Archives of Clinical Neuropsychology 27 (2012) 165–175

Montreal Cognitive Assessment: Influence of Sociodemographic

and Health Variables
Sandra Freitas 1, *, Mário R. Simões 1, Lara Alves 2, Isabel Santana 3,4

Downloaded from https://academic.oup.com/acn/article/27/2/165/4917 by guest on 07 November 2024

1
Centro de Investigação do Núcleo de Estudos e Intervenção Cognitivo Comportamental (CINEICC), Faculty of Psychology and Educational Sciences,
University of Coimbra, Coimbra, Portugal
2
Faculty of Psychology and Educational Sciences, University of Coimbra, Coimbra, Portugal
3
Neurology Department, Coimbra University Hospital, Coimbra, Portugal
4
Faculty of Medicine, University of Coimbra, Coimbra, Portugal
*Corresponding author at: Psychological Assessment Department, Faculty of Psychology and Educational Sciences, University of Coimbra, Rua do Colégio
Novo, Apartado 6153, 3001-802 Coimbra, Portugal. Tel.: +351-918-414-941; fax: +351-256-084-386.
E-mail address: [email protected] (S. Freitas).
Accepted 23 December 2011

Abstract
The Montreal Cognitive Assessment (MoCA) is a brief cognitive instrument for screening milder forms of cognitive impairment. The
present study aimed to analyze the influence of sociodemographic (age, gender, educational level, marital and employment status, geographic
region, geographic localization, and residence area) and health variables (subjective memory complaints of the participant and evaluated by
the informant, depressive symptoms, and family history of dementia) on the participants’ performance on the MoCA. The investigation was
carried out in a Portuguese community-based sample of 650 cognitively healthy adults, who were representative of the distribution observed
in the Portuguese population. Educational level and age significantly contributed to the prediction of the MoCA scores, explaining 49% of the
variance. Regarding health variables, only the subjective memory complaints of the participant showed a small contribution (9%) to the vari-
ance on the MoCA scores. This study contributes a useful approach to understanding MoCA performance, stressing the great impact of edu-
cation and age on scores.

Keywords: Assessment; Elderly; Norms; Mild Cognitive Impairment; Alzheimer’s disease; Dementia

Introduction

Several studies have demonstrated that performance on screening tests is influenced by sociodemographic variables. It has
been widely reported that age and educational level have a significant effect on cognitive screening test performance. With
regard to age, older age has been found to significantly increase the probability of obtaining lower scores (Bravo & Hébert,
1997; Gallacher et al., 1999; Han et al., 2008; Langa et al., 2009; Matallana et al., 2011; Mathuranath et al., 2007; Moraes,
Pinto, Lopes, Litvoc, & Bottino, 2010; Rossetti, Lacritz, Cullum, & Weiner, 2011). Regarding the educational level, the
worst performance has been found among those with lower education levels, and ceiling effects have been observed among
highly educated individuals. The magnitude of this effect is strong; therefore, education is invariably considered a criterion
for the establishment of normative data for cognitive tests (Bravo & Hébert, 1997; Guerreiro, 1998; Han et al., 2008;
Lieberman et al., 1999; Mathuranath et al., 2007; Measso et al., 1993; Moraes et al., 2010; Morgado, Rocha, Maruta,
Guerreiro, & Martins, 2010; Nguyen, Black, Ray, Espino, & Markides, 2002; Rossetti et al., 2011). The results regarding
gender are more controversial. Some studies have suggested that gender contributes significantly to the explanation of variance
on scores of cognitive screening tests (Han et al., 2008; Measso et al., 1993; Mı́as, Sassi, Masih, Querejeta, & Krawchik, 2007;
Ribeiro, Oliveira, Cupertino, Neri, & Yassuda, 2010; Scazufca, Almeida, Vallada, Tasse, & Menezes, 2009), whereas
others have not supported this hypothesis (Bertolucci et al., 2001; Lieberman et al., 1999; Mathuranath et al., 2007;

# The Author 2012. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: [email protected].
doi:10.1093/arclin/acr116 Advance Access publication on 24 January 2012
166 S. Freitas et al. / Archives of Clinical Neuropsychology 27 (2012) 165–175

Morgado et al., 2010). There has also been a lack of consensus regarding marital status, as some studies have reported greater
performances among married persons (Fratiglioni, Wang, Ericsson, Maytan, & Winblad, 2000; Moraes et al., 2010; Nguyen
et al., 2002; Ribeiro et al., 2010; Wu, Lan, Chen, Chiu, & Lan, 2011), whereas others have found no influence on cognitive state
assessment (Bertolucci et al., 2001; Mı́as et al., 2007). Information regarding employment status is relatively scarce. One study
found better scores among individuals who are currently employed (Moraes et al., 2010) and another reported worse scores
among individuals with occupations with low intellectual demands (Anderson, Sachdev, Brodaty, Trollor, & Andrews,
2007). Investigations of geographic variables are complicated, and international inter-study comparison is meaningless due
to the specificities of the populations and territories. Some research has reported an IQ discrepancy in different geographical
regions of a country (Kaufman, McClean, & Reynolds, 1988; Lynn, 1979), which could be associated with average regional
incomes (Almeida, Lemos, & Lynn, unpublished manuscript; McDaniel, 2006). Because there are no Portuguese studies on the
influence of geographic variables on cognitive test performance, the inclusion of these variables in the current study assume an
exploratory nature.

Downloaded from https://academic.oup.com/acn/article/27/2/165/4917 by guest on 07 November 2024

