Mild cognitive impairment
Mild cognitive impairment is a syndrome defined as cognitive decline greater than expected for an individual's age and education level but that does not interfere notably with activities of daily life. Prevalence in population-based epidemiological studies ranges from 3% to 19% in adults older than 65 years. Some people with mild cognitive impairment seem to remain stable or return to normal over time, but more than half progress to dementia within 5 years. Mild cognitive impairment can thus be regarded as a risk state for dementia, and its identification could lead to secondary prevention by controlling risk factors such as systolic hypertension. The amnestic subtype of mild cognitive impairment has a high risk of progression to Alzheimer's disease, and it could constitute a prodromal stage of this disorder. Other definitions and subtypes of mild cognitive impairment need to be studied as potential prodromes of Alzheimer's disease and other types of dementia.
The Canadian Alzheimer Disease Review • June 2002 • 15
he study of cognitive changes
with aging has grown expo-
nentially over the past few years.
At one end of the spectrum is Alz-
heimer’s disease (AD), for which
the best estimates of lifetime risk
range from 14.5% to 26.2% of the
The Canadian Study
of Health and Aging (CSHA) has
estimated the prevalence of dem-
entia to be 8% in the population
over the age of 65 years.
other end of the spectrum are
changes in cognitive performance
that now are classified as normal.
Over a two-year period, retesting
elderly patients using memory
tests revealed that approximately
one third of them had subtle but
measurable memory decline over
On the WAIS-R IQ scale, an
85-year-old can achieve a score of
100 by correctly answering only
half as many questions as a
21-year-old. Hence, before claim-
ing that an elderly person shows
“cognitive impairment,” it is best
to compare that person to a group
of individuals similar in age.
Mild memory problems, falling
between the two poles of “normal”
and “dementia,” are a common phe-
nomenon in older people. The
CSHA documented a 16.8% preva-
lence of cognitive impairment with-
out dementia in the elderly.
Currently, the most widely used
term to characterize this group is
“mild cognitive impairment” (MCI),
derived from the World Health
Organization (WHO) and adapted
by a number of centres.
MCI is a
clinical label which includes elderly
subjects with short-term or long-
term memory impairment and with
no significant daily functional dis-
ability. The original diagnosis of
MCI required a subjective report of
cognitive decline from a former
level, gradual in onset, and present
for at least six months. This subjec-
tive complaint required supplemen-
tation with objective evidence of
memory and learning decline, with
other cognitive domains remaining
There was no
clear delineation as to how the pres-
ence of memory loss was to be
established. In all cases, the term
MCI excluded individuals with sig-
nificant depression, delirium, men-
tal retardation, or other psychiatric
disorders likely responsible for the
impairment. If the memory loss was
severe and accompanied by signifi-
Mild Cognitive Impairment
by Howard Chertkow, MD, FRCPC
Dr. Chertkow is a Neurologist in the
Departments of Neurology &
Neurosurgery and Geriatric Medicine
at the Sir Mortimer B. Davis-Jewish
General Hospital. He also is Director
of the Bloomfield Centre for Research
in Aging, Lady Davis Institute for
Medical Research, McGill University
and co-directs the Jewish General
Hospital/McGill Memory Clinic,
Mr. C is a 63-year-old executive who presents to
you complaining of memory loss. The problem
began insidiously over the past year and he
believes it has been getting worse. He previously
had difficulty remembering the names of
occasionally-seen employees, but now even the
details of important current accounts and recent
meetings escape him. He has started keeping
copious notes and lists, and double-checking
details with his secretary and wife (who confirms
the memory decline). There has been no impact
on his functional abilities at work and, according
to him and his wife, he remains otherwise
Mr. C denies experiencing depression. He admits
to a high level of stress at work, poor sleep,
occasional use of sedatives, but no alcohol
consumption. He is in good health, with controlled
hypertension. There is no history of transient
ischemic attack (TIA) or cerebral infarction.
