COGNITIVE

IMPAIRMENT AMONG
OLDER ADULTS
WITH LATE-LIFE
SCHIZOPHRENIA OR
BIPOLAR DISORDER
Barton W. Palmer, Casey I. Loughran, Thomas W. Meeks

ABSTRACT
Neurologists are increasingly faced with the daunting task of disentangling dementia
from primary psychiatric conditions or recognizing their coexistence in older patients.
Both schizophrenia and bipolar disorder are characterized by substantial intergroup
cognitive heterogeneity among older and younger patients. In schizophrenia, deficits
in many cognitive domains are common; however, rapid forgetting, loss of crystallized knowledge, and greater than age-normal declines in cognitive function are rare
and warrant careful evaluation for secondary causes. The cognitive deficits associated
with bipolar disorder tend be most severe during acute affective episodes, but some
deficits tend to persist even during periods of relative euthymia. Lifetime number of
affective episodes in bipolar disorder may adversely affect cognitive functions in bipolar
disorder, but severe deficits and/or substantive declines over a period of a few years are
unusual and warrant careful evaluation for secondary causes.
Continuum Lifelong Learning Neurol 2010;16(2):135152.

INTRODUCTION
When most individuals think of neuropsychiatric conditions in late life, the
first disorders that come to mind are
primary dementias. In contrast, schizophrenia and bipolar disorder are more
commonly thought of as disorders affecting individuals in the early or mid-adult
years. Yet, as part of the aging baby-boom

generation (the first of whom turn 65 in

2011) and the general graying of the
industrialized world population, and because better treatments for severe mental illness have reduced excess diseaseassociated mortality, the coming decades
are expected to bring a rapid expansion
in both the number and proportion of
older adults with schizophrenia, bipolar

Relationship Disclosure: Dr Palmer and Ms Loughran have nothing to disclose. Dr Meeks has received personal
compensation for serving as the assistant to the editor of American Journal of Geriatric Psychiatry.
Unlabeled Use of Products/Investigational Use Disclosure: Dr Palmer, Ms Loughran, and Dr Meeks discuss the
unlabeled use of medication for cognitive impairment and the use of psychotropics in the treatment of psychotic
symptoms secondary to dementia or a general medical condition. Dr Palmer, Ms Loughran, and Dr Meeks
discuss the unlabeled use of cholinesterase inhibitors for treatment of Lewy body dementia, quetiapine for
treatment of psychosis secondary to Lewy body dementia, and olanzapine for treatment of psychosis secondary
to a medical condition.

Copyright # 2010, American Academy of Neurology. All rights reserved.

Copyright @ American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

135

" COGNITIVE IMPAIRMENT

KEY POINTS

136

It is essential that
neurologists
disentangle a
primary
psychiatric
diagnosis
from dementia
when a person
presents
with psychotic
or other
noncognitive
neurobehavioral
symptoms in
later life, as
such symptoms
are frequent
in many
diagnoses.
Late-life
schizophrenia
occurs in
both (1) elderly
people with
schizophrenia
who
experienced
the onset in
adolescence
or early
adulthood,
and (2) those
whose onset
was after
age 40.

disorder, and other chronic primary psychiatric conditions.1 For neurologists, the
rising number of older adults living with
schizophrenia or bipolar disorder may be
particularly relevant because the cognitive
deficits associated with these two conditions2,3 may compound the cognitive
changes associated with normal aging.4 As
psychotic or other neuropsychiatric symptoms are common among persons with
primary dementias5 and may emerge
prior to the manifestation of obvious
cognitive symptoms (as apparently occurred with Auguste D, the woman in
Alzheimers original case study),6 neurologists and other clinicians will increasingly face the task of disentangling
dementia from primary psychiatric conditions or recognizing their coexistence.
In this chapter we provide an overview of late-life schizophrenia and bipolar disorder, with an emphasis on
the cognitive aspects of these conditions. We begin with a focus on late-life
schizophrenia, including a description
of the typical level, pattern, and course
of cognitive deficits over the life span,
the influence of these deficits (relative
to the influence of psychotic symptoms)
on independent functioning, and a description of atypical deficits that warrant
follow-up assessment to rule out the
presence of a new-onset neurodegenerative condition. Our comments in regard
to cognitive deficits in schizophrenia also
apply to schizoaffective disorder7; however, recent reports on plans for the
Diagnostic and Statistical Manual of
Mental Disorders (Fifth Edition) (DSM-V),
scheduled for publication in 2012, indicate schizoaffective disorder may be
deleted as a formal diagnostic entity
separate from schizophrenia.8 After that
overview, we provide some recommendations in regard to cognitive screening tools that can be applied in everyday clinical practice. In addition, we
will illustrate key points, dilemmas,
and possible solutions via several case
vignettes.

COGNITION IN OLDER ADULTS

WITH SCHIZOPHRENIA
Late-Life Schizophrenia
The term late-life schizophrenia encompasses two groups: (1) currently older
individuals who experienced the onset
of the illness (generally defined as the
time at which the first prodromal symptoms of schizophrenia emerged) in adolescence or early adulthood (comprising
about three-quarters of older schizophrenia patients), and (2) older adults with
late-onset schizophrenia (those whose
clinical symptoms first emerged after
age 40).
An international consensus statement
issued by some of the leading experts
in late-life schizophrenia suggested that
the term late-onset schizophrenia be
applied to persons whose clinical symptoms first manifest between the ages
of 40 and 60, whereas the term very
late-onset schizophrenia-like psychosis should be applied to persons whose
symptoms of schizophrenia first emerge
after age 60.9 The prevalence of the
latter is unknown, but one rough estimate is that 3% of older adults with
what appears to fit the schizophrenia
syndrome may experience onset after
age 60.10
Cognitive deficits in schizophrenia and late-life schizophrenia. Many,
perhaps most, contemporary experts
consider schizophrenia to be fundamentally a neurocognitive disorder with
neurodevelopmental origins.11 However,
in the early and mid-20th century the
deficits in cognitive test performance
among persons with schizophrenia were
frequently thought to be secondary
effects of the psychopathologic symptoms.3 The Diagnostic and Statistical
Manual of Mental Disorders (Fourth
Edition, Text Revision) (DSM-IV-TR) criteria for schizophrenia focus exclusively
on the psychopathologic symptoms and
do not formally include or code for
severity of cognitive dysfunction.12 This

Continuum Lifelong Learning Neurol 2010;16(2)

Copyright @ American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

is unfortunate because, as reviewed

later in this chapter, cognitive deficits
are increasingly recognized as a core
manifestation of this disorder that has
particular relevance for predicting and
understanding varying levels of functional independence among older as
well as younger adults with this disorder. Although the inclusion of cognitive
deficits among the diagnostic criteria for
schizophrenia has been proposed,13 it
appears that the forthcoming DSM-V,
may instead provide for coding of cognition along with other dimensions
(such as severity of positive or negative
symptoms) as an alternative to characterizing patients in terms of the traditional clinical schizophrenia subtypes.8
(One of the reviewers of an earlier
version of this chapter suggested we
define and distinguish the terms cognition versus neurocognition. These
terms are used interchangeably in the
present chapter to refer to the constructs that are targeted in standard
neuropsychological assessment, such as
attention, working memory, declarative
episodic learning and memory, executive functions, visuospatial functions,
and psychomotor/processing speed.14
It should be noted, however, that the
phrase deficits in cognitive processes
has, in some of the schizophrenia literature, been used in reference to the
structure or content of thinking in schizophrenia, particularly the construct of
thought disorder.3 The latter may
be less implicitly grounded in a neurobiologic model of psychosis.)
Heterogeneity, level, and pattern
of cognitive deficits. Although considerable interpatient heterogeneity in
the level and pattern of neurocognitive deficits exists, approximately 75%
to 80% of persons with schizophrenia
manifest cognitive impairment.3 The severity of deficits tends to be in the mild
to moderate range with an average
global cognitive level that is approximately 1 standard deviation (SD) below