The influence of health variables on performance at this level of cognitive screening has also been reported in the lit-
erature. Co-morbidity of depressive symptoms and cognitive decline is common, and several studies have aimed to clarify
the complex interaction between these conditions (Chen, Ganguli, Mulsant, & DeKosky, 1999; Emery & Oxman, 1992;
Rovner, Broadhead, Spencer, Carson, & Folstein, 1989). The general tendency toward a poor cognitive performance in
the presence of depressive symptoms is well-documented in the literature (Gallacher et al., 1999; Moraes et al., 2010;
Nguyen et al., 2002). The manifestation of subjective memory complaints is one of the diagnostic criteria for Mild
Cognitive Impairment (Petersen, 2000, 2007), but such complaints are also frequent among healthy elderly populations
(Reid & MacLullich, 2006). Additionally, also there is a significant association between anxiety and depressive disorders,
which further raises questions about the complex relationship between affective symptoms and memory. Reviewing the
literature on the influence of subjective memory complaints on cognitive tests and their predictive value of conversion
to dementia is a complex task due to the diversity of methodologies used to evaluate subjective memory complaints
and cognitive function. As expected, the results of this research are conflicting. Whereas some studies have reported
worse performance on cognitive tests among subjects with memory complaints, other investigations have evidenced that
memory complaints are a poor indicator of cognitive function (Reid & MacLullich, 2006). Family history of dementia
is a well-known risk factor for Alzheimer’s disease; however, few studies have examined the influence of this genetic
trait on the cognitive screening tests of healthy subjects. The limited evidence that is available points to a lack of associ-
ation between these variables (Mı́as et al., 2007).
The Montreal Cognitive Assessment (MoCA; Nasreddine et al., 2005) is a recent screening test that was specifically devel-
oped to screen for milder forms of cognitive impairment. Although it was initially designed to assess the cognitive states of
patients with Mild Cognitive Impairment and Alzheimer’s disease, the MoCA is now an extensively validated screening
tool for many disorders (e.g., Parkinson’s disease, Huntington’s disease, HIV, multiple sclerosis). This measure overcomes
the limitations of the Mini-Mental State Examination (MMSE; Folstein, Folstein, & McHugh, 1975), because it allows a
more comprehensive assessment of the major cognitive domains, including the executive functions assessment, and utilizes
more complex tasks to measure short-term memory (and with longer delay), language, attention, concentration, working
memory, and visuospatial skills. Several studies have reported the good psychometric properties and excellent sensitivity of
the MoCA to cognitive impairment, which has driven its rapid international dissemination and recommendation as a cognitive
screening tool in various guidelines (Arnold et al., 2007; Chertkow et al., 2007; Gauthier et al., 2011; Hachinski et al., 2006;
Ismail, Rajji, & Shulman, 2009; Jacova, Kertesz, Blair, Fisk, & Feldman, 2007; Lonie, Tierney, & Ebmeier, 2009; Zhao, Liu,
Shen, & Zhao, 2011). To expand upon this research, new population studies are needed to explore the modulation of MoCA
scores by the most significant sociodemographic and health variables, as this analysis has not been done in detail in any
country.
Studies of the translation, adaptation, validation, and normative of the MoCA for the Portuguese population were performed
by our group (Freitas, Santana, & Simões, 2010; Freitas, Simões, Alves, & Santana, 2011, 2012; Freitas, Simões, Marôco,
Alves, & Santana, 2012; Freitas, Simões, Martins, Vilar, & Santana, 2010; Simões et al., 2008). In the present investigation,
we aim to analyze the influence of sociodemographic variables (age, gender, educational level, marital status, employment
status, geographic region, geographic localization, and residence area) and health variables (depressive symptoms, subjective
memory complaints, and family history of dementia) on the participants’ performance on the MoCA using a large Portuguese
community-based sample, stratified according to the main sociodemographic variables of the population.
 S. Freitas et al. / Archives of Clinical Neuropsychology 27 (2012) 165–175 167

Methods

Participants and Procedures

The investigation was carried out in a community-based sample of volunteers who were recruited at national health and
social security services and resided in all geographic regions of the Portuguese continental territory. This sample is represen-
tative of the Portuguese population and was used in a recent MoCA normative study published by our group (Freitas et al.,
2011). The inclusion criteria were as follows: age 25 years and older; native speaker of Portuguese and schooling in
Portugal; and the absence of significant motor, visual, or auditory deficits, all of which may influence performance on tests.
To ensure that participants were cognitively healthy adults, we also defined the following exclusion criteria: evidence of
loss of autonomy in daily living activities; history of alcoholism or substance abuse; relevant neurological or psychiatric dis-
eases or chronic unstable systemic disorders that impact cognition; significant depressive complaints; and medication with a

Downloaded from https://academic.oup.com/acn/article/27/2/165/4917 by guest on 07 November 2024

possible impact on cognition (e.g., psychotropic or psycho-active drugs). A psychologist confirmed these general criteria in an
interview that included a complete sociodemographic questionnaire, an inventory of current clinical health status, and past
habits and medical history. For older participants, this information was also confirmed by a general practitioner, community
centre directors, and/or an informant, typically an individual in co-habitation or a close relative. Due to the extensive use of the
MoCA in clinical populations, we also established normative data for people 25 years old to allow the use of this instrument
with younger patients with other diseases.
After this initial selection, all subjects were required to display normal performance on the assessment battery used in this
study (see “Materials”), considering the Portuguese cut-off points, for effective inclusion in the study. Each participant was
assessed in a single session by one of two psychologists with expertise in neuropsychological assessment.
From the initial community-based sample of 936 volunteers, 194 subjects (20.73%) were excluded in light of data collected
in the interview (most frequent reasons: history of neurological or psychiatric disorder, history of alcohol abuse, and subjective
self-evaluation that memory complaints significantly influence day-to-day activities), 58 subjects (6.20%) were excluded due
to the presence of significant depressive symptoms (according to the criteria: Geriatric Depression Scale [GDS] score over 20
points), and 34 subjects (3.63%) were excluded due to suspected cognitive impairment based on their performance on the as-
sessment battery and respective Portuguese cutoff points.
Only the subjects who met all of the defined inclusion and exclusion criteria were eligible for the study. The final sample
comprised 650 cognitively healthy adults, and stratification according to the sociodemographic variables confirmed that this
sample was representative of the distribution observed in the Portuguese population (Table 1).
Informed consent was obtained from all of the participants after the aims and the procedures of the investigation, and con-
fidentiality requirements were fully explained by a member of the study group. The present research complied with the ethical
guidelines for human experimentation stated in the Declaration of Helsinki and was approved by the Portuguese Foundation for
Science and Technology and by the Faculty of Psychology and Educational Sciences Scientific Committee.

Materials

The assessment battery for the global assessment of each participant was composed of the following instruments:

(i)Complete sociodemographic questionnaire.

