Dementia: Behavioral and Cognitive Aspects

Barbara L. Malamut

Laurie M. Ryan

Recognition that a person is experiencing a decline in mental status can often be obvious, especially to the physician who is well acquainted with the patient over a period of years; however, determining the cause of the mental decline is often more challenging and frequently requires a multidisciplinary approach. Some mental status changes such as depression, anxiety, metabolic disorders, and adverse drug interactions are reversible if properly diagnosed and treated, but they can resemble dementia; therefore, if they are not carefully assessed, they can lead to misdiagnosis and improper management. Similarly, there are many different forms of dementia, each with its own cognitive profile and behavioral manifestation, and as new medications that mediate the progression of illness are developed, diagnostic accuracy will become essential to inform treatment decisions. Neuropsychology contributes greatly to the diagnosis of dementia by helping to determine the subtype, nature, and severity of cognitive changes and by characterizing the impact of these deficits on functional aspects of a person’s behavior.

In this chapter, the neuropsychological and behavioral aspects of the most common neurodegenerative dementing illnesses are discussed. Particular emphasis is placed on Alzheimer’s disease (AD) because of the prevalence of this disorder.

According to the American Psychiatric Association’s Diagnostic and Statistical Manual of Mental Disorders, 4th Edition (DSM-IV) (7), the cognitive manifestations of dementia must include impairment in at least two or more of the following domains: short- and long-term memory, abstract thinking, impaired judgment, language or other disturbances of higher cortical functioning, or personality change. Dementia affects an individual’s social and occupational skills as well as the ability for self-care. The cognitive and behavioral disturbances must not be caused by psychiatric mental disorders. Once the physician suspects mental status changes, performs a screening test of mental status, and runs the appropriate battery of medical tests to rule out a reversible process, the patient should be referred for a neuropsychological assessment.

THE ROLE OF THE NEUROPSYCHOLOGIST

Distinguishing between age-related decline, mild cognitive impairment (MCI), and dementia such as AD can be difficult, requiring careful evaluation within and across many cognitive functions. There is significant overlap in symptoms among conditions, and it is only by looking at the pattern of functioning within various cognitive and behavioral domains that one can begin to understand the etiology of the disorder. For example, impairment in memory is not specific to any one dementing disorder and should be viewed as a sign of cognitive dysfunction rather than a diagnostic certainty. Problems in free recall can be characteristic of confusional states, depression, or attentional disorders, thereby reflecting a deficit in activating encoding processes. Other memory problems observed in depression, frontal lobe dementias, or a frontal lobe-related deficit can be caused by impairment in retrieval of information. Alternatively, poor performance on a free recall measure could be a symptom of a pure amnestic disorder caused by hippocampal or temporal lobe damage.

Because different cognitive functions subserve functional activities, it is also the role of the neuropsychologist to address functional issues related to a person’s ability to safely carry out instrumental activities of daily living (IADL), such as driving, medication management, and handling finances, as well as basic activities of daily living (BADL), such as bathing, toileting, and dressing. By definition, dementia interferes with a person’s ability to function normally in everyday life. Skills needed for self-care, social and occupational functioning, and financial management are greatly affected, but the degree and pattern of functional impairment vary widely according to the type and stage of dementia.

In the early and middle stages of a dementing process, evaluation of cognitive status is critical to help ensure the individual’s day-to-day safety, as IADLs and BADLs are strongly associated with changes in executive functioning and memory (72). When a cognitively impaired person lives alone,
they can be at even greater risk for harm due to self-neglect or disorientation. In an 18-month prospective study examining cognitive predictors of harm in 130 elderly people, 21% of the people had an incident of harm. The study found that three neuropsychological tests that measured executive functioning, recognition memory, and conceptualization were independent risk factors for harm (146). In the later stages of dementia, a neuropsychological evaluation is usually requested to help with problematic behaviors such as wandering, screaming, or verbal abuse to determine what, if any, behavioral or environmental interventions would be effective in arresting or minimizing such behavior, thereby reducing the use of psychotropic medication to control these behaviors.

