This article was downloaded by: [FU Berlin]

On: 07 May 2015, At: 00:09

Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered
office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Aging, Neuropsychology, and Cognition:

A Journal on Normal and Dysfunctional
Development
Publication details, including instructions for authors and
subscription information:
http://www.tandfonline.com/loi/nanc20

Age- and education-adjusted normative

data for the Montreal Cognitive
Assessment (MoCA) in older adults age
70–99
a a a
Michael Malek-Ahmadi , Jessica J. Powell , Christine M. Belden ,
b b c b
Kathy O’Connor , Linda Evans , David W. Coon & Walter Nieri
a
Banner Sun Health Research Institute Cleo Roberts Center for
Click for updates Clinical Research, Sun City, AZ, USA
b
Banner Sun Health Research Institute Center for Healthy Aging,
Sun City, AZ, USA
c
College of Nursing and Health Innovation, Arizona State
University, Phoenix, AZ, USA
Published online: 05 May 2015.

To cite this article: Michael Malek-Ahmadi, Jessica J. Powell, Christine M. Belden, Kathy
O’Connor, Linda Evans, David W. Coon & Walter Nieri (2015): Age- and education-adjusted
normative data for the Montreal Cognitive Assessment (MoCA) in older adults age 70–99, Aging,
Neuropsychology, and Cognition: A Journal on Normal and Dysfunctional Development, DOI:
10.1080/13825585.2015.1041449

To link to this article: http://dx.doi.org/10.1080/13825585.2015.1041449

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the
“Content”) contained in the publications on our platform. However, Taylor & Francis,
our agents, and our licensors make no representations or warranties whatsoever as to
the accuracy, completeness, or suitability for any purpose of the Content. Any opinions
and views expressed in this publication are the opinions and views of the authors,
and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content
should not be relied upon and should be independently verified with primary sources
of information. Taylor and Francis shall not be liable for any losses, actions, claims,
proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or
 howsoever caused arising directly or indirectly in connection with, in relation to or arising
out of the use of the Content.

This article may be used for research, teaching, and private study purposes. Any
substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,
systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &
Conditions of access and use can be found at http://www.tandfonline.com/page/terms-
and-conditions
Downloaded by [FU Berlin] at 00:09 07 May 2015
 Aging, Neuropsychology, and Cognition, 2015
http://dx.doi.org/10.1080/13825585.2015.1041449

Age- and education-adjusted normative data for the Montreal

Cognitive Assessment (MoCA) in older adults age 70–99
Michael Malek-Ahmadi a*, Jessica J. Powella, Christine M. Beldena, Kathy O’Connorb,
Linda Evansb, David W. Coonc and Walter Nierib
a
Banner Sun Health Research Institute Cleo Roberts Center for Clinical Research, Sun City, AZ,
USA; bBanner Sun Health Research Institute Center for Healthy Aging, Sun City, AZ, USA; cCollege
of Nursing and Health Innovation, Arizona State University, Phoenix, AZ, USA
(Received 22 September 2014; accepted 11 April 2015)

The original validation study for the Montreal Cognitive Assessment (MoCA) suggests
Downloaded by [FU Berlin] at 00:09 07 May 2015

a cutoff score of 26; however, this may be too stringent for older adults, particularly for
those with less education. Given the rapidly increasing number of older adults and
associated risk of dementia, this study aims to provide appropriate age- and education-
adjusted norms for the MoCA. Data from 205 participants in an ongoing longevity
study were used to derive normative data. Individuals were grouped based on age
(70–79, 80–89, 90–99) and education level (≤12 Years, 13–15, ≥16 Years). There were
significant differences between age and education groups with younger and more
educated participants outperforming their counterparts. Forty-six percent of our sample
scored below the suggested cutoff of 26. These normative data may provide a more
accurate representation of MoCA performance in older adults for specific age and
education stratifications.
Keywords: cognitive impairment; cognitive screening; cognitive decline; Alzheimer’s
disease; dementia

