Archives of Gerontology and Geriatrics 59 (2014) 450–456

Contents lists available at ScienceDirect

Archives of Gerontology and Geriatrics

journal homepage: www.elsevier.com/locate/archger

Geriatric depression and its relation with cognitive impairment and

dementia
Carol Dillon a,c,*, Marı́a Florencia Tartaglini b, Dorina Stefani b, Pablo Salgado a,
Fernando E. Taragano c, Ricardo F. Allegri a
a
Memory Research Center, Department of Neurology, Hospital General Abel Zubizarreta (GCABA), Buenos Aires, Argentina
b
Instituto de Investigaciones Cardiológicas ‘‘Prof. Dr. Alberto C. Taquini’’ (ININCA-UBA-CONICET), Argentina
c
CEMIC University Institute (CONICET), Buenos Aires, Argentina

A R T I C L E I N F O A B S T R A C T

Article history: Different subtypes of depressive syndromes exist in late life; many of them have cognitive impairment
Received 11 July 2013 and sometimes it is difficult to differentiate them from dementia. This research aimed to investigate
Received in revised form 10 April 2014 subtypes of geriatric depression associated with cognitive impairment, searched for differential
Accepted 24 April 2014
variables and tried to propose a study model. A hundred and eighteen depressive patients and forty
Available online 9 May 2014
normal subjects matched by age and educational level were evaluated with an extensive
neuropsychological battery, scales to evaluate neuropsychiatric symptoms and daily life activities
Keywords:
(DLA). Depressive patients were classified in groups by SCAN 2.1: Major Depression Disorder (MDD) (n:
Geriatric depression
Cognitive impairment
31), Dysthymia Disorder (DD) (n: 31), Subsyndromal Depression Disorder (SSD) (n: 29), Depression due
Dementia to Dementia (n: 27) (DdD). Neuropsychological significant differences (p < 0.05) were observed between
depressive groups, demonstrating distinctive cognitive profiles. Moreover, significant differences
(p < 0.05) were found in DLA between DdD vs all groups and MDD vs controls and vs SSD. Age of onset
varied in the different subtypes of depression. Beck Depression Inventory (BDI) and Mini Mental State
Examination (MMSE) were significant variables that helped to differentiate depressive groups.
Significant correlations between BDI and Neuropsychological tests were found in MDD and DD groups.
Depressive symptoms and its relation with neuropsychological variables, MMSE, cognitive profiles, DLA
and age of onset of depression should be taken into consideration for the study of subtypes of geriatric
depression.
ß 2014 Elsevier Ireland Ltd. All rights reserved.

1. Introduction Late onset depression and cognitive impairment often occur

together, suggesting a close association between them (Migliorelli
Depression and cognitive impairment are among the most et al., 1995; Zubenko et al., 2003). It is not known, however,
important mental health problems in elderly people (Dillon et al., whether depression leads to cognitive decline or vice versa (Jorn,
2013; Solhaug, Romuld, Romild, & Stordal, 2012; Zhao et al., 2001; Schweitzer, Tuckwell, O’Brien, & Ames, 2012).
2012). Both conditions have severe consequences for the patients, The literature on the interplay of depression and dementia has
including diminished quality of life, functional decline, increased always reported contradictory results. Three main hypotheses to
use of services, and high mortality (Macdonald, 1997). Further- explain the association were formulated: (1) depression may be
more, these diseases impact health of caregivers. The World considered a psychological reaction to eroding cognitive capacities
Health Organization (2008) considers the care of these patholo- early in the course of dementia. (2) A common underlying central
gies a risk factor for the development of mental disorders and nervous system disorder may cause depression as well as cognitive
burden. decline in elderly persons: it has been shown that elderly depressed
people have more frequent and more severe white matter and
other subcortical abnormalities on brain magnetic resonant images.
(3) Depression may be associated with high levels of cortisol, which
may lead to neuronal death and dysregulation of the hypothalamic-
* Corresponding author at: CEMIC University Institute, Av E. Galvan 4102, 1431
Buenos Aires, Argentina. Tel.: +54 11 52990372; fax: +54 11 52990372. pituitary-adrenal axis with, as a consequence, hippocampal atrophy
E-mail addresses: [email protected], [email protected] (C. Dillon). and cognitive decline (Bagulho, 2002).

http://dx.doi.org/10.1016/j.archger.2014.04.006
0167-4943/ß 2014 Elsevier Ireland Ltd. All rights reserved.
 C. Dillon et al. / Archives of Gerontology and Geriatrics 59 (2014) 450–456 451