(ii)Inventory of current clinical health status.
(iii)Inventory of past habits and medical history.
(iv) MoCA (Nasreddine et al., 2005; Simões et al., 2008), which is a brief cognitive screening instrument that was
developed for the screening of milder forms of cognitive impairment. The tool is a one-page test with
paper-and-pencil format, and the application time is approximately 10– 15 min. A manual provides explicit
instructions for administration and an objectively defined scoring system. The maximum score is 30 points,
with higher scores indicating better cognitive performance. It evaluates the following eight cognitive domains:
executive functions; visuospatial abilities; short-term memory; language; attention, concentration, and working
memory; and temporal and spatial orientation. In the current study, the MoCA was not used as a diagnostic
tool. Furthermore, the MoCA total score refers to the raw score without the correction point for educational
effects proposed in the original study (Nasreddine et al., 2005) because this correction point is not used in the
Portuguese population (Freitas et al., 2011).
(v) MMSE (Folstein et al., 1975; Guerreiro, 1998), which is the most widely used brief screening instrument for
detecting cognitive deficits and, therefore, is not described in detail here.
168 S. Freitas et al. / Archives of Clinical Neuropsychology 27 (2012) 165–175

(vi) Clinical Dementia Rating scale (CDR; Garret et al., 2008; Hughes, Berg, Danziger, Coben, & Martin, 1982),
which is a global staging tool for dementia that is based on the assessment of cognitive function and functional
capacity (in six cognitive-behavioral categories: memory, orientation, sense and problem-solving, community ac-
tivities, home activities and hobbies, and personal care). The scale is administered to the adult/elderly patients and
an informant through a semi-structured interview. The CDR was only administered to participants over 49 years of
age and when a close informant was available. In the current study, a global score of zero was used as a criterion
for inclusion.
(vii) Irregular Word Reading Test (Teste de Leitura de Palavras Irregulares; Alves, Simões, & Martins, 2009), which is
a tool for estimating premorbid intelligence that consists of a list of 46 irregular words that the participant reads.
(viii) GDS-30 (Barreto, Leuschner, Santos, & Sobral, 2008; Yesavage et al., 1983), which is a brief scale to assess de-
pressive symptoms in adults. It is composed of 30 dichotomous response questions that assess emotional and be-
havioral symptoms of depression (score range ¼ 0 – 30).

Downloaded from https://academic.oup.com/acn/article/27/2/165/4917 by guest on 07 November 2024

(ix) Subjective Memory Complaints scale (SMC; Ginó et al., 2008; Schmand, Jonker, Hooijer, & Lindeboom, 1996).
This scale consists of 10 multiple choice items that assess the presence of subjective memory complaints (score
range ¼ 0 – 21). It was administered under two conditions: (i) SMC participants: answered by the participants to
evaluate their own subjective memory complaints, and (ii) SMC informants: answered by informants to assess
their opinion about the memory capacity of the participant (when a close informant was available).

Variable Definitions and Sample Stratification

To enhance the representativeness of the observed distribution in the Portuguese population, the sample of 650 participants
was stratified according to the following sociodemographic variables:

Table 1. Sociodemographic characterization and stratification of the sample

Levels Sample Portugal
n Percent n Percent

Sociodemographic stratification of sample

Age 25–49 214 33.0 —
50–64 218 33.5 —
≥65 218 33.5 —
Gender Female 408 62.8 3946 52.6
Male 242 37.2 3559 47.4
Educational level Primary 256 39.4 2426 36.6
Middle 170 26.2 2280 34.4
High 112 17.2 960 14.5
University 112 17.2 956 14.5
Geographic region North 251 38.6 2722 36.0
Center 174 26.8 1794 24.0
Lisbon 164 25.2 2091 28.0
Alentejo 44 6.8 577 8.0
Algarve 17 2.6 321 4.0
Geographic localization Coast 546 84.0 6379 85.0
Inland 104 16.0 1126 15.0
Residence area PUA 446 68.6 5103 68.0
MUA 112 17.2 1200 16.0
PRA 92 14.2 1200 16.0
Others sociodemographic variables
Marital status Married 489 75.2 —
Single 161 24.8 —
Employment status Active 330 50.8 —
Inactive 320 49.2 —

Notes: PUA ¼ predominantly urban areas; MUA ¼ moderately urban areas; PRA ¼ predominantly rural areas.
The values (n) of the Portuguese population are expressed in thousands and represent the data of the resident population in continental Portugal aged over 24
years (Instituto Nacional de Estatı́stica, 2010).
 S. Freitas et al. / Archives of Clinical Neuropsychology 27 (2012) 165–175 169

(I) age (age intervals were: 25– 49 [“young adults”: mean age ¼ 38.12 + 8.086], 50– 64 [“adults”: mean age ¼
57.12 + 4.199], and 65 and over [“elderly”: mean age ¼ 71.96 + 5.433]);
(II) gender (women and men);
(III) educational level (four educational levels were considered, according to the number of school years successfully
completed in the Portuguese education system: 1 – 4 [primary education], 5 – 9 [middle school], 10– 12 [high
school], and over 12 years of education [university/college]; these categories match the divisions in the
Portuguese school system);
(IV) geographic region (Portuguese continental territory is divided into five geographic regions [NUTS-II classifica-
tion; Instituto Nacional de Estatı́stica, 2010]: North, Centre, Lisbon, Alentejo and Algarve);
(V) geographic localization (two geographic localizations were considered: coast and inland);
(VI) residence area (according to the Types of Urban Areas [Instituto Nacional de Estatı́stica, 2010], categorized into
predominantly urban areas, moderately urban areas, and predominantly rural areas).

Downloaded from https://academic.oup.com/acn/article/27/2/165/4917 by guest on 07 November 2024

In this study, we also included the following sociodemographic and health variables that were not criteria for sample
stratification:

(VII) marital status (categorized into “single” [single, divorced, or widowed participants] or “married” [married or
living in union participants]);
(VIII) employment status (categorized into “active” [participants with an active work situation] or “inactive” [partici-
pants unemployed, retired, or domestic]);
(IX) family history of dementia (only the information about first-degree relatives was considered relevant, and classi-
fication was dichotomized into “positive” or :negative”);
(X) depressive symptoms (operationalized by the total score on the GDS-30). Once the participants with severe de-
pressive symptoms were excluded, this study analyzed the influence of depressive symptom levels among non-
depressed to mildly depressed individuals on MoCA performance;
(XI) subjective memory complaints (operationalized by the total score on SMC; two conditions were considered: (a)
SMC participants and (b) SMC informants).