Introduction
The Montreal Cognitive Assessment (MoCA) is a brief cognitive screening tool that has
demonstrated good diagnostic accuracy in identifying amnestic Mild Cognitive Impairment
(aMCI) as well as dementias such as Alzheimer’s disease (AD) (Nasreddine et al., 2005).
The original MoCA validation study (sample mean age 74.92 years; mean education
11.88 years) suggested using a cutoff score of 26 to identify “normal” versus “impaired”
performance (Nasreddine et al., 2005). This original cutoff score has been contested by
several researchers who believe that it is too stringent and may misclassify cognitively
normal individuals as being impaired (Coen, Cahill, & Lawlor, 2011; Damian et al., 2011;
Luis, Keegan, & Mullan, 2009; Rossetti, Lacritz, Munro Cullum, & Weiner, 2011). In fact,
Damian et al. (2011) found that a cutoff score of 24 provided optimal diagnostic accuracy
for populations with a high prior probability of cognitive impairment (i.e., elderly popula-
tions) and a cross-validation study by Luis et al. (2009) found that scores below 23 were
most useful. Further, a recent study by Gluhm et al. (2013) found that cognitively normal
older adults aged 70–89, had MoCA scores that ranged from 19 to 30. Taken together, these
studies support the notion that normal performance on the MoCA is relatively wide and that
scores below 26 are quite common among cognitively normal older adults.

*Corresponding author. Email: [email protected]

© 2015 Taylor & Francis

 2 M. Malek-Ahmadi et al.

In addition to a stringent cutoff score, another potential downfall of the MoCA norms
is the lack of attention to individual education status. The original validation study
suggested a one-point education correction for those with ≤12 years; however, this may
not adequately capture the positive correlation between education and cognitive abilities
(Jefferson et al., 2011). Specifically, Freitas and colleagues (2012) found that 49% of
MoCA score variability was attributed to age and education.
The above studies highlight the need for more accurate MoCA cutoff scores that
incorporate various age groups as well as education levels. Surprisingly, there appears to
be only one study (Rossetti et al., 2011) that has attempted to stratify MoCA norms by age
and education and these researchers found clinically significant differences between
education groups. For example, they found the mean MoCA score for individuals aged
70–80 with less than 12 years of education was 16.07 compared to 20.35 for those with
12 years of education and further compared to 23.60 for those with more than 12 years of
education, resulting in an approximate 7.5 point difference between those with less than
and greater than 12 years of education. Additionally, for this sample, the suggested 26
point cutoff would imply that well over half of the participants were impaired, with
Downloaded by [FU Berlin] at 00:09 07 May 2015

significant impairment for those with less education.

While Rossetti et al. (2011) normative data are certainly beneficial, a limitation is that
the norms only extend to age 80. The need for normative data in older age groups is
crucial as the fastest growing age segment of the US’ population is for those aged 85 and
above (National Institute on Aging, 2008). Further, utilization of screening measures such
as the MoCA increases for these older adults as the concern of neurodegenerative disease
also increases with age. Though many of the previous studies include samples of
individuals well into their ninth and tenth decades, they do not also account for individual
education differences.
Given the lack of normative data for MoCA performance for the oldest of old and
education status, the aim of this study is to provide both age- and education-adjusted
normative data on the MoCA for older adults aged 70–99.

Materials and methods

Study participants
Data from 205 participants in an ongoing study on longevity were used for this analysis.
Participants are assessed annually on a number of cognitive, medical, psychosocial, and
demographic measures to identify factors associated with increased longevity. Participants
were recruited from the Phoenix, Arizona metropolitan area through advertisements,
community talks, and referrals from individuals already participating in the study. All
participants signed an informed consent form prior to participating. Approval for the study
was granted by the Institutional Review Board. The initial query of the database yielded
data for 324 participants age 70 and older, however 119 were removed based on the
inclusion/exclusion criteria listed below.

Inclusion/exclusion criteria
Individuals between the ages 70 and 99 were included in the study. Data from each
individual’s first MoCA administration was used. Participants were excluded if they had a
history of stroke, transient ischemic attack, dementia, or other neurological condition that
could affect their performance. Additional exclusion criteria included scores below 26 on
 Aging, Neuropsychology, and Cognition 3

the Mini Mental State Exam (MMSE) (Folstein, Folstein, & McHugh, 1975), and scores
equal to or higher than 16 on the Center for Epidemiologic Studies Depression scale
(CES-D) (Radloff, 1977). Of note, a score of 26 on the MMSE was chosen as this is a
typical standard for aMCI clinical trials (range from 23 to 26; Stephan et al., 2013), and in
an effort to ensure that our sample only included cognitively normal individuals, this score
was deemed most appropriate.

