What are hallucinations?

/Home/Alzheimer's Symptoms/Alzheimer's and Hallucinations

1. WHAT ARE HALLUCINATIONS?

Hallucinations have been defined as, "A sensory experience which occurs in the absence of corresponding external stimulation of the relevant sensory organ, has a sufficient sense of reality to resemble a veridical perception, over which the subject does not feel s/he has direct and voluntary control, and which occurs in the awake state." (David AS. Cogn Neuropsychiatry 2004;9:107.)

As far as caregivers are concerned, that typically translates into a loved one seeing and/or hearing something that simply isn't there. Hallucinations differ from misidentifications or misinterpretations of something that actually is there, such as the loved one thinking that his reflection in a mirror is a real person, or that people on TV are real.

Hallucinations may involve any of the senses, including visual, auditory, gustatory, tactile, and olfactory. However, visual and auditory hallucinations are by far the most common. Hallucinations may be unimodal, i.e., involve only one of the senses; or they may be multimodal, involving two or more senses simultaneously.

Patients with many different disorders can experience hallucinations, including schizophrenia and schizoaffective disorder, other psychiatric illnesses (including major depressive disorder, bipolar disorder, affective disorders, dissociative disorders, borderline personality disorder, post-traumatic stress disorder, multiple personality disorder, and post-partum psychosis), lesions of the brainstem, certain sleep disorders, sensory impairment, delirium, seizures of all sorts, migraines, tumors, and some inborn errors of metabolism (e.g., homocysteine remethylation defects, urea cycle defects, GM2 gangliosidosis, Neimann-Pick disease type C, and α-mannosidosis.)

Many -- but not all -- of the neurodegenerative disorders that can cause dementia, including Alzheimer's disease, posterior cortical atrophy, Lewy body dementia, Parkinson's disease, vascular dementia, and Creutzfeldt-Jakob disease, can cause hallucinations. Interestingly, hallucinations are relatively rare in some types of dementia, such as behavioral variant frontotemporal lobe dementia (bvFTD) and progressive supranuclear palsy.

Even perfectly healthy people can hallucinate. For example, after the death of a spouse, one-third to one-half of bereaved spouses report hallucinations of the deceased. It should be noted, however, that it is not known whether hallucinations are generated by similar mechanisms in healthy people as in any of the many disorders that can cause them.

Historically, hallucinations have been lumped together with delusions under the umbrella of "psychosis". From the time of Alois Alzheimer’s first description of psychotic symptoms in a patient with "presenile dementia" in 1907, psychosis has been recognized as a major clinical syndrome in this illness. Psychotic symptoms have been found to be a significant predictor of functional decline and institutionalization. Psychotic Alzheimer’s patients are also more likely to have worse general health, and tend to have more frequent and problematic behaviors, including agitation, verbal and physical aggression, and anxiety.

Lumping hallucinations in with delusions under the umbrella of "psychosis" is very unfortunate, however, since more recent studies have shown that there are very significant differences between the two, including:

- Prevalence. (The median prevalence of delusions in 55 studies on Alzheimer's patients was 36%, whereas visual hallucinations were seen in 18.7% and auditory hallucinations in 9.2% of Alzheimer's patients.)

- Associated risk factors. (Hallucinations and misidentifications correlate with cognitive impairment, whereas persecutory delusions relate more closely to mood disturbances or other factors, such as delirium; and delusions, but not hallucinations, have been found to be closely associated with depression.)

- Response to treatments. (E.g., hallucinations were significantly reduced by treatment with tacrine, whereas delusions were unchanged.)

- The neurotransmission systems involved. (E.g., muscarinic M1 receptors are only increased in Lewy body dementia patients with delusions, whereas M4 receptors are only increased in LBD patients with visual hallucinations.)

- Clinical findings. (Plaques and tangles staging have been reported to be associated with visual hallucinations but not with delusions in LBD patients.)

- Rate of decline. (The rate of cognitive decline is significantly associated with hallucinations, but not delusions, in Alzheimer's patients.)

Therefore, to the extent possible, only studies that clearly differentiated between hallucinations and delusions have been considered in writing this article.

Hallucinations are common in Alzheimer disease. Studies have found that their presence is associated with increased risk for cognitive decline (especially if both visual and auditory hallucinations are present), functional decline, and institutionalization. One large, recent study also found that the presence of hallucinations was associated with an ~1.5-times higher risk of death. The authors speculated that hallucinations may reflect higher burden or more biologically detrimental localization of neuropathology; and may also be associated with more risk-taking behavior or lower attention to medical problems that could eventually lead to mortality. Another study also concluded that hallucinations, especially visual ones, were associated with increased mortality, and that the association with mortality increased with higher level of education. Delusions and misperceptions were not strongly related to cognitive decline or mortality.


1.1. Different types of hallucinations

Fénelon (2008) noted: "Hallucinations cannot be directly studied or quantified. The examiner is dependent on what the patient or his/her caregiver reports. When the patient is cognitively impaired, one may assume that only striking hallucinations, or those associated with behavioral disorders, will be recorded by the caregiver." In addition, the scales for evaluating the prevalence and/or phenomenology of hallucinations, even in patients who do not have dementia (e.g., Parkinson's disease patients) have been criticized as not having "the basic content, or mechanistic and psychometric properties that seemed necessary for evaluating" hallucinations. Ergo, keep in mind that much of what has been reported may be deeply flawed.

Since most studies of hallucinations in dementia patients have focused on visual hallucinations, they will be the primary focus of this article. I suspect that other types of hallucinations may also be common in dementia patients, but since they may be more difficult to discern (how do you tell if your loved one is experiencing an olfactory hallucination?) and/or may have minimal impact on the caregiver, are significantly under-reported.

Visual hallucinations, simply put, are seeing something that isn't there. They may range from "simple" visual hallucinations, e.g., small, brightly colored spots or shapes that flash, may suddenly change shape, and/or move across the field of vision, to "complex" visual hallucinations that may include colorful, vivid scenes with people, animals, and other images.

As noted earlier, patients with a very wide range of disorders can experience visual hallucinations. Because these patient populations differ (e.g., in terms of the presence of sensory deficits, brain anomalies, environmental factors, traumatic events, genetic factors, etc.), it is not surprising that the phenomenological characteristics of their hallucinations also vary.

Simple hallucinations far outnumber complex ones in patients who have restricted visual input, e.g., from acquired eye disease, occipital stroke (or any other visual-pathway lesion), and sensory deprivation. Conversely, complex hallucinations far outnumber simple ones in Parkinson’s disease, certain types of dementia (e.g., Lewy body dementia, Parkinson’s disease with dementia, Alzheimer's disease, and vascular dementia), and delirium.

