/clinical/,/clinical/cckm-tools/,/clinical/cckm-tools/content/,/clinical/cckm-tools/content/questionnaires/,/clinical/cckm-tools/content/questionnaires/related/,

/clinical/cckm-tools/content/questionnaires/related/name-97092-en.cckm

201606168

page

100

UWHC,UWMF,

Tools,

Clinical Hub,UW Health Clinical Tool Search,UW Health Clinical Tool Search,Questionnaires,Related

Montreal Cognitive Assessment Validation Study for Mild Cognitive Impairment and Alzheimer Disease

Montreal Cognitive Assessment Validation Study for Mild Cognitive Impairment and Alzheimer Disease - Clinical Hub, UW Health Clinical Tool Search, UW Health Clinical Tool Search, Questionnaires, Related


Montreal Cognitive Assessment
Validation Study for Mild Cognitive Impairment
and Alzheimer Disease
Sandra Freitas, PhD,* Ma´rio Rodrigues Simo˜es, PhD,*
Lara Alves, PhD,* and Isabel Santana, PhD, MDwz
Abstract: The Montreal Cognitive Assessment (MoCA) was recently
proposed as a cognitive screening test for milder forms of cognitive
impairment, having surpassed the well-known limitations of the Mini-
Mental State Examination (MMSE). This study aims to validate the
MoCA for screening Mild Cognitive Impairment (MCI) and Alzheimer
disease (AD) through an analysis of diagnostic accuracy and the pro-
posal of cut-offs. Patients were classified into 2 clinical groups ac-
cording to standard criteria: MCI (n=90) and AD (n=90). The 2
control groups (C-MCI: n=90; C-AD: n=90) consisted of cogni-
tively healthy community dwellers selected to match patients in sex,
age, and education. The MoCA showed consistently superior psycho-
metric properties compared with the MMSE, and higher diagnostic
accuracy to discriminate between MCI (area under the curve=0.856;
95% confidence interval, 0.796-0.904) and AD patients (area under the
curve=0.980; 95% confidence interval, 0.947-0.995). At an optimal
cut-off of below 22 for MCI and below 17 for AD, the MoCA achieved
significantly superior values in comparison with MMSE for sensitivity,
specificity, positive predictive value, negative predictive value, and
classification accuracy. Furthermore, the MoCA revealed higher sen-
sitivity to cognitive decline in longitudinal monitoring. This study
provides robust evidence that the MoCA is a better cognitive tool than
the widely used MMSE for the screening and monitoring of MCI and
AD in clinical settings.
Key Words: MoCA, neuropsychological test, cognitive screening,
Mild Cognitive Impairment, Alzheimer disease
(Alzheimer Dis Assoc Disord 2013;27:37–43)
C
ognitive impairment and dementia are the major health
issues among older people. Alzheimer disease (AD) is the
most common neurodegenerative disorder with a prevalence
of 4.4% for those older than 65 years and represents at least
60% of all dementia cases.
1
The serious impact of AD in
health care systems worldwide
2,3
and the dramatic projec-
tions for the coming years
4,5
stress the need for new effective
strategies that are able to slow down or stop the disease
progression. It is now generally accepted that prodromal AD
is the ideal time window for disease-modifying therapies.
Mild Cognitive Impairment (MCI) is considered a
transitional stage between normal cognitive aging and im-
paired cognition caused by several pathologies, most fre-
quently AD. This state of continuum is characterized by a
deterioration in cognitive functioning greater than expected
for the person’s age and educational level; however, it does
not cause significant functional disability and is insufficient
to establish the diagnosis of dementia.
6–9
Longitudinal
studies show that these patients progress to overt dementia
at a rate of 10% to 15% per year, compared with a rate of
1% to 2% in control subjects.
9
This explains why MCI is
now the focus of prediction studies and the target of clinical
trials of new disease-modifying therapies.
The early screening of cognitive impairment and its
differentiation from age-related decline is thus extremely
important. A brief and sensitive cognitive screening tool is
indispensable in dealing with this gray boundary area of
normality between normal aging, MCI, and mild dementia.
The Montreal Cognitive Assessment (MoCA)
10
is a novel
international brief cognitive screening tool developed for the
detection of MCI and mild AD that may be suitable for this
purpose. Previous studies have shown that the MoCA is
useful and accurate in identification of milder forms of cog-
nitive impairment, having revealed a high sensitivity in the
detection of MCI and AD patients.
11–18
One of the reasons
for the good sensitivity of the test is that it allows a more
comprehensive assessment of the major cognitive domains,
compared with other screening tests. These domains include
executive function, short-term memory, language skills, and
visuospatial processing. Furthermore, it has been shown that
the MoCA’s total score is an accurate quantitative estimate
of the global cognitive ability in mild and moderate
stages.
19,20
Thus, beyond routine screening, the MoCA scores
can be used in longitudinal studies as an indicator of the
global cognitive decline during progression of the disease.
21
The aim of the present study was to validate the
MoCA
10,22
for cognitive screening of MCI and AD pa-
tients. This was carried out by the analysis of its diagnostic
accuracy and the by the establishment of optimal cut-off
points to detect MCI and AD patients. The data from
a longitudinal study with MCI and AD patients have
also been analyzed to establish the MoCA’s sensitivity for
cognitive decline in a short period of time.
METHODS
Design
In the current study, 3 groups of participants were
considered: (I) the MCI group; (II) the AD group; and (III)
the control group. Patients were recruited from the
Received for publication September 2, 2011; accepted November 12,
2011.
From the *Faculty of Psychology and Educational Sciences, University
of Coimbra, Coimbra, Portugal; wFaculty of Medicine, University of
Coimbra; and zNeurology Department of the Coimbra University
Hospital, Coimbra, Portugal.
Supported by the Fundac¸a˜o para a Cieˆncia e Tecnologia [Portuguese
Foundation for Science and Technology] through of a PhD fellowship
(SFRH/BD/38019/2007) and PIC/IC/83206/2007.
The authors declare no conflicts of interest.
Reprints: Sandra Freitas, PhD, Faculty of Psychology and Educational
Sciences, University of Coimbra, Rua do Cole´ gio Novo, Apartado
6153, 3001-802 Coimbra, Portugal (e-mail: sandrafreitas0209@gmail.
com).
Copyright
r
2013 by Lippincott Williams & Wilkins
ORIGINAL ARTICLE
Alzheimer Dis Assoc Disord