Statistical Analysis

All data analyses were conducted using the Statistical Package for the Social Sciences, version 17.0 (SPSS, v.17.0).
Descriptive statistics were computed for all sociodemographic and health variables. The observed correlations (using the
Pearson correlation coefficient; Cohen, 1988), Cronbach’s a coefficients, and corrected correlations among measures were
also calculated. The differences on the MoCA scores among subgroups stratified according to sociodemographic variables
were examined using Student’s t-test, analysis of variance, Tukey’s HSD, the Bonferroni post hoc test, and analysis of covari-
ance. Partial eta-squared (h2p ) was used as an estimate of the effect size (Cohen, 1988). The contribution of the variables age
and educational level was the subject of further statistical analysis. The correlation between MoCA scores, age, and education
was performed using the Pearson correlation coefficient (r) (Cohen, 1988). To investigate the significance of age (in years) and
education (years of schooling successfully completed) as influencing factors of the MoCA, multiple linear regression (MLR)
analyses were performed using the enter method. The multicollinearity was examined through tolerance and variance inflation
factor statistics (Meyers, Gamst, & Guarino, 2006), and the coefficient of determination (R2) was considered in the analysis of
effect size in the regressions (Cohen, 1988). Finally, the influence of health variables on the MoCA performance was inves-
tigated using MLR analysis, stepwise method, for variables with significant Pearson correlations. All data for this investigation
were collected by two neuropsychologists only, which allowed rigorous data collection and procedures. Therefore, the only
variable with missing data was SMC informants, as not all participants had a close informant available (n ¼ 156; 24%).
For this variable, the results correspond to 494 participants who had a close informant available.

Results

The final study sample included 650 cognitively healthy participants (mean age ¼ 55.84 + 15.12, age range ¼ 25– 91;
mean education ¼ 8.16 + 4.72, education range ¼ 2 –27). The sociodemographic characteristics of the sample are presented
in detail in Table 1, taking into account the stratification variables as well as the other sociodemographic variables considered
170 S. Freitas et al. / Archives of Clinical Neuropsychology 27 (2012) 165–175

in the study. The distribution of the study sample in several strata is comparable with the distribution of the target Portuguese
population.
The observed correlations, Cronbach’s a coefficients, and corrected correlations among the different measures are presented
in Table 2. Table 3 summarizes the analysis of differences of the MoCA mean scores between the subgroups.
The sociodemographic variables that showed significant group differences (age, educational level, geographic region, area
of residence, marital status, and employment status) were targeted for further data analysis. Significant differences in mean age

Table 2. Observed correlations, Cronbach’s a coefficients, and corrected correlations among measures
Variables MoCA MMSE TeLPI GDS SMC
Participants Informants

MoCA (.78) 1.10 .73 .27 .39 .07

Downloaded from https://academic.oup.com/acn/article/27/2/165/4917 by guest on 07 November 2024

MMSE .65** (.45) .64 .24 .35 .07
TeLPI 2.62** 2.43** (.92) .16 .24 .10
GDS 2.22** 2.15** .14** (.86) .65 .35
SMC-participants 2.31** 2.21** .21** .54** (.80) .42
SMC-informants 2.05 2.04 2.08 .28** .32** (.73)
Notes: MoCA ¼ Montreal Cognitive Assessment; MMSE ¼ Mini-Mental State Examination; TeLPI ¼ Irregular Word Reading Test; GDS ¼ Geriatric
Depression Scale; SMC ¼ Subjective Memory Complaints scale.
Alpha coefficients are presented on the diagonal, observed correlations below the diagonal, and correlations corrected for attenuation above the diagonal.
**p , .01.

Table 3. Analysis of group differences on the MoCA scores (without control of the effect of covariates)
Variables MoCA M + SD t/F Post hoc

Age
25–49 26.98 + 2.548 F(2,647) ¼ 95.130, p , .001 All groups differ.
50–64 24.46 + 3.432
≥65 22.71 + 3.668
Gender
Female 24.50 + 3.798 t ¼ 1.80, p ¼ .072 —
Male 25.04 + 3.419
Educational level
Primary 21.73 + 3.185 F(3,646) ¼ 194.996, p , .001 All groups differ. Significant linear effect:
Middle 25.65 + 2.501 F ¼ 435.895, p , .001
High 26.77 + 2.153
University 28.04 + 1.942
Geographic region
A. North 24.22 + 3.644 F(4,645) ¼ 8.765, p , .001 A = C, D
B. Center 24.28 + 3.841 B = C, D
C. Lisbon 25.76 + 3.395 C = A, B, E
D. Alentejo 26.11 + 2.713 D = A, B, E
E. Algarve 22.41 + 3.658 E = C, D
Geographic localization
Coast 24.80 + 3.715 t ¼ 1.546, p ¼ .122 —
Inland 24.19 + 3.382
Residence area
PUA 24.93 + 3.728 F(2,647) ¼ 4.175, p ¼ .016 PUA = PRA
MUA 24.60 + 3.362
PRA 23.73 + 3.604
Marital status
Married 24.40 + 3.569 t ¼ 3.652, p ≤ .001 —
Single 25.61 + 3.823
Employment status
Active 26.22 + 2.988 t ¼ 11.781, p ≤ .001 —
Inactive 23.13 + 3.649
Notes: PUA ¼ predominantly urban areas; MUA ¼ moderately urban areas; PRA ¼ predominantly rural areas; M: mean; SD: standard deviation; t: Student’s
t-test values; F: analysis of variance values; Post hoc: Tukey HSD and Bonferroni post hoc test analyses.
 S. Freitas et al. / Archives of Clinical Neuropsychology 27 (2012) 165–175 171

and mean educational level between the subgroups were verified (Table 4). Based on this result, we proceeded with the analysis
of covariance to examine whether differences in the MoCA scores remained significant after controlling for the effect of cov-
ariates (age and/or educational level) and to estimate the effect size of each variable (Table 5).
The results indicated that after controlling for the effects of covariates, only the variables age, educational level, and geo-
graphic region contributed significantly to the explanation of variance of the MoCA scores. In the subsequent analysis, we only
considered variables whose effect size was medium—age: F(2, 646) ¼ 34.098, p , .001, h2p ¼ 0.095—or large—educational
level: F(3, 645) ¼ 117.459, p , .001, h2p ¼ 0.353. The variable geographic region revealed a small effect size—F(4,643) ¼
4.972, p ¼ .030—and, therefore, was not further examined.
Statistically significant correlations were observed between the MoCA scores and age (r ¼ 2.522, p , .01) and educational
level (r ¼ .652, p , .01). To examine the contributions of these variables and their interactions to the explanation of variance
of the MoCA scores, a MLR analysis was performed using the enter method. This analysis resulted in two significant regression
models. The first model—F(1, 648) ¼ 480.093, p , .001—included only the variable educational level (b ¼ 0.652,

