symptoms of delirium has been reported to range from less than 1 week to more than 2 months (9–14). Typically the symptoms of delirium resolve within 10–12 days; however, up to 15% of patients with delirium have symptoms that persist for up to 30 days and beyond (10). Elderly patients with delirium may be more likely to have a prolonged course, with symptom durations frequently exceeding 1 month (11, 12).

While the majority of patients recover fully, delirium may progress to stupor, coma, seizures, or death, particularly if untreated. Full recovery is less likely in the elderly, with estimated rates of full recovery by the time of discharge varying from 4% to 40% (9, 15). Persistent cognitive deficits are also quite common in elderly patients recovering from delirium, although such deficits may be due to preexisting dementia that was not fully appreciated (9). Similarly, in a study of delirium in AIDS patients Fernandez et al. (16) found that only 27% had complete recovery of cognitive function, possibly because of underlying AIDS dementia.

Delirium in the medically ill is associated with significant morbidity. Medically ill patients, particularly the elderly, have a significantly increased risk of developing complications, such as pneumonia and decubitus ulcers, resulting in longer hospital stays (17, 18). In postoperative patients, delirium is a harbinger of limited recovery and poor long-term outcome. Patients who develop delirium, particularly after orthopedic surgery, are at increased risk for postoperative complications, longer postoperative recuperation periods, longer hospital stays, and long-term disability (19, 20). Seizures may occur in delirium, particularly among patients with alcohol or sedative-hypnotic withdrawal, cocaine intoxication, head trauma, hypoglycemia, strokes, or extensive burns (21).

Delirium in the medically ill is also associated with an increased mortality rate (22, 23). Elderly patients who develop delirium during a hospitalization have been estimated to have a 22%–76% chance of dying during that hospitalization (22, 24). Patients who develop delirium during a hospitalization also have a very high rate of death during the months following discharge. Several studies suggest that up to 25% of patients with delirium die within 6 months and that their mortality rate in the 3 months after diagnosis is 14 times as high as the mortality rate for patients with affective disorders (25, 26).

The disorders included in the delirium section of DSM-IV have a common symptom presentation but are differentiated according to presumed etiology (see Table 1 for a list of common etiologies).

1. Due to a general medical condition

In determining that delirium is due to a general medical condition, the clinician must first establish the presence of a general medical condition and then establish that the delirium is etiologically related. A careful and comprehensive assessment is necessary to make this judgment. A temporal association between the onset, exacerbation, or remission of the general medical condition and that of the delirium is a helpful guide. Evidence from the literature that suggests the condition in question can be directly associated with the development of delirium is also useful. Delirium can be associated with many different general medical conditions, each of which has characteristic physical examination and laboratory findings. When these are present they may help confirm the relationship between delirium and the general medical condition. General medical conditions commonly causing delirium are shown in Table 1.

2. Due to substance use or withdrawal

Delirium is frequently due to substance use or withdrawal (27). Substances with the potential to cause delirium include both agents that are not usually regarded as having psychoactive properties and those with established psychoactive properties. Delirium that occurs during substance intoxication may arise within minutes to hours after ingestion of high doses of drugs such as cocaine or hallucinogens; other drugs, such as alcohol, barbiturates, or meperidine, may cause delirium after intoxication is sustained for several days. During substance intoxication, the potential for additional agents with anticholinergic activity to cause delirium is increased. Usually the delirium resolves as the intoxication ends or within hours to days thereafter. Delirium associated with substance withdrawal develops as fluid and tissue concentrations of the substance decrease after reduction of sustained, high-dose use of certain substances. Substance-withdrawal delirium can also occur after the reduction of lower doses in patients having poor clearance, experiencing drug interactions, or taking combinations of drugs. The duration of the delirium usually varies with the half-life of the substance involved. Longer-acting substances usually are associated with less severe but more protracted withdrawal and may not have an onset of withdrawal symptoms for days or weeks after use of the substance is discontinued. Substance-withdrawal delirium may continue for only a few hours or may persist for as long as 2–4 weeks.

Table 2 lists substances associated with delirium, including substances of abuse, prescription medications, and toxins.