Depression is one of the most studied symptom in mild 2.3. Exclusion criteria
cognitive impairment (MCI) and dementia. There is a high
prevalence of depressive symptoms in vascular dementia Patients with drug or alcohol abuse, with moderate or severe
(31.4%) and in dementia of Alzheimer’s type (DAT) (19.8%) when dementia by the CDR (2 = moderate dementia or 3 = severe
compared with the cognitively normal elderly (13.2%) (Barca, Knut, dementia), and with schizophrenia or schizoaffective disorder
Lacks, & Selback, 2012). Depressive symptoms are often present were excluded from the study.
during early stages of dementia (Dillon et al., 2013), due to this
sometimes it is difficult to differentiate them with elderly patients 2.4. Screening procedure
with Major Depression (Barca et al., 2012).
Late-life depression, defined as a major depressive episode Depressive patients were divided into four different
occurring in older adults (usually after the age of 65 years), is a groups according to DSM IV (APA, 1994) and ICD 10 (1990)
heterogeneous mood disorder frequently associated with cognitive criteria with the SCAN 2.1 (WHO: World Health Organization,
impairment. Late-life depression encompasses both late onset Wing et al., 1990) schedules for clinical assessment in
cases as well as early onset cases that recur or continue into later neuropsychiatry.
years of life. The temporal association between cognitive and SCAN is a set of instruments and manuals aimed at assessing,
depressive symptoms in elderly patients varies widely, yet measuring, and classifying psychopathology and behavior
evidence suggests that depressive illness contributes to the associated with the major psychiatric disorders in adult life.
development of persistent or progressive cognitive deficits in It can be used for clinical, research, and training purposes, and
some individuals (Butters et al., 2008). was developed within the WHO framework. SCAN has a bottom-
Butters et al. (2008), propose that depression alters an up approach where no diagnosis-driven frames are applied in
individual’s risk of cognitive dysfunction, shortening the latent grouping the symptoms. Each symptom is assessed in its own
period between the development of Alzheimer’s disease (AD) right. SCAN has a proven stability. The method used is that of a
neuropathology and the onset of clinical dementia, thus increasing semi-structured standardized clinical interview, with cross-
the incidence and prevalence of AD among older adults with examination of the subject. Rating is done on the basis of
depression. matching the answers of the respondent against the definitions
In our recent study about different subtypes of geriatric of the symptoms in the Glossary, which is an integral part of
depression (MDD, DD, SSD and DdD of Alzheimer type) we found SCAN. All the symptoms and signs and classification items are
clinical and neuropsychological variables that are useful for defined in this Glossary, which is largely based on the
differential diagnosis (Dillon et al., 2011). phenomenology of Jaspers. With SCAN the interviewer decides
This research aimed to analyze these depressive subtypes what to rate on the basis of the subject’s information, always
(MDD, DD, SSD, DdD), which are associated with cognitive bearing in mind the definitions and rating rules (Wing et al.,
impairment in geriatric patients, and tried to propose a study 1990).
model. The study participants were grouped as follows:

2. Materials and methods Group 1: MDD (n: 31);

Group 2: DD (n: 31);
A cross-sectional analytical study was performed. The study Group 3: SSD (n: 29);
took place in a Memory Clinic from a Buenos Aires community- Group 4: DdD: only mild Alzheimer dementia; CDR 1 (CDR
based outpatient hospital, the Hospital General Zubizarreta (public Scale) were recruited (n: 27);
health system). Group 5: Controls (C) (n: 40).
Following this previous study, different variables associated to Screen failures: 12 patients, 5 controls.
Geriatric Depression were studied in this investigation such as
psychiatric symptoms, level of depression, cognitive variables, DLA All of the recruited patients were assessed using a semi-
and age of onset of depression (late onset depression-depressive structured neuropsychiatric interview. Different psychiatric scales
symptoms that began over 65 years of age-, early onset depression- were used, including the Beck Depression Scale, the Hamilton
depressive symptoms that began under the age of 65 years–). In Depression Scale, and the Hamilton Anxiety Scale.
addition, we analyzed which variables had more weight to Patients and normal controls were matched by age, education,
differentiate depressive groups. and overall cognitive status using the MMSE (Folstein, Folstein, &
McHugh, 1975).
2.1. Study population Each patient underwent an extensive neuropsychological
battery to evaluate the following areas of cognitive ability:
A total of 118 depressive patients with cognitive complaints
and 40 controls (those lacking signs and symptoms of depression Orientation: MMSE (Folstein et al., 1975).
or cognitive impairment) from the general population, matched by Attention: Digit span (forward and backward) (Wechsler,
age and educational level (age 63  8.28 years, educational level 1988); Trail making test ‘‘A’’ (Reitan, 1958).
9.8  4.3 years), were recruited. Language: Boston naming test (BNT) (local version adapted by
Allegri, Mangone, Rymberg, Fernandez, & Taragano, 1997),
2.2. Inclusion criteria Semantic fluency (SF) (Benton, Hannay, Varney, & Spreen,
1983), Verbal fluency (VF) (Benton et al., 1983).
Included in the study were patients who consulted or were Memory: Signoret memory battery (Signoret & Whiteley,
referred to our Memory Clinic, presenting depressive symptoms 1979): episodic memory (immediate logic memory (ILM),
that were due to psychiatric causes or were related to with mild delayed logic memory (DLM); verbal serial learning (VSL),
DAT (CDR 1: mild dementia; CDR: Clinical Dementia Rating delayed serial memory (DSM), cued recall (CR), recognition
Scale), who were more than 55 years old but less than 80 years (Recog).
old, and with a Hamilton Depression Scale rating >9 points and Visuospatial abilities: Clock drawing test (Freedman et al.,
BDI >9. 1994).
452 C. Dillon et al. / Archives of Gerontology and Geriatrics 59 (2014) 450–456

Executive functions: Trail making test ‘‘B’’ (Reitan, 1958), VF language. DdD group showed a cortical profile in memory
(Benton et al., 1983). (impairment in learning and recognition) while the other
DLA were determined by Lawton and Brody (1969) self- depressive groups (MDD, DD, SSD) had a subcortical profile in
maintaining and instrumental activities of daily living scale. memory (impairment in memory with good recognition). More-
over, DdD showed impairment in language tests (cortical profile)
Written informed consent was obtained from each subject and SSD had a tendency toward language impairment, while MDD
after they had been given a full explanation of the study. The and DD did not show significant differences in BNT vs controls. See
research was performed in accordance with the ICH Good Clinical Table 2.
Practice guidelines, the latest revision of the 1964 Helsinki
Declaration (as amended in Tokyo 1975, Venice 1983, Hong Kong 3.4. Correlations between neuropsychological tests and BDI
1989, Republic of South Africa 1996, Edinburgh 2000, Washington
2002, Tokyo 2004, Seoul 2008) and the Buenos Aires Government Correlations were found in MDD group between Beck and:
Health Authorities. MMSE, Verbal and SF, BNT and Clock Drawing Test, TMB, see
Table 4. Also, DD group had correlations between Beck and ILM,
2.5. Data analysis TMA, TMB (see Table 3).

Demographic variables for both populations (patients and 3.5. DLA

controls) as well as the results of neuropsychiatric and neuropsy-
chological general global tests were expressed as means, standard Significant differences (p < 0.05) were found in DLA between
deviation and medians. Quantitative variables were compared different depressive groups (DdD vs all depressive groups; MDD vs
using ANOVA. Schefeé test was also performed for multiple SSD) and between depressive groups and controls (DdD vs
comparisons. The relationship between late onset and early onset Controls; MDD vs Controls). Patients with DdD have moderate
depression was compared by the chi-square test. Predictive factors to severe impairment in DLA. MDD patients have mild to moderate
for depression were analyzed using the odds ratio (OR), with 95% impairment of DLA. DD and SDD are independent or have mild
confidence intervals (95% CI). A multivariate analysis (multiple impairment in DLA (see Table 4).
discriminant analysis) was performed using the neuropsychiatric
and neuropsychological variables. 3.6. Late vs early onset depression
Data was analyzed using the SSPS 15 statistical package.
Table 5 shows age of onset of the different subtypes of
3. Results depression. Note that MDD and DD had more cases with age of
onset before 65 years and DdD and SSD had a tendency to begin
3.1. Demographic data with depression after 65 years.
To be able to find the OR, we grouped MDD with DD and DdD
There were no significant differences in age and educational with SSD and found that late onset depression (age onset before 65
level among the five groups. Significant cognitive level (MMSE) years) was greater in patients with DdD and SSD (x2 0.006; OR 3.09
differences (p < 0.05) between the DdD and the other depressive (1.3-7) (see Table 5).
groups (MDD, DD, SSD) and between DdD and normal controls are
depicted in Table 1. 3.7. Multiple discriminate analyses (MDA)