Statistical analysis
Descriptive statistics were derived to provide demographic characteristics of the sample.
A two-sample t-test was conducted to examine gender differences on MoCA scores.
MoCA normative data for age and education was derived by dividing the sample into
nine groups; first by stratifying the sample into three age groups (70–79, 80–89, 90–99)
and then further classifying them into three education groups (≤12, 13–15, and
≥16 Years). A 3 × 3 analysis of variance (ANOVA) was used to examine the relationships
between the age and education groups on MoCA total score. Post-hoc groupwise compar-
Downloaded by [FU Berlin] at 00:09 07 May 2015

isons were carried out using the Tukey honestly significant difference test. The suggested
one point correction for those with ≤12 years of education was not applied as the aim of
the study was to generate normative data rather than a cutoff score. A post-hoc compar-
ison of the percent of individuals who fell below the cutoff 26, both with and without the
one point education adjustment, was also carried to examine the degree of misclassifica-
tion using the recommended cutoff.

Results
Demographic characteristics of the study sample are presented in Table 1. The mean age
of the study sample was 84.67 years (SD = 7.88). Of the 205 participants, there were 65
males and 140 females. MoCA scores were not significantly different between males and
females, t(203) = −0.84, P = 0.40. MoCA scores correlated moderately with MMSE
scores, r = 0.48, P < 0.001. MoCA scores showed no correlation with CES-D scores,
r = −0.02, P = 0.74 indicating a lack of significant association between MoCA scores and
the presence of depressive symptoms.
The participants were divided into nine groups as described above and a 3 × 3
ANOVA was conducted to evaluate the effects of age and education on MoCA perfor-
mance. The means and standard deviations of MoCA scores as a function of the two
factors (age and education) are presented in Table 2. The ANOVA indicated no significant
interaction between age and education, F(4, 196) = 0.46, P = 0.77, partial η2 = 0.01, but

Table 1. Demographic characteristics by age group with mean CES-D, MMSE, and MoCA scores.

70–79 80–89 90–99 Total

n 53 89 63 205
Males/Females (n) 12/41 28/61 25/38 65/140
CES-D 3.47 (3.36) 4.00 (3.82) 3.81 (3.65) 3.81 (3.64)
MMSE 29.25 (0.96) 28.60 (1.50) 28.28 (1.42) 28.91 (1.18)
MoCA 26.49 (2.46) 25.09 (3.04) 23.73 (3.01) 25.03 (3.06)

Notes: CES-D = Center for Epidemiologic Studies Depression scale; MMSE = Mini Mental State Exam;
MoCA = Montreal Cognitive Assessment; Mean (standard deviation).
 4 M. Malek-Ahmadi et al.

Table 2. MoCA normative data stratified by age and education.

≤12 Years 13–15 Years ≥16 Years

70–79 25.25 (4.11) 27.78 (2.24) 27.59 (2.04)

n=4 n = 27 n = 22
80–89 23.47 (2.97) 25.08 (3.13) 25.82 (2.75)
n = 15 n = 40 n = 34
90–99 23.00 (2.63) 23.35 (3.43) 24.61 (2.59)
n = 14 n = 26 n = 23
Note: Mean (SD).

yielded significant main effects for age, F(2, 196) = 11.76, P < 0.001, partial η2 = 0.10,
and education, F(2, 196) = 3.22, P = 0.004, partial η2 = 0.05. Post-hoc analyses to the
main effect for age indicated that older participants tended to have lower scores on the
MoCA than their younger counterparts (70–79 > 80–89, P = 0.02; 70–79 > 90–99,
Downloaded by [FU Berlin] at 00:09 07 May 2015