3. Due to multiple etiologies

Delirium, particularly in the critically ill and in elderly hospitalized patients, often has multiple etiologies (25). Francis and Kapoor (28) found that while 56% of elderly patients with delirium had a single definite or probable etiology for delirium, the remaining 44% had an average of 2.8 etiologies per patient.

4. Due to unspecified etiology

Occasionally, no clear etiology is immediately apparent. Often, unrecognized medication use or substance abuse is the cause of an intoxication or withdrawal delirium, and sometimes a rare cause of delirium, such as disseminated intravascular coagulation, is eventually revealed. There has been some controversy as to whether particular settings can themselves cause delirium (e.g., there has been speculation that the intensive care environment can cause "intensive care unit psychosis"). Koponen et al. (11) found a clear organic etiology in 87% of patients with delirium, and they found relatively little evidence that delirium was caused primarily by environmental factors.

Although standard psychiatric, general medical, and neurological histories and examinations are usually sufficient to diagnose and evaluate the severity of delirium, they can be supplemented by assessments using formal instruments. A large number of delirium assessment methods have been designed, some intended for clinical evaluations and others for research. Detailed reviews of the psychometric properties of instruments, as well as suggestions for choosing among instruments for particular clinical evaluations or research purposes, are available (29–31). Four types of instruments are briefly mentioned in the following sections: tests that screen for delirium symptoms, delirium diagnostic instruments, delirium symptom severity ratings, and some experimental laboratory tests.

1. Screening instruments

Several tools have been developed to screen for delirium symptoms among patients, and most have been designed to be administered by nursing staff. These may aid in the recognition of delirium, especially in nursing homes, where physician visits are less frequent. The number of delirium symptoms covered, the specificity of items for delirium, and the complexity of administration all vary. Screening instruments include the Clinical Assessment of Confusion–A (CAC-A) (32), the Confusion Rating Scale (CRS) (33), the MCV Nursing Delirium Rating Scale (MCV-NDRS) (34), and the NEECHAM Confusion Scale (35).

2. Diagnostic instruments

Investigators have designed a variety of instruments to make a formal diagnosis of delirium. These instruments consist of operationalized delirium criteria from a variety of diagnostic systems, often in the form of a checklist incorporating information from patient observation and the medical record (e.g., DSM-III-R, DSM-IV, ICD-9, and ICD-10). The rate of delirium diagnosis obtained by using these diagnostic instruments varies according to both the diagnostic system that was used and the particular way in which the authors chose to operationalize the criteria. One structured diagnostic interview schedule, the Delirium Symptom Interview (DSI), can be administered by lay interviewers and used in epidemiological studies (36). Other delirium diagnostic instruments include the Confusion Assessment Method (CAM) (37), Delirium Scale (Dscale) (38), Global Accessibility Rating Scale (GARS) (39), Organic Brain Syndrome Scale (OBS) (40), and Saskatoon Delirium Checklist (SDC) (41).

3. Delirium symptom severity rating scales

Several instruments have been developed to rate the severity of delirium symptoms. Ratings are generally based both on behavioral symptoms and on confusion and cognitive impairment. Rating the severity of delirium over time may be useful for monitoring the effect of an intervention or plotting the course of a delirium over time. These scales have also been used to make the diagnosis of delirium by considering patients with scores above a specified cutoff to have the diagnosis. Such rating scales include the Delirium Rating Scale (DRS) (42) and the Memorial Delirium Assessment Scale (MDAS) (43).

4. Laboratory tests

Several laboratory evaluations have been investigated for possible use in evaluating delirium. With the exception of the EEG, these tests are experimental and currently appropriate only for research purposes. For several decades, investigators have observed EEG changes in patients with delirium (44). EEG changes consist mainly of generalized slowing, although low-voltage fast activity is seen in some types of delirium, such as delirium tremens (45). The presence of EEG abnormalities has fairly good sensitivity for delirium (in one study, the sensitivity was found to be 75%), but their absence does not rule out the diagnosis; thus, the EEG is no substitute for careful clinical observation. Among the experimental laboratory tests that have been investigated for use in delirium, those that appear to show some promise include brain imaging (46, 47) and measures of serum anticholinergic activity (48).