Volume 27, Number 1, January–March 2013 www.alzheimerjournal.com
|
37

Dementia Clinic, Neurology Department of the Coimbra
University Hospital (Coimbra University Hospital, Coim-
bra, Portugal). Control subjects were selected from the
database of the MoCA’s normative study for the Portu-
guese population
23
to match patients in sex, age, and edu-
cational level. Two subgroups of patients belonging to both
clinical groups (MCI and AD) were assessed at a second
time point for preliminary longitudinal analysis.
Participants
The total study sample comprised 360 participants
distributed between 3 subgroups: (I) the MCI group with 90
patients; (II) the AD group with 90 patients; and (III) the
control group with 180 cognitively healthy adults. The
demographic data of the participants in each group are
provided in Table 1.
To exclude other causes of cognitive decline apart
from a degenerative process, all patients were examined by
a neurologist (I.S.), and a standard investigation was
always carried out, including routine laboratory examina-
tions/analyses (apolipoprotein E genotyping) and imaging
studies [structural (computed tomography and/or magnetic
resonance imaging) and functional (single-photon emission
computed tomography)]. Positron emission tomography
and cerebrospinal fluid analysis were carried out more
restrictively, although these techniques were considered
in younger patients. All patients underwent a battery of
comprehensive neuropsychological assessment tests com-
prising at least the following tools: Mini-Mental State
Examination (MMSE) (M. Guerreiro, unpublished doctoral
dissertation),
24
Alzheimer Disease Assessment Scale,
25,26
Clinical Dementia Rating scale (CDR),
27,28
Irregular Word
Reading Test (TeLPI)
29
for premorbid intelligence estima-
tion, Subjective Memory Complaints scale,
30,31
and Geri-
atric Depression Scale.
32,33
The MoCA was never used for
diagnostic purposes. The diagnosis was established by a
multidisciplinary team consensus considering the results of
the comprehensive assessment and based on international
criteria for MCI of the Petersen workgroup
7
and probable
AD.
34,35
The MCI group included patients classified as
“amnestic MCI” (single or multidomain)
8
with a classi-
fication of 0.5 in the CDR. The AD group included only
those patients with mild-to-moderate severity (classified
with CDRr2 and MMSEZ12 points).
Control group participants were selected, as referred
above, from the database of the MoCA’s normative study
for the Portuguese population.
23
Each patient was matched
to a cognitively healthy adult on the basis of variables
shown to affect the MoCA’s performance (educational level
and age)
23
and also on sex, resulting in a perfect match
between MCI and associated controls (then designated as
the C-MCI group) and between AD and associated controls
(C-AD group). Details regarding the controls’ recruitment
procedure, inclusion and exclusion criteria, and neuro-
psychological assessment have been described in the
previous study.
23
Procedures
All participants were recruited between September
2008 and July 2010, and each participant was assessed in a
single session by an expert in neuropsychology. Only pa-
tients with a stable clinical condition (without significant
comorbidities), a complete clinical evaluation, and already
with a well-established diagnosis, according to the above
international criteria, were considered eligible for this
study. For each patient considered suitable for the study
and at the time of data collection, a diagnosis was recorded
by the neurologist in the clinical file. These restrictive cri-
teria imposed the exclusion of 30 patients who were still
waiting for data considered essential in the differential di-
agnosis between AD and other dementias and those whose
classification between MCI and AD was not fully estab-
lished by the multidisciplinary team. In addition, at the
outset of this study, the exclusion criteria taken into
account in the patients’ selection were: higher dementia
severity (CDR>2 and MMSE<12 points), recent psychi-
atric comorbidities or therapeutic changes (6mo before the
current neuropsychological evaluation), and significant
motor, visual, or auditory deficits, all of which may influ-
ence the neuropsychological assessment results.
For the preliminary analysis of the MoCA’s sensitivity
to global cognitive decline in longitudinal monitoring, we
assessed 2 subgroups of patients (35 with MCI and 40 with
AD) at a second time point, on average 176.81±67.09
days apart (minimum=63; maximum=340).
The present research complied with the ethical guide-
lines for human experimentation stated in the Declaration
of Helsinki and was approved by the Ethics Board of
Coimbra University Hospital, by the “Fundac¸a˜o para a
Cieˆncia e Tecnologia” (Portuguese Foundation for Science
and Technology), and by the Faculty of Psychology and
Educational Sciences Scientific Committee. An informed
consent was obtained from all the participants after the
aims and research procedures were fully explained by a
member of the study group. For the AD patient who was
incapable of providing consent on his/her behalf, a legal
representative provided it.
TABLE 1. Descriptive Statistics for the Sample’s Subgroups
n Education Age Sex MMSE MoCA
MCI 90 6.50±4.565 70.52±7.950 55 (61.1) 27.08±2.395 18.31±3.868
C-MCI 90 6.53±4.498 69.59±7.053 55 (61.1) 28.88±1.297 23.64±3.223
AD 90 6.23±4.119 74.22±8.212 52 (57.8) 20.88±4.091 10.06±4.410
C-AD 90 6.24±4.128 73.10±7.539 52 (57.8) 28.09±1.577 22.33±3.471
Clinical group 180 6.37±4.338 72.37±8.270 107 (59.4) 23.98±4.565 14.18±5.851
Control group 180 6.39±4.307 71.34±7.490 107 (59.4) 28.48±1.493 22.99±3.404
Total 360 6.38±4.316 71.86±7.895 214 (59.4) 26.23±4.073 18.59±6.503
Sex is characterized by female’s n and respective percentage (%). Data of other variables are presented as mean±SD.
AD indicates Alzheimer disease; C-AD, subgroup of controls matched with AD patients; clinical group, all patients with MCI and AD; C-MCI, subgroup of
controls matched with MCI patients; control group, all controls; MCI, Mild Cognitive Impairment; MMSE, Mini-Mental State Examination (maximum
score=30); MoCA, Montreal Cognitive Assessment (maximum score=30).
Freitas et al Alzheimer Dis Assoc Disord