Several researchers have suggested that this "double dissociation" between simple and complex hallucinations means that the causes are different for each. The focus here will be complex visual hallucinations.

Even when focusing on complex visual hallucinations, there are differences between the hallucinations produced by different disorders. For example, complex hallucinations such as figures in patients with eye disease tend to be bizarre and unfamiliar, often wearing elaborate costumes and hats. In contrast, the figures hallucinated in Parkinson’s disease and the dementias tend to be mundane and familiar.

Manford and Andermann (1998), describe complex visual hallucinations this way: "The content is dramatic; they are usually in vivid colors… Sufferers often see human figures or faces (sometimes torsos without heads or vice versa); animals (real or bizarre), sometimes in miniature (Lilliputian hallucinations) or scenery of outstanding beauty. The images may be static or moving and their detail and vividness seem to exceed anything experienced in real life. …Modalities other than vision may be involved, especially faces may talk, but this is relatively uncommon. The emotional reaction to the images may be neutral, pleasant fascination, or extreme fear." It should be noted that this description is a general one, of complex hallucinations experienced by patients with a variety of disorders. It does not necessarily hold up when talking about dementia patients (see below).

Manford and Andermann describe a number of case reports, from patients with different disorders causing their hallucinations (which make for fascinating reading). For example, a patient with Parkinson's disease developed "bimodal" hallucinations (i.e., hallucinations involving two senses, in this case, vision and touch): "…The furniture appeared to be skewed and for 30 min there seemed to be numerous cats in the room… They were black or brown and moved silently around the room. One jumped onto his knee and he was able to stroke it; he later recalled the sensation of its fur. The same hallucination recurred almost daily, at any time of day, but it was most common in the early morning or evening. During subsequent similar episodes, his wife noted that he appeared to be petting something at his feet. This hallucination was associated with reduced awareness and responsiveness, but not unconsciousness, and usually lasted 1-3 h."

Collerton et al (2005) note that, in dementia patients, bizarre or incongruous hallucinations are "less common than commonplace images -- a man sitting on a chair or a dog in the corner of the room, for example. Hallucinations of people tend to be more common than are those of animals. Images of objects such as tables or cars are the least frequent. Unrecognised images are hallucinated as frequently as or more so than familiar ones. There is often a stereotyped or repetitive quality to the images. Commonly, the same image repeats itself on different occasions, though over time, many patients will experience a range of hallucinatory images. There is generally movement, although this is often stereotyped and restricted. The images rarely interact with or respond to the environment. The image is usually whole and sharply focused. It is normal sized or unusually small with a normal or unusually vivid colour. If there is distortion, this is usually of the face with a consistent exaggeration of the mouth and eye areas."

Hallucinatory images occur in the focus of the visual field of a dementia patient, and are usually seen against the background of the existing visual scene. In contrast, panoramic hallucinations (images of people and background filling the entire visual field) are sometimes reported by patients with eye disease. The reason for the latter may be that when a person has no effective vision, there is no existing visual scene to act as the background to a nonpanoramic hallucination.


According to Collerton et al, typical visual hallucinations last for minutes, rather than seconds or hours. (Not everyone agrees with this. Manford and Andermann, for example, cite cases where complex visual hallucinations lasted a few hours. My husband -- who had postmortem biopsy-confirmed Alzheimer's -- was known to hallucinate for hours at a time, as well.) Onset is abrupt, with no apparent trigger. The hallucinations disappear just as suddenly, sometimes due to changes in the visual environment, but often due to no apparent cause. They are more rare when the loved one's eyes are closed (and so asking the loved one to close his/her eyes can often stop the hallucinating). Visual hallucinations are often associated with low light; and so, keeping a room well-lighted can help prevent visual hallucinations.

A given dementia patient may hallucinate consistently at the same time of day, although most hallucination episodes are not predictable. They do, however, tend to occur in the same location, mostly in the house or looking out of the house. For example, dementia patients often report visitors who only appear in their living room. This may be partially due to the amount of time the patient spends in each location. However, it is striking that once the patient moves to a new environment, the hallucinations usually disappear.

Collerton et al (2005) state that the image usually appears in a contextually correct location (a person who is sitting in a chair rather than floating on the ceiling) and with the correct orientation (an upright rather than inverted face, for example). Others agree that the orientation is usually correct, but location is not. In their commentary on the Collerton et al paper, Morrison and David state that "though the content of [complex visual hallucinations] tends to be of people and animals, they are more often unfamiliar and just as often appear in inappropriate positions or contexts as they do in appropriate ones." They cite many papers in which hallucinations were reported as being in inappropriate positions (people/animals on the wall or ceiling, people floating outside the window, children under the bed), inappropriate size (Lilliputian figures, “pixies” running along the window, “miniature policemen guiding a midget villain”), or inappropriate context (birds flying in a hospital, a dragon, an angel). One paper concluded that only 22% of hallucinations “fit in well” with the environment. Again, my husband sided with Morrison and David; he often chatted with visitors who were perched on top of a bureau or floating near the ceiling, and once startled me by telling me there was a horse behind me ... when I was in the bedroom with him, just a foot or two away from the wall. He was known to give lectures -- including question and answer sessions, in which he clearly heard his audience's questions -- to apparently large audiences (judging from the way he projected his voice, and had to ask people in the back to speak up), while lying in bed in the middle of the night. He was aware of me, in bed beside him, and did not find my presence there incongruous. In fact, he sometimes tried to draw me into the conversations, or asked me to read his vugraphs, which were apparently being projected onto the ceiling or the wall behind the audience.

1.2. Hallucinations versus illusions

Hallucinations occur in the absence of external stimuli. In contrast, "illusions" or "misperceptions" or "misidentifications" are incorrect perceptions of external stimuli.

A classic example of a visual illusion in Alzheimer's patients is "the man in the mirror", i.e., the loved one sees his own reflection and interprets it as another person being in the room with him. Illusions may also involve distortions in the shape (metamorphopsia), size (micropsia), and axis (tilt) of the stimulus, while the stimulus itself is correctly recognized.

Some situations may not be readily identified as being an illusion rather than an hallucination, e.g., when patterns on walls or cushions metamorphose into faces. It may be particularly difficult to determine which symptom a dementia patient is exhibiting, since they often have difficulty expressing themselves and/or have visual agnosia. (Visual agnosia is an extremely common symptom, especially in Alzheimer's. Recent studies indicate it may develop even before short-term memory loss, expressed as visuospatial problems. In loved ones with visual agnosia, the eyes function normally, but the brain interprets the signals incorrectly.)