DIAGNOSIS

A first step in determining a diagnosis is to understand the nature of the cognitive disturbance and to narrow the causes for the change in mental status. Other than medical and psychiatric history, four key features are important when forming a diagnosis: information regarding when the problems began (i.e., time of onset of initial symptoms); the course of the symptomatology (i.e., insidious, acute presentation with little change, or step-wise); rate of progression of the disorder (i.e., slow or rapidly progressing); and presenting symptoms. Demographic information (e.g., education, occupation) and social history (e.g., living situation, hobbies, alcohol and drug consumption) are also pertinent variables so that test results may be placed within a context. Tables 21-9 and 21-10 illustrate differences between the pattern of presentation and rate of progression among the various dementing syndromes.

Table 21-9. Patterns of Onset and Progression of Dementing Disorders

	Disease Progression
Onset	Progressive (Slow)	Step-Wise	Rapid
Insidious
	AD VaD FTD-lv (PPA); -bv; -mv CSH Alcoholism Parkinson’s disease NPH, HD, PSP	AD + VaD	Pick’s disease Jakob-Creutzfeldt disease Lewy body disease
Acute
	Vitamin B deficiency	MID, stroke Head trauma Encephalitis Anoxia	Herpes encephalitis Meningitis
AD, Alzheimer’s disease; CSH, chronic subdural hematoma; FTD-lv, -bv, -mv, frontal temporal dementia language variant, behavioral variant, motor variant; HD, Huntington’s disease; MID, multi-infarct dementia; NPH, normal pressure hydrocephalus; PPA, primary progressive aphasia; PSP progressive supranuclear palsy; VaD, vascular dementia.

When obtaining the history, even in the early stages of dementia, it is general practice to obtain independent collateral information about cognitive and functional capabilities from a person who knows the patient well. Individuals with dementia often are unaware of their deficits, and self-reports are often not consistent with cognitive performance on testing (28,44). Despite the inconsistency between a patient’s self-report and objective test data, it is still important to interview patients because, in the early phase of AD, the patient’s lack of awareness of their functional deficits has been shown to strongly predict a future diagnosis of AD (144). However, Loewenstein et al. (88) found that informant reports were not always reliable with regard to functional capacity. They found that caregivers were accurate in predicting the functional performance of patients with AD who were not impaired on objective tests of functional status, but they overestimated the functional abilities of AD patients who were impaired on objective testing. Higher Mini-Mental State Examination (MMSE) (47) scores were associated with caregivers’ overestimation of functional capacities (e.g., ability to tell time, identify currency, make change for a purchase, and use eating utensils).

Table 21-10. Comparison of Neuropsychological Findings in the Early Stages of Different Dementing Disorders and Depression