Volume 27, Number 1, January–March 2013
38
|
www.alzheimerjournal.com
r
2013 Lippincott Williams & Wilkins

Neuropsychological Testing and Materials
In the clinical interview, the demographic and clinical
data were collected through a complete sociodemographic
questionnaire and an inventory of past habits, current clinical
health status, and medical history. Following this, the same
neuropsychologist administered the MMSE (M. Guerreiro,
unpublished doctoral dissertation)
24
and the MoCA,
10,22
in
that order for all the subjects. The MMSE is a widely rec-
ognized and used brief screening tool for cognitive decline;
therefore, it is not described in detail here. Both the MMSE
and the MoCA are in paper-and-pencil format and are scored
out of a possible total score of 30 points, with higher scores
indicating better cognitive performance. The MoCA was
developed to screen milder forms of cognitive impairment
through the assessment of 6 cognitive domains: executive
functions; visuospatial abilities; short-term memory; lan-
guage; attention, concentration, and working memory; and
temporal and spatial orientation.
10
It is composed of a 1-page
test, with an application time of approximately 10 to 15 mi-
nutes, and of a manual in which explicit instructions on its
administration and scoring system are available. The cultural
adaptation process of the MoCA for the Portuguese pop-
ulation involved the translation, retroversion, and linguistic
improvement of the tool and of the administration and
scoring instruction manual, studies with the MoCA’s Portu-
guese experimental version, the revision and adjustments re-
quired to finalize the MoCA’s Portuguese final version, and
an analysis of the equivalence between the original and the
Portuguese final version, as described by Freitas et al.
36
In the
current study, the MoCA’s total score refers to the raw score
without correction point for education effects, considered in
the original study,
10
because this correction is not used in the
Portuguese population.
23
Statistical Analysis
Statistical analyses were carried out using the Stat-
istical Package for the Social Sciences (version 19.0, IBM
SPSS, Chicago, IL). Descriptive statistics were used for the
samples’ characterization, and the w
2
test and the 2-sample
t test allowed comparisons between the groups. Cronbach a
was considered as an index of internal consistency. To as-
sess test-retest reliability, intraclass correlation coefficients
between scores at baseline and at follow-up after 3 and 18
months for the control patients were calculated. Interrater
reliability was calculated using the Pearson correlation co-
efficient between the scoring of 2 independent evaluators.
The convergent validity was determined using Pearson
correlation coefficients between the MoCA scores and
MMSE scores. The group differences were examined using
the 2-sample t test and analysis of covariance. The
preliminary data of the longitudinal study were analyzed
using a paired-sample t test.
The diagnostic accuracy of the MoCA and the MMSE
for the prediction of the clinical diagnosis of MCI and AD
was assessed through the receiver operating characteristics
(ROC) curve analysis implemented in MedCalc (version
11.6, MedCalc Software, Mariakerke). In this analysis, the
areas under the curve (AUC) can vary between 0.5 and 1,
with larger AUC indicating better diagnostic accuracy. The
ROC curves were compared according to the AUC com-
parison method of Hanley and McNeil.
37
The optimal cut-
off points for each screening instrument that yielded the
highest Youden index were selected, with higher Youden
index indicating maximization of the sensibility and spe-
cificity. For the analysis of the predictive value of these
tests, we calculated, for each cut-off point, the sensitivity
(the probability for subjects with cognitive impairment to
have a positive test), specificity (the probability for subjects
without cognitive impairment to have a negative test),
positive predictive value (PPV, the probability of disease in
subjects who have a positive test), negative predictive value
(NPV, probability of the classification “lack of disease”
in subjects who have a negative test), and classification
accuracy (probability of correct classification of subjects
with or without cognitive impairment).
RESULTS
Sample Characterization
Characteristics of the study sample, and in more detail
of all the subgroups, are provided in Table 1. For this de-
scription, we considered the following variables: sample size,
educational level, age, sex, MMSE score, and MoCA score.
As mentioned above, the control participants were
selected from the database of MoCA’s normative study for
the Portuguese population
23
to match the educational level,
age, and sex of patients of the clinical groups. No statisti-
cally significant differences were found in educational level
[t(178)=0.049, P=0.961], age [t(178)=0.833, P=0.406],
and sex [w
2
(1)=0.000, P=1.0] between the MCI and
C-MCI groups. Similarly, the AD and C-AD groups did
not differ in educational level [t(178)=0.018, P=0.986],
age [t(178)=0.955, P=0.341], and sex [w
2
(1)=0.000,
P=1.0]. The MCI group and the AD group did not differ
in educational level [t(178)=0.411, P=0.681] and sex
[w
2
(1)=0.092, P=0.761]; however, the AD patients were
significantly older than MCI patients [t(178)=3.071,
P=0.002], because of which the average onset of symp-
toms of MCI precedes the onset of AD.
Psychometric Properties
Cronbach a of the MoCA as an index of internal
consistency was 0.903 for the total study sample, and the
respective value for the MMSE was 0.856. Regarding the
analysis carried out to determine which of the MoCA items
could be eliminated to increase consistency, the results in-
dicate that none should be excluded. Cronbach a values for
the subgroups are provided in Table 2. The test-retest reli-
ability was measured through the intraclass correlation co-
efficient between baseline and follow-up data. This analysis
was carried out only for the subsample of the control group
in 2 follow-up settings: 3 months (n=30; on average
146.87±42.937d apart; minimum=68d and maximum=
200d) and 18 months (n=30; on average 515.04±
154.195d apart; minimum=101d and maximum=676d).
The obtained MoCA values were, respectively, 0.909
and 0.877 and the corresponding values for MMSE were,
respectively, 0.755 and 0.665 (Table 2). Interrater reliability
data were collected from a subsample of 60 tested partic-
ipants of all groups, and the obtained intraclass correlation
index for the MoCA was 0.988. Another observation
was that MoCA scores were highly and positively associated
with MMSE scores (total study sample, r=0.849,
P<0.001), which is indicative of convergent validity.
The correlation values for the subgroups are presented
in Table 2.
Group Differences
When analyzing the total sample, the MoCA scores were
lower in the AD group than in all other groups and were lower
Alzheimer Dis Assoc Disord