3.2. Depressive symptoms Lambda Wilks’ method was administered introducing the
variables step by step that allows us to evidence which are the
Significant differences were observed in BDI p < 0.05 variables that better discriminate the groups of the dependent
between controls and depressive groups (MDD, DD, DdD) and variables.
between SSD and the other depressive groups (MDD, DD and The dependent variables were the four depressive groups
DdD) p < 0.05 (Beck, Ward, Mendelson, Mock, & Erbaugh, 1961; (MMD, DD, SSD, DdD) and the controls. Significant variables from
see Table 2). the univariate analyses (ANOVA) were used (neuropsychological
test and BDI).
3.3. Neuropsychological variables The variables selected by Lambda Wilks’ method were in first
place MMSE (tolerance 1.000; F 21.643) and in second place BDI
From all the neuropsychological variables studied, the ones that (tolerance 0.977; F 4.708). Using these variables (MMSE and Beck
best differentiated between depressive groups were memory and Inventory 44.3% of the cases were grouped.

Table 1
Demographic data.

MDD DD SSD DdD Controls p

N 31 31 29 27 40
Age of onset 48.6  15.3 55.3  15.2 59.2  15.4 60.3  17.0
Sex (F/M) 23/8 24/7 23/6 20/7 32/8
Age (years) 64.9  6.9 66.6  9.7 66.2  9.7 70.7  7.7 66.0  6.7 ns
Education (years) 9.6  3.7 8.8  4.5 9.4  4.4 8.0  3.4 10.7  3.1 ns

MMSE 27.2  3.1 26.9  3.3 27.1  2.1 21.8  4.4 29.0  0.9 DdD vs control < 0.0001
DdD vs MDD < 0.0001
DdD vs DD = 0.001
DdD vs SSD < 0.0001

All depressives: MDD, DD, SSD, DdD.

 C. Dillon et al. / Archives of Gerontology and Geriatrics 59 (2014) 450–456 453

Table 2
BDI and neuropsychological battery (memory and language).

MDD DD SSD DdD Control p (<0.5)

BDI 23.4  9.0 22.2  9.5 14.2  7.5 22.1  10.4 4.4  3.1 Control vs all depressives p < 0.05
SSD vs (MDD, DD, DdD) p < 0.05

Memory
DLM 5.3  3.0 4.6  2.0 4.9  2.2 1.9  1.8 7.4  2.1 Control vs all depressives p < 0.05
DdD vs (MDD, DD, SSD) p < 0.05
VSL 7.5  2.3 7.5  2.1 7.4  2.1 4.9  1.7 9.4  1.4 Control vs all depressives p < 0.05
DdD vs (MDD, DD, SSD) p < 0.05
DSM 6.1  2.5 5.6  2.8 5.2  2.5 2.3  2.4 8.1  1.5 Control vs all depressives p < 0.05
DdD vs (MDD, DD, SSD) p < 0.05
CR 8.4  3.2 8.8  3.1 8.5  3.5 5.5  3.1 11.1  1.0 Control vs (MDD, SSD, DdD) p < 0.05
DdD vs (MDD, DD, SSD) p < 0.05
Recognition 10.5  2.1 10.6  2.3 10.8  1.5 8.2  3.3 11.7  0.4 Control vs DdD p < 0.05
DdD vs (MDD, DD, SSD) p < 0.05

Language
BNT 45.8  6.8 46.2  7.7 44.5  7.2 37.4  10.1 51.6  4 Control vs (SSD and DdD)
DdD vs (MDD, DD, SSD) p < 0.05
SF 14.4  4 14.6  5.3 13.7  4 9.7  3.5 19.6  6 Control vs all depressives p < 0.05
DdD vs (MDD and DD) p < 0.05

All depressives: MDD, DD, SSD, DdD.

Table 3 4.2. Cognitive functions

Correlations. Significant correlations between Beck and Neuropsychological tests.