Downloaded from https://academic.oup.com/acn/article/27/2/165/4917 by guest on 07 November 2024

t ¼ 21.911, p , .001), which significantly explained 42.5% of total variance of the MoCA scores. In the second regression

Table 4. Analysis of covariates: group differences in mean age and educational level
Variables Age Educational level

M + DP t/F M + DP t/F

Age
25–49 — — 10.64 + 4.688 F(2,647) ¼ 62.757, p , .001
50–64 7.75 + 4.289
≥65 6.08 + 3.989
Educational level
Primary 63.99 + 10.843 F(3,646) ¼ 57.631, p , .001 — —
Middle 53.69 + 14.454
High 48.69 + 14.523
University 47.50 + 16.008
Geographic region
North 53.53 + 13.741 F(4,645) ¼ 6.517, p , .001 6.91 + 3.891 F(4,645) ¼ 15.675, p , .001
Center 60.09 + 14.092 8.50 + 5.382
Lisbon 54.41 + 16.454 10.20 + 4.627
Alentejo 54.82 + 18.746 7.59 + 4.299
Algarve 63.12 + 10.624 4.94 + 2.703
Residence area
PUA 55.63 + 14.879 F(2,647) ¼ 2.094, p ¼ .124 8.60 + 4.871 F(2,647) ¼ 8.450, p , .001
MUA 54.43 + 16.408 7.79 + 4.298
PRA 58.62 + 14.456 6.46 + 4.064
Marital status
Married 57.24 + 13.431 t ¼ 3.527, p ¼ .001 7.65 + 4.551 t ¼ 4.842, p ≤ .001
Single 51.60 + 18.789 9.70 + 4.917
Employment status
Active 45.84 + 11.549 t ¼ 23.147, p ≤ .001 9.81 + 4.739 t ¼ 9.673, p ≤ .001
Inactive 66.16 + 10.829 6.46 + 4.065
Notes: M ¼ mean; SD ¼ standard deviation; t ¼ Student’s t-test values; F ¼ analysis of variance values.

Table 5. Analysis of group differences in the MoCA scores while controlling for the effect of covariates and estimation of the effect sizes
Variables Covariates ANCOVA Effect size

Age Educational level F(2,646) ¼ 34.098, p , .001 Medium h2p = 0.095

Educational Level Age F(3,645) ¼ 117.459, p , .001 Large h2p = 0.353
Geographic Region Age and Educational level F(4,643) ¼ 4.972, p ¼ .001 Small h2p = 0.030
Residence Area Educational level F(2,646) ¼ .122, p ¼ .885 Null h2p = 0.000
Marital Status Age and Educational level F(1,646) ¼ .014, p ¼ .907 Null h2p = 0.000
Employment Status Age and Educational level F(1,646) ¼ 3.469, p ¼ .063 Null h2p = 0.005
Notes: F ¼ analysis of covariance (ANCOVA) values; h2p ¼ partial eta-squared value.
According to Cohen (1988), h2p values of 0.01, 0.06, and 0.14 are considered small, medium, and large effect sizes, respectively.
172 S. Freitas et al. / Archives of Clinical Neuropsychology 27 (2012) 165–175

model, the two variables were combined, and no evidence of multicollinearity was detected. In this model, F(2, 647) ¼ 317.016,
p , .001, both variables significantly contributed to the prediction of the MoCA scores (educational level:
b ¼ 0.524, t ¼ 16.871, p , .001; age: b ¼ 2.293, t ¼ 29.426, p , .001. The b weights indicated that educational level
was the major contributor to the prediction of the MoCA scores, but age also contributed to the prediction. The adjusted R 2
value was .49, which signifies that 49% of the variance on the MoCA scores was explained by this model.
The health variables considered in the study were as follows: family history of dementia (16% of participants had a positive
family history), depressive symptoms (GDS mean ¼ 7.34 + 5.371, range ¼ 0 – 20), and subjective memory complaints (two
conditions: (a) SMC participants [mean ¼ 5.66 + 3.592, range ¼ 0 – 18] and (b) SMC informants [mean ¼ 4.15 + 2.735,
range ¼ 0 – 11]). The results of the intercorrelations among the MoCA scores and these health variables are provided in
Table 2, except for the family history of dementia, which showed no significant correlation with MoCA scores (r ¼ .00,
p , .001) or SMC informants scores (r ¼ .095, p ¼ .239) and a significant correlation with GDS scores (r ¼ .092,
p ¼ .019) and SMC participants scores (r ¼ .127, p , .01).

Downloaded from https://academic.oup.com/acn/article/27/2/165/4917 by guest on 07 November 2024

We observed that MoCA scores only showed a statistically significant and negative correlation with depressive symptoms
and the subjective memory complaints of the participants. The influence of these health variables on MoCA performance was
investigated using MLR analysis, a stepwise method. The resulting model—F(2,647) ¼ 64.860, p , .001—only included the
subjective memory complaints of the participant, which explained 9% of the total variance on the MoCA scores.
Depressive symptoms did not reveal a significant contribution to the model (b ¼ 2.082, t ¼ 1.845, p ¼ .065).