Volume 27, Number 1, January–March 2013 MoCA in MCI and AD
r
2013 Lippincott Williams & Wilkins www.alzheimerjournal.com
|
39

in the MCI group than in both control groups. The MoCA
scores did not differ between the control groups
[t(178)=2.626, P=0.225] (Table 1). Furthermore, we can
observe that there were statistically significant differences when
MoCA scores were compared between MCI and C-MCI
groups [t(178)=10.050, P<0.001, mean difference=5.333
±0.531] and between AD and C-AD groups [t(178)=
20.756, P<0.001, mean difference=12.278±0.592]. Because
AD patients were significantly older than MCI patients, the
analysis of differences in scores between clinical groups was
carried out using an analysis of covariance to control for the
effects of age. It can be observed that the differences between
MCI and AD patients’ scores [F (1177)=160.052, P<0.001,
Z
2
=0.48, mean difference=7.930±0.627] were in fact sig-
nificant. The corresponding values for the MMSE were as
follows: (I) the MCI and C-MCI group: t(178)=6.270,
P<0.001, mean difference=1.800±0.287; (II) the AD
and C-AD group: t(178)=15.603, P<0.001, mean differ-
ence=7.211±0.462; and (III) the MCI and AD group:
F(1177)=146.899, P<0.001, Z
2
=0.45, mean difference=
6.231±0.514. These results indicate that, although the dif-
ferences in the MMSE scores are statistically significant, the
score differences obtained with the MoCA are more pro-
nounced. A more detailed analysis reveals that there were
statistically significant differences in all cognitive domains of
the MoCA in the 3 comparisons: (I) the MCI and C-MCI
group; (II) the AD and C-AD group; and (III) the MCI and
AD group. Table 3 summarizes the results.
Cut-off Points
The ROC curve analysis was carried out and the pre-
dictive values were determined to evaluate the diagnostic
accuracy of MoCA to discriminate MCI and AD patients
from cognitively healthy adults. Graphic representations of
the ROC curves are provided in Figure 1.
It can be observed that both ROC curves that were
referred to the MoCA fully include the curve for the
MMSE, which is a clear indication that there is always a
cut-off for the MoCA with higher sensitivity and specificity
for any cut-off chosen for the MMSE. The discriminant
potential of the MoCA for MCI was high, with an AUC of
0.856 [95% confidence interval (CI), 0.796-0.904], and that
for AD was excellent, with an AUC of 0.980 (95% CI,
0.947-0.995). In contrast, corresponding values for MMSE
were 0.745 (95% CI, 0.674-0.807) and 0.957 (95% CI,
0.916-0.981). The AUCs for MCI are significantly different
(z=3.372, P=0.0007), according to the AUC comparison
method of Hanley and McNeil,
36
indicating different classi-
fication accuracies of the tools for milder cognitive impair-
ment. No statistically significant differences were found
between the AUCs for AD (z=1.636, P=0.1018). The op-
timal cut-off point for maximum accuracy (Youden
index) and the respective values of sensitivity, specificity, PPV,
NPV, and classification accuracy are described in Table 4.
The cut-off point of below 22 yielded the greatest
Youden index for the MoCA in discrimination between
MCI and controls. With this cut-off point, MoCA had a
good sensitivity (81%), specificity (77%), PPV (78%), NPV
(80%), and classification accuracy (80%), and all these
values were significantly superior compared with the re-
spective values for the MMSE. Furthermore, with respect
to the capacity of discrimination between AD patients and
controls, once again the MoCA demonstrated excellent
sensitivity (88%), specificity (98%), PPV (98%), NPV
(89%), and classification accuracy (93%) at the optimal
TABLE 3. Group Differences in Cognitive Domains of the MoCA
Cognitive Domains MCI and C-MCI AD and C-AD MCI and AD
Executive functions t(178)=4.975, P<0.001 t(178)=9.766, P<0.001 t(178)=7.073, P<0.001
Visuospatial skills t(178)=5.564, P<0.001 t(178)=9.616, P<0.001 t(178)=7.006, P<0.001
Short-term memory t(178)=9.773, P<0.001 t(178)=20.732, P<0.001 t(178)=6.581, P<0.001
Language t(178)=2.964, P=0.003 t(178)=8.800, P<0.001 t(178)=7.010, P<0.001
Attention, concentration, and working memory t(178)=5.199, P<0.001 t(178)=11.123, P<0.001 t(178)=7.217, P<0.001
Temporal and spatial orientation t(178)=2.974, P=0.003 t(178)=13.886, P<0.001 t(178)=12.038, P<0.001