2. WHAT CAUSES HALLUCINATIONS?

Many studies utilizing many different techniques have been conducted in an effort to determine what, exactly, causes hallucinations in dementia patients. Current consensus is that neuropathological factors alone are unlikely to be sufficient to explain them. Instead, many diverse factors, such as environmental changes, neurochemical abnormalities, past psychiatric history (including premorbid personality), social history (e.g., intellectual achievement and life-long learning), family history, and genetic susceptibility, among others, may influence their pathogenesis. And, as mentioned earlier, there is a distinct possibility that different factors are involved in causing hallucinations in different types of disorders. Even different types of visual hallucinations may have different causes in the same patient.

For example, in Parkinson's patients, the mechanisms that may be involved in causing visual hallucinations include central dopaminergic overactivity and an imbalance with cholinergic neurotransmission (caused by the disease itself and/or exacerbated by the use of dopaminergic and/or anticholinergic drugs); dysfunction of the visual pathways, including specific Parkinson's-associated retinopathy and functional alterations of the extrastriate visual pathways; alterations of brainstem sleep-wake and dream regulation; and impaired attentional focus. Other factors, often noted in epidemiological studies, that may play a role include clinical risk factors such as age, the duration and severity of the Parkinson's disease, depression, cognitive impairment and dementia, and sleep disturbances. Parkinson's patients who hallucinate exhibit worse cognitive function overall, including executive, attentional, and visuospatial abilities, and also have greater sleep disturbances and REM behavior disorder. Neuroimaging studies also suggest that the brains of Parkinson's patients who hallucinate have distinct structural, functional, and metabolic abnormalities than those of patients who do not.

It has long been thought that the drugs used to treat Parkinson's disease itself were the cause of visual hallucinations. However, this hypothesis is becoming controversial, since more recent analyses of the data indicate that (a) Parkinson's patients developed hallucinations before levodopa (L-dopa) was introduced, and (b) increasing the use of this drug has not been found to increase the prevalence of the symptom. Some studies suggest that dopamine agonists are more likely than levodopa to make a patient susceptible to developing hallucinations; and that anticholinergics may also make them more susceptible, especially elderly Parkinson's patients. It has been suggested that dopaminergic hyperfunction plays a role if and only if cholinergic hypoactivity is already present.


In Alzheimer's, visual hallucinations and misidentifications generally correlate with the severity of cognitive impairment. Functional studies suggest that visual hallucinations are associated with decreased function of cortical networks rather than localizing the origin of visual hallucinations to individual specific brain regions. A certain degree of deterioration is required for hallucinations to appear; then, once a certain amount of additional deterioration has occurred in a brain region where the hallucination is originating, the visual hallucinations will end. Parkinsonism and exaggerated cognitive decline are significant predictors of visual hallucinations in Alzheimer’s patients, with prospective, longitudinal studies showing that the presence of cortical Lewy bodies is associated with more persistent and severe hallucinations, independently of any association with severity of cognitive decline.

Other findings suggest that disturbances in neurotransmission systems contribute to the development of hallucinations. For example, pharmacologic, biochemical, and structural observations implicate cholinergic deficiencies. One strong argument in support of the cholinergic dysfunction playing a role (in Lewy body dementia and Parkinson's, as well as in Alzheimer's) is that hallucinations may be reduced or eliminated if the loved one is given a cholinesterase inhibitor such as Aricept/donepezil, Razadyne/galantamine, or Exelon/rivastigmine. A neurochemical study in patients with Lewy body dementia showed that the activity of acetyltransferase (the enzyme that synthesizes acetylcholine) in the parietal and temporal cortex of patients with visual hallucinations was less than 20% of healthy control values, whereas patients who did not experience visual hallucinations had values around 50% of the normal range. Adrenergic and serotoninergic systems may also play a role. One study found that dopamine receptor (D3) density was selectively increased in Alzheimer's disease patients with a history of psychosis, independent of the presence or absence of Lewy bodies.

Visual acuity also appears to play a key role, with impaired vision due to cataracts being highly predictive of visual hallucinations in Alzheimer's patients, and optical correction and cataract surgery being the most helpful interventions. Poor eyesight contributes to the severity, but not the persistence, of the hallucinations.

In addition to more severe cognitive impairment, the risk factors for Alzheimer's patients developing hallucinations include a more rapid cognitive decline, and, although there have been only a limited number of studies to date, African American or black ethnicity. Some studies found a significant association with age, age at onset of Alzheimer's disease, and illness duration. Gender, education, and family history of dementia or psychiatric illness showed weak or inconsistent relationships with hallucinations.

Interestingly, one paper commented that, "visual illusions and hallucinations are most common in delirium and [Lewy body dementia], compared with Alzheimer’s disease where careful interviewing is needed to differentiate complex visual hallucinations from delusions or confabulations." I originally thought that my husband was exhibiting bimodal (visual and auditory) hallucinations that could sometimes last for hours. Certainly, he was both seeing things that weren't there and hearing sounds when there were none. I observed him carrying on spirited conversations with thin air on dozens of occasions. However, the more I read about hallucinations, the more I wonder if he was actually "re-living" very vivid memories, e.g., was caught up in spontaneous confabulations, instead. And yes, some researchers hypothesize that hallucinations -- like confabulations -- are caused, at least in part, by faulty memory retrieval; I have seen discussions on, e.g., "hallucinations of childhood memories." Dominic ffytche (discussions following Collerton, Perry, and McKeith, 2005) argues convincingly of the need to split visual hallucinations into two separate syndromes, instead of considering all complex visual hallucinations as belonging to a single pathophysiological entity.


2.1. Causes Unrelated to Dementia

The study, diagnosis, and treatment of hallucinations is often complicated by the fact that dementia patients may develop hallucinations due to causes that are not directly related to their dementia.

For example, many different drugs induce hallucinations, either during their use or upon their withdrawal, especially in the elderly demented. Often, the mechanism is by inducing delirium. Drugs that carry a high risk of causing delirium with hallucinations include benzodiazepines (diazepam, temazepam, chlordiazepoxide), antidepressants (amitriptyline, doxepin, trazodone), antiparkinsonian drugs (levodopa, bromocriptine, selegeline, trihexyphenidyl, orphenadrine, pergolide), analgesics (opiates, aspirin, NSAIDs -- especially indometacin), lithium, and steroids.