Discussion

A reliable evaluation of an individual’s cognitive performance must be based on robust normative data stratified according
to the sociodemographic variables most influential on and predictive of performance. The current study is essential for adapting
this instrument for a specific population. Furthermore, the study is relevant due to the widespread international use of the
MoCA, the lack of international studies that analyze a wide variety of variables that may influence one’s performance on
this test, and the absence of studies using stratified community-based samples in Portugal to examine the influence of socio-
demographic and health variables on cognitive screening instruments. The few studies available were limited by small samples
within restricted regional areas and only focused on specific variables. The use of a sample stratified by different levels of
sociodemographic variables and with a distribution close to that observed in the Portuguese population enhances the equiva-
lence with the target population and the confidence of conclusions drawn.
Our results confirm that age and educational level significantly contribute to the prediction of the MoCA scores, explaining
49% of the results variance. This is considered a large effect, according Cohen (1988), and a respectable result, according to
Pallant (2007). As expected, and according to previous studies of cognitive screening tests (Anderson et al., 2007; Bravo &
Hébert, 1997; Gallacher et al., 1999; Langa et al., 2009; Matallana et al., 2011; Mathuranath et al., 2007; Moraes et al.,
2010), our results confirm that older age and lower educational level have a significant effect on MoCA performance, increas-
ing the likelihood of obtaining a lower total score.
The influence of other sociodemographic variables on screening tests is further conflicting in the literature. In the present
study, gender, marital status, and employment status did not reveal a significant effect on the MoCA results. Regarding geo-
graphical variables, our results indicated no statistically significant differences between subjects living in the coastal and inland
areas. The differences observed among residents in predominantly urban or rural areas were not significant after controlling for
education. Finally, the observed differences between residents in different geographic regions showed a small magnitude after
controlling for age and education. Of note, these regional subgroups were not completely matched for age and education, which
may explain the results obtained.
Regarding the influence of health variables on MoCA performance, similar to a previous study (Mı́as et al., 2007), the
results suggest no significant association between family history of dementia or memory complaints evaluated by the informant
and MoCA performance. On the other hand, both depressive symptoms and subjective memory complaints of the participant
presented significant and negative correlations with MoCA total scores. Moreover, these variables also showed a significant
correlation between them, which is consistent with the well-documented association between these symptoms (Reid &
MacLullich, 2006). Considering the results of the MLR analysis performed, only the subjective memory complaints of the
participant showed a small contribution (9%) to the explanation of the variance on the MoCA scores. However, these data
regarding health variables should be interpreted with caution. Since they result from the analysis of performance of cognitively
healthy and non-depressed-to-mildly-depressed individuals, these findings should not be generalized to individuals with clin-
ical conditions. Additionally, the findings about the depressive symptoms and subjective memory complaints would benefit
from better operationalization of these variables. The inclusion of more specific and descriptive instruments may have shed
 S. Freitas et al. / Archives of Clinical Neuropsychology 27 (2012) 165–175 173

greater light on the influence of these variables on MoCA performance. Furthermore, these findings must be complemented
with studies that consider patients with cognitive impairment and patients with depression.
Another limitation of the present study was the inability to completely match all of the age subgroups in terms of education
due to the higher education of the younger group. However, the observed discrepancy is, in fact, representative of the demo-
graphic profile of the country. This can be explained by the change in the school system in the last decades, namely the im-
position of higher levels of obligatory education, which has had a selective impact on the younger generations. The older strata
were characterized by a very low mean education. Another issue involves the classification of participants as cognitively
healthy subjects. To ensure cognitive health, we established strict criteria for inclusion and exclusion in the sample, as previ-
ously explained, and these criteria were confirmed in the clinical interview and neuropsychological evaluation. Furthermore,
for older participants, confirmatory information was also obtained through a general practitioner, community centre directors,
and/or an informant. However, given the sample size and geographical distribution of the participants, it was not possible to
perform a neurological consultation or additional diagnostic tests such as neuroimaging, which would have further ensured the

Downloaded from https://academic.oup.com/acn/article/27/2/165/4917 by guest on 07 November 2024

normal cognitive status of participants. Finally, due to the lack of international studies analyzing the influence of sociodemo-
graphic and health variables on MoCA performance, there is no comparison for these results.
This study is a useful approach for better understanding MoCA performance in a community population. The influence of
education and age on MoCA scores was clearly demonstrated, and therefore, these variables are the optimal criteria for the
establishment of MoCA normative data for the Portuguese population.

Funding

This work was supported by the Fundação para a Ciência e Tecnologia [Portuguese Foundation for Science and Technology]
(SFRH/BD/38019/2007) to SF; and the Fundação Calouste Gulbenkian [Calouste Gulbenkian Foundation] (Proc.74569,
SDH-22 Neurociências).

References

Almeida, L. S., Lemos, G. C., & Lynn, R. Regional differences in intelligence and per capita incomes in Portugal. Unpublished manuscript, University of
Minho, Portugal.
Alves, L., Simões, M. R., & Martins, C. (2009). Teste de Leitura de Palavras Irregulares (TeLPI) [Reading Word Irregular Test]. Coimbra, Portugal: Serviço
de Avaliação Psicológica da Faculdade de Psicologia e de Ciências da Educação da Universidade de Coimbra [Psychological Assessment Department,
Faculty of Psychology and Educational Sciences, University of Coimbra].
Anderson, T. M., Sachdev, P. S., Brodaty, H., Trollor, J., & Andrews, G. (2007). Effects of sociodemographic and health variables on Mini-Mental State Exam
scores in older Australians. American Journal of Geriatric Psychiatry, 15 (6), 467–476.
Arnold, J. M., Howlett, J. G., Dorian, P., Ducharme, A., Giannetti, N., Haddad, H., et al. (2007). Canadian Cardiovascular Society Consensus Conference
recommendations on heart failure update 2007: Prevention, management during intercurrent illness or acute decompensation, and use of biomarkers.
Canadian Journal of Cardiology, 23 (1), 21–45.
Barreto, J., Leuschner, A., Santos, F., & Sobral, M. (2008). Escala de Depressão Geriátrica [Geriatric Depression Scale]. In Grupo de Estudos de
Envelhecimento Cerebral e Demências [Study Group on Brain Aging and Dementia] (Ed.), Escalas e testes na demência [Scales and tests in dementia]
(pp. 69–72). Lisbon: GEECD.
Bertolucci, P. H., Okamoto, I. H., Brucki, S. M., Siviero, M. O., Toniolo Neto, J., & Ramos, L. R. (2001). Applicability of the CERAD neuropsychology battery
to Brazilian elderly. Arquivos de Neuropsiquiatria, 59, 532–536.
Bravo, G., & Hébert, R. (1997). Age and education specific reference values for the Mini-Mental and Modified Mini-Mental State Examination derived from a
non-demented elderly population. International Journal of Geriatric Psychiatry, 12, 1008–1018.
Chen, P., Ganguli, M., Mulsant, B. H., & DeKosky, S. T. (1999). The temporal relationship between depressive symptoms and dementia: A community based
prospective study. Archives of General Psychiatry, 56, 261–266.
Chertkow, H., Nasreddine, Z., Joanette, Y., Drolet, V., Kirk, J., Massoud, F., et al. (2007). Mild cognitive impairment and cognitive impairment, no dementia:
Part A, concept and diagnosis. Alzheimer’s and Dementia, 3, 266–282.
Cohen, J. (1988). Statistical power analysis for the behavioural sciences (2nd ed.). Hillsdale: Lawrence Erlbaum Associates.
Emery, V. O., & Oxman, T. E. (1992). Update on the dementia spectrum of depression. American Journal of Psychiatry, 149, 305 –317.
Folstein, M., Folstein, S., & McHugh, P. (1975). Mini-Mental State: A practical method for grading the cognitive state of patients for the clinician. Journal of
Psychiatric Research, 12 (3), 189–198.
Fratiglioni, L., Wang, H. X., Ericsson, K., Maytan, M., & Winblad, B. (2000). Influence of social network on occurrence of dementia: A community-based
longitudinal study. Lancet, 355 (9212), 1315–1319.
Freitas, S., Santana, I., & Simões, M. R. (2010). The sensitivity of the MoCA and MMSE to cognitive decline: A longitudinal study. Alzheimer’s and Dementia,
6 (4), S353–S354.
Freitas, S., Simões, M. R., Alves, L., & Santana, I. (2011). Montreal Cognitive Assessment (MoCA): Normative study for the Portuguese population. Journal of
Clinical and Experimental Neuropsychology, 33 (9), 989– 996.
174 S. Freitas et al. / Archives of Clinical Neuropsychology 27 (2012) 165–175