AD indicates Alzheimer disease; C-AD, subgroup of controls matched with AD patients; C-MCI, subgroup of controls matched with MCI patients; MCI,
Mild Cognitive Impairment.
TABLE 2. Psychometric Properties
Internal Consistency Reliability
Cronbach a Test-Retest Convergent Validity
MoCA MMSE MoCA MMSE Interrater Correlations MoCA/MMSE
MCI (n=90) 0.723 0.617 3mo: 0.909
18mo: 0.877
3mo: 0.755
18mo: 0.665
0.988 0.601
AD (n=90) 0.824 0.771 0.700
C-MCI (n=90) 0.648 0.457 0.637
C-AD (n=90) 0.677 0.402 0.600
Total (n=360) 0.903 0.856 0.849
Correlation values at a significant level, P<0.01.
AD indicates Alzheimer disease; C-AD, subgroup of controls matched with AD patients; C-MCI, subgroup of controls matched with MCI patients; MCI, Mild
Cognitive Impairment; MMSE, Mini-Mental State Examination (maximum score=30); MoCA, Montreal Cognitive Assessment (maximum score=30).
Freitas et al Alzheimer Dis Assoc Disord

Volume 27, Number 1, January–March 2013
40
|
www.alzheimerjournal.com
r
2013 Lippincott Williams & Wilkins

cut-off of below 17 points, and again all these values were
more favorable than the respective values for the MMSE.
Preliminary Analysis of the Longitudinal Study
For the preliminary analysis of the MoCA’s sensitivity
to global cognitive decline during longitudinal monitoring,
2 clinical subgroups of patients (35 with MCI and 40 with
AD) were assessed at a second time point, on average
176.81±67.09 days apart (minimum=63; maximum=
340). When considering all patients (n=75), statistically
significant differences in MoCA scores were observed be-
tween both assessments [t(74)=4.278, P<0.001], in contrast
to what was found with the MMSE [t(74)=1.871, P=065].
A similar analysis for each clinical subgroup showed stat-
istically significant differences on MoCA scores for both MCI
[t(34)=2.612, P=0.014] and AD patients [t(39)=0.5651,
P<0.001]. An equivalent analysis using the MMSE revealed
that the differences were significant for the AD group
[t(39)=2.824, P=0.008], whereas for MCI the MMSE
showed no sensitivity to cognitive decline [t(34)=1.873,
P=0.070]. A more detailed and parceled analysis of the
cognitive domains of the MoCA also revealed interesting
results. When considering the total sample, the differences
between the 2 evaluations were significant for visuospatial
skills [t(74)=2.487, P=0.015]; short-term memory [t(74)=
2.669, P=0.009]; attention, concentration, and working
memory [t(74)=2.213, P=0.030]; and temporal and spatial
orientation [t(74)=4.449, P<0.001]; the differences were
without significance for language and executive functions.
When considering the clinical subgroups, an isolated sig-
nificant difference was found for MCI patients in the short-
term memory domain [t(34)=2.390, P=0.023], whereas the
same analysis for the AD subgroup revealed statistical sig-
nificance for attention, concentration, and working memory
[t(39)=2.071, P=0.045] and also for orientation [t(39)=
5.244, P<0.001].
DISCUSSION
The main objective of this study was to validate the
MoCA as a cognitive screening tool for MCI and AD. The
results confirm its tremendous potential and provide robust
evidence that the MoCA is a better tool for this purpose in
FIGURE 1. ROC curve analysis of the MoCA (dark gray) and MMSE (medium gray) to detect MCI (left) and AD (right). AD indicates
Alzheimer disease; MCI, Mild Cognitive Impairment; MMSE, Mini-Mental State Examination; MoCA, Montreal Cognitive Assessment;
ROC, receiver operating characteristics.
TABLE 4. Diagnostic Classification Accuracy
Cut-off AUC Sensitivity Specificity PPV NPV Classification Accuracy
MCI
MoCA <22 0.856 81 77 78 80 80
MMSE <29 0.745 67 72 71 48 69
AD
MoCA <17 0.980 88 98 98 89 93
MMSE <26 0.957 85 93 93 87 89
Sensitivity, specificity, PPV, NPV, and classification accuracy values were expressed in percentage.
Cut-off values indicate the minimum score required for absence of signal.
AD indicates Alzheimer disease; AUC, area under the operating characteristic curve; MCI, Mild Cognitive Impairment; MMSE, Mini-Mental State
Examination (maximum score=30); MoCA, Montreal Cognitive Assessment (maximum score=30); NPV, negative predictive value; PPV, positive
predictive value.
Alzheimer Dis Assoc Disord