Physical examination is normal and his
Folstein Mini-Mental State Examination (MMSE)
score is 28 (he only recalled one of three words
after a delay). His basic bloodwork results—
including B12, folate, thyroid-stimulating
hormone (TSH) and electrolytes—are normal, as
is a computed tomography (CT) scan. He asks you
for prognosis and treatment, and wonders whether
he should retire.
16 • The Canadian Alzheimer Disease Review • June 2002
cant functional impairment, the
individual met clinical criteria for
dementia, as opposed to MCI. The
intent behind the concept of MCI
was to capture and classify patients
who seem to have a cognitive prob-
lem that one would hesitate to label
as “normal,” but that is not severe
enough to qualify as dementia
MCI currently is the focus of
natural history studies and bio-
marker studies, and is an important
target group for future AD preven-
tion studies. Since the majority of
MCI individuals (who, after all, are
complaining of memory troubles)
are showing progressive memory
decline due to the presence of AD
pathology in its earliest stages, this
may be the optimum stage at
which to intervene with preventive
MCI will be used increasingly
as a label by neurologists, geriatri-
cians, and family physicians who
treat elderly cognitively impaired
individuals. It remains, however, a
“fuzzy”, elusive and controversial
concept, for reasons that will be
Current interest is predicated large-
ly on the fact that patients classified
under MCI have a high rate of con-
version to dementia, particularly
AD. In most clinic-based studies of
MCI, about 44% of individuals
meeting the criteria for diagnosis of
MCI progress to AD over three to
five years, equating to a conversion
rate of about 15% per year.
The controversial issue is
whether all MCI individuals pro-
gress to AD with time. In an
analysis by this author of 90 MCI
subjects, it appears that about
25% will not progress to AD, even
10 years after onset of memory
In population-based studies, the
prognosis of mild cognitive deficits
seems much less ominous. Ritchie
found that only 22% of such
a group went on to a degenerative
dementia, over an eight-year
follow-up. In contrast, Morris et al
have argued that, with sufficient
time, most or all subjects in this
category deteriorate to dementia,
Disagreement may centre on the
different diagnostic criteria being
used to assemble cohorts of sub-
jects for study. There are at least
seven different operational defini-
tions of groups presented or dis-
cussed as having MCI. The CSHA
study mentioned earlier utilized the
general term “cognitive impair-
ment, no dementia,” or CIND,
which included numerous subjects
that did not meet the MCI criteria
(e.g., some had alcoholism, suf-
fered strokes, or were mentally
It is estimated that 50% to 66%
of CIND subjects meet criteria for
MCI. On Reisberg’s Global Deteri-
oration Scale, most MCI individu-
als score as Stage 3—“Earliest sub-
tle deficits”—but it is not clear
whether the two concepts (CIND
and MCI) are interchangeable.
Some studies have utilized the
Clinical Dementia Rating (CDR)
a semi-structured interview
with patient and caregiver, which
assigns a rating of 0.5 (defined as
“questionable dementia”) to sub-
jects who are roughly equivalent to
those with MCI. It has been pointed
out, however, that a patient can be
CDR 0.5 and also meet diagnostic
criteria for AD and dementia, thus
making this grouping significantly
different from MCI. It is notable
that Morris et al (who have the most
pessimistic view of MCI prognosis)
have based their conclusions on
groups assembled using a CDR 0.5
rating as the inclusion criteria.
Another issue of disagreement
surrounds the criteria for determin-
ing objective memory loss—the
The intent behind the concept of MCI was to capture
and classify patients who seem to have a cognitive
problem that one would hesitate to label as “normal,”
but that is not severe enough to qualify as dementia.
General Criteria for Mild Cognitive Impairment
• Subjective complaint of memory loss
• Objective impairment of memory
• Other cognitive abilities generally preserved
• Preserved basic, day-to-day functioning
• No other obvious medical, neurologic or psychiatric explanation for the memory
• Criteria for dementia not met
Adapted from references 8 and 45.