Drugs that pose a low or moderate risk include antihypertensives (α-blockers, β-blockers, ACE inhibitors, calcium channel blockers, diuretics), antiarrhythmics (digoxin, amiodarone, disopyramide, lignocaine), antipsychotics (chlorpromazine, trifluoperazine, haloperidol), anticonvulsants (primidone, phenytoin, carbamazepine, valproate), anticholinergic antihistamines and antispasmodics including over-the-counter medicines (chlorpheniramine, atropine), histamine blockers (cimetidine, ranitidine, amotidine), respiratory drugs (aminophylline), and all antibiotics.

Oral hypoglycaemic agents, such as tolbutamide and glibenclamide, may cause hypoglycaemia and hyponatraemia, both of which can cause delirium. Since delirium frequently occurs in malignant disease, it is difficult to attribute symptoms to the antineoplastic drugs used to treat it. However, some of these drugs have been associated with confusion more often than others, e.g., methotrexate, vinca, alkaloids, fluorouracil, altretamine,, asparginase, procarbazine, carmustine, dacabazine, and interferon-α.

Sensory deprivation may cause hallucinations. One of the most widely studied examples is Charles Bonnet syndrome (CBS), which is the occurrence of visual hallucinations in the visually impaired. Any cause of visual impairment (including macular degeneration, glaucoma, cataracts, cerebrovascular disease, and tumors) may be associated with CBS. The visual hallucinations in CBS have been described as clear and detailed; they often involve Lilliputian images of people, faces, animals, and inanimate objects. Complex hallucinations of figures tend to be bizarre and unfamiliar, often wearing elaborate costumes and hats (unlike the mundane and familiar figures reported by dementia patients.) They tend to be short in duration (seconds or minutes), silent, and occur more in dim light. CBS patients may stop the hallucinations by closing or moving their eyes or by increasing the light. While patients may initially be unaware that these images are indeed hallucinations, one of the hallmarks of CBS is the eventual development of insight regarding them. While CBS can, and often does, develop in cognitively normal people, there is growing evidence that preexisting dementia or cognitive impairment may predispose a patient to the development of CBS. The strongest risk factors for CBS include bilateral visual impairment, declining visual acuity, cerebral damage, cognitive defects, social isolation, and sensory deprivation. The majority of visual hallucinations resolve spontaneously or with treatment of the underlying visual disorder (e.g., removal of cataracts), but some cases are chronic.

Similarly, people with auditory impairment may develop musical hallucinations, either choral or instrumental. All musical hallucinations reported in the literature had onset after age 60. In contrast to visual hallucinations, musical hallucinations are continuous rather than lasting just a short while, and tend to appear in the setting of chronic deafness rather than during a sudden change in hearing. All reported cases of musical hallucinations were chronic.

Dementia patients may develop delirium; and this superimposed delirium may be the actual cause of their hallucinations. Delirium may be caused by physical illness, even one that is mild, or any process which interferes with the normal metabolism or function of the brain. For example, electric shock, fever, pain, poisons (including carbon monoxide, and toxic drug reactions), brain injury, hypoxia, anoxia, surgery, traumatic shock, withdrawal symptoms of certain drug and alcohol dependent states, too little food or water or sleep, and even physical restraints, immobility, and a stressful environment such as a hospital environment are all known to cause delirium. Moreover, there is an interaction between acute and chronic symptoms of brain dysfunction; delirious states are more easily produced in people already suffering with underlying chronic brain dysfunction such as a dementia. The infamous urinary tract infection (UTI), which is so well-known for causing sudden changes in the symptoms of dementia patients, does so by inducing delirium. Tactile hallucinations and formication (a sensation that resembles that of insects crawling on or under the skin) are highly suggestive of delirium.

It should be noted that Alzheimer's patients can develop any type of seizure, at any point in the progression. Seizures are well-known for causing hallucinations. By far, the best-studied hallucinations are those associated with epilepsy. I won't touch on seizure-related hallucinations further in this blog, but for those who are interested in learning more on hallucinations in epilepsy, see the excellent, two-part review by Elliott, Joyce and Shorvon 2009.


3. HOW COMMON ARE HALLUCINATIONS IN DEMENTIA PATIENTS?

Studies of hallucinations in dementia patients are often incomplete, inaccurate, and/or difficult to compare. First of all, the findings largely reflect the tools that are used for assessing the symptom. Some studies rely on caregiver reports to identify patients experiencing hallucinations; others rely on patient interviews. Few studies on dementia patients report any characteristics of the hallucinations other than prevalence or Neuropsychiatric Inventory score. In addition, the groups of patients in a particular study may be heterogeneous, e.g., for disease severity and clinical expression. Misdiagnosis is a particularly serious problem. The use of standard diagnostic criteria for patient selection does not guarantee a correct diagnosis; however, few studies have taken the precaution of confirming the diagnoses via autopsy. For example, patients who have Lewy body dementia are often misdiagnosed as having Alzheimer's, and so their inclusion in Alzheimer's groups is likely to produce an over-estimation of the frequency of hallucinations in Alzheimer's dementia. This is particularly true of earlier studies, conducted before the existence of Lewy body dementia and the criteria for its diagnosis were established in the late 1990s.

Nevertheless, it is reasonably well-established that the prevalence and incidence of hallucinations depends on the type(s) of disorder(s) that the dementia patient has.


3.1. Alzheimer's disease

A 2005 review of 55 studies, including a total of 9,749 subjects, concluded that 18.7% of Alzheimer's patients develop visual hallucinations and 9.2% develop auditory hallucinations at some point in the progression of the disease. (Similar rates of visual and auditory hallucinations have been observed in subsequent studies.) A higher prevalence tended to occur in inpatient settings (e.g., acute care hospitals, nursing homes, neurobehavioral units), since hallucinations increase the probability that an Alzheimer's patient may be placed, whereas lower rates were noted in patients referred to outpatient memory or research clinics.

Visual and auditory hallucinations are most common, but somatic, olfactory, and tactile hallucinations have been reported. For example, data from a large, multicenter study that followed patients with probable Alzheimer's from the early stages of the disease for up to 14 years found that, overall, 7% had hallucinations at baseline: 2.4% auditory, 2% visual, 2% olfactory, 0.4% tactile, and 0.7% "other". Among the 130 patients (33%) who had hallucinations at any time during the follow-up, 23.8% had visual, 18.7% auditory, 4.4% olfactory, 3.3% tactile, and 2.6% "other".

The visual hallucinations that are reported in Alzheimer's are most often of people (88.9%), including images of familiar people and dead relatives; then of animals (44.4%) and least commonly of objects such as machinery (27.8%).

Hallucinations are often reported as typically involving only a single sensory mode; however, one study found that 54% of Alzheimer's patients with visual hallucinations also experience an auditory component.