P < 0.001; 80–89 > 90–99, P = 0.02). Post-hoc analyses for the education main effect
indicated that the ≤12 and the 13–15 groups had significantly lower scores than the ≥16
group (13–15 lower than ≥16 Years, P = 0.01; ≤12 lower than ≥16 Years, P = 0.01). The
≤12 and 13–15 groups were not significantly different (P = 0.89).
In general, the MoCA performance for the entire study sample was relatively low
compared to the standard cutoff of 26. The mean MoCA score for the entire sample was
25.03 (SD = 3.06) which is approximately one point lower than the cutoff score suggested
by Nasreddine et al. (2005). Using the suggested one-point correction for those with
education levels equal to or less than 12 years (Nasreddine et al., 2005), our sample’s
MoCA mean rose slightly to 25.29 (SD = 2.99). When unadjusted scores were used 51%
of the study sample scored below the recommended cutoff of 26 and when using the one-
point correction 46% of our sample still scored below 26.

Discussion
The purpose of this study was to enhance the utility of the MoCA by creating expanded
age- and education-adjusted normative data for older adults aged 70–99. Participants with
more years of education performed better than their less-educated counterparts. Similarly,
there were significant differences on MoCA performances between age groups, with
younger participants outperforming older. When compared to the MMSE there were
larger differences between each of the age-groups on the MoCA, which could suggest
that the MoCA is more sensitive to age-related changes in cognition than the MoCA.
Although a significant interaction between the age and education was not found, our
normative data suggest clinically significant relevance. For example, our data showed that
those aged 70–79 with ≥16 years of education had a mean MoCA score of 27.59
compared to those aged 90–99 with ≤12 years of education who attained a mean score
of 23.00. This difference of over four points emphasizes the need for stratified norms that
include age and education. While cutoff scores are useful for quickly determining whether
an individual might require a more extensive evaluation, age- and education-corrected
normative scores provide a more refined estimate of cognitive performance and may
reduce the occurrence of misclassification at screening. Given that older adults often
worry that subjective memory complaints may be indicative of dementia onset (Ossher,
Flegal, & Lustig, 2013), the use of these normative data will allow clinicians to be more
 Aging, Neuropsychology, and Cognition 5

confident in deciding whether reported cognitive changes maybe age-associated or

whether they warrant further assessment to determine their etiology.
Consistent with previous studies (Damian et al., 2011; Freitas et al., 2012; Luis et al.,
2009; Rossetti et al., 2011), our findings also suggest that lower thresholds for classifying
cognitive impairment on the MoCA are needed particularly when older and less-educated
individuals are being assessed. Even with the suggested 1-point education correction, a
significant proportion of our participants (46%) scored below the recommended cutoff
score of 26. Not surprisingly, this difference is most prominent with the oldest age group
(90–99) and those with less education (≤12 Years). This implies that the cutoff of 26 and
one-point correction may not be adequate when assessing older adults with varying
education levels. We propose utilizing the normative data presented in Table 2, which
provides means and standard deviations of MoCA scores stratified by age and educational
attainment.
The primary advantage of utilizing a normative approach versus a cutoff score
approach is that means and standard deviations may be stratified by a number of different
factors (i.e., age and education) which allows for a more accurate estimate of cognitive
Downloaded by [FU Berlin] at 00:09 07 May 2015