the mean of demographically matched

peers.15 No single pathognomonic deficit or pattern of deficits characterizes
persons with schizophrenia,16 but some
of the most frequently impaired cognitive abilities are immediate recall on
tests of auditory learning or memory
(such as the word list learning task) and
slowed processing speed. Deficits in
other dimensions such as attention,
working memory, and executive functions are also common.15 Tasks of immediate verbal recall tend to be among
the tests with the largest effect size
differences (comparing schizophrenia
patients to healthy controls), but these
performance deficits usually reflect difficulties with initial learning/encoding of
new information, not actual retention
of learned information. While antipsychotic medications have been shown
overall to have very modest benefits
in various cognitive domains for some
patients with schizophrenia,17 parkinsonian side effects of antipsychotics (especially medications with a high ratio
of dopamine-2 receptor:serotonin-2 receptor antagonism, such as high potency typical antipsychotics and, among
newer medications, risperidone) can affect motor/processing speed and hence
results on timed tests.
Potential red flag deficits. In stark
contrast to the common difficulties in information acquisition, retention of newly
acquired information (eg, as evidenced
by comparing delayed free or cued recall to the last learning trial on a test
such as the California Verbal Learning
Test)18 is among the least affected cognitive abilities in schizophrenia. That is,
rapid forgetting, one of the hallmarks
of Alzheimer disease, is unusual even
among older patients with schizophrenia,19 so its presence in a clinical situation
warrants further investigation of a possible comorbid dementia or other condition. Another dimension that tends to be
relatively (although not completely)
spared is crystallized verbal knowledge.

KEY POINTS

Although typically
considered a
disorder of
psychopathologic
symptoms,
schizophrenia
also entails
cognitive deficits,
which can
predict and aid
in understanding
levels of
functional
independence.
No single
pathognomonic
deficit or
pattern of
deficits
characterizes
persons with
schizophrenia,
but some of the
most frequently
impaired
cognitive
abilities are
immediate
recall on tests
of episodic
learning or
memory and
slowed
processing
speed. Deficits
in other
dimensions
such as
attention,
working
memory, and
executive
functions are
also common.

Continuum Lifelong Learning Neurol 2010;16(2)

Copyright @ American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

137

" COGNITIVE IMPAIRMENT

KEY POINT

138

Although
learning of new
information is
often impaired,
people with
schizophrenia
do not generally
have difficulty
retaining
information
once it has been
acquired or
difficulty with
long-term
knowledge.
Patients with such
difficulties should
be investigated
for secondary
causes.

If a person with schizophrenia starts to

show consistent difficulties with longterm/previously documented verbal
knowledge, this would be another potential red flag to consider possible
secondary causes beyond schizophrenia.
For persons with reliable collateral
historians, changes in functional abilities
(independent or even basic activities of
daily living) in the absence of prominent
psychopathology are a clue to investigate the possibility of the onset of a
comorbid, distinct cognitive disorder. It
is important to remember that although
positive symptoms (such as hallucinations or delusions) tend to decline with
age, negative symptoms (such as alogia,
avolition, flattened affect) often persist.
Thus, collateral information about longstanding lack of spontaneous independent activities may also signal persistent
negative symptoms of apathy and avolition rather than functional decline
caused by cognitive deficits. Also, psychotic decompensation or baseline suboptimal recovery may result in quite
idiosyncratic and/or disorganized patterns of speech that could be construed
as signs of aphasia. It is apparent from
these suggestions that establishing a
reliable collateral historian (be it family
or hired caregivers) is often the best tool
in deciding when to become concerned
about the onset of cognitive disorders
in persons with severe mental illness, as
seen in Case 7-1.
Course and effects of aging. With
the exception of persons who have been
chronically institutionalized for schizophrenia, aging itself appears to have no
discernible influence on cognitive functioning in schizophrenia beyond that
which characterizes normal aging.20 This
stability in function is a relatively rare
example of contemporary data contradicting assertions by Emil Kraepelin a
century ago. As implied by his preferred
term for this condition, dementia praecox, Kraepelin held a rather pessimistic
view of the long-term course of this dis-

order as being one of progressive mental

decline. This does not appear to be the
case for most persons with schizophrenia.
As illustrated in Figure 7-1, the typical
course of cognitive deficits in schizophrenia is: (1) mild (on average) premorbid
cognitive deficits in childhood and adolescence,21 (2) cognitive decline in the
peri-onset period,22 and (3) a stable
pattern of cognitive functions across all
neurocognitive domains over the longterm chronic course among noninstitutionalized patients.20 There is one possible exception to the preceding assertion
about the general course of cognitive deficits in schizophrenia. Specifically, studies
of geriatric patients with schizophrenia
who have lived in long-term institutional
settings for much of their adult lives report higher rates of progressive cognitive
decline than expected from age alone.23
Of note, the cognitive decline among
these institutionalized patients does not
appear to be attributable to any increased
occurrence of Alzheimer disease or other
identified primary dementia.24
The clinical implications of the preceding information on the course of
cognitive deficits are clear; if one sees
a progressive greater than age-normal
decline in cognitive functions among a
community-dwelling older adult with
schizophrenia, consideration of a secondary cause of the cognitive deficits
(such as a comorbid primary dementia)
should be investigated. When in doubt,
referral for neuropsychological testing
should be considered. Although it may
not give definitive answers suggestive
of the presence or absence of a cooccurring dementia, it does establish a
baseline performance that can be compared with repeat testing months to
years later and thus aid clinical observations of cognitive stability versus decline.
Cognitive deficits in late-onset schizophrenia. As noted earlier in this chapter, late-onset schizophrenia is generally
defined as that in which the symptoms
of schizophrenia first emerged between

Continuum Lifelong Learning Neurol 2010;16(2)

Copyright @ American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

KEY POINTS

Case 7-1
A 70-year-old man with a history of undifferentiated schizophrenia
dating back to at least his thirties and probably even earlier presented to
the geriatric psychiatry clinic to establish care. He lived in a skilled nursing
facility and had only intermittent contact with a brother and two nieces.
While his hallucinations and delusions were well controlled with a regimen
of low-dose quetiapine and some depressive symptoms had responded to
paroxetine 20 mg daily, he remained generally socially withdrawn, although
complacent, and moderately disorganized in his speech and behavior.
Luckily, a case manager who had known him for several years accompanied
him to all visits. The case manager described the patients organization of
thoughts as baseline. However, in recent months a new behavior had
emerged in which the patient would hoard food in his room in case they
forgot to serve a meal. He would forget that he had the food in the room
until it became spoiled enough to be malodorous. He denied paranoid
thoughts about being poisoned or anyone trying to starve him, but insisted
he had not been served several meals despite evidence to the contrary,
stating, They must have forgotten to tell me the meal was ready. He also
apparently forgot it was his birthday and seemed perplexed when his
brother and nieces came to the nursing home with a birthday cake. The case
manager felt these were abnormal cognitive symptoms for the patient, who
scored a 24/30 on the Mini-Mental State Examination (MMSE), missing most
points for orientation and recall. Formal neuropsychological testing was
ordered, and the patient displayed deficits in rapid forgetting and
confrontational naming more than 2.5 SD below demographically matched
norms. The tester did not feel any psychopathologic symptoms interfered
with the testing, and a diagnosis of probable Alzheimer disease was given.
Nursing staff was made aware that the patient may need extra reminders
and prompts for activities and daily grooming beyond those called for
by the negative symptoms of schizophrenia. A logbook was created for
him to sign every time he ate a meal, although he was allowed to keep some
nonperishable food items in his room. Donepezil 5 mg daily was initiated,
and his antidepressant was changed from paroxetine to citalopram to lower
anticholinergic burden.
Comment. This patients cognitive functioning and food hoarding
gradually improved to a moderate degree, especially with staff now
attending to his new dementia diagnosis, with further titration of donepezil
to full therapeutic dose of 10 mg and close attention to decrease
anticholinergic medication burden. The patient continued to have residual
negative symptoms of schizophrenia, some thought disorganization, and
gradually progressive loss of short-term recall. This case illustrates how having
objective records or reliable subjective collateral history of baseline cognition
(often more easily obtained by focusing on functional abilities or changes
therein) can prove crucial. It also emphasizes that persons with long-standing
mental illness are still susceptible to develop new neurologic illnesses.