Roughly one-third of Alzheimer's patients who hallucinate do so on a daily basis, whereas the remaining two-thirds hallucinate less than daily but more than once per month.

Alzheimer's patients can manifest hallucinations at any stage of the disease, although they are most common in the moderate stages. In a given individual, symptoms typically persist for several months, but rarely persist for more than a year. As noted earlier, this suggests that while some deterioration is needed for development of visual hallucinations, severe deterioration can no longer support their existence. On the other hand, since some hallucinations do occur in earlier or later stages, and since some patients never develop hallucinations despite passing through moderate stages of illness, there must be other risk factors associated with their development.

Note that more recent studies have concluded there is a "Lewy body variant" of Alzheimer's, also known as Alzheimer's with Lewy body pathology (AD-LBP). Some estimates are as high as half of all Alzheimer's patients have AD-LBP. Studies of the data from autopsy-confirmed cases of LBD, AD, and AD-LBP indicate that Lewy body dementia patients differed from Alzheimer's at initial evaluation, with more frequent hallucinations and delusions, EPSs, and seizures, and longitudinally in neuroleptic sensitivity, but the data did not distinguish AD-LBP from Lewy body dementia.


3.2. Posterior cortical atrophy (PCA)

Posterior cortical atrophy (PCA), also known as Benson's syndrome, is a rare degenerative condition in which damage is initially focused at the back (posterior) region of the brain. In the vast majority of cases, the cause of PCA is Alzheimer's disease, and PCA is often considered to be an atypical variant of Alzheimer's. In rare cases, a few other diseases, including Lewy body dementia and Creutzfeldt–Jakob disease, can cause similar symptoms, at least initially. The first symptoms of PCA tend to occur when people are in their mid-50s or early 60s. However, the first signs are often subtle and so it may be some time before a formal diagnosis is made.

Despite being caused by the same disease process, the initial effects of PCA and typical Alzheimer’s disease are very different. Typical Alzheimer’s disease is most commonly associated with deterioration in memory, language, perception and a host of other skills and abilities. By contrast, people with PCA tend to have relatively well-preserved memory and language, but show a progressive, dramatic and relatively selective decline in vision and/or literacy skills such as spelling, writing and arithmetic; the ability to make skilled movements may also be affected. As damage in the brain spreads and the PCA progresses, people develop the typical symptoms of Alzheimer's disease, such as memory loss and confusion.

Visual hallucinations have been reported to occur in up to 25% of patients who meet the criteria for PCA. The hallucinations are very similar to the type encountered in patients with Lewy body dementia, being well formed, recurrent, spontaneous, and non-fleeting. They occur at least one year after PCA symptom onset, and in many cases are present for many years. Visual hallucinations in PCA patients are associated with parkinsonism, rapid eye movement sleep behavior disorder (RBD), and myoclonic jerks. One study found that the onset of the hallucinations occurred on average four years after the onset of the typical symptoms of PCA, as did the onset of parkinsonism (four years), RBD (three years) and myoclonic jerks (two years). The parkinsonism was less than or equal to a modified Hoehn and Yahr stage of 1.5 (i.e., unilateral plus axial involvement).


3.3. Lewy body dementia

Lewy body dementia is often misdiagnosed as probable Alzheimer's disease. Two of the differences between the two diseases are that patients with Lewy body are much more likely to hallucinate ... and they are often very sensitive to the antipsychotics that are often prescribed for treating hallucinations.

Visual hallucinations occur in roughly 25% of patients with mild-stage Lewy body dementia, in comparison with less than 5% of patients with mild-stage Alzheimer's. The prevalence of hallucinations over the course of Lewy body dementia have been estimated to be as low as 46 to as high as 93%, the latter having been reported in studies that involved autopsy-confirmed diagnoses. Although visual hallucinations are the most frequent, auditory, olfactory and tactile hallucinations are also reported. Auditory hallucinations may occur in 20% of Lewy body patients over the course of their illness.

The visual hallucinations in Lewy body dementia are frequently complex, consisting of colorful, three-dimensional images of anonymous people/soldiers, friends and family, children/babies, body parts, animals, and machines. Inanimate objects are also seen and abstract perceptions such as writing on walls and ceiling are not unusual. They may involve seeing objects move when they are actually still, and seeing complex scenarios of people and items that are not present. These visual hallucinations typically occur at least once a day, and last for minutes; and tend to persist and be stable over time. The loved one's emotional response may vary through amusement, indifference, and fear, although hallucinations are usually not perceived as threatening. Some degree of insight into their unreal nature is often either present during their occurrence or gained afterward.

In comparison with Alzheimer's patients, Lewy body patients are more likely to hear their visual hallucinations speak, but studies have found no significant differences in the other phenomenological characteristics, including whether the hallucinations moved, the time of day that they were experienced, their size, the degree of insight, and whether they were complete.

There is considerable overlap between true visual hallucinatory symptoms (i.e., in the absence of an adequate external stimulus) and other perceptual disorders, including misidentification syndromes and visual agnosias. Lewy body patients may describe such experiences occurring simultaneously, for example, seeing faces emerging out of the patterns on chair cushions or hidden among trees and flowers at the same time as figures are observed against a blank background.


3.4. Parkinson's disease and Parkinson's disease dementia

Hallucinations have been studied more extensively in patients with Parkinson's than in patients with other diseases that can cause dementia, presumably because they develop early enough in the course of the Parkinson's that the patients are able to describe them clearly and accurately.

Visual hallucinations occur in approximately 25% of patients with Parkinson's disease, and in up to half of patients with Parkinson's disease dementia. These hallucinations are similar to those in patients with Lewy body dementia, and can range from seeing a person or animal to seeing more complex, formed, and mobile people, animals, or objects.

Visual hallucinations usually appear several years after disease onset and usually contain five or fewer images. Various studies report inanimate objects (leaves on the wall), animated figures (children playing), normal sized figures, Lilliputian (miniature) people and animals, images from daily life experience, and images from TV programs. Fragmented figures of beloved familiar persons or animals are often seen. Many studies report that the typical hallucination occurs while the patient is alert and with eyes open, in dim surroundings, often in the evening or when the patient is alone in a quiet environment. It is present for a few seconds and then suddenly vanishes.

Insight (i.e., knowledge of the non-real nature of the images) is common, but often accompanied by complex interaction with the hallucinations, with the Parkinson's patient inviting the hallucinatory presence to dinner or trying to push away his hallucinatory bedside companion. Interestingly, in one study, more than half the Parkinson's patients who had dementia maintained insight into the hallucinatory nature of the phenomenon.