The Canadian Alzheimer Disease Review • June 2002 • 17
key feature of MCI. In the clinic,
one might accept impaired three-
word recall as sufficient evidence
of memory impairment in a com-
plaining individual, as in the case
presented above. A number of MCI
studies have insisted, however, that
there must be abnormal perfor-
mance at least one standard devia-
tion (SD)—or even 1.5 SD—below
the norm for age-appropriate popu-
lations on standardized verbal
This attempt to
establish neuropsychological crite-
ria for MCI, while well-motivated,
is fraught with difficulty and would
restrict the diagnosis to centres
having easy access to neuro-
psychological evaluation. Petersen
has suggested that non-memory
domains be “generally intact” in
MCI, and in terms of bedside men-
tal status testing, this is a reason-
able criterion to apply.
It is possible, however, to
impose more stringent criteria for
MCI than the “fuzzy” clinical eval-
uations above. Ritchie and Touch-
suggested that for patients to
be considered as having MCI, all
other domains besides verbal de-
layed memory should be within
1.5 SD of the age-adjusted norm.
Unfortunately, when such con-
straints are firmly applied, few
MCI subjects continue to qualify as
In fact, detailed cognitive
and neuropsychological testing
shows that a range of subtle “low-
normal” test results is the most
common finding in groups of MCI
abnormalities, however, are not so
evident with the much cruder bed-
side testing carried out by medical
clinicians. Since one major goal of
using the terms “MCI” and “dem-
entia” is to allow general physic-
ians to better assess, diagnose and
treat patients, the diagnostic criteria
are more meaningful when viewed
with respect to clinical mental sta-
tus assessment and “fuzzy” bedside
testing, rather than when viewed as
being dependent on more accurate
(and often unavailable) neuropsy-
There are other problems with
over-reliance on neuropsychologi-
cal criteria for diagnosing MCI. For
example, many elderly individuals
are rated as “cognitively intact” but
are substantially impaired on tasks
that tap into a range of cognitive
there are “normal” patients who,
for their entire lives, fall at the bot-
tom of any Gaussian curve of neu-
ropsychological results. These indi-
viduals must not be falsely labelled
as having MCI.
Cultural, educational, and atten-
tional factors (i.e., impaired atten-
tion due to stress, anxiety or
depression) also can impact on
neuropsychological testing. Thus,
clinical judgement from a physi-
cian or psychologist is essential in
assessing whether an MCI label is
tests can be used, however, to sup-
port such clinical decision-making.
For the above reasons, it is typ-
ical to encounter subtle but signif-
icant heterogeneity in MCI
Perhaps MCI would be best
viewed as a general syndrome, akin
to the status of the term “dementia.”
Petersen describes most MCI
patients as showing only memory
impairment. These “amnesic” MCI
patients are more likely to go on to
Rarely, MCI subjects
show a non-memory domain af-
fected, while memory remains nor-
mal. Presumably, these subjects are
more likely to go on to another
diagnostic entity, such as Primary
This author, and others, have
found that most MCI patients show
subtle deficits in other neuropsy-
chological and reaction-time tests.
Clinicians should not be surprised
when an MCI individual performs
slightly abnormally on, for exam-
ple, placement of hands on clock
drawing. Similarly, many MCI sub-
jects (as in the case study above)
demonstrate subtle functional
changes; having to maintain writ-
ten lists of things might be consid-
ered a functional alteration. Such
changes should affect only “higher
functions,” and should not repre-
sent “significant impairment,” al-
though this is crudely defined.
Another important question is
whether patients with mild de-
pression should be excluded from
consideration. Studies have found
depression in up to 60% of MCI
patients that go on to develop AD,
and the presence of depression
may, in fact, be a useful prognostic
sign in MCI individuals.
As can be seen, there are many
diagnostic issues that arise in char-
Statistically speaking, there are “normal” patients who,
for their entire lives, fall at the bottom of any Gaussian
curve of neuropsychological results. These individuals
must not be falsely labelled as having MCI.
18 • The Canadian Alzheimer Disease Review • June 2002
When is MCI Just an
Earlier Stage of AD?
Numerous attempts have been
made to delineate prognostic
markers in MCI, most involving
rather small samples of subjects
followed for limited periods of
time. Some of these markers are
CDR scale. Daly et al
123 subjects, recruited through the
Boston media, who all met the cri-
teria of rating 0.5 on the CDR scale.