ages 40 and 60 years.9 The contention

that real schizophrenia can emerge after age 40 is generally accepted by most
contemporary schizophrenia researchers,9 particularly when distinguished from

A collateral
historian is a
person who
can provide
information
about the
patients lifetime
course of
psychopathology
or cognitive
abilities, which
can be useful
when trying to
make a
differential
diagnosis.
Aging appears
to have no
significant
effect on
cognitive
functioning,
which
contradicts Emil
Kraepelins
assertion that
dementia
praecox is
characterized
by progressive
mental decline.
Neurocognitive
deficits among
people with
late-onset
schizophrenia
(characterized
by symptoms
first emerging
between ages
40 and 60) do
not differ
greatly from
those in
patients with
early-onset
schizophrenia.

very-late-onset schizophrenia-like psychosis. One of the historical objections to the

notion of late-onset schizophrenia was
that it must reflect neurodegenerative
processes given the absence of symptoms
Continuum Lifelong Learning Neurol 2010;16(2)

Copyright @ American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

139

" COGNITIVE IMPAIRMENT

FIGURE 7-1

Pictorial summary of general course of

cognitive deficits in schizophrenia over the
life span.

through many decades. But, in contrast

to what might be expected from a neurodegenerative model, the overall level,
pattern, and postonset course of neurocognitive deficits do not markedly differ
among patients with late-onset versus
earlier-onset schizophrenia.19,25 If anything, those with late-onset schizophrenia
tend to have less-severe impairment in
processing speed, certain aspects of executive function, and verbal memory
tasks.26,27

TABLE 7-1

140

Summary in regard to late-life

schizophrenia. Some of the major
conclusions that can be drawn about
cognitive deficits in late-life schizophrenia are summarized in Table 7-1. Overall, aging and late-onset diagnosis have
little discernible influence on cognitive
functions beyond those associated with
schizophrenia in general. Thus, onset of
progressive decline in cognitive or functional skills among people with schizophrenia should also alert the clinician to
the need for further evaluation of possible secondary causes. Indeed, most clinicians, although realizing schizophrenia
may have a late onset, are more aggressive in medical and neurologic workups
for persons presenting with primary psychotic symptoms after age 40, and even
more so after age 60. A variety of medical
and neurologic conditions could of
course masquerade, at least temporarily,
as schizophrenia without many abnormal findings on a standard physical
examination (eg, thyroid disorders, frontotemporal dementia, HIV-associated
CNS disorders, nonmotor/nonsensory
strokes, frontal lobe tumors).

Summary of Late-Life Schizophrenia

"

Seventy percent to 80% of older adults with schizophrenia manifest

cognitive deficits

"

Interpatient heterogeneity exists in the level and pattern of cognitive

impairment, but the average deficit is approximately 1 SD below the
normative mean of demographic-matched peers in the general population

"

Cognitive deficits are common in immediate recall, processing speed,

attention/working memory, and executive functions, but rapid forgetting
and substantive loss of crystallized knowledge are rare and warrant further
evaluation for possible comorbid conditions

"

Aging and late-onset diagnosis have a minor influence on cognitive

functioning; if anything, those with later onset (age > 40 years) tend to have
slightly better performance in certain aspects of cognitive functioning than
those with earlier onset

"

Decline in cognitive or functional skills, or severe deficits, are unusual and

should be evaluated for possible secondary causes

SD = standard deviation.

Continuum Lifelong Learning Neurol 2010;16(2)

Copyright @ American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

COGNITION IN OLDER ADULTS

WITH BIPOLAR DISORDER
Late-Life Bipolar Disorder
The body of published empirical data on
cognitive deficits and other aspects of
late-life bipolar disorder is substantially
smaller than that for schizophrenia. Available findings and/or clinical recommendations that most clearly differ for bipolar
disorder relative to those for schizophrenia are summarized in this section.2,28
Age of Onset in
Bipolar Disorder
Most older patients with bipolar disorder have been living with the condition
for several decades; the peak incidence
of new-onset bipolar disorder occurs in
the late twenties.29 The most commonly
employed cutoff for what constitutes
late onset is an onset of bipolar disorder
at age 40 or later.30 Prevalence data are
limited, but it appears that approximately 20% of patients with bipolar
disorder would meet this cutoff for late
onset30; approximately 2% to 5% of
older adults with bipolar disorder experienced first onset at age 60 or
later.29,30 In fact, most older adults
who manifest new-onset symptoms of
mania do not have bipolar disorder but
rather mania secondary to an underlying neurologic or other medical condition.31 New-onset manic symptoms in
older adults thus warrant thorough
medical workup; some of the secondary
causes to rule out include neurologic
conditions such as stroke or tumor,
metabolic problems, HIV or other infectious conditions, and medication side
effects (eg, dopaminergic antiparkinsonian medications, corticosteroids, antidepressants).31,32 Some researchers comparing bipolar patients with onset before
or after age 40 have reported that those
with a later age of onset have better
overall functioning, but few large-scale
systematic studies have been done, and
the results in the available reports are

KEY POINTS

somewhat inconsistent in regard to the

effects of age of onset. For example, Depp
and colleagues30 found no evidence of a
relationship between age of onset and
cognitive performance among patients
with bipolar disorder, whereas Schouws
and colleagues found that patients with
late-onset bipolar disorder had worse
performance on a number of cognitive
measures relative to that among those
with an earlier onset.33 Current age and
age of onset were confounded in both
studies; definitive reconciliation of such
results awaits additional large-scale studies comparing carefully matched older
adult patients with early-onset and lateonset bipolar disorder.

Cognitive Deficits in Younger

and Older Patients With
Bipolar Disorder
Although the effects of age of onset on
cognition are unclear at present, bipolar
disorder itself (among younger and older
adults) is clearly associated with cognitive deficits, which, while worse during
manic or depressed episodes, also persist during euthymic periods.2 The partial effects of psychopathologic state on
cognitive functions in bipolar disorder represent an interesting contrast to
schizophrenia, in which neither symptom severity nor fluctuations of symptoms of schizophrenia have a discernible
influence on the severity of cognitive
deficits, although symptoms could affect
the degree of validity of test results if they
affect cooperation with testing.25 Moreover, these deficits are not just apparent
as a statistical deviation from healthy
comparison subjects; they affect everyday functioning, including work performance and quality of life.3436
Few large-scale studies of neurocognitive functioning among older adults
with bipolar disorder have been reported. In their comprehensive search
of the empirical literature on cognitive
functioning in older patients with bipolar
disorder, Young and colleagues found

It is relatively
unclear
whether
cognitive
deficits are
greater for
persons with
late-onset
bipolar disorder
(with an onset
at age 40 years
or older) than
for persons
with an onset
before age 40,
as study
findings of this
topic have been
inconsistent.
The cognitive
deficits
associated with
bipolar disorder
are worse
during manic or
depressed
episodes but
persist during
euthymic
periods. This
contrasts with
the pattern
seen in
schizophrenia
wherein neither
symptom
severity nor
fluctuation
appears to
influence
severity of
cognitive
deficits.