Patients with significantly preserved cognitive functions report that the hallucinatory images disappear when they try to focus their vision on the object. Focusing attention on visual hallucinations in order to let them disappear is therefore a "coping strategy" useful for non-pharmacological management of the symptom.

Initially, hallucinations are neutral and non-threatening, and may be amusing or entertaining; and the patient retains insight. Without intervention, however, as insight declines, the content of the hallucinations may change to frightening images (e.g., insects, rats), inducing anxiety and panic attacks. Some studies report that loss of insight is unrelated to cognitive status.

In patients with cognitive decline, the incidence/prevalence of hallucinations is higher and hallucinations are also qualitatively different, involving complex interactions with hallucinatory perceptions with lack of insight of the unreal nature of perceptions. In patients with Parkinson's dementia, night-time hallucinations can assume the form of confusional states, in which the patient interacts with hallucinations and wandering behavior is common.

Several lines of evidence suggest that peripheral ocular and retinal dysfunction, as well as central dysfunction of visual processing can facilitate visual hallucinations in Parkinson's patients. Progressive deficits in color discrimination and contrast sensitivity are significantly worse in Parkinson's patients with than in those without visual hallucinations.


In one recent study that involved 70 Parkinson's patients, 31 of whom hallucinate, the typical hallucinatory experience was that of a non-threatening, familiar or unfamiliar solid, real sized, colored or colorless human body that appeared suddenly anytime during the day and disappeared a few seconds later. Little animals and insects were also common. A variety of other non-typical hallucinatory sensations were very frequently described. These researchers concluded that simplistic classifications such as benign-malignant, simple-complex or minor-major should be avoided until a scaling method has been developed and validated. They also found that hallucinations were unrelated to light cycle in the majority of patients.

Single mode hallucinations were reported in 20 patients (visual/14, auditory/4, olfactory/2), whereas multiple modalities were reported in 11 patients. Hallucinations lacking a visual component were reported in 7 patients. The most common hallucinatory experience was a whole person followed by small animals, insects and reptiles. Visual hallucinations were usually of normal size, with only 3 patients reporting size distortion. Eleven out of 24 patients experiencing visual hallucinations reported that their hallucinations were considered "achromatic". Only 2 reported transparent images.

Of the 31 patients, more than half (56%) experienced hallucinations once per week or more. Hallucinations were instantaneous (<1 sec) in 10 (32%), of medium duration (<10 sec) in 18 (58%) and of prolonged duration (>10 sec) in 1 patient.

Spontaneous remission of hallucinations was seen in 9 of the 31 (29%) of patients.

Although cognitive impairment and dementia have been frequently reported as risk factors for hallucinations in Parkinson's, two interview-based prospective studies did not find any association between MMSE scores and hallucinations.


The content of non-visual hallucinations may vary considerably. Auditory hallucinations may be elementary (ringing, knocks, etc) or, more often, complex. They typically consist of whispers that are neutral or sometimes incomprehensible, as opposed to the threatening voices reported in schizophrenia; however, some cases of threatening auditory hallucinations have been reported in Parkinson's. They may also be musical, in which case they are commonly associated with deafness. A 2000 paper on the study of 216 Parkinson's patients was one of the few that looked into auditory hallucinations in more detail. Auditory hallucinations occurred in 21 patients but were isolated (i.e. the hallucination consisted only of auditory phenomena) in only five cases (2.3% of the patients). In most cases, auditory hallucinations were combined with visual hallucinations, either at different moments or simultaneously (in seven cases), like the sound-track of the scene (e.g. hallucinated persons were heard speaking). They could also be combined with "presence hallucinations" (see below). Auditory hallucinations were verbal in 13 cases (62%), musical in three (14%) and consisted of various other noises (often the sound of steps) in 11 patients (48%). Deafness was present in two cases of musical hallucinations.

Tactile, olfactory and gustatory hallucinations are less common, and they tend to co-occur with visual hallucinations. Tactile hallucinations, for example, typically involve contact with small animals (although they do not generate the belief of infestation) or the feeling of being touched by someone.

Parkinson's hallucinations may also include "a false sense of presence" (i.e., a vivid sensation that people or animals are present, just out of eyesight); "passage" hallucinations that comprise brief visions of a person or animal passing to the side; or a feeling of floating. Passage hallucinations are visual phenomena, but presence hallucinations cannot be related to a specific sensory modality, although they were often described by the patients in terms referring to imagery (e.g. "the image is behind me"). In most cases, the sensation is precisely located, behind or to the side, or occasionally in another room. Visual illusions, consisting of seeing an inanimate object as a living being, are also frequently reported. For example, a branch of a tree is seen as a cat, a stone or a trunk as a face, or dust on the floor as an insect.

Most Parkinson's patients experiencing visual hallucinations also experience non-visual hallucinations and/or minor phenomena. One study found that older patients were more likely to have non-visual or mixed hallucinations than purely visual hallucinations, suggesting that age may influence the phenomenology.


3.5. Vascular dementia (VaD)

There are many different types of vascular dementia, distinguished by what has caused the damage in the brain -- such as stroke, small vessel disease, or amyloid angiopathy -- and which part of the brain has been damaged. Since the symptoms that a person experiences depend on the particular areas of the brain that are damaged, it is difficult to draw generalities about the prevalence and characteristics of specific symptoms such as hallucinations.

Despite the fact that it is generally agreed that patients with many types of vascular dementia may develop hallucinations, I found very little information about them in the literature. Most studies compared the prevalence of symptoms in VaD patients with those in Alzheimer's patients, and concluded that either there are no significant differences between the two cohorts, or the VaD cohort tends to have lower rates of psychoses.

Both visual and auditory hallucinations have been described for VaD patients. Visual are more common. On occasion, hallucinations are difficult to clearly distinguish from delusions, confabulations, and/or misperceptions.

The percentage of VaD patients reported as having experienced visual hallucinations has ranged from 14 to 60%. Some of the variability may be due to the composition of the VaD cohort, which may include differing percentages of patients with multi-infarct, strategic infarct, and/or subcortical dementia. Studies comparing different types of VaD have reported hallucinations more often in patients with small-vessel than with large-vessel VaD. Several studies found hallucinations to be more common in multi-infarct than subcortical vascular dementia. Some of the variability seen in the prevalence of hallucinations may be due to the severity of the dementia involved, although the evidence that hallucinations are more common in VaD patients with moderate or severe dementia is less consistent than it is in Alzheimer's patients. And some variability may be due to the setting in which the study was conducted; higher rates of psychoses are seen in Alzheimer's patients in inpatient settings (e.g., acute care hospitals, nursing homes, neurobehavioral units) than in patients referred to outpatient memory or research clinics, and one might suspect the same would hold true for VaD patients.