Volume 27, Number 1, January–March 2013 MoCA in MCI and AD
r
2013 Lippincott Williams & Wilkins www.alzheimerjournal.com
|
41

comparison with the widely used MMSE. In fact, it was
verified that the correlation coefficient between the 2 cog-
nitive screening tools was moderate to good, suggesting
convergent validity. Nevertheless, the psychometric prop-
erties of the MoCA examined both in the total sample and
in each subgroup showed good properties and were re-
vealed to be consistently superior to those of the MMSE.
As was discussed previously, we believe that the 2 main
reasons for the higher results of the MoCA at this level were
as follows: first, the inclusion of the executive function as-
sessment; and second, the consideration of more complex
tasks to measure short-term memory, language, attention,
concentration, working memory, and visuospatial skills.
Moreover, the analysis of group differences indicates
that both instruments are able to distinguish between the
clinical and control groups. However, the differences between
the groups were much more pronounced when the MoCA
was used, in comparison with the MMSE, which is reflected
in the consistently higher mean differences of the MoCA.
Furthermore, we observed statistically significant differences
in all cognitive domains of the MoCA and in all group
comparisons. These results confirm the higher capacity of the
MoCA to discriminate between normal aging and pathologic
cognitive decline, as well as between MCI and dementia.
The ROC curve analysis of the MoCA compared with
the MMSE also showed that the MoCA exhibits a better
diagnostic accuracy to discriminate MCI and AD patients
from cognitively healthy adults. In our sample, the ideal cut-
off point reached was lower than the original cut-off of 26
proposed by the authors,
10
as in other published re-
sults.
14,15,17,18
We observed that at an optimal cut-off point
below 22 for MCI, the MoCA had values significantly su-
perior to the MMSE for sensitivity (81%), specificity (77%),
PPV (78%), NPV (80%), and classification accuracy (80%).
With an optimal cut-off of below 17 points for AD, the
MoCA once again showed better results than the MMSE on
sensitivity (88%), specificity (98%), PPV (98%), NPV (89%),
and classification accuracy (93%). These results confirm that
the MoCA is a better cognitive screening tool for the de-
tection of MCI and AD conditions compared with the
MMSE, showing overall superior discrimination validity.
The capacity of the MoCA to identify different severity levels
of cognitive decline justifies the pertinence of considering
different cut-off points for MCI and dementia. This approach
seems to be more useful and informative than a single cut-off
point for cognitive decline as suggested in other studies,
particularly in the original work of Nasreddine et al.
10
An additional observation based on the present study
regards the extremely poor diagnostic accuracy of the
MMSE to identify MCI, which is reflected in overall low
results, mainly in poor sensitivity (67%), classification ac-
curacy (69%), and very poor NPV (48%). This is a clear
indication that, whenever the MMSE is used to screen for
milder forms of cognitive decline, the probability of false-
negative cases is very high. This is especially critical because
of the current emphasis placed upon the early detection of
cognitive impairment. Nevertheless, the MMSE remains the
most commonly used screening tool despite the widely re-
ferred limitations in the literature. Our results are a clear
argument in favor of these opinions.
Finally, considering our analysis of the sensitivity of
the MoCA to cognitive decline in patients that were
monitored longitudinally, we could demonstrate evidence
of decline in a short period of time. Furthermore, beyond
its capacity to quantify cognitive decline, the MoCA also
provides comprehensive information on the differential
profile of clinical deterioration in MCI and AD.
We believe that the added value of the present study is
the rigorous methodology used. It included the following:
(I) well-validated study samples (patients with mis-
classification and more advanced dementia were excluded,
both characteristics capable of compromising the analysis
of the discriminant capacity of the tools); (II) homogeneity
of the clinical groups; (III) a control sample with subjects
recruited from the community and well-characterized
as cognitively healthy adults; (IV) equivalent sample sizes
(which reduces the possible biases of sample sizes in
statistical analysis); (V) perfect matching between groups
regarding sociodemographic characteristics that have a
significant influence on the performance of MoCA; and
(VI) rigorous application of MoCA with no interrater
variability (all participants were assessed by the same
experienced neuropsychologist).
However, some limitations of the current study must be
addressed. First of all, because only the amnestic subtype of
MCI (single or multidomain) was considered, the general-
ization of the results to other forms of MCI should be done
cautiously. Similarly, although the Portuguese final version of
the MoCA resulted in a rigorous process that followed the
methodological guidelines for cultural adaptation studies, and
the maximum equivalence between the original tool and the
Portuguese final version of MoCA was pursued,
36
the gen-
eralization of these results to other target populations should
be done cautiously. In contrast, the present study compares
people with a clear diagnosis of AD/MCI with healthy people
who do not present health and cognitive difficulties, like the
majority of the clinical validation studies of screening tools.
However, in the context of clinical applicability of a cognitive
screening tool, such as the MoCA, the most common diag-
nostic challenge is to identify clinical conditions among people
with complaints of memory impairment or other cognitive
difficulties or psychological disorders. Hence, we believe that
such a question represents a very interesting challenge with a
clear practical utility that should be a part of future efforts
within this field of research. Finally, despite being a promising
tool, the results of the preliminary analysis of the longitudinal
evaluation require the corroboration by an ongoing study with
longer follow-up and more robust samples.
In conclusion, this study produced considerable evi-
dence of the overall superiority of the MoCA in compar-
ison with the MMSE as a global cognitive assessment tool
with respect to discriminant validity and diagnostic accu-
racy. This was confirmed by the identification of MCI and
AD and by the discrimination between both forms of
cognitive decline and normal cognitive aging. Furthermore,
the results suggest that the MoCA is sensitive to cognitive
decline in a short period of time and may capture profiles of
cognitive deterioration along the evolution of the disease.
Thus, this study shows a clear advantage in the use of the
MoCA compared with the use of the MMSE and brings
together arguments for the use of the MoCA as a reliable
brief cognitive tool, which should be recommended both for
screening and follow-up in primary clinical setting and
geriatric health care.
REFERENCES
1. Lobo A, Launer LJ, Fratiglioni L, et al. Prevalence of dementia
and major subtypes in Europe: a collaborative study of
population-based cohorts. Neurology. 2000;54(suppl 5):S4–S9.
Freitas et al Alzheimer Dis Assoc Disord