Continuum Lifelong Learning Neurol 2010;16(2)

Copyright @ American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

141

" COGNITIVE IMPAIRMENT

KEY POINTS

142

Patients with
bipolar disorder
have worse
cognitive
functioning than
demographically
similar healthy
comparison
subjects and
somewhat
better overall
cognitive
functioning than
people with
schizophrenia.
Cognitive
functioning in
people with
bipolar disorder
seems to
decline with
longer duration
of illness and/or
higher numbers
of lifetime
episodes.

only seven reports published between

1970 and 2005.28 However, our research
group recently published two reports
of comprehensive cross-sectional and
longitudinal neuropsychological data
from middle-aged and older (45 to 80
years of age) community-dwelling persons with bipolar disorder.37,38 We found
that the patients with bipolar disorder had significantly worse performance
than healthy comparison subjects in overall cognitive functioning and within each
of the six specific cognitive domains assessed (crystallized verbal knowledge,
attention/working memory, verbal memory, visual memory, processing speed,
and reasoning/problem solving). On the
other hand, the bipolar disorder group
performed better than a schizophrenia comparison group on four domains
(crystallized verbal knowledge, attention/working memory, verbal memory,
processing speed), but not on visual
memory or reasoning/problem solving
(a form of executive functions). Using a
previously established cut-off score, we
also classified patients in terms of overall
cognitive impairment (impaired versus
intact) and found that 56% of the bipolar
disorder group, 65% of the schizophrenia group, and 12% of the healthy
comparison group were classified as cognitively impaired. These deficits among
middle-aged and older patients with bipolar disorder are somewhat broader
than those reported by other investigators who focused on younger adults,
among whom deficits in verbal memory
and executive functioning appear particularly common.2 On the other hand, the
finding that patients with bipolar disorder
tend to have worse cognitive functioning
than demographically similar healthy comparison subjects, but somewhat better
than patients with schizophrenia, is broadly
consistent with the findings among
younger adults with bipolar disorder.
In contrast to the static/stable nature
of postonset cognitive deficits across
the life span of most patients with

schizophrenia, some investigators have

tentatively suggested that cognitive functioning in people with bipolar disorder
may decline with longer duration of
illness and/or higher numbers of lifetime episodes.2 On the other hand, the
volume of large-scale, methodologically
sound studies employing comprehensive neurocognitive batteries has been
quite limited, particularly in regard to
longitudinal studies of older patients
with bipolar disorder. In our longitudinal
study,38 we compared the trajectories of
overall cognitive function among middle-aged and older patients with bipolar
disorder, schizophrenia, and healthy
comparison subjects followed for a
period of 1 to 3 years. All three groups
showed a slight mean improvement over
time (likely reflecting practice effects);
no significant differences were found in
the mean slope of change among bipolar
disorder, schizophrenia, or healthy comparison groups. On the other hand,
significantly greater variability occurred
around the mean slope for the bipolar
disorder group than for the other two
groups.

TABLE 7-2

Summary of
Late-Life Bipolar
Disorder

"

Interpatient heterogeneity
exists in severity of
cognitive deficits

"

A long-term neurotoxic
effect of affective disorders
may exist

"

Neurocognitive functioning
in middle-aged or older
patients remains relatively
stable over 1 to 3 years

"

Substantive fluctuation in
cognitive performance, or
severe cognitive deficit,
warrants further evaluation
for secondary causes of
cognitive decline

Continuum Lifelong Learning Neurol 2010;16(2)

Copyright @ American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

Summary of Implications
in Regard to Late-Life
Bipolar Disorder
Some of the major conclusions that can be
drawn about cognitive deficits in late-life
schizophrenia are summarized in Table 7-2.
Key among these is that while, as with
schizophrenia, considerable interpatient
heterogeneity exists in severity of cogni-

tive deficits associated with bipolar disorder, severe impairment and/or substantive decline in cognitive functions outside
the context of acute affective episodes are
unusual enough to warrant a more comprehensive assessment of possible secondary
causes of cognitive decline (Case 7-2).
Two key clinical implications can be
drawn from these results: (1) Considerable

Case 7-2
An 85-year-old woman was in routine follow-up treatment for bipolar
disorder that had begun with depressive episodes in her twenties. Her first
manic episode was not until years later, around age 40. No medical or
neurologic origin for the mania was identified, and manic symptoms occurred
even in the absence of antidepressants. She had suffered at least 10 distinct
mood episodes, mostly depressive, but had been stabilized on a regimen of
olanzapine 7.5 mg daily, bupropion 200 mg daily, and venlafaxine 150 mg
daily for several years. Her daughter, a psychologist, began to report that the
patient seemed spacey and had been hypersomnolent for several weeks. On
her next visit, the patient did appear to have difficulty maintaining attention,
and a parkinsonian tremor was evident in both upper extremities, along with
sialorrhea. The patient herself reported her thinking had felt increasingly
foggy over the past months. After laboratory workup to exclude delirium, it
was felt that with advancing age she may not have been tolerating the
olanzapine dose, causing possible anticholinergic, hypersomnolent, and
extrapyramidal symptoms. Olanzapine was slowly tapered, and divalproex
was substituted at 500 mg daily for mood stabilization. Her mood remained
stable, and her extrapyramidal symptoms improved but did not disappear.
She continued to have periods of confusion and later damaged her car
by backing into a lamppost. She was thus referred for neuropsychological
testing, which revealed some moderate impairment with recall helped
significantly with cueing, but most prominently deficits in visuospatial and
executive functions. These results, combined with her persistent parkinsonism
and fluctuating level of consciousness, supported a diagnosis of Lewy body
dementia, and she was started on rivastigmine titrated to 9 mg/d. Later,
congruent with this new diagnosis she developed visual hallucinations of
children running through her house, a symptom very dissimilar from any
previous symptoms of her bipolar disorder. These hallucinations were
successfully treated with quetiapine 25 mg at bedtime.
Comment. The patient remained stable with her mood symptoms. A trial
of carbidopa/levodopa failed to improve her parkinsonian symptoms and
increased visual hallucinations, and thus this medication was discontinued.
The patient was relatively stable in cognitive abilities for 6 months after
starting rivastigmine, but then began a slowly progressive loss of global
cognition and decline in ability to perform instrumental activities of daily
living. Once again, as in Case 7-1 involving a person with early-onset
schizophrenia, this patient with lifelong bipolar disorder began to experience
cognitive and motor neurologic symptoms atypical for the general course of
late-life bipolar disorder. Additionally, the case highlights that neurologic
medications may have psychiatric side effects and vice versa.