MDD Global cognitive impairment was found in depressive patients

MMSE SF VF BNT TMB Clock drawing in comparison to controls. Nevertheless, different profiles were
test demonstrated between groups. MDD and DD had greater
impairment in attention (DD) and executive functions (MDD)
Beck
Pearson correlation 0.386 0.436 0.42 0.412 0.446 0.382 with a subcortical profile in memory (impairment in memory with
Significant (p) 0.035 0.016 0.021 0.024 0.014 0.037 good recognition) while DdD had a cortical profile with more
DD impairment in memory (learning and recognition) and language
(BNT and SF). SSD showed a mixed profile, subcortical in memory
ILM TMA TMB
but with impairment in language (cortical), this group shows mild
Beck level of depression with moderate cognitive impairment. There-
Pearson correlation 0.374 0.361 0.377
fore, SSD group could be considered in risk of dementia (MCI
Significant (p) 0.038 0.05 0.04
associated to depressive symptoms).
MCI represents an intermediate state of cognitive function
between the changes observed in aging and the dementia
diagnosis. It is currently defined as a syndrome with impairment
4. Discussion of memory or another cognitive domain that does not interfere
substantially with personal autonomy (Petersen et al., 2001). In
Different variables associated to Geriatric Depression were DSM-5 classification MCI is recognized as an entity and is named
analyzed in this investigation such as psychiatric symptoms: level ‘‘Minor Neurocognitive Disorder’’.
of depression, cognitive variables and DLA. Depression is the most studied and common symptom in MCI
(or Minor Neurocognitive Disorder) and dementia (Dillon et al.,
4.1. Depressive symptoms 2013).

Level of depression was in general moderate. However the SSD 4.3. Correlations between cognitive variables and depressive
group showed a significant difference in level of depression with symptoms
the other 3 depressive groups (MDD, DD and DdD). Scores of BDI
were used to correlate levels of depression and cognitive functions When depressive symptoms (BDI) were correlated with
in depressive patients; moreover, it was also useful in the multiple cognitive functions in the different depressive groups, MDD and
discriminant analyses where BDI and MMSE were the most DD were the only ones that had significant correlations between
important variables to distinguish between groups. these variables. Depression in these patients had an impact on

Table 4
Daily life activities.

DLA MDD DD SSD DdD Controls p

Independ/dependent (%) 45%/55% 71%/29% 79%/21% 0%/100% 98%/2% DdD vs all groups p < .05
Frequency (I/D) 14/17 22/9 22/6 0/27 39/1 MDD vs (Controls and SSD) p < .05
Mean (SD) 3 (2.6) 1.7 (1.4) 1.3 (0.8) 6.7 (2.3) 1.0 (0.1)
Media 2 1 1 6.5 1

References: Independ = independent; I = Independent; D = Dependent.

All groups = MDD, DD, SSD and Controls.
454 C. Dillon et al. / Archives of Gerontology and Geriatrics 59 (2014) 450–456