Freitas, S., Simões, M. R., Alves, L., & Santana, I. (2012). Montreal Cognitive Assessment (MoCA): Validation study for Mild Cognitive Impairment and
Alzheimer’s disease. Alzheimer Disease and Associated Disorders, in press, doi: 10.1097/WAD.0b013e3182420bfe.
Freitas, S., Simões, M. R., Marôco, J., Alves, L., & Santana, I. (2012). Construct validity of the Montreal Cognitive Assessment (MoCA). Journal of
International Neuropsychology Society, 18, 1– 9.
Freitas, S., Simões, M. R., Martins, C., Vilar, M., & Santana, I. (2010). Estudos de adaptação do Montreal Cognitive Assessment (MoCA) para a população
portuguesa [Montreal Cognitive Assessment (MoCA): Portuguese adaptation studies]. Avaliação Psicológica, 9 (3), 345– 357.
Gallacher, J. E., Elwood, P. C., Hopkinson, C., Rabbitt, P. M., Stollery, B. T., Sweetnam, P. M., et al. (1999). Cognitive function in the Caerphilly study:
Associations with age, social class, education and mood. European Journal of Epidemiology, 15 (2), 161– 169.
Garret, C., Santos, F., Tracana, I., Barreto, J., Sobral, M., & Fonseca, R. (2008). Avaliação Clı́nica da Demência [Clinical Dementia Rating Scale]. In Grupo de
Estudos de Envelhecimento Cerebral e Demências [Study Group on Brain Aging and Dementia] (Ed.), Escalas e testes na demência [Scales and tests in
dementia] (pp. 17–32). Lisbon: GEECD.
Gauthier, S., Patterson, C., Gordon, M., Soucy, J., Schubert, F., & Leuzy, A. (2011). Commentary on “Recommendations from the National Institute on
Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease”: A Canadian perspective. Alzheimer’s and Dementia, 7,
330–332.
Ginó, S., Mendes, T., Ribeiro, F., Mendonça, A., Guerreiro, M., & Garcia, C. (2008). Escala de Queixas de Memória [Memory Complaints Scale]. In Grupo de

Downloaded from https://academic.oup.com/acn/article/27/2/165/4917 by guest on 07 November 2024