After a three-year follow-up, only
19% progressed to dementia (AD),
2% progressed due to strokes, and
5% became “normal.” It was found
that most of the group suffered
milder cognitive deficits compared
to those in other studies, and that
this explained the lower rate of pro-
gression/conversion to AD.
The CDR score is a single-
number rating derived from six
CDR categories, all of which can
be rated separately. A global CDR
of 0.5 can exist with a “sum of the
six categories” between 0.5 and
3.5. The categories represent lev-
els of increasing impairment
across a range of questions in a
semi-structured interview. Daly et
al found that, while 4% of subjects
scoring a CDR sum of 1.0 pro-
gressed to AD, 67% of those with
a total score of 3.0 progressed to
AD. Not surprisingly, these results
showed that, within the broad
range of individuals labelled with
mild memory loss, those most
impaired at onset were more like-
ly to progress over follow-up.
Unfortunately, the CDR is time-
consuming to administer (at least
60 to 90 minutes of clinician time),
and requires such sophistication to
use that it will remain limited to
ApoE status. There have been
suggestions that an individual’s
apolipoprotein (apoE) status might
be useful in determining risk for
progression to AD. Petersen et al
found a major effect of apoE-4 load
on progression to AD. However, in
this author’s study group of
90 MCI subjects,
three to five years, apoE-4 failed to
emerge as a useful predictor.
Homocysteine. One recent
study targets high homocysteine
levels as another biological risk
factor for progression to AD, but
this remains to be confirmed.
Imaging markers. Imaging
markers may be useful in the prog-
nosis of MCI, but they have been
utilized mainly in academic re-
search settings. Using magnetic
resonance image (MRI) scans, the
formal measurement of hippocam-
and the rate of
change in global or hippocampal
are more powerful pre-
dictors of progression to AD.
While visual inspection of stan-
dard single photon emission com-
puted tomography (SPECT) brain
images is not useful in prognos-
the sophisticated quantifica-
tion of SPECT images may allow
fairly accurate prognosis.
ing reports have emerged using
positron emission tomography
and magnetic res-
onset of AD in patients with MCI.
Imaging remains a very active
field of diagnostic investigation
which should provide helpful indi-
cators in the coming years.
Neuropsychological and cogni-
tive measures also have been
assessed for their predictive capac-
ity. This author and colleagues
found that MCI subjects with any
disorientation to time, or even sub-
tle problems with clock drawing,
are more likely to progress to AD.
In fact, the presence of any quan-
tifiable abnormalities, beyond
memory, constitute a risk for pro-
gression from MCI to dementia.
Batteries of different formal neu-
ropsychological tests have been
used together with the objective of
stratifying risk of progression to
Tests of delayed verbal re-
call and executive function appear to
have the best discriminating power
for prediction. While promising,
these tests are not sufficiently robust
to prognosticate in all MCI cases.
Other biomarkers. Currently, a
long series of other biomarkers
are under investigation. These
include smell testing,
heme oxygenase 1,
brospinal fluid (CSF) levels of tau
and beta-amyloid protein ending
at amino acid 42 (Abeta42),
note only a few.
Ideally, simple office tests are
needed that can accurately stratify
most MCI individuals, thereby
reserving the more sophisticated
testing for uncertain cases. This
author and colleagues have devel-
Using magnetic resonance image (MRI) scans, the
formal measurement of hippocampal volumes
rate of change in global or hippocampal volumes
are more powerful predictors of progression to AD.
The Canadian Alzheimer Disease Review • June 2002 • 19
oped a combination of low-tech and
high-tech markers with moderate
predictive power for MCI patients,
based on retrospective analysis of
our 90 MCI subjects. This approach
is being validated in a new set of
MCI subjects to determine whether
it is sufficiently robust for use in
prognosticating MCI cases.