Continuum Lifelong Learning Neurol 2010;16(2)

Copyright @ American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

143

" COGNITIVE IMPAIRMENT

interpatient heterogeneity exists in severity of cognitive deficits, but the overall

implications for clinical settings from the
findings discussed here are that severely
impaired cognitive functions among elderly persons with bipolar disorder warrant further evaluation for secondary
causes. (2) Although it is possible that a
long-term neurotoxic effect of affective
disorders (or their treatment) associated
with bipolar disorder exists, neurocognitive functioning in middle-aged and older
patients appears relatively stable over a
period of 1 to 3 years. Thus, substantive
fluctuations in performance, while perhaps reflecting the variance in slopes seen
in individual patients, is unusual enough
to warrant more comprehensive assessment of possible secondary causes of
cognitive decline.

144

CLINICAL EVALUATION
As with most of medical practice, a thorough history of illness time course, symptoms sequence, past medical/neurologic/
psychiatric histories (in most cases obtained from a reliable collateral historian and/or medical records), review of
current/recent medications, and thorough neurologic examination usually
determine whether any neurologic condition is high in the differential diagnosis. Many primary neurologic illnesses
will have onset after age 40 (and are
especially suspect if after age 60) versus
the mean age of onset of schizophrenia and bipolar disorder being late
adolescence/early adulthood. While not
pathognomonic, nonauditory hallucinations raise suspicion for a medical or
neurologic etiology. Also, many neurologically driven presentations include
more confusion/delirium than primary
psychiatric conditions; but clear sensorium may occur in neurologic illnesses,
and deliriumlike confusion with poor
cognitive performance is not unusual
in acute psychosis of schizophrenia or
in acute mania of bipolar disorder.
Routine laboratory tests that are gener-

ally reasonable to obtain in any late-onset

psychiatric condition include complete
blood counts, serum chemistries, renal
function, liver function, urinalysis with
culture, thyrotropin, red blood cell folic
acid, vitamin B12, and rapid plasma reagin). Other specialized tests suggested
by examination or history may include
HIV antibody test, urine toxicology for
medication overdose or substance abuse,
neuroimaging (CT or MRI), lumbar puncture, blood cultures, chest x-ray, EKG,
EEG, or EMG with nerve conduction
studies, among other specialized testing
in unusual or rare cases. Table 7-3 lists
some examples of neurologic disorders
that can mimic late-life schizophrenia or
bipolar disorder.
Cognitive Functions in Patients
With Late-Life Schizophrenia or
Bipolar Disorder
Two key questions in regard to management of cognitive deficits among patients with late-life schizophrenia or
bipolar disorder are: (1) What measures
should be given priority in screening of
cognitive deficits, and (2) When should
a referral for neuropsychological assessment be initiated?
Screening cognitive deficits. In
regard to screening measures, it is useful to consider that if a patient with a
known primary psychiatric condition,
such as schizophrenia or bipolar disorder, has been referred for evaluation by
a neurologist, it is likely someone already suspects that the patients deficits
extend beyond those characteristic of
schizophrenia or bipolar disorder (although the level of sophistication in knowing when to refer will certainly vary). Thus,
a primary focus of the differential diagnosis will be to use a neurologic examination to discern the presence of other
signs of neurologic disease, ie, beyond
cognitive deficits. To the extent that
neurocognitive functions are key to a
differential diagnosis, a referral for comprehensive neuropsychological testing

Continuum Lifelong Learning Neurol 2010;16(2)

Copyright @ American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

KEY POINT

TABLE 7-3

Examples of Neurologic Illnesses That May Mimic

Late-Life Schizophrenia or Bipolar Disorder

"

Stroke (especially frontal or temporal for schizophrenia and right

frontal for bipolar)

"

Neoplasm (primary or secondary), particularly those involving

frontal/temporal lobes

"

Other mass lesions (eg, frontal or temporal subdural hematoma,

arteriovenous malformation, abscess, or meningioma)

"

Seizures (in particular, frequent complex partial seizures of

frontal/temporal lobes)

"

Paraneoplastic syndromes (substances produced by the tumor that

affect the CNS)

"

Systemic lupus erythematosus disease effects or reactions to

high-dose steroid treatments

"
"

Multiple sclerosis (especially if extensive frontal white matter damage)

"

HIV (eg, lymphomas, toxoplasmosis, cryptococcal meningitis, primary

neurologic effects of the HIV virus infecting neurons)

"

Neurosyphilis (check rapid plasma reagin followed by confirmatory

lumbar puncture)

"

Frontotemporal degeneration (may present with apathy, odd social

behavior, disinhibition)

"

Neurologic effects of endocrinopathy (eg, hyper-/hypothyroidism

or parathyroidism, adrenal disease)

"

Syndrome of inappropriate secretion of antidiuretic hormone

(SIADH)/hyponatremia (either primary or iatrogenic, eg, most
antidepressants, diuretics, oxcarbazepine)

"

Iatrogenic (eg, corticosteroids, antidepressant-induced mania,

antineoplastics, anticholinergics, stimulants)

Differential
diagnosis (or a
secondary cause
of uncommonly
severe cognitive
deficits) in
patients with
schizophrenia
or bipolar
disorder may
consist of
neurologic
examination
and
comprehensive
neuropsychological
testing.

CNS infection (eg, herpes simplex virus encephalitis with tendency

for temporal lobe involvement)

CNS = central nervous system; HIV = human immunodeficiency virus.

145
should be considered a top priority as
part of the overall clinical workup. As
detailed by the authors of the chapter
Neurocognitive Assessment, a full neuropsychological evaluation can yield a
wealth of data that is simply not possible
to obtain via bedside screening measures. Those caveats aside, a few potential
screening measures may aid the questions directed to the neuropsychologist,
reveal circumstances in which neuropsychological testing is clearly unnecessary,
or assist neurologists who work in set-

tings in which access to neuropsychologists is limited.

As noted above, the ubiquitous presence of rapid forgetting among patients with Alzheimer disease contrasts
starkly with the pattern seen even among
older patients with schizophrenia, who
do actually maintain learned material.
Thus, clinically useful screening measures will tend to be those that include
assessment of learning and delayed recall skills. One measure that might be
considered for this purpose is the
Continuum Lifelong Learning Neurol 2010;16(2)

Copyright @ American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

" COGNITIVE IMPAIRMENT

KEY POINT

146

The Montreal
Cognitive
Assessment is a
clinically useful
screening
measure for
cognitive
impairment
that assesses
several common
neuropsychological
abilities. The MoCA
is sometimes
preferred over
other screening
measures, such
as the Mini-Mental
State Examination,
because of its
sensitivity to
mild cognitive
impairment.

Montreal Cognitive Assessment (MoCA)

test.39 Similar to the more widely known
Folstein Mini-Mental State Examination
(MMSE),40 the MoCA is a brief (approximately 10-minute) cognitive screening
measure. It includes brief assessments
of several common neuropsychological
tasks that are evaluated more extensively in neuropsychological testing but
are not part of the MMSE, including the
Trail Making Test part B (a test of executive function among other things),
three-dimensional drawings (a test of
visuospatial skills), clock drawing (an assessment of multiple cognitive domains,
including executive functioning and
visuospatial skills), a three-item confrontational naming task (a task commonly
impaired in mild-moderate Alzheimer
disease), a digit span task (to assess
attention), a go-no-go task (to assess
behavioral inhibition, an executive function), an abstraction (similarities) task,
and a test of letter verbal fluency (in part
mediated by executive functions as well
as language abilities). Items shared in common with the more well-known MMSE
include some items assessing orientation to place and time, serial 7 subtraction
to assess concentration (as well as mathematic skills), and a short-term recall task
(although it consists of five words in contrast to three on the MMSE). The MoCA
seems to have some key advantages over
the MMSE in regard to screening cognitive
deficits among nondemented patients.
Foremost, the MoCA was designed to be
sensitive to mild cognitive impairment,
whereas the MMSE was designed to
measure severity of dementia. Second,
whereas the memory task on the MMSE
involves only three items, with a brief
delayed interval, the MoCA includes five
words (with two learning trials), several
intervening tasks (resulting in a longer
recall interval), followed by free recall,
cued recall, and multiple-choice recall.
Comparison of free recall to cued or multiple choice performance can be vital to
discerning loss of the memory trace (ie,