Table 5 Ávila-Funes, Melano-Carranza, Hélène Payette, and Amieva

Age of onset. Depression onset after 65 years.
(2007) demonstrated that the presence of depressive symptoms is
Yes No x2 (p) a risk factor for the development of impairment in instrumental
MDD 4 27 DLA. However, this risk was not observed for basic life activities.
DD 8 21 0.02 Functional impairment is multifactor, starts and develops in an
SSD 11 18 insidious way and manifests in a jerarquic order. As a consequence,
DdD 21 13 instrumental DLA are affected in the first place followed by basic
Yes No x2 (p) OR (IC) life activities.
SSD and DdD 24 31 0.006 3.09 (1.3–7) Depressive symptoms as well as cognitive performance (MMSE)
MDD and DD 12 48 are related with DLA in the elderly population.
DLA are an important disability factor in patients with AD;
moreover, they determine the transition from MCI to dementia.
cognition. DdD and SSD had no correlations between depressive
symptoms and neuropsychological variables; this means that 4.4.1. Late and early onset depression
cognitive variables are not significantly affected by depressive On one hand, different studies (Alexopoulos, 2003; Butters
symptoms. et al., 2008; Dillon et al., 2009) had demonstrated that late onset
In the multiple discriminate analyses (MDA), BDI and MMSE depression was associated with cognitive impairment, cerebro-
were significant variables that helped to differentiate depressive vascular disease and higher risk to develop degenerating process.
groups. On the other hand, early onset depression is also related to
Balash et al. (2013) found that presence of Subjective structural changes in specific brain regions (hippocampal-entorr-
memory complaints (SMC) was inversely correlated with MMSE inal degeneration) (Bell-McGinty et al., 2002; Sheline, 2011) and
scores and that depressive symptoms were higher among cognitive impairment.
subjects with SMC. Subjective memory complaints are associat- In the present study age of onset of depressive episodes was
ed with sub-syndromal depression and anxiety in healthy analyzed and significant differences were found between the
cognitively normal elders. different depressive groups. Depression due Dementia (DdD) and
Milian et al. (2013) in an investigation where they evaluated the SSD patients tended to begin with depressive symptoms since 65
specificity of separation dementia from depression of Mini-Cog, years old.
Clock Drawing Test (CDT), and the MMSE, found that concerning In a recent investigation about late vs early onset depression
the depressed patients, the MMSE demonstrated significant higher (Dillon et al., 2009), late-onset depression patients scored lower
specificity than the Mini-Cog and the CDT, but also showed the and exhibited more severe impairment in memory domains than
lowest sensitivity for the detection of dementia. Surprisingly, only early-onset depression patients (p < 0.05). Late onset geriatric
the MMSE was susceptible for the depression severity. depression may be a manifestation of brain degeneration, and the
On one hand, in Veiel (1997) about neuropsychological deficit initial symptom of a dementia.
associated with Major Depression concludes that a global cognitive From the data obtained in this investigation and the variables
impairment exists. On the other hand, Elliot (1998) suggests that in studied, a study model was developed for patients older than 55
spite of the fact that there is impairment in multiple cognitive years old that have depressive symptoms and cognitive im-
domains the executive function is predominantly affected in pairment.
depressive patients associated with dysfunction of frontal lobe.
Several investigations (Atre-Vaidya et al., 1998; Backman, Hill, 4.4.2. Study model
& Forsell, 1996; Bearden, Hoffman, & Cannon, 2001) demonstrated It postulates the possibility to study different cognitive and
that Major Depressive patients show a subcortical impairment in functional variables that can help physicians in their general
verbal memory with problems in memory recall but with good practice (see Table 6).
response in CR and recognition. Depressive patients have
impairment in executive functions and hence fail in searching 4.4.3. DdD
strategy and generate low fluency (verbal and phonologic). In Patients with depressive symptoms and moderate to severe
geriatric depressive patients, sustained and selective attention as impairment in DLA (more than 4 points in Lawton’s scale).
well as forced attention and inhibitory control (working memory) Presence of a cortical cognitive profile (impairment in memory
are impaired (Bearden et al., 2001; Martinez-Aran et al., 2002). In (poor learning and recognition) and language), generally MMSE
general terms, language domain is not impaired (Taragano, Allegri, score <24 points and without correlation between depressive
Mangone, & Paz, 1998). symptoms and neuropsychological variables. Late age of onset:
presence of depressive symptoms beginning since 65 years old (see
4.4. DLA Table 6).

This research demonstrated that approximately a 50% of 4.4.4. Major depressive disorder (MDD)
depressive patients have impairment in DLA. MDD and DdD Patients with depressive symptoms and mild to moderate
patients were the most affected ones. impairment in DLA (more than 2 points in Lawton’s scale). It is

Table 6
Differential variables. Proposed study model.

Depressive group Depressive symptoms MMSE Impairment in daily life activities Age of onset Cognitive profile Correlations (DS and NPS)

MDD Moderate >24 Points Mild to moderate Early Subcortical Yes

DD Moderate >24 Points Independent to mild Early Subcortical Yes
SSD Mild >24 Points Independent to mild Late Subcortical No
DdD Moderate <24 Points Moderate to severe Late Cortical No

References: Depressive Symptoms (Beck Depression Inventory Score).

Impairment in Activities: DLA (Lawton’s Scale); Correlations (DS and NPS): Correlations between depressive symptoms and neuropsychological symptoms or variables.
 C. Dillon et al. / Archives of Gerontology and Geriatrics 59 (2014) 450–456 455

frequent to find correlations between depressive symptoms and Dillon were responsible for the statistical design of the study and
neuropsychological variables (especially memory and executive for carrying out the statistical analysis. All authors have read and
functions). These patients present a subcortical cognitive profile approved the final manuscript.
with MMSE >24 points. They tend to have early onset depression
(presence of depressive symptoms after 65 years old; see Table 6). Role of funding source