Current Treatments for MCI
The three available cholinesterase
inhibitors in Canada are approved
for treatment of mild to moderate
AD—not MCI; in provinces that
reimburse such therapy, a diagno-
sis of MCI will lead to rejection of
Symptomatic treatment of the
memory complaints in MCI is, in
fact, generally disappointing. This
author’s anecdotal experience with
MCI patients, using cholinesterase
inhibitors, Ginkgo biloba, or stimu-
lants such as methylphenidate
), is that none of these
agents make a real impact on
patients’ mild memory loss. Since
there are no proven preventive ther-
apies for AD at the present time, it
follows that such therapies cannot
honestly be prescribed for MCI
patients. Having said this, it should
be noted that cholinesterase inhibi-
tors, various anti-inflammatories,
estrogen, statins, drugs to block
amyloid production, various anti-
oxidants, and vitamin E could all,
theoretically, delay or prevent pro-
gression to AD. Studies testing all
of these agents are underway. These
studies generally are several years
in duration, and results will begin to
reach the literature in the next 36
months (Table 2).
Given the lack of proven pharmaco-
logic therapies to prevent cognitive
decline, what therapeutic interven-
tions should clinicians take when
confronted with a patient such as the
one in the case study above?
Stress, lack of sleep, and use of
sedatives certainly need to be con-
sidered as contributing factors to
the memory problems of the patient
in the case study. Counselling
regarding these issues should be
given. Given our recent knowledge
of vascular factors interacting with
sense to aggressively treat vascular
risk factors in MCI individuals.
There is increasing evidence that
physical, leisure, and mentally
stimulating activities all have
effects in decreasing cognitive
decline and reducing AD risk.
These activities should be recom-
mended to patients with MCI.
Given the lack of clear prognos-
tic markers, an uncertain natural
history and, as discussed above, the
lack of proven therapies to prevent
cognitive decline, there is little con-
sensus on the management of MCI
Such patients should be
told that they meet criteria for MCI,
which has a real statistical risk (but
by no means any certainty) of pro-
gressing to dementia. This is vastly
preferable to giving no diagnosis or
to giving false assurances that “it’s
just normal aging.” Physicians also
should acknowledge their current
lack of certainty regarding progno-
sis, as well as the lack of useful tests
or proven biomarkers that should be
administered or considered.
One important intervention for
MCI patients is close follow-up.
Baseline mental status (preferably
with formal tools, such as the
MMSE) should be carried out.
Reversible causes of memory loss
should be sought using laboratory
tests or even brain imaging; these
also could be applied to AD
assessment. The patient should be
seen at six-month intervals, over a
long term. In previous work, this
Preventing Progression to AD: Drugs Under Investigation
• Cholinesterase inhibitors
• Cholesterol-lowering drugs
• Anti-amyloid drugs (beta-secretase inhibitors, gamma-secretase inhibitors,
glycosaminoglycan [GAG] mimetics, amyloid immunotherapy)
• Various antioxidants, including vitamin E
• Nootropics and psychic stimulants (e.g., piracetam)
Tests of delayed verbal recall and executive function
appear to have the best discriminating power for
prediction. While promising, these tests are not
sufficiently robust to prognosticate in all MCI cases.
20 • The Canadian Alzheimer Disease Review • June 2002
author and colleagues found that,
for reversible dementias, the
chances of reversal were higher if
the causes were detected early.
It probably is not appropriate
to advise patients with MCI to
stop work immediately, or to
insist that they stop driving. The
studies of Daly and Ritchie both
noted that a subset of MCI indi-
viduals improved to normal
over follow-up, and that MCI
alone usually will not affect
functioning at most work-
Physicians also should acknow-
ledge to patients that there currently
is no sufficiently substantiated,
government-approved treatment for
MCI. Treating MCI individuals (or
the healthy elderly) to prevent sub-
sequent AD, using cholinesterase in-
hibitors, anti-inflammatories, estro-
gen, statins, various antioxidants or
even vitamin E, represent prescrip-
tion beyond proven therapies.
In this author’s opinion, it is
advisable to refer eager MCI
patients to a research clinic where
randomized, controlled, clinical tri-
als of the above agents and other
preventive medications are under-
way. Only when the results of MCI
drug studies begin to reach the lit-
erature will we be in a position to
make informed scientific recom-
mendations for such therapies
(alone or in combinations) for the
benefit of our patients.