3.6. Creutzfeldt-Jakob disease (CJD)

Creutzfeldt-Jakob disease (CJD) is a very rare and very rapidly progressing disorder, and, therefore, little information is available about the symptoms, such as hallucinations, that may be seen. Some papers discuss the disorder in general, and report that visual hallucinations are relatively rare at onset. However, the "Heidenhain variant" is characterized by visual problems that develop within the first week after onset, often as the only symptom, and persist throughout the course of the disease. These problems can include disturbed perceptions of objects or colors (metamorphopsia, micropsia, and dyschromatopsia), optical hallucinations, visual field defects, visual agnosia, cortical blindness, and/or denial of visual loss (optical anosognosia, Anton syndrome).

Visual hallucinations may develop in as many as 50% of sporadic CJD patients throughout the course of the illness. They are usually composed of complex visual phenomena (many different forms, visions of people, images of animals, inanimate objects) and scenic hallucinations. Auditory hallucinations have also been reported but are much more rare; one study found that 6% of patients had them at onset, and no additional patients developed them over the course of the illness. A review of 126 patients evaluated at the Mayo Clinic stated that the patients showed a "diffuse range" of hallucinatory phenomena, including "vivid and interactive visual and auditory hallucinations."


3.7. Frontotemporal lobe dementia (FTD)

Our knowledge and understanding of dementias is rapidly changing. As of 2009, frontotemporal dementia (FTD) was divided into three main syndromic variants. Two of the three are associated with left-sided frontotemporal atrophy, and disorders of speech and language figure predominantly in these two. Hallucinations are rare to nonexistent in the left-sided FTDs.

Very little is known about the variant that is primarily associated with right temporal lobe atrophy (RTLA). The most prominent cognitive deficits were impairment of episodic memory and getting lost. Symptoms particular to the RTLA patient group included complex visual hallucinations of animate objects, and cross-modal sensory experiences in response to various sensory stimuli, each of which was seen in 10% of patients. The fearful aspect of the hallucinations recounted by the RTLA patients, involving visions of snakes and headless figures, contrasts with the typically undisturbing hallucinations associated with Lewy body dementia. It has been reported that reduced serum vitamin B12 may predispose people with FTD to develop hallucinations.

Some researchers report that hallucinations are more common in FTD patients with fused in sarcoma (FUS) protein pathology (18%), and in those who later develop amyotrophic lateral sclerosis (ALS). However, in one study comparing FTD patients with and without ALS, there were no significant differences in the number who exhibited hallucinations (5% of each). FTD/ALS patients had visual hallucinations, including insects, spies, witches, gorillas, demons, a lion, relatives (dead and living), and a boyfriend.

A case study of an unusual patient with TDP-43 FTD reported nonverbal auditory and olfactory hallucinations.


3.8. Progressive Supranuclear Palsy (PSP)

Progressive supranuclear palsy (PSP) is a rare brain disorder, belonging to the family of frontotemporal lobe dementias, that causes serious and progressive problems with control of gait and balance. Difficulty coordinating eye movement is the hallmark of PSP, the symptom that usually develops first, and the feature that is most useful in distinguishing PSP from similar neurodegenerative disorders (such as Parkinson's). Patients cannot coordinate their eyes to look up or down. Some patients experience this visual deficit as blurring. This progressive inability to coordinate eye movements is the cause of the physical instability.

Roughly 10% of PSP patients experience visual hallucinations. They are typically images of familiar people or fragmented faces, or animals, and occur during twilight or nighttime more often than not. There may be an auditory component. As in Parkinson's, the hallucinations may become frightening as their complexity increases and the disease progresses.


3.9. Mild cognitive impairment (MCI)

Hallucinations are very rare among patients with mild cognitive impairment (MCI). Several major studies reported prevalence rates of 0-4%. The Swedish Kungsholmen project was the only one to compare the frequency of psychosis in MCI patients and normal elderly people, and found no significant difference.

A 2008 study found a significantly higher prevalence of hallucinations and sleep disorders in patients with nonamnestic MCI than in patients with amnestic MCI, however, and concluded that such symptoms may be the earliest manifestation of different diseases, each one associated with different clinical profiles at the stage of MCI.



4. WHAT CAN BE DONE TO TREAT, MINIMIZE, OR COPE WITH HALLUCINATIONS?

First, the symptom must be verified as actually being hallucinations, rather than, e.g., confabulations, illusions, misinterpretations, or delusions. Different pharmacological and nonpharmalogical interventions are needed for preventing or treating these different types of symptoms.

Second, the impact of hallucinations on the patient needs to be considered. Some hallucinations are actually enjoyable or entertaining to the loved one. For example, talking to deceased family and friends may be comforting. Unless the hallucinations are distressing to the loved one, or leading him/her to do something dangerous, drugs for treating them are usually not warranted. If caregivers are disturbed, they should consider what the appropriate goals of an intervention really are; comfort and positive life experience for the loved one are often much more important than experiencing reality in a way that matches that of the caregivers.


4.1. Non-pharmacological interventions for hallucinations

When your loved one appears to be hallucinating, assess the situation and determine whether the hallucination is a problem for your loved one. Even if he is upset, there are often ways to calm him without resorting to drugs. Stay calm yourself -- or at least, try to act as if you are. (Easier said than done, sometimes...) Reassure him that he is safe and you will take care of him. Offer to hold his hand, or give him a hug. Use distraction. You might try to turn his attention to a favorite activity, such as listening to music, drawing, looking at a photo album or counting coins. Hallucinations can be very susceptible to suggestion; so try offering an alternative explanation for something that is frightening your loved one (e.g., “Those aren’t enemy soldiers, those are Uncle Arthur’s hunting buddies”) or calmly reassure your loved one that they will be gone soon (“Don’t worry, he’s just leaving”.)

Since visual hallucinations are more rare when the eyes are closed, asking the loved one to close his eyes can often stop the hallucination. Also, encouraging the loved one to focus on the hallucination -- for example, to describe it to you -- will typically make it disappear. (That approach worked particularly well with my husband.)

Hallucinations tend to occur in the same location -- I can't remember a time when my husband hallucinated anywhere other than the bedroom. Often, moving to a new environment will make the hallucinations disappear, so suggest that your loved one come with you on a walk or sit next to you in another room. Even frightening hallucinations often go away in well-lit areas where other people are present.