lack of being helped with cues or multiplechoice format being more typical of
Alzheimer disease) versus difficulties with
efficient retrieval (which could reflect the
executive deficits common in schizophrenia, bipolar disorder, as well as normal
aging). The authors of the MoCA have
made it available free of charge in multiple
languages at www.mocatest.org/.
It is also important to consider factors
that could affect cognitive test scores in
the absence of a neurodegenerative
cognitive disorder. Language of origin
and dominant lifetime language should
be assessed. Even when English fluency
seems adequate to the clinician, it can
cause difficulties in testing. Other items
to account for are (obviously) age and
educational attainment, for which increases
and decreases, respectively, may artificially lower cognitive test scores. The
MMSE has fairly well-established norms
adjusted for age and educational level.
Also, a school history may reveal areas that
have historically been difficult for the
patient (eg, they may recall being in a
remedial class for a certain subject),
which alters test interpretation. Conversely, high levels of education and
intelligence (certainly still possible in
schizophrenia and bipolar disorder)
make short screening cognitive tests less
sensitive in these persons.
Motivation by the patient is sometimes clearly lacking during cognitive
testing for a variety of possible reasons.
Persons may not want to learn they have
dementia, may feel embarrassment or
irritation during the testing, may feel
insulted by the relative simplicity of
some items, or may be savvy enough to
know that failing cognitive tests could
strip them of certain freedoms (eg, independent financial management, ability to live independently, or ability to
drive, as dementia is a mandatory or
voluntary reportable illness to departments of motor vehicles in certain
states). Thus, the examiner should establish initial rapport, explain that these

Continuum Lifelong Learning Neurol 2010;16(2)

Copyright @ American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

tests are rarely definitive, that results

obtained in 5 to 10 minutes almost never
prompt the clinician to recommend
permanent changes in the patients life,
and that the test is standardized (not
designed for any one persons apparent
abilities), so some questions may seem
silly or obvious while others seem difficult in a way that seems to have
nothing to do with everyday life. One
can often respond to a person who
immediately says, I cant do that with
a response such as, Well give it your
best try. I am surprised by how often
people underestimate their own abilities. If you reach a point where you are
obviously stuck, well just move on, and
no harm in having tried. Other questions that may derail an assessment
include, What has this got to do with
why I am here? to which one may
respond, I think, in your shoes, I would
be asking myself that same question, but
believe it or not, it helps me see how
certain parts of your brain are working.
Other patients see this as a test of
whether they are losing it or how
dumb they are. In this case, one may
explain in truth that we are seeing if
some things like your memory are
declining; but the phrase losing it is
just a common but incorrect way people
express certain changes with aging.
Additionally, one may easily reassure
patients this is not a test of intelligence.
It may also be somewhat reassuring to
patients that missing some items is not
always abnormal. Finally, testing the
person alone (without family members)
is usually best (unless there are absolutely no other means of having
someone else act as an interpreter for
non-English speakers). This may lessen
patient embarrassment and certainly
eliminates the possibility of some family
members trying to help the patient in
subtle (or not so subtle) ways. Yet,
honesty is usually also respected by
patients in statements such as, Well
the purpose of this test is to see if you

might benefit from more precise testing

to determine whether you may have
something like a memory disorder. If
not, then you can feel reassured; and if
so, then you and your family (or caregivers) can work together at an early
stage to best plan for what services you
may need in the future. You can decide
for yourself important things like who
would act as your health care or financial
decision maker if you could not do so for
some reason. Also, if you are just having
mild to moderate memory problems,
you could put your affairs in order as you
see fit, such as making your will the exact
way you want it and deciding if you want
aggressive medical treatments, such as
breathing tubes, feeding tubes, or electrical shocks to your heart if it stops
beating, were you to be hospitalized or
taken to the emergency department.
The goal is not to make you feel
incompetent. In fact, we want you to
be in control of these decisions as much
as possible now in case your thinking or
memory starts to slip over time. Many
people have this testing and we still
dont know for sure if they have a
memory disorder like dementia, so the
results are not always conclusive. With
regard to making advanced directives
and other health care decisions, many
people with severe mental illness, such
as schizophrenia or bipolar disorder
(and even persons with early dementia),
still can retain decisional capacity for
some important issues alluded to; but
because they are at increased risk of lack
of decisional capacity compared with
the general public, encouraging advanced directives is recommended.
For a somewhat more extensive
screening battery, the Mattis Dementia
Rating Scale (DRS)41 and the Repeatable
Battery for the Assessment of Neuropsychological Status (RBANS)42 might also
be considered. The DRS and RBANS
each requires about 20 minutes to administer, but they are easy to administer
and score, are well tolerated by patients,

KEY POINTS

Patients may
become
uncomfortable,
anxious, or
frustrated
when
completing
cognitive tests,
making such
assessments
difficult; but
there are
many ways
to manage
those risks,
such as building
rapport or
reassuring
them that
questions are
supposed to
be difficult.
Being honest
with patients
about the
reasons for
cognitive
testing is the
best way to
ameliorate their
concerns and/
or paranoid
thoughts and
more easily and
comfortably
complete the
assessment.

Continuum Lifelong Learning Neurol 2010;16(2)

Copyright @ American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

147

" COGNITIVE IMPAIRMENT

KEY POINT

148

The Dementia
Rating Scale
(DRS) and
Repeatable
Battery for the
Assessment of
Neuropsychological
Status
(RBANS) are
comprehensive
assessments
that are short,
easy to
administer and
score, and well
tolerated by
patients. The
DRS is sensitive
to cognitive
and functional
deficits in older
patients, and the
RBANS enables
longitudinal
assessments.

and provide for somewhat more comprehensive cognitive assessment than

the MoCA or MMSE. The DRS is widely
used in studies of Alzheimer disease and
other dementia, but has also proven
sensitive to cognitive deficits and functional impairment in middle-aged and
older patients with schizophrenia43 and
bipolar disorder.28 It yields a total score
(range 0 to 144) reflecting overall cognitive functioning,28 as well as five subscale scores: attention (range 0 to 37),
initiation/perseveration (range 0 to 37),
construction (range 0 to 6), conceptualization (range 0 to 39), and memory
(range 0 to 25). The RBANS has also
proven useful for assessment of cognitive status in schizophrenia.44 Like the
DRS, it provides an overall total score, as
well as subscale scores reflecting the
following ability areas: attention (digit
span and coding), immediate memory
(list recall and story recall), visuospatial
constructional skills (figure copy and
line orientation tasks), delayed memory
(delayed list, story, and figure recall),
and language (confrontational naming,
and category fluency). One advantage of
the RBANS over the DRS is the availability of a validated alternate form, which
facilitates retesting/longitudinal assessment by decreasing practice effects. At
least initially (until the clinician becomes
familiar with these tools), the DRS and
RBANS are best used by or in consultation with a neuropsychologist or other
person with expertise in administration
and interpretation of standardized psychometric measures. When properly
employed, these tools can be key in
distinguishing the relatively static pattern of cognitive performance expected
in schizophrenia and bipolar disorder
from the decline that would typify a
secondary neurodegenerative process.
When to refer out. The question
When is it appropriate to refer patients
for primary treatment or collaborative
treatment with a psychiatrist? may also
arise in many clinical situations with older