1. Drachman, D. If we live long enough, will we all be demented?
Neurology 1994; 44:1563-5.
2. Canadian Study of Health and Aging (CSHA) Working Group. Study meth-
ods and prevalence of dementia. Can Med Assoc J 1994; 150:899-913.
3. Collie A, Maruff P, et al. Memory decline in healthy older people: impli-
cations for identifying mild cognitive impairment. Neurology 2001;
4. Graham JE, Rockwood K, et al. Prevalence and severity of cognitive
impairment with and without dementia in an elderly population. Lancet
5. Ebly EM, Hogan DB, et al. Cognitive impairment in the nondemented
elderly. Results from the Canadian Study of Health and Aging (CSHA).
Arch Neurol 1995; 52(6):612-9.
6. Smith GE, Petersen RC, et al. Definition, course, and outcome of mild
cognitive impairment. Aging Neuropsychology & Cognition 1996;
7. Petersen RC, Stevens JC, et al. Practice parameter: early detection of
dementia: mild cognitive impairment (an evidence-based review). Report
of the Quality Standards Subcommittee of the American Academy of
Neurology. Neurology 2001; 56(9): 1133-42.
8. Petersen RC, Smith GE, et al. Mild cognitive impairment: Clinical charac-
terization and outcome. Correction. Arch Neurol 1999; 56(3):303-8.
9. Petersen RC, Smith GE, et al. Apolipoprotein E status as a predictor of the
development of Alzheimer’s disease in memory-impaired individuals
[published erratum appears in JAMA 1995; 274(7):538]. JAMA 1995;
10. Grundman M, Petersen RC, et al. ADCS Cooperative Study. Rate of
dementia of the Alzheimer type (DAT) in subjects with mild cognitive
impairment. Neurology 1996; 46:A403.
11. Chertkow H, Verret L, et al. Predicting progression to dementia in elderly
subjects with mild cognitive impairment—a multidisciplinary approach.
2001 Contemporary Clinical Issues Plenary Session. 53rd Annual Meeting
American Academy of Neurology, Philadelphia, 2001.
12. Ritchie K, Leibovici D, et al. A typology of subclinical senescent cogni-
tive disorder. Br J Psychiatry 1996; 168(4):470-6.
13. Morris JC, Storandt M, et al. Mild cognitive impairment represents early-
stage Alzheimer disease. Arch Neurol 2001; 58(3):397-405.
14. Kluger A, Ferris SH, et al. Neuropsychological prediction of decline to
dementia in nondemented elderly. Journal of Geriatric Psychiatry &
Neurology 1999; 12(4):168-79.
15. Hughes C, Berg L, et al. A new clinical scale for the staging of dementia.
Br J of Psychiatry 1982; 140:566-72.
16. Ritchie K, Touchon J. Mild cognitive impairment: conceptual basis and
current nosological status. Lancet 2000; 355(9199):225-8.
17. Ritchie K, Artero S, et al. Classification criteria for mild cognitive impair-
ment: A population-based validation study. Neurology 2001; 56(1):37-42.
18. Chertkow HM, Bergman H, et al. Standard neuropsychological tests do
not predict development of Alzheimer’s Disease in individuals with “Age-
associated Cognitive Decline.” Can J Neurol Sciences 1998;
19. Craik F. Memory changes in normal aging. Current Directions in
Psychological Science 1994; 3(5):155-8.
20. Lautenschlager NT, Riemenschneider M, et al. Primary degenerative mild
cognitive impairment: Study population, clinical, brain imaging and bio-
chemical findings. Dementia & Geriatric Cognitive Disorders 2001;
21. Visser P, Verhey F. Distinction between preclinical Alzheimer’s disease and
depression. Journal of the American Geriatric Society 2000; 48:479-84.
22. Daly E, Zaitchik D, et al. Predicting conversion to Alzheimer disease
using standardized clinical information. Arch Neurol 2000; 57(5):675-80.