4.4.5. Dysthymic disorder (DD) Funding for this study was provided by scientific research
Patients have DLA which are not affected or are affected but in a grants from the CONICET of Argentina (Carol Dillon and Ricardo F.
mild level (less than 3 points in Lawton’s Scale). These patients Allegri), from the Secretary of Health of Buenos Aires Government
present a subcortical cognitive profile with correlations between (GCABA) and from Carrillo-Oñativia Scholarship 2005–2006 –
depressive symptoms and neuropsychological variables (especial- Carol Dillon and Ricardo F. Allegri); the CONICET and the Secretary
ly memory and attention). Early onset depression: see Table 6. of Health of Buenos Aires Government had no further role in study
design; in the collection, analysis and interpretation of data; in the
4.4.6. SSD writing of the report; and in the decision to submit the paper for
Patients with DLA which are not affected or are affected but in a publication.
mild level (less than 3 points in Lawton’s Scale). These patients
present a subcortical cognitive profile without correlations Conflict of interest
between depressive symptoms and neuropsychological variables
(mild depressive symptoms with important cognitive im- All authors declare that they have no conflicts of interest.
pairment). Late onset depression: see Table 6.
Acknowledgments
5. Conclusions
This research was supported by scientific research grants from
Depressive symptoms and its relation with neuropsychological the CONICET of Argentina (C.D. and R.F.A), from the Secretary of
variables, MMSE, cognitive profiles, DLA and age of onset of Health of Buenos Aires Government (GCABA) and from Carrillo-
depression should be taken into consideration for the study of Oñativia Scholarship 2005-2006 – CD and R.F.A).
subtypes of geriatric depression. The views expressed in the publication are those of the authors
and not necessarily those of the Ministry of Health of Argentina or
5.1. Limitations the Secretary of Health of Buenos Aires Government.

The cohort referred to a memory clinic with memory

References
complaints is a biased sample and the results cannot be
generalized to other non-memory symptomatic cohorts. Many Alexopoulos, G. (2003). Role of executive function in late-life depression. Journal of
variables were examined in this work. It is known that this can Clinical Psychiatry, 64, 18–23.
inflate the probability of erroneously finding statistically signifi- Allegri, R. F., Mangone, C. A., Rymberg, S., Fernandez, A., & Taragano, F. E. (1997).
Spanish version of the Boston naming Test in Buenos Aires. Clinical Neuropsycholo-
cant differences. Correction for multiplicity was made for all tests gy, 1, 416–420.
of a given variable to preserve the overall type I error at 5%. American Psychiatric Association. (1994). Diagnosis and statistical manual of disorders
However, correction for multiple test taking into account that (DSM-IV). Washington, DC: American Psychiatric Association.
Atre-Vaidya, N., Taylor, M. A., Seidenberg, M., Reed, R., Perrine, A., & Glick-Oberwise, F.
several outcome variables were analyzed was not done because it (1998). Cognitive deficits, psychopathology, and psychosocial functioning in
is difficult to treat those variables as common multivariate bipolar mood disorder. Neuropsychiatry Neuropsychology and Behavioral Neurol-
distribution. In a sensitivity analysis, the overall type I error for all ogy, 11, 120–126.
Ávila-Funes, J. A., Melano-Carranza, E., Hélène Payette, H., & Amieva, H. (2007).
test for a given variable was reduced to 0.2% before applying the
Depressive symptoms as a risk factor for dependence in elderly people. Salud
Dunnet or Schefé method. The majority of the statistically Pública Méxicana, 49, 5.
significant results versus controls were retained. In this alterna- Backman, L., Hill, R. D., & Forsell, Y. (1996). The influence of depressive symptomatol-
ogy on episodic memory functioning among clinically no depressed older adults.
tive analysis, differences among depressive were more difficult to
Journal of Abnormal Psychology, 105, 97–105.
be detected, however, this sensitivity analysis applied an Bagulho, F. (2002). Depression in older people. Current Opinion in Psychiatry, 15,
additional correction for type I error that is known to be overly 417–422.
conservative. Balash, Y., Mordechovich, M., Shabtai, H., Giladi, N., Gurevich, T., & Korczyn, A. D.
(2013). Subjective memory complaints in elders: Depression, anxiety, or cognitive
The model was based in variables studied in a reduced decline? Acta Neurologica Scandinava, 127, 344–350.
population. Only four depressive subgroups were considered. Of Barca, M. L., Knut, E., Lacks, J., & Selback, G. (2012). Factors associated with Depression
these four groups the smallest one was one of patients with DdD Disorder in Alzheimer’s Disease are different from those found for other Dementia
Disorders. Journal of Dementia and Geriatric Cognitive Disorders Extra, 2, 19–28.
( = 27 patients), patients that had only mild Alzheimer dementia. Beck, A. T., Ward, C. H., Mendelson, M., Mock, J., & Erbaugh, J. (1961). An inventory for
However, the difference of patients with the other groups is not measuring depression. Archives of General Psychiatry, 4, 53–63.
significant. Moreover, this group is only made up of with Bearden, C., Hoffman, K., & Cannon, T. (2001). The neuropsychology and neuroanatomy
of bipolar affective disorder: A critical review. Bipolar Disorders, 3, 106–150.
Alzheimer’s dementia in a mild state (CDR 1), this is less prevalent Bell-McGinty, S., Butters, M. A., Meltzer, C. C., Greer, P. J., Reynolds, C. F., & Becker, J. T.,
that the pathologies of the other three groups. 3rd. (2002). Brain morphometric abnormalities in geriatric depression: Long-term
As this study was developed with a reduced population, authors neurobiological effects of illness duration. American Journal of Psychiatry, 159,
1424–1427.
cannot exclude that some patients can suffer cognitive impairment
Benton, A. L., Hannay, H. J., Varney, N. R., & Spreen, O. (1983). Contributions to
(in an initial phase). neuropsychological assessment. New York, NY: Oxford University Press.
Butters, M. A., Young, J. B., Lopez, O., Aizenstein, H. J., Mulsant, B. H., Reynolds, C. F., et al.
(2008). Pathways linking late-life depression to persistent cognitive impairment
Authors’ contribution
and dementia. Dialogues in Clinical Neuroscience, 10, 345–357.
Dillon, C., Allegri, R. F., Serrano, C. M., Iturry, M., Gaser, F., Salgado, P., et al. (2009). Late
C. Dillon and R. Allegri conceived the project, designed the versus early onset geriatric depression in a memory research center. Neuropsychi-
study, supervised data collection and wrote the paper. C. Dillon, atric Disease and Treatment, 5, 517–526.
Dillon, C., Machnicki, G., Serrano, C. M., Rojas, G., Vazquez, G., & Allegri, R. F. (2011).
M.F. Tartaglini, D. Stefani and R. Allegri and F. Taragano involved in Clinical manifestations of geriatric depression: Toward a proposed subtyping of
data collection and writing assistance. R. Allegri, P. Salgado and C. geriatric depression. Journal of Affective Disorders, 134, 177–187.
456 C. Dillon et al. / Archives of Gerontology and Geriatrics 59 (2014) 450–456