adults presenting with co-occurring psychiatric and cognitive symptoms, The

use of standardized cognitive tests in the
absence of specific didactic and experiential supervised training in their use
and interpretation can be problematic.
Even common and widely used bedside
tests, such as clock drawing, may be
misinterpreted. An ideal situation would
allow specialists to examine the patients,
obtain neuropsychological testing and interpretation from an appropriately trained
and experienced neuropsychologist, and
then reach a consensus on probable
or possible diagnosis and treatment
plan. Unfortunately, such collaborative
care is not common, and fragmentation
of health care among busy specialists
is more likely the norm. As such, our
recommendations for less than ideal
circumstances are as follows: for persons
with known preexisting schizophrenia
or bipolar disorder with onset before
age 40, a neurologist should (1) ensure
ongoing psychiatric care and discuss what
prompted the visit with the patients
psychiatrist; (2) attempt as best as possible to learn from collateral historians
whether notable changes have occurred
in cognition or functional independence
in the absence of florid affective or psychotic episodes; (3) attempt an office
screening of cognition and refer for
neuropsychological testing when there
is a need for ambiguous cases and/or when
more than quick/gross cognitive screening is involved and/or interpretation of
scores is dependent on the specific knowledge obtained via neuropsychological
training; (4) conduct a thorough neurologic examination and follow up any abnormalities with appropriate diagnostic
tests (Case 7-3); and (5) assess for possible
iatrogenic causes of cognitive impairments.
For persons with new-onset signs of
schizophrenia or bipolar disorder after
age 40, clinical suspicion for underlying medical or neurologic conditions
must increase. Before referral to a psychiatrist, the neurologist likely would

Continuum Lifelong Learning Neurol 2010;16(2)

Copyright @ American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

Case 7-3
A 68-year-old woman was admitted to a geriatric psychiatry unit because
of 2 to 3 months of marked decompensation in functioning. According
to her daughters, 6 months earlier she was still working in a postal office.
However, because of bizarre changes in her behavior, including paranoia
toward a supervisor, she was forced to retire. Other symptoms included
mood swings from odd laughter to uncontrollable crying, as well as bizarre
notions that terrorists were stealing her mail. She had been seen by two
neurologists prior to admission and had been screened with a battery
of laboratory tests, structural MRI, and two PET scans, none of which
revealed any evidence of medical or neurodegenerative diseases. Her
daughters noted she was not caring for her home or paying her bills
and seemed somewhat forgetful, but any cognitive changes were
overshadowed in their minds by her behavioral changes. During
hospitalization, neuropsychological testing was attempted, and she
finally completed the Mattis DRS, scoring in the mildly impaired range
(130/144). Basic laboratory panels were repeated and were unremarkable
except for elevated serum calcium of 11.0 mg/dL. Follow-up tests
revealed elevated levels of ionized calcium and parathyroid hormone. A
technetium parathyroid scan revealed results consistent with a right lobe
parathyroid adenoma. Her psychiatric symptoms were partially controlled
with olanzapine 10 mg/d,3 but 6 to 7 weeks later she underwent surgery
for a parathyroidectomy. One month after surgery she had no residual
psychotic or affective symptoms, and she scored 30/30 on the MMSE.
Comment. Follow-up by a geriatric internist revealed no recurrence
of parathyroid, psychiatric, or cognitive symptoms. This case clearly
emphasizes the importance of a clear time history of symptoms and
symptom patterns (ie, subacute onset of diffuse symptoms after the age
of 65 raising high suspicion for an organic cause of illness), although
it should be noted that late-onset psychiatric disorders without any clear
medical/neurologic precipitant have been reported. Nonetheless,
evaluating for possible reversible causes of neuropsychiatric symptoms
should remain the first priority.

want to (1) conduct a thorough neurologic examination; (2) obtain laboratory

results that could explain new-onset
psychiatric illness (eg, thyroid functions,
complete blood counts, serum chemistries [including calcium, liver function
tests, fasting glucose, renal function, and
sodium], infectious disease screens as
indicated [HIV, syphilis, Lyme disease,
encephalitis, and meningitis], vitamin
B12, and folic acid), ensure cardiorespiratory stability (eg, EKG, pulse oximetry,
blood pressure, and pulse), and screen
for alcohol and illicit or prescription
drugs (alcohol and benzodiazepine dependence are underrecognized among

older adults, and thiamine may prove

quite helpful in emergent amnestic syndromes); (3) strongly consider neuroimaging such as MRI or CT scan to evaluate for occult tumors, strokes, bleeds,
or other mass lesions; (4) if symptoms
appear atypical and paroxysmal, consider EEG, which may prove useful to
evaluate for seizures or in rapidly progressive dementias may aid in the diagnosis of Creutzfeldt-Jacob disease; (5) reserve lumbar puncture for very specific
suspicions (eg, meningitis, normal pressure hydrocephalus, herpes simplex encephalitis, and neurosyphilis) or atypical
cases that defy any known diagnosis; and
Continuum Lifelong Learning Neurol 2010;16(2)

Copyright @ American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

149

" COGNITIVE IMPAIRMENT

(6) again assess for iatrogenic causes of

cognitive impairment (eg, benzodiazepines,
opiates, cumulative anticholinergic burden, including the popular over-thecounter antihistamine sleep aids). If all
such evaluations fail to provide a definitive physiologic/neurologic basis for newonset symptoms, referral to a psychiatrist
would be quite appropriate. In fact, even
in the case of known etiologies of a cog-

nitive change among a person without a

known past psychiatric history, psychiatric referral may help provide palliative
relief of symptoms while awaiting definitive treatment or in cases in which the
prognosis is poor for ultimate recovery.
ACKNOWLEDGMENT
This work was supported, in part, by
NIMH ROI MH064722.

REFERENCES
1. Jeste DV, Alexopoulos GS, Bartels SJ, et al. Consensus statement on the upcoming
crisis in geriatric mental health: research agenda for the next 2 decades. Arch Gen
Psychiatry 1999;56(9):848853.
2. Goldberg JF, Burdick KE. Cognitive dysfunction in bipolar disorder: a guide for
clinicians. Washington DC: American Psychiatric Press Inc, 2008.
3. Palmer BW, Dawes SE, Heaton RK. What do we know about neuropsychological
aspects of schizophrenia? Neuropsychol Rev 2009;19(3):365384.
4. Palmer BW, Dawes SE. Cognitive aging: from basic skills to scripts and schemata.
In: Jeste DV, Depp CD, eds. Handbook of successful cognitive and emotional aging.
Washington, DC: American Psychiatric Publishing, Inc, 2010:3754.
5. Jeste DV, Meeks TW, Kim DS, Zubenko GS. Research agenda for DSM-V:
diagnostic categories and criteria for neuropsychiatric syndromes in dementia.
J Geriatr Psychiatry Neurol 2006;19(3):160171.
6. Alzheimer A, Stelzmann RA, Schnitzlein HN, Murtagh FR. An English
translation of Alzheimers 1907 paper, Uber eine eigenartige Erkankung der
Hirnrinde. Clin Anat 1995;8(6):429431.