23. Lehmann M, Gottfries CG, Regland B. Identification of cognitive impair-
ment in the elderly: Homocysteine is an early marker. Dementia &
Geriatric Cognitive Disorders 1999; 10:12-20.
24. Jack C, Petersen R, et al. Prediction of AD with MRI-based hippocampal
volume in mild cognitive impairment. Neurology 1999; 52(7):1397-403.
25. Fox NC. Increased rates of atrophy in early and preclinical AD: studies
with registration of serial MRI. Neurobiology of Aging 2000; 21:S74.
26. Jack CR Jr., Petersen RC, et al. Rates of hippocampal atrophy correlate with
change in clinical status in aging and AD. Neurology 2000; 55(4):484-9.
27. McKelvey R, Bergman H, et al. Lack of prognostic significance of SPECT
abnormalities in non-demented elderly subjects with memory loss. Can J
Neuroscience 1999; 26:23-8.
28. Johnson KA, Jones K, et al. Preclinical prediction of Alzheimer’s disease
using SPECT. Neurology 1998; 50:1563-71.
29. Reiman EM, Caselli RJ, et al. Preclinical evidence of Alzheimer’s disease
in persons homozygous for the epsilon 4 allele for apolipoprotein E.
NEJM 1996; 334(12):752-8.
30. Kantarci K, Jack CR Jr., et al. Regional metabolic patterns in mild cogni-
tive impairment and Alzheimer’s disease: A 1H MRS study. Neurology
31. Bozoki A, Giordani B, Heidebrink J, et al. Mild cognitive impairments
predict dementia in non-demented elderly patients with memory loss.
Arch Neurol 2001; 58:411-6.
32. Tierney MC, Szalai JP, et al. Prediction of probable Alzheimer’s disease in
memory-impaired patients: A prospective longitudinal study. Neurology
33. Devanand DP, Folz M, et al. Questionable dementia: clinical course and
predictors of outcome. Journal of the American Geriatrics Society 1997;
34. Devanand DP, Michaels-Marston K, Liu X, et al. Olfactory deficits in
patients with mild cognitive impairment predict Alzheimer’s Disease at
follow-up. American Journal of Psychiatry 2000; 157:1399-1405.
35. Albert MS, Moss MB, et al. Preclinical prediction of AD using neuropsy-
chological tests. Journal of the International Neuropsychological Society
36. Graves AB, Bowen JD, et al. Impaired olfaction as a marker for cognitive
decline: Interaction with apolipoprotein E epsilon 4 status. Neurology
37. Schipper H, Chertkow H, Mehindate K, et al. Evaluation of heme oxyge-
nase-1 as a systemic biological marker of sporadic AD. Neurology 2000;
38. Andreasen N, Minthon L, et al. Evaluation of CSF-tau and CSF-Abeta42
as diagnostic markers for Alzheimer disease in clinical practice. Arch
Neurol 2001; 58(3):373-9.
39. Snowdon DA, Greiner LH, et al. Brain infarction and the clinical expres-
sion of Alzheimer disease: The nun study. JAMA 1997; 277(10):813-7.
40. Forette F, Seux ML, et al. Prevention of dementia in randomised, double-
blind, placebo-controlled Systolic Hypertension in Europe (Syst-Eur) trial.
Lancet 1998; 352(9137):1347-51.
41. Launer LJ, Ross GW, et al. Midlife blood pressure and dementia: the
Honolulu-Asia aging study. Neurobiology of Aging 2000; 21(1):49-55.
42. Laurin D, Verreault R, Lindsay J, et al. Physical activity and risk of cognitive
impairment and dementia in elderly persons. Arch Neurol 2001; 58:498-504.
43. Scarmeas N, Levy G, Tang M, et al. Influence of leisure activity on the
incidence of Alzheimer’s disease. Neurology 2001; 57:2236-42.
44. Freter S, Bergman H, et al. Prevalence of potentially reversible dementias
and actual reversibility in a memory clinic cohort. Can Med Assoc J
45. Golomb J, Kluger A, Garrard P, and Ferris S. Clinicians Manual on Mild
Cognitive Impairment, Science Press, London, U.K., 2001.