Domain	AD	DLB	VaD	FTD-lv (PPA)	DFT-bv	PD	Depression
Attention	Distractible	Fluctuating	Distractible	Normal	Distractible	Normal	Poor
Digit Span Forward	Normal	Unknown	Variable	Normal	Impaired	Normal	Variable
Executive	Mild imp	Variable	Impaired	Intact	Impaired	Impaired	Slowed
Reasoning	Mild imp	Impaired	Variable	Intact	Impaired	Impaired	Slowed
Info Processing Speed	Variable	Variable	Slow	Variable	Fast	Impaired	Slow
Language
Naming	Impaired	Intact	Intact	Impaired	Intact	Intact	Intact
Fluency	Impaired	Intact	Slow	Impaired	Mild, echolalia	Slow	Slow
Comprehension	Normal	Intact	Intact	Impaired	Intact	Intact	Intact
Praxis	Mild imp	Intact	Intact	Impaired	Unknown	Intact	Intact
Memory
Acquisition	Impaired	Intact	↓Initiating strategies	Impaired	Mild	Impaired	Poor
Storage	Impaired	Intact	Mild	Impaired	Intact	Impaired	Intact
Free Recall	Impaired	↓ Retrieval	Variable	Impaired	Variable	Impaired	Impaired
Recognition	Impaired	Preserved	Nml/mild	Impaired	Variable	Impaired	Intact
Remote	Variable	Unknown	Intact	Unknown	Intact	Intact	Intact
Motor	Intact	EPS	Impaired	Intact	Intact	Impaired	Slow
Handwriting	Intact	Impaired	Intact/tremor	Intact	Intact	Small/tremor	Intact
Visuoperception	Normal	Agnosia	Variable	Intact	Intact	Intact	Intact
Visual Constructional	Impaired	Severely impaired	Variable	Intact	Intact	Impaired	Intact
Affect	Variable, apathetic	Variable	Blunted, apathetic	No change	Hyperoral, blunt	Flattened	Blunted, apathetic
Psychiatric	Paranoid, depressed, anxious	Visual hallucinations, depression, delusions	Depression	None	↓ Insight ↓ Social conduct	Depressed	Depressed
Symptoms contained in a box represent the most common presenting signs for the type of dementia.
AD, Alzheimer’s disease; DLB, dementia with Lewy bodies; VaD, vascular dementia due to small-vessel disease; FTD-lv, frontotemporal degeneration language variant; PPA, primary progressive aphasia; FTD-bv, frontotemporal dementia behavioral variant; PD, Parkinson’s disease; EPS, extrapyramidal signs; Imp, impaired; Nml, normal.

SCREENING MEASURES

The most commonly used clinical screening tool for the presence of dementia in medical and psychological settings is the Folstein MMSE (133), which is easily administered in about 5 to 10 minutes. Although it has proved to be highly sensitive to changes in mental status in people over 65, it is most helpful for amnestic forms of dementia. The MMSE is not a good screening test for individuals with other illnesses that primarily involve language dysfunction. One reason for its lack of specificity is that the MMSE does not measure long-term memory, recognition memory, or executive functions (91). Another drawback to the MMSE is that it is not informative about the level of care needed for patients with mild to moderate levels of dementia. The MMSE has a total score of 30 points, with a score of 23 or lower usually considered indicative of organic dysfunction. However, when age (>80 years) and education (<9 years) are taken into account, the cutoff score suggestive of dementia decreases. It has been suggested that, for people with fewer than 9 years of education, a score of 17 should be considered as the cutoff (148), whereas a cutoff score of 26 or less has been suggested as optimum to detect mild to moderate dementia in people with higher education (151). Cultural background has also been shown to affect MMSE scores; Hispanics performed worse than non-Hispanics on the serial subtraction subtest when both groups were matched on education, age, and overall level of dementia. Similar results were found when spelling a word in reverse was substituted for serial subtraction (67).

Figure 21-2. Top row from left to right, clocks drawn by: a normal elderly woman and a patient with mild AD. Second row from left to right, clocks drawn by: a patient with moderate AD and a patient with moderate to severe AD.

Another quick screening test that is often administered along with the MMSE is the Clock Drawing Test. Together with MMSE scores, Clock Drawing Test performance has been shown to be sensitive to the presence of dementia, especially in those with very mild disease, but, on its own, it is not useful in discriminating between different forms of dementia (61,133). There are multiple steps involved in executing this task, and several diverse skills are required, including planning, visual attention, spatial conceptualization, and graphomotor control. In this test, the person is asked to draw the face of a clock, put all the numbers in the correct location, and set the clock to a specific time, either 10 minutes after 11:00 or 20 minutes after 8:00 (Fig. 21-2). These two times are most sensitive to dementia because they are sensitive to hemi-inattention and spatial conceptualization (49). Several different scoring systems of the clock have been devised to objectively quantify the presence and level of impairment (49,86,120), one of which (132) was found to be the most effective in detecting AD when combined with the MMSE.