150

7. Palmer BW, Savla GN. Neuropsychological deficits in schizoaffective disorder. In:

Yip KS, ed. Schizoaffective disorders: international perspectives on understanding,
intervention and rehabilitation. New York: Nova Science, 2009:115134.
8. Gever J. APA: major changes loom for bible of mental health. MedPage Today
www.medpagetoday.com/MeetingCoverage/APA/14270. Published May 19, 2009.
Updated July 27, 2009. Accessed February 10, 2010.
9. Howard R, Rabins PV, Seeman MV, Jeste DV; The International Late-Onset
Schizophrenia Group. Late-onset schizophrenia and very-late-onset schizophrenialike psychosis: an international consensus. Am J Psychiatry 2000;157(2):172178.
10. Harris MJ, Jeste DV. Late-onset schizophrenia: an overview. Schizophr Bull
1988;14(1):3955.
11. Rapoport JL, Addington AM, Frangou S, Psych MR. The neurodevelopmental
model of schizophrenia: update 2005. Mol Psychiatry 2005;10(5):434449.

Continuum Lifelong Learning Neurol 2010;16(2)

Copyright @ American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

12. American Psychiatric Association. Diagnostic and statistical manual of mental

disorders: fourth editiontext revision (DSM-IV-TR). Washington DC: American
Psychiatric Association, 2000.
13. Keefe RS, Fenton WS. How should DSM-V criteria for schizophrenia include cognitive
impairment? Schizophr Bull 2007;33(4):912920.
14. Palmer BW, Savla GN. Neuropsychological evaluation [Geriatric Psychiatry 54.2e].
In: Sadock BJ, Sadock VA, Ruiz P, editors. Kaplan and Sadocks comprehensive
textbook of psychiatry. 9th ed. Baltimore, MD: Lippincott Williams & Wilkins,
2009:39893994.
15. Dickinson D, Ramsey ME, Gold JM. Overlooking the obvious: a meta-analytic
comparison of digit symbol coding tasks and other cognitive measures in
schizophrenia. Arch Gen Psychiatry 2007;64(5):532542.
16. Palmer BW, Savla GN, Fellows IE, et al. Do people with schizophrenia have
differential impairment in episodic memory and/or working memory relative to
other cognitive abilities? Schizophr Res 2010;116(23):259265.
17. Davidson M, Galderisi S, Weiser M, et al. Cognitive effects of antipsychotic drugs
in first-episode schizophrenia and schizophreniform disorder: a randomized,
open-label clinical trial (EUFEST) [published erratum appears in Am J Psychiatry
2009;166(6):731]. Am J Psychiatry 2009;166(6):675682.
18. Delis DC, Kramer JH, Kaplan E, Ober BA. California Verbal Learning Test (CVLT)
manual. San Antonio, TX: The Psychological Corporation, 1987.
19. Heaton R, Paulsen JS, McAdams LA, et al. Neuropsychological deficits in
schizophrenics: relationship to age, chronicity, and dementia. Arch Gen Psychiatry
1994;51(6):469476.
20. Kurtz MM. Neurocognitive impairment across the lifespan in schizophrenia: an
update. Schizophr Res 2005;74(1):1526.
21. Woodberry KA, Giuliano AJ, Seidman LJ. Premorbid IQ in schizophrenia:
a meta-analytic review. Am J Psychiatry 2008;165(5):579587.
22. Mesholam-Gately RI, Giuliano AJ, Goff KP, et al. Neurocognition in first-episode
schizophrenia: a meta-analytic review. Neuropsychology 2009;23(3):315336.
23. Harvey PD. Cognitive impairment in elderly patients with schizophrenia: age related
changes. Int J Geriatr Psychiatry, 2001;16(suppl 1):S78S85.
24. Friedman JI, Harvey PD, Kemether E, et al. Cognitive and functional changes with
aging in schizophrenia. Biol Psychiatry 1999;46(7):921928.
25. Heaton RK, Gladsjo JA, Palmer BW, et al. The stability and course of
neuropsychological deficits in schizophrenia. Arch Gen Psychiatry 2001;58(1):2432.
26. Jeste DV, Harris MJ, Krull A, et al. Clinical and neuropsychological characteristics of
patients with late-onset schizophrenia. Am J Psychiatry 1995;152(5):722730.
27. Vahia IV, Palmer BW, Depp C, et al. Is late-onset schizophrenia a subtype of
schizophrenia? Acta Psychiatr Scand. In press.

Continuum Lifelong Learning Neurol 2010;16(2)

Copyright @ American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

151

" COGNITIVE IMPAIRMENT

28. Young RC, Murphy CF, Heo M, et al. Cognitive impairment in bipolar disorder in
old age: literature review and findings in manic patients. J Affect Disord
2006;92(1):125131.
29. Depp CA, Jeste DV. Bipolar disorder in older adults: a critical review. Bipolar Disord
2004;6(5):343367.
30. Depp CA, Jin H, Mohamed S, et al. Bipolar disorder in middle-aged and elderly
adults: is age of onset important? J Nerv Ment Dis 2004;192(11):796799.
31. Brooks JO 3rd, Hoblyn JC. Secondary mania in older adults. Am J Psychiatry
2005;162(11):20332038.
32. Krauthammer C, Klerman GL. Secondary mania: manic syndromes associated with
antecedent physical illness or drugs. Arch Gen Psychiatry 1978;35(11):13331339.
33. Schouws SN, Comijs HC, Stek ML, et al. Cognitive impairment in early and late
bipolar disorder. Am J Geriatr Psychiatry 2009;17(6):508515.
34. Green MF. Cognitive impairment and functional outcome in schizophrenia and
bipolar disorder. J Clin Psychiatry, 2006;67(suppl 9):38;discussion 3642.
35. Martinez-Aran A, Vieta E, Colom F, et al. Cognitive impairment in euthymic
bipolar patients: implications for clinical and functional outcome. Bipolar Disord
2004;6(3):224232.
36. Martinez-Aran A, Vieta E, Torrent C, et al. Functional outcome in bipolar disorder:
the role of clinical and cognitive factors. Bipolar Disord 2007;9(12):103113.
37. Depp CA, Moore DJ, Sitzer D, et al. Neurocognitive impairment in middle-aged
and older adults with bipolar disorder: comparison to schizophrenia and normal
comparison subjects. J Affect Disord 2007;101(13):201209.
38. Depp CA, Savla GN, Moore DJ, et al. Short-term course of neuropsychological
abilities in middle-aged and older adults with bipolar disorder. Bipolar Disord
2008;10(6):684690.
39. Nasreddine ZS, Phillips NA, Bedirian V, et al. The Montreal Cognitive Assessment,
MoCA: a brief screening tool for mild cognitive impairment. J Am Geriatr Soc
2005;53(4):695699.

152

40. Folstein MF, Folstein SE, McHugh PR. Mini-mental state: a practical method for
grading the cognitive state of patients for the clinician. J Psychiatr Res
1975;12(3):189198.
41. Mattis S. Dementia rating scale. Odessa, FL: Psychological Assessment Resources, Inc, 1973.
42. Randolph C. RBANS. Manual: repeatable battery for the assessment of
neuropsychological status. San Antonio, TX: The Psychological Corporation, 1998.
43. Twamley EW, Doshi RR, Nayak GV, et al. Generalized cognitive impairments, ability
to perform everyday tasks, and level of independence in community living situations
of older patients with psychosis. Am J Psychiatry 2002;159(12):20132020.
44. Gold JM, Queern C, Iannone VN, Buchanan RW. Repeatable battery for the
assessment of neuropsychological status as a screening test in schizophrenia I:
sensitivity, reliability, and validity. Am J Psychiatry 1999;156(12):19441950.

Continuum Lifelong Learning Neurol 2010;16(2)

Copyright @ American Academy of Neurology. Unauthorized reproduction of this article is prohibited.