Volume 27, Number 1, January–March 2013
42
|
www.alzheimerjournal.com
r
2013 Lippincott Williams & Wilkins

2. Comas-Herrera A, Northey S, Wittenberg R, et al. Future
costs of dementia-related long-term care: exploring future
scenarios. Int Psychogeriatr. 2011;23:20–30.
3. Langa KM, Chernew ME, Kabeto MU, et al. National
estimates of the quantity and cost of informal caregiving for
the elderly with dementia. J Gen Intern Med. 2001;16:770–778.
4. Ferri CP, Prince M, Brayne C, et al. Global prevalence of
dementia: a Delphi consensus study. Lancet. 2005;366:2112–2117.
5. Wimo A, Winblad B, Aguero-Torres H, et al. The magnitude
of dementia occurrence in the world. Alzheimer Dis Assoc
Disord. 2004;17:63–67.
6. Petersen RC. Mild cognitive impairment: transition between
aging and Alzheimer’s disease. Neurology. 2000;15:93–101.
7. Petersen RC. Mild cognitive impairment as a diagnostic entity.
J Intern Med. 2004;256:183–194.
8. Petersen RC. Mild cognitive impairment. Contin Lifelong
Learn Neurol. 2007;13:15–38.
9. Petersen RC, Smith GE, Waring SC, et al. Mild cognitive
impairment: clinical characterization and outcome. Arch
Neurol. 1999;56:303–308.
10. Nasreddine Z, Phillips NA, Be´ dirian V, et al. The Montreal
Cognitive Assessment, MoCA: a brief screening tool for Mild
Cognitive Impairment. J Am Geriatr Soc. 2005;53:695–699.
11. Smith T, Gildeh N, Holmes C. The Montreal Cognitive
Assessment: validity and utility in a memory clinic setting. Can
J Psychiatry. 2007;52:329–332.
12. Rahman TT, Gaafary MM. Montreal Cognitive Assessment
Arabic version: reliability and validity prevalence of mild
cognitive impairment among elderly attending geriatric clubs
in Cairo. Geriatr Gerontol Int. 2009;9:54–61.
13. Wen HB, Zhang ZX, Niu FS, et al. The application of
Montreal Cognitive Assessment in urban Chinese residents of
Beijing. Chin J Intern Med. 2008;47:36–39.
14. Lee J, Lee DW, Cho S, et al. Brief screening for mild cognitive
impairment in elderly outpatient clinic: validation of the
Korean version of the Montreal Cognitive Assessment.
J Geriatr Psychiatry Neurol. 2008;21:104–110.
15. Luis CA, Keegan AP, Mullan M. Cross validation of the
Montreal Cognitive Assessment in community dwelling older
adults residing in the Southeastern US. Int J Geriatr
Psychiatry. 2009;24:197–201.
16. Fujiwara Y, Suzuki H, Yasunaga M, et al. Brief screening toll
for mild cognitive impairment in older Japanese: validation of
the Japanese version of the Montreal Cognitive Assessment.
Geriatr Gerontol Int. 2010;10:225–232.
17. Zhao S, Guo C, Wang M, et al. A clinical memory battery for
screening for amnestic mild cognitive impairment in an elderly
Chinese population. J Clin Neurosci. 2011;18:774–779.
18. Damian AM, Jacobson SA, Hentz JG, et al. The Montreal
Cognitive Assessment and the Mini-Mental State Examination
as screening instruments for cognitive impairment: item
analyses and threshold scores. Dement Geriatr Cogn Disord.
2011;31:126–131.
19. Koski L, Xie H, Finch L. Measuring cognition in a geriatric
outpatients clinic: rash analysis of the Montreal Cognitive
Assessment. J Geriatr Psychiatry Neurol. 2009;22:151–160.
20. Koski L, Xie H, Konsztowicz S. Improving precision in the
quantification of cognition using the Montreal Cognitive
Assessment and the Mini-Mental State Examination. Int
Psychogeriatr. 2011;1:1–9.
21. Freitas S, Santana I, Simo˜es MR. The sensitivity of the MoCA
and MMSE to cognitive decline: a longitudinal study. Alz-
heimers Dement. 2010;6:S353–S354.
22. Simo˜es MR, Freitas S, Santana I, et al. Montreal Cognitive
Assessment (MoCA): Versa˜o portuguesa [Portuguese version].
Coimbra, Servic¸o de Avaliac¸a˜o Psicolo´ gica da Faculdade de