Hallucinations may be very real to your loved one. Do not argue about what your loved one claims to see or hear -- that can just cause or increase fear or confusion. At the same time, respond honestly if your loved one asks you about the hallucination. My husband was never upset when I told him I did not see or hear what he did -- in fact, he never seemed very surprised that I didn't, and sometimes even seemed reassured -- but I never told him there was nothing there. If your loved one doesn't try to involve you, and doesn't seem upset, it's often best simply to ignore the hallucinating.

If hallucinations keep recurring, investigate possible causes other than the dementia itself, and determine whether they can be minimized or eliminated. These causes may include sensory deprivation (e.g., poor vision and/or hearing), excessive sensory stimulation (e.g., shiny surfaces or too much noise), and delirium and other medical causes (e.g., antibiotics and opiates, infections, dehydration, vitamin B12 deficiency, etc.)

A dementia patient often has problems with his sensory systems (e.g., "agnosias"); and because of that, sensory deprivation should be one of the first things to suspect when our loved ones hallucinate. If they have visual hallucinations, have their vision checked. Make sure they wear good eyeglasses. Cataracts, in particular, are very common in Alzheimer's patients, since they can be caused both by the Alzheimer's itself and by some of the drugs that are prescribed for Alzheimer's. Consider surgery if their cataracts are significantly interfering with their vision. (My husband was stage 6 when he had cataract surgery. It went surprisingly well, and he couldn't get over how beautiful all the colors were afterward.)

Create an environment that makes it easier for them to see. Visual hallucinations are often associated with dim light; and so, keeping a room well-lighted can help prevent visual hallucinations. Check for lighting that casts shadows, which may be confusing or difficult to interpret. Light that is too bright can also be a problem, so reduce glare by using soft light or frosted bulbs, partially closing blinds or curtains, and using adequate globes or shades on light fixtures. Check for glare, reflections, or distortions from the surfaces of floors, walls and furniture. Use light colors, e.g., on walls, to reflect more light, but use a matt finish, rather than a shiny one. Large, bold prints (for example, florals in wallpaper or drapes) may cause confusing illusions, so use solid colors instead.

Asking your loved one to point to the area where he sees or hears something may help you identify one particular aspect of the environment that is being misinterpreted. Glare from a window may look like snow to your loved one, and dark squares on tiled floor may look like dangerous holes. Dementia patients often don’t recognize their own reflections; if your loved one is distressed by seeing a strange person in the bathroom, cover up the mirror or take it down. (Note, however, that some dementia patients actually enjoy chatting with the Man in the Mirror.) Art, such as statues, may appear real; and so can images on a TV screen.

Similarly, if your loved one has auditory hallucinations, have his hearing checked and make sure he wears his hearing aids. Try to minimize or eliminate noises that might cause confusion. Noise from a radio or traffic outside may sound like someone else is present in the house, for example.

Because overall sensory deprivation can lead to hallucinations, an increase in auditory and/or tactile sensory input, such as music, therapeutic touch, and massage may sometimes be beneficial.

At the same time, be aware that sensory overload may cause hallucinations. For example, it is becoming very well-established that the sensory overload of a hospital environment (bright lights at night, intercoms, nurses and aides entering the room at all hours, etc) can cause delirium and hallucinations. Even something as simple as too much noise (from the loved one's perspective), e.g., from loud conversations, too many people talking, or traffic sounds, may trigger hallucinations.


4.2. Drugs for treating hallucinations

The first approach to pharmacologic intervention is to determine whether any of the drugs that the loved one is taking might be causing the hallucinations. For example, virtually all anti-Parkinson's drugs may produce psychotic symptoms, including hallucinations. Discontinuing a suspect drug, changing its dosage, or switching to another drug in the same category may alleviate the problem.

Caregivers need to realize that, aside from the use of clozapine and cholinesterase inhibitors in patients with synucleinopathies, there is very little evidence that the administration of drugs will be generally beneficial for treating hallucinations. Not only are antipsychotics (neuroleptics) often ineffective, but they carry substantial risk for any elderly dementia patient. If there is any possibility that the loved one has a synucleinopathy (Lewy body dementia, Parkinson's disease, Parkinson's disease dementia, or the Lewy body variant of Alzheimer's), antipsychotics should be avoided if at all possible. Studies have concluded that as many as half (53%) of such patients suffer a life-threatening "severe neuroleptic sensitivity" reaction to antipsychotics, even when using low doses and/or the newer atypical antipsychotics.

If hallucinations are extremely distressing to the loved one, however, and non-pharmacological interventions are ineffective, it may become necessary to attempt to find a drug that will help. The choice of drugs to try in a given dementia patient depends on the disorder(s) and co-morbidities involved; and may depend as much on avoiding drugs that may be harmful to the patient as on selecting candidates that may be helpful. Unfortunately, there is little data to go on when it comes to identifying drugs that may be beneficial.

Traditionally, antipsychotics have been the most commonly used treatment for psychotic symptoms, including hallucinations. Their usefulness in treating schizophrenia, both chronic and late onset, is well established. The atypical antipsychotics are often thought to be more suitable for use in elderly patients, since they tend to have a better side-effect profile in young schizophrenia patients. However, elderly people show variable responses and increased sensitivity to medications in general, and to antipsychotics in particular. Age-related bodily changes affect the pharmacokinetics and pharmacodynamics of antipsychotic drugs, which have numerous side-effects that can be more persistent and disabling in older people. Tardive dyskinesia, for example, can lead to a number of physical and psychological complications, including difficulty in eating and swallowing, weight loss, falls, difficulty in keeping balance and depression. In theory, atypical antipsychotics are less likely to cause extrapyramidal syndrome (EPS) than the typical antipsychotics. Recently, however, the focus has shifted from EPS toward adverse effects on body weight and related metabolic consequences that seem to be more common with atypical than with typical antipsychotics, and which are associated with greatly increased risks for diabetes, dyslipidemia, and hypertension (the metabolic syndrome) and are related to increased coronary heart disease morbidity and mortality.

Moreover, there has been increasing concern about the safety of atypical antipsychotics ever since the FDA issued "black box" warnings in 2005 on the even higher risks of serious cardiac, metabolic, cerebrovascular, and mortality risks associated with their use in elderly dementia patients.

A series of recent studies has shown that Alzheimer's patients who take antipsychotics (including the newer atypical antipsychotics) are more likely to have a much faster rate of deterioration than those who do not, and an even higher rate of deterioration if taking sedatives as well.

On top of this, meta-analyses of the efficacy of antipsychotics for treating dementia-related problems such as hallucinations showed no significant difference compared to placebo. This is, of course, "on