performance to be made. Although cutoff scores may provide good indications of

sensitivity and specificity for differentiating clinical groups from control groups, they
are not able to account for significant confounding factors which increases the likelihood
of misclassification. The results of this study underscore this point as significant MoCA
score differences were noted for both age group and education group. Thus, the use of
normative data allows clinicians to make a more refined estimate of an individual’s
performance which reflects contribution of factors such as age and education. Dai and
colleagues (2013) found that both age and education are responsible for significant
sources of variation on MMSE scores in older adults, which highlights the importance
of the need to account for age and education on cognitive screening instruments.
O’Bryant and colleagues (2008) found that in a sample of highly educated older adults
a cutoff score of either 27 or 28 on the MMSE provided optimal diagnostic accuracy in
identifying cognitive impairment. The findings of O’Bryant et al. (2008) may be proble-
matic given that the normal range for MMSE scores is between 24 and 30. Since many
clinicians are familiar with and have likely established the 24–30 range as being normal
for their patient populations, it might be difficult for them utilize a “normal” score as a
cutoff for impairment in a specific segment of their patient population. In this case, the use
of an age- and education-adjusted normative score approach is advantageous and would
provide the clinician with a more accurate assessment.
It is important to note that decreased performance on cognitive tests does not
necessarily equate to decreased functional status. Although the MoCA correlates well
with the Clinical Dementia Rating Sum of Boxes (r = −0.62) (Malek-Ahmadi, Davis,
Belden, & Sabbagh, 2014), premorbid educational and intellectual status will significantly
impact cognitive test performance (Jefferson et al., 2011). The result is that older adults
with differing educational backgrounds, but who are similar in terms of functional status,
may have very different MoCA scores which also underscores the importance of account-
ing for age and education in cognitive test performance.
A recommended area of future research is to establish cutoff scores in addition to
normative data as such scores are practical in situations that necessitate rapid and
accurate estimations of cognition. Related to this, another potential direction would be
to determine whether classification of impairment using these normative data corre-
sponds with specific clinical diagnoses (i.e., aMCI, AD, etc.) in an independent group.
Another recommendation would be to complete a study with more ethnic diversity as
 6 M. Malek-Ahmadi et al.

all participants in this study self-identified as non-Hispanic white. Rossetti et al. (2011)
demonstrated that Caucasians had significantly higher MoCA scores than other ethnic
groups, so our normative data should be used with caution when working with
ethnically diverse populations. Related to this, caution should be used when applying
these normative data to individuals whose native tongue is not English.
Another possible limitation of this study is that our sample, as a whole, has higher
educational attainment than the general population, which may limit the generalizability
of these normative data. Nevertheless, approximately half of our highly educated sample
still scored below the original suggested cutoff, underscoring the need for specific age and
education corrected norms. A final potential weakness of these normative data is the small
cell sizes for some the age- and education-stratified groups. Previous studies have noted
similar standard deviations on MoCA scores in older adults (Freitas et al., 2012; Larner,
2012) so it is likely that these data still accurately reflect MoCA performance in this
population.
Taken together, these results highlight the need for precise MoCA normative data,
particularly in light of the widespread use of the instrument with the elderly population. It
Downloaded by [FU Berlin] at 00:09 07 May 2015

is recommended that caution be used when applying the established cutoff scores as our
data, as well as others, have shown that they may be too stringent and therefore
misclassify cognitively normal people as impaired. Further, the use of demographically
adjusted MoCA scores may provide clinicians with a more accurate estimate of the
severity of cognitive impairment (or lack thereof), thus improving the clinical utility of
the instrument.

Disclosure statement
No potential conflict of interest was reported by the authors.

Funding
Funding for this study was provided by the Arizona Alzheimer’s Research Center [AZDHS
agreement # AGR2009-017] and Arizona State University, College of Human Services
Centenarian Database contract, 2007-2008.

ORCID
Michael Malek-Ahmadi http://orcid.org/0000-0001-9901-3650

References
Coen, R. F., Cahill, R., & Lawlor, B. A. (2011). Things to watch out for when using the Montreal
Cognitive Assessment (MoCA). International Journal of Geriatric Psychiatry, 26, 107–108.
doi:10.1002/gps.v26.1
Dai, T., Davey, A., Woodard, J. L., Miller, L. S., Gondo, Y., Kim, S.-H., & Poon, L. W. (2013).
Sources of variation on the mini-mental state examination in a population-based sample of
centenarians. Journal of the American Geriatrics Society, 61, 1369–1376. doi:10.1111/
jgs.2013.61.issue-8
Damian, A. M., Jacobson, S. A., Hentz, J. G., Belden, C. M., Shill, H. A., Sabbagh, M. N., . . .
Adler, C. H. (2011). The Montreal Cognitive Assessment and the mini-mental state examination
as screening instruments for cognitive impairment: Item analyses and threshold scores.
Dementia and Geriatric Cognitive Disorders, 31, 126–131. doi:10.1159/000323867
 Aging, Neuropsychology, and Cognition 7