Psicologia e de Cieˆncias da Educac¸a˜o da Universidade de
Coimbra [Psychological Assessment Department, Faculty of
Psychology and Educational Sciences, University of Coimbra],
2008.
23. Freitas S, Simo˜es MR, Alves L, et al. Montreal Cognitive
Assessment (MoCA): normative study for the Portuguese
population. J Clin Exp Neuropsychol . 2011;33:989–996. doi:
10.1080/13803395.2011.589374.
24. Folstein M, Folstein S, McHugh P. Mini-Mental State: a
practical method for grading the cognitive state of patients for
the clinician. J Psychiatr Res. 1975;12:189–198.
25. Rosen WG, Mohs RC, Davis KL. A new rating scale for
Alzheimer’s disease. Am J Psychiatry. 1984;141:1356–1364.
26. Guerreiro M, Fonseca S, Barreto J, et al. Escala de avaliac¸a˜o
da Doenc¸a de Alzheimer [Alzheimer’s disease assessment
scale]. In: Grupo de Estudos de Envelhecimento Cerebral e
Demeˆncias [Study Group on Brain Aging and Dementia], eds.
Escalas e testes na demeˆncia [Scales and Tests in Dementia].
Lisbon: GEECD; 2008:42–68.
27. Hughes CP, Berg L, Danziger WL, et al. A new clinical
scale for the staging of dementia. Br J Psychiatry. 1982;140:
566–572.
28. Garret C, Santos F, Tracana I, et al. Avaliac¸a˜o Clı´ nica da
Demeˆncia [Clinical Dementia Rating Scale]. In: Grupo de
Estudos de Envelhecimento Cerebral e Demeˆncias [Study
Group on Brain Aging and Dementia], eds. Escalas e testes
na demeˆncia [Scales and Tests in Dementia]. Lisbon: GEECD;
2008:17–32.
29. Alves L, Simo˜es MR, Martins C. Teste de Leitura de Palavras
Irregulares (TeLPI) [Reading Word Irregular Test]. Coimbra,
Servic¸o de Avaliac¸a˜o Psicolo´ gica da Faculdade de Psicologia e
de Cieˆncias da Educac¸a˜o da Universidade de Coimbra
[Psychological Assessment Department, Faculty of Psychology
and Educational Sciences, University of Coimbra], 2009.
30. Schmand B, Jonker C, Hooijer C, et al. Subjective memory
complaints may announce dementia. Neurology. 1996;46:121–125.
31. Gino´ S, Mendes T, Ribeiro F, et al. Escala de Queixas de
Memo´ ria [Memory Complaints Scale]. In: Grupo de Estudos de
Envelhecimento Cerebral e Demeˆncias [Study Group on Brain
Aging and Dementia], eds. Escalas e testes na demeˆncia [Scales
and Tests in Dementia]. Lisbon: GEECD; 2008:117–120.
32. Yesavage JA, Brink TL, Rose TL, et al. Development and
validation of a geriatric depression screening scale: a prelim-
inary report. J Psychiatr Res. 1983;17:37–49.
33. Barreto J, Leuschner A, Santos F, et al. Escala de Depressa˜o
Geria´ trica [Geriatric Depression Scale]. In: Grupo de Estudos
de Envelhecimento Cerebral e Demeˆncias [Study Group on
Brain Aging and Dementia], eds. Escalas e testes na demeˆncia
[Scales and Tests in Dementia]. Lisbon: GEECD; 2008:69–72.
34. American Psychiatric Association. Diagnostic and Statistical
Manual of Mental Disorders (DSM-IV-TR). 4th ed. Wash-
ington, DC: American Psychiatric Association; 2000.
35. McKhann G, Drachman D, Folstein M, et al. Clinical
diagnosis of Alzheimer’s disease: report of the NINCDS-
ADRDA Work Group under the auspices of the Department
of Health and Human Services Task Force on Alzheimer’s
disease. Neurology. 1984;34:939–944.
36. Freitas S, Simo˜es MR, Martins C, et al. Estudos de adaptac¸a˜odo
Montreal Cognitive Assessment (MoCA) para a populac¸a˜o
portuguesa [Montreal Cognitive Assessment (MoCA): Portuguese
adaptation studies]. Avaliac¸a˜oPsicolo´gica. 2010;9:345–357.
37. Hanley JA, McNeil BJ. A method of comparing the areas
under receiver operating characteristic curves. Radiology.
1983;148:839–843.
Alzheimer Dis Assoc Disord

Volume 27, Number 1, January–March 2013 MoCA in MCI and AD
r
2013 Lippincott Williams & Wilkins www.alzheimerjournal.com
|
43