Estudos de Envelhecimento Cerebral e Demências [Study Group on Brain Aging and Dementia] (Ed.), Escalas e testes na demência [Scales and tests in
dementia] (pp. 117– 120). Lisbon: GEECD.
Guerreiro, M. (1998). Contributo da Neuropsicologia para o estudo das demências [Contribution of neuropsychology to the study of dementia]. Unpublished
doctoral dissertation, University of Lisbon, Lisbon.
Hachinski, V., Iadecola, C., Petersen, R. C., Breteler, M. M., Nyenhuis, D. L., Black, S. E., et al. (2006). National Institute of Neurological Disorders and
Stroke—Canadian stroke network vascular cognitive impairment harmonization standards. Stroke, 37, 2220– 2241.
Han, C., Jo, S. A., Jo, I., Kim, E., Park, M. H., & Kang, Y. (2008). An adaptation of the Korean Mini-Mental State Examination (K-MMSE) in elderly Koreans:
Demographic influence and population-based norms (the AGE study). Archives of Gerontology and Geriatrics, 47, 302– 310.
Hughes, C. P., Berg, L., Danziger, W. L., Coben, L. A., & Martin, R. L. (1982). A new clinical scale for the staging of dementia. The British Journal of
Psychiatry, 140, 566–572.
Instituto Nacional de Estatı́stica. (2010). Portal de Estatı́sticas Oficiais [Portuguese Official Statistics]. Retrieved from http://www.ine.pt.
Ismail, Z., Rajji, T. K., & Shulman, K. I. (2009). Brief cognitive screening instruments: An update. International Journal of Geriatric Psychiatry, 25 (2),
111–120.
Jacova, C., Kertesz, A., Blair, M., Fisk, J. D., & Feldman, H. H. (2007). Neuropsychological testing and assessment for dementia. Alzheimer’s Dementia, 3 (4),
299–317.
Kaufman, A. S., McClean, J. E., & Reynolds, C. R. (1988). Sex, race, region and education differences on the 11 WAIS-R subtests. Journal of Clinical
Psychology, 44, 231–248.
Langa, K. M., Llewellyn, D. J., Lang, I. A., Weir, D. R., Wallace, R. B., Kabeto, M. U., et al. (2009). Cognitive health among older adults in the United States
and in England. BMC Geriatrics, 9, 23. doi:10.1186/1471-2318-9-23.
Lieberman, D., Galinsky, D., Fried, V., Grinshpun, Y., Mytlis, N., Tylis, R., et al. (1999). Factors affecting the results of the Clock Drawing Test in elderly
patients hospitalized for physical rehabilitation. International Journal of Geriatric Psychiatry, 14, 325–330.
Lonie, J. A., Tierney, K. M., & Ebmeier, K. P. (2009). Screening for Mild Cognitive Impairment: A systematic review. International Journal of Geriatric
Psychiatry, 24 (9), 902– 915.
Lynn, R. (1979). The social ecology of intelligence in the British Isles. British Journal of Social and Clinical Psychology, 18, 1– 12.
Matallana, D., Santacruz, C., Cano, C., Reyes, P., Samper-Ternent, R., Markides, K. S., et al. (2011). The relationship between educational level and
Mini-Mental State Examination domains among older Mexican Americans. Journal of Geriatric Psychiatry and Neurology, 24 (1), 9 –18.
Mathuranath, P. S., Cherian, J. P., Mathew, R., George, A., Alexander, A., & Sarma, S. P. (2007). Mini Mental State Examination and the Addenbrooke’s
Cognitive Examination: Effect of education and norms for a multicultural population. Neurology India, 55 (2), 106– 110.
McDaniel, M. A. (2006). State preferences for the ACT versus SAT complicates inferences about SAT-derived state IQ estimates: A comment on Kanazawa
(2006). Intelligence, 34, 601– 606.
Measso, G., Cavarzeran, F., Zappalà, G., Lebowitz, B. D., Crook, T. H., Pirozzollo, F. J., et al. (1993). The Mini-Mental State Examination: Normative study of
an Italian random sample. Developmental Neuropsychology, 9 (2), 77– 85.
Meyers, L. S., Gamst, G., & Guarino, A. J. (2006). Applied multivariate research: Design and interpretation. Thousand Oaks: Sage Publications.
Mı́as, C. D., Sassi, M., Masih, M. E., Querejeta, A., & Krawchik, R. (2007). Deterioro cognitivo leve: Estúdio de prevalência y factores sociodemográficos en la
ciudad de Córdoba, Argentina [Mild cognitive impairment: A prevalence and sociodemographic factors study in the city of Córdoba, Argentina]. Revista de
Neurologı́a, 44 (12), 733–738.
Moraes, C., Pinto, J. A., Lopes, M. A., Litvoc, J., & Bottino, C. M. (2010). Impact of sociodemographic and health variables on Mini-Mental State Examination
in a community-based sample of older people. European Archives of Psychiatry and Clinical Neuroscience, 260 (7), 535 –542.
Morgado, J., Rocha, C. S., Maruta, C., Guerreiro, M., & Martins, I. P. (2010). Cut-off scores in MMSE: A moving target? European Journal of Neurology, 17
(5), 692–695.
Nasreddine, Z., Phillips, N. A., Bédirian, V., Charbonneau, S., Whitehead, V., Collin, I., et al. (2005). The Montreal Cognitive Assessment, MoCA: A brief
screening tool for Mild Cognitive Impairment. American Geriatrics Society, 53 (4), 695– 699.
Nguyen, H. T., Black, S. A., Ray, L. A., Espino, D. V., & Markides, K. S. (2002). Predictors of decline in MMSE scores among older Mexican Americans.
Journal of Gerontology, 57A (3), M181–M185.
Pallant, J. (2007). SPSS Survival Manual: A step by step guide to data analysis using SPSS for Windows (3rd ed.). New York: McGraw-Hill.
Petersen, R. C. (2000). Mild Cognitive Impairment: Transition between aging and Alzheimer’s disease. Neurology, 15 (3), 93– 101.
Petersen, R. C. (2007). Mild Cognitive Impairment. Continuum: Lifelong Learning in Neurology, 13 (2), 15– 38.
Reid, L. M., & MacLullich, A. M. (2006). Subjective memory complaints and cognitive impairment in older people. Dementia and Geriatric Cognitive
Disorders, 22, 471– 485.
 S. Freitas et al. / Archives of Clinical Neuropsychology 27 (2012) 165–175 175

Ribeiro, P. C., Oliveira, B. H., Cupertino, A. P., Neri, A. L., & Yassuda, M. S. (2010). Desempenho de idosos na bateria cognitiva CERAD: Relações com
variáveis sociodemográficas e saúde percebida [Performance of the elderly in the CERAD Cognitive Battery: Relations with socio-demographic variables
and perceived health]. Psicologia, Reflexão e Cri?tica, 23 (1), 102– 109.
Rossetti, H. C., Lacritz, L. H., Cullum, C. M., & Weiner, M. F. (2011). Normative data for the Montreal Cognitive Assessment (MoCA) in a population-based
sample. Neurology, 77, 1272– 1275.
Rovner, B. W., Broadhead, J., Spencer, M., Carson, K., & Folstein, M. F. (1989). Depression and Alzheimer’s disease. American Journal of Psychiatry, 146,
350–353.
Scazufca, M., Almeida, O. P., Vallada, H. P., Tasse, W. A., & Menezes, P. R. (2009). Limitations of the Mini-Mental State Examination for screening dementia
in a community with low socioeconomic status. European Archives of Psychiatry and Clinical Neurosciences, 259, 8 –15.
Schmand, B., Jonker, C., Hooijer, C., & Lindeboom, J. (1996). Subjective memory complaints may announce dementia. Neurology, 46 (1), 121– 125.
Simões, M. R., Freitas, S., Santana, I., Firmino, H., Martins, C., Nasreddine, Z., et al. (2008). Montreal Cognitive Assessment (MoCA): Versão portuguesa
[Montreal Cognitive Assessment (MoCA): Portuguese version]. Coimbra, Portugal: Serviço de Avaliação Psicológica da Faculdade de Psicologia e de
Ciências da Educação da Universidade de Coimbra [Psychological Assessment Department, Faculty of Psychology and Educational Sciences,
University of Coimbra].
Wu, M., Lan, T., Chen, C., Chiu, H., & Lan, T. (2011). Sociodemographic and health-related factors associated with cognitive impairment in the elderly in

Downloaded from https://academic.oup.com/acn/article/27/2/165/4917 by guest on 07 November 2024

Taiwan. BMC Public Health, 11, 22. doi:10.1186/1471– 2458-11-22.
Yesavage, J. A., Brink, T. L., Rose, T. L., Lum, O., Huang, V., Adey, M., et al. (1983). Development and validation of a geriatric depression screening scale: A
preliminary report. Journal of Psychiatric Research, 17 (1), 37– 49.
Zhao, S., Liu, G., Shen, Y., & Zhao, Y. (2011). Reasonable neuropsychological battery to identify mild cognitive impairment. Medical Hypotheses, 76, 50–53.