Folstein, M. F., Folstein, S. E., & McHugh, P. R. (1975). “Mini-mental state”. A practical method
for grading the cognitive state of patients for the clinician. Journal of Psychiatric Research, 12,
189–198. doi:10.1016/0022-3956(75)90026-6
Freitas, S., Simoes, M. R., Alves, L., & Santana, I. (2012). Montreal Cognitive Assessment:
Influence of sociodemographic and health variables. Archives of Clinical Neuropsychology,
27, 165–175. doi:10.1093/arclin/acr116
Gluhm, S., Goldstein, J., Loc, K., Colt, A., Van Liew, C., & Corey-Bloom, J. (2013). Cognitive
performance on the mini-mental state examination and the Montreal Cognitive Assessment
across the healthy adult lifespan. Cognitive and Behavioral Neurology, 26, 1–5. doi:10.1097/
WNN.0b013e31828b7d26
Jefferson, A. L., Gibbons, L. E., Rentz, D. M., Carvalho, J. O., Manly, J., Bennett, D. A., &
Jones, R. N. (2011). A life course model of cognitive activities, socioeconomic status, educa-
tion, reading ability, and cognition. Journal of the American Geriatrics Society, 59, 1403–1411.
doi:10.1111/j.1532-5415.2011.03499.x.
Larner, A. J. (2012). Screening utility of the Montreal Cognitive Assessment (MoCA): In place of –
or as well as – the MMSE?. International Psychogeriatrics, 24, 391–396. doi:10.1017/
S1041610211001839
Lezak, M., Howieson, D. B., Loring, D. W., Hannay, H. J., & Fischer, J. S. (2004).
Neuropsychologial assessment (3rd ed.). New York: Oxford University Press.
Downloaded by [FU Berlin] at 00:09 07 May 2015

Luis, C. A., Keegan, A. P., & Mullan, M. (2009). Cross validation of the Montreal Cognitive
Assessment in community dwelling older adults residing in the Southeastern US. International
Journal of Geriatric Psychiatry, 24, 197–201. doi:10.1002/gps.v24:2
Malek-Ahmadi, M., Davis, K., Belden, C. M., & Sabbagh, M. N. (2014). Comparative analysis of
the Alzheimer Questionnaire (AQ) with the CDR sum of boxes, MoCA, and MMSE. Alzheimer
Disease & Associated Disorders, 28, 296–298. doi:10.1097/WAD.0b013e3182769731
Nasreddine, Z. S., Phillips, N. A., Bedirian, V., Charbonneau, S., Whitehead, V., Collin, I., . . .
Chertkow, H. (2005). The Montreal Cognitive Assessment, MoCA: A brief screening tool for
mild cognitive impairment. Journal of the American Geriatrics Society, 53, 695–699.
doi:10.1111/jgs.2005.53.issue-4
National Institute on Aging. (2008, March 1). An aging world: 2008. Retrieved from http://www.
census.gov/prod/2009pubs/p95-09-1.pdf
O’Bryant, S. E., Humphreys, J. D., Smith, G. E., Ivnik, R. J., Graff-Radford, N. R., Petersen, R. C.,
& Lucas, J. A. (2008). Detecting dementia with the mini-mental state examination in highly
educated individuals. Archives of Neurology, 65, 963–967. doi:10.1001/archneur.65.7.963
Ossher, L., Flegal, K. E., & Lustig, C. (2013). Everyday memory errors in older adults. Aging,
Neuropsychology, and Cognition, 20, 220–242. http://dx.doi.org/10.1080/13825585.2012.690365
Radloff, L. S. (1977). The CES-D scale: A self report depression scale for research in the general
population. Applied Psychological Measurement, 1, 385–401. doi:10.1177/014662
167700100306
Rossetti, H. C., Lacritz, L. H., Munro Cullum, C., & Weiner, M. F. (2011). Normative data for the
Montreal Cognitive Assessment (MoCA) in a population-based sample. Neurology, 77,
1272–1275. doi:10.1212/WNL.0b013e318230208a
Stephan, B. C. M., Minett, T., Pagett, E., Siervo, M., Brayne, C., & McKeith, I. G. (2013).
Diagnosing mild cognitive impairment (MCI) in clinical trials: A systematic review. BMJ
Open, 3. doi:10.1136/bmjopen-2012-001909