The Mattis Dementia Rating Scale-2 (DRS-2) (74) is another frequently used screening measure that takes somewhat longer to administer than the MMSE and Clock Drawing Test but is useful for early detection, differential diagnosis, and staging of dementia for individuals at the lower end of functioning. The DRS-2 is more comprehensive than the MMSE and looks at attention, initiation/perseveration, construction, conceptualization, and memory in greater depth. The second version of the DRS uses the same test
questions as the first version, but in collaboration with the Mayo Older Americans Normative Studies, the normative data were revised to take age and education into account when interpreting test results. Because the normative data were drawn mostly from white samples in rural or suburban settings, another set of normative data has been developed for older African-Americans to allow for greater diagnostic accuracy (121).

The Alzheimer’s Disease Assessment Scale (ADAS) is used mostly in research settings, but it is mentioned here because of its wide use in drug trials (127). It is designed to evaluate the severity of both cognitive and noncognitive functions that are characteristic of AD. The scale is comprised of 21 items tapping memory, language and praxis, mood, and behavioral changes (e.g., depression, agitation, psychosis, and vegetative symptoms). It is administered in an interview format and takes about 45 minutes.

NEUROPSYCHOLOGICAL PROFILES OF DEMENTING DISORDERS

ALZHEIMER’S DISEASE

AD is the most common form of dementia, affecting an estimated 7% to 8% of people over 65 years of age and 30% to 40% of people over the age of 85 (138). The initial presentation for most patients with AD often involves memory lapses with an insidious onset. Families often report that a holiday meal was burned or a parent became lost while driving a familiar route. Often, relatives of the affected individual will realize in hindsight that certain episodes that occurred several years earlier were actually the beginning signs of the dementia.

Three diagnostic categories for AD are definite, probable, and possible. Table 21-5 in Chapter 21.2 outlines the National Institute of Neurological and Communicative Disorders and Stroke and the Alzheimer’s Disease and Related Disorders Association (NINCDS-ADRDA) criteria for diagnosing probable, possible, and definite AD. A diagnosis of definite AD can be made only postmortem because, in addition to meeting criteria for probable AD, specific histopathologic findings are required.

Neuropsychological Profile of Presymptomatic Alzheimer’s Disease Patients

Although patients with AD most often present to their physicians in the mild to moderate stages of the disease, more and more individuals complain of memory loss before any other signs of disease are evident, including normal performance on the MMSE. It is now understood that AD can have a long preclinical period during which subtle difficulties are detectable on neuropsychological testing and a diagnosis of MCI is given (see Chapter 21.4). On neuropsychological testing, problems in delayed recall memory of a word list and executive functioning as measured by the Trail Making Test B (118) in individuals who were clinically presymptomatic at the time of neuropsychological testing could discriminate those who would manifest AD 18 months later from those who would remain normal (29). In another population-based prospective study, recall memory distinguished those who would remain healthy from those individuals who would develop AD from 3 to 6 years in advance of a diagnosis (9). No differences were noted in digit span forward and reverse between those who remained healthy and those who developed the disease.

Olfactory dysfunction also has been found in patients diagnosed with AD very early in the disease process (4,103,104). Patients with normal MMSE scores who were unaware of their deficit were more likely to develop AD than those with olfactory deficits who were aware of their poor performance on the test (40). Research is ongoing to determine whether impairment in olfactory identification is a reliable indicator of early AD.

Imaging has also been helpful in diagnosing AD. Magnetic resonance imaging (MRI) studies in AD show atrophy in the same brain areas where AD neuropathology is seen, and over time, these atrophic changes mirror the progression of the neuropathology. Specifically, in MCI/early AD, atrophy is observed in the entorhinal cortex and hippocampus, progressing to the temporal and parietal lobes and finally the frontal lobes in the late stages (101,136). Neuron loss and atrophy are not specific to AD and are found in other neurodegenerative disorders and normal aging. However, both longitudinal and cross-sectional studies have demonstrated differences in the pattern and rate of atrophy between AD and normal aging, particularly in the hippocampus and entorhinal cortex (42,71,101). Despite some overlap in brain areas affected, the degree of atrophy and the pattern of affected brain areas is different enough to allow discrimination between AD and other types of dementia including dementia with Lewy bodies (DLB) and frontotemporal dementia (FTD) (19,24,101). For example, in contrast to AD, marked atrophy of the amygdala has been found in FTD (155).

An advanced MRI technique, magnetic resonance spectroscopy (MRS), measures brain metabolites including N-acetylaspartate (NAA), which is a marker of neuronal activity; myoinositol (mI), which is a marker of gliosis; and creatine (Cr), which is a marker of intracellular energy use. Studies are very limited but suggest abnormalities in these metabolites in patients with AD in many of the same regions that show atrophy (8,101). Single photon emission computed tomography (SPECT) and positron emission tomography (PET)
findings correlate with the degree of cognitive dysfunction and the progression of cognitive dysfunction in patients with AD (5). SPECT, however, has a lower spatial resolution than PET, and studies comparing both techniques show greater sensitivity with PET in terms of correlation with dementia severity and in differentiating controls and AD patients (31,101).

Presenting Neuropsychological Signs

The core features of individuals who were shown to have only AD pathology determined at autopsy are a progressive decline in explicit memory, anomia, executive function difficulties, visuospatial deficits, and a variety of other behavioral features discussed below. The rate and order in which these functions decline are variable among patients and reflect the differential competencies premorbidly and differences in the progressive accumulation of neuritic plaques and NFTs within the various association cortices (139).

Most patients with AD present initially with the greatest problems in episodic memory, but great variability is seen in the pattern of decline among the other neuropsychological domains such that some skills remain preserved longer than others. Consistent with the pattern of cognitive changes, studies with PET, specifically 18F-flurodeoxyglucose (FDG) PET, show a related pattern of glucose hypometabolism in parietal, temporal, and posterior cingulate areas with later involvement of frontal areas (101,137). In the largest PET study of mildly impaired patients with dementia to date, not only was this pattern of regional brain metabolism a highly sensitive indicator of AD, but the initial pattern was also strongly associated with the subsequent overall course of disease progression (134).

Memory

Verbal and visual episodic memory loss is the hallmark of AD. On tests of episodic memory (e.g., story recall or list learning), AD patients demonstrate a rapid rate of forgetting and often do not recall or recognize the new information after 10 to 20 minutes (16,81,147). Furthermore, the normal pattern of learning a list of words known as the serial position effect (i.e., better immediate recall of items from the beginning and end of a word list, with poorest recall of words in the middle of the list) is also disrupted early in the disease process (15,135). In contrast to normal elderly people, AD patients tend to recall the last few words of a supraspan word list immediately after it is read aloud. Patients with AD also tend to make more intrusion and perseverative errors on memory tests when compared with normal elderly controls (26). In addition to new learning, the disease also affects memories from the remote past but to a much lesser degree. In the early stages of the disease, patients perform worse than normal control subjects in their ability to recall and recognize famous faces, public events, and autobiographical information, but some memories from the past remain preserved.

Executive Functioning

The term “executive functioning” refers to the capacity to plan and carry out complex goal-oriented behaviors and allows a person to independently adapt his or her behavior to changing contingencies, which is necessary for the performance of all independent goal-directed activities including IADLs. Due to the variety of cognitive processes involved in executive functioning, it must be evaluated with a series of tasks looking at different variables. The most commonly used tests involve problem solving (e.g., Wisconsin Card Sorting) (63), planning and organization, working memory (e.g., digit span backward), mental flexibility (e.g., Trail Making Test B) (118), and response inhibition (e.g., Color Word Interference Test) (37). Changes in executive functioning appear very early in the disease process and have even been found in nondemented older adults who were at genetic risk for AD by having the apolipoprotein E ε4 allele (153). Although executive dysfunction is evident in the early stages of AD, it is usually not as prominent as episodic memory loss. In patients with mild or moderate AD, deficits in executive functioning have been related to a decline in IADL and, therefore, have important implications for safety issues such as medication management and driving (22), as well as medical decision making and financial management capacities.

Language and Semantic Memory

Early in the course of AD, patients demonstrate impairments in the linguistic aspects of speech. Conversations are often vague, and discourse is filled with circumlocutions and overlearned phrases. On formal testing, AD patients have deficits on tests of word list generation (52) and naming, and they tend to make more lexical and semantic naming errors on fluency and naming tests compared with normal elderly controls (13,26,36,143). With disease progression, paraphasic errors and impairment in written and oral comprehension develop. In the late stages, dysarthria and mutism may occur.

Typically, the linguistic aspects of speech are examined through category fluency tests (i.e., rapidly name items within a specific category) and letter fluency tests (i.e., rapidly name words beginning with specific letters). Many neuropsychological and imaging studies have shown that differences in performance between these tests can be of diagnostic significance. At one time, having better category than letter fluency was thought to be evidence of cortical dementia, such as AD, whereas the reverse pattern was thought
to indicate a subcortical dementia such as in Parkinson’s disease, Huntington’s disease, or vascular dementia (VaD) (25,126). However, several studies have disputed these findings and found no difference in performance on category and letter fluency tests between patients with dementia due to different causes (130,143). It is now well recognized that both fluency tasks involve executive functions mediated by the frontal lobe such as attention, search strategies, initiation, and retrieval, and differences in test results may be due to differential impairment in functions other than linguistic processes. The impact of executive dysfunction on different verbal fluency tasks was illustrated in a study looking at two cortical dementia groups: autopsy-confirmed AD and FTD (115). In this study, letter fluency was more impaired in the FTD group compared to the AD group, whereas semantic fluency was more impaired in patients with AD. For both groups, the disparity between fluency tests increased with increasing severity of dementia.

Semantic memory refers to knowledge of words (vocabulary) and factual information and is affected in the beginning stages of AD as noted by consistent impairments in naming and fluency tasks. The level of the semantic memory impairment is related to the level of dementia. Semantic memory for generic knowledge (e.g., how many days are there in a week?) is impaired relatively early in the course of the disease and declines further as the disease progresses (105). One possible explanation for the semantic memory impairment is that a degradation of the organization and content of semantic knowledge causes the impairment. An alternative argument is that the patient has problems in accessing information from an intact semantic store (14,48,129,145). In either case, studies have shown that task difficulty that requires active searches of semantic memory and greater attentional resources is one of the intervening variables.

Praxis

Ideomotor apraxia is a disorder of skilled movement or gestural behavior to verbal command or imitation in the absence of motor impairment. It is thought to be caused by disruption in the access to stored memories of familiar action patterns. Apraxia usually emerges in AD after memory and language disturbances, but it can be seen in the early stages of the disease (33). Ideomotor praxis can be evaluated in many ways. When the four types of movements listed in Table 21-11 were evaluated in a group of patients with moderate to severe disease, transitive limb movements to command were found to decline before intransitive limb, buccofacial, and axial actions (113). The most common errors on transitive limb praxis testing involved patients using their body part as the object (e.g., using a finger to demonstrate teeth brushing). Deficits are also found in ideational praxis where a person is asked to perform a series of actions using objects to accomplish a goal, such as, “Fold a paper, put it in the envelope, seal it, and address it.” In ideational praxis, the deficits are usually a result of sequencing errors (e.g., sealing the envelope before putting the paper in). Evaluation of praxis is an important part of a dementia evaluation in patients suspected of having AD because it has been shown to differentiate depressed elderly patients with cognitive impairment from those with probable disease (33).

Table 21-11. Sample Apraxia Battery

Buccofacial
Show me how you …
1.	Cough
2.	Stick out your tongue
3.	Suck through a straw
4.	Blow out a match
5.	Sniff a flower
Limb intransitive
Show me how you …
1.	Wave goodbye
2.	Beckon “come here”
3.	Hitchhike
4.	Salute
5.	Signal “stop”
Limb transitive
Show me how you …
1.	Brush your teeth
2.	Stir coffee with a spoon
3.	Comb your hair
4.	Saw a board
5.	Use a screwdriver
Ideational
1.	Fold the paper, place it in an envelope, and seal and address it
2.	Prepare a cup of instant coffee with sugar
3.	Put a candle in the holder, light it, and blow it out