Dillon, C., Serrano, C. M., Castro, D., Leguizamon, P. P., Heisecke, S., & Taragano, F. E. Sheline, Y. I. (2011). Depression and the hippocampus: Cause or effect? Biological
(2013). Behavioral symptoms related to cognitive impairment. Neuropsychiatric Psychiatry, 70, 308–309.
Disease and Treatment, 9, 1443–1455 PMID:24092982. Signoret, J. L., & Whiteley, A. (1979). Memory battery scale. International Neuropsycho-
Elliot, R. (1998). The neuropsychological profile in unipolar depression. Trends in logical Society, 2–26.
Cognitive Sciences, 11, 447–454. Solhaug, H. I., Romuld, E. B., Romild, U., & Stordal, E. (2012). Increased prevalence of
Folstein, M. F., Folstein, S. E., & McHugh, P. R. (1975). ‘‘Mini-mental state’’ A practical depression in cohorts of the elderly: An 11-year follow-up in the general population
method for grading the cognitive state of patients for the clinician. Journal of – the HUNT study. International Psychogeriatrics, 24, 151–158.
Psychiatric Research, 12, 189–198. Taragano, F. E., Allegri, R. F., Mangone, C. A., & Paz, J. (1998). Similitudes y diferencias
Freedman, M., Learch, K., Kaplan, E., Winocur, G., Shulman, K. I., & Delis, D. (1994). Clock entre la depresión geriátrica y la demencia con depresión. Alcmeon, 9, 250–260.
Drawing: SA neuropsychological analysis. New York, NY: Oxford University Press. Schweitzer, I., Tuckwell, V., O’Brien, J., & Ames, D. (2012). Is late onset depression a
Jorn, A. F. (2001). History of depression as a risk factor for dementia: An updated prodrome to dementia? Internal Journal of Geriatric Psychiatry, 17, 997–1005.
review. Australian and New Zealand Journal of Psychiatry, 35, 776–781. Veiel, H. O. F. (1997). A preliminary profile of neuropsychological deficits associated
Lawton, M. P., & Brody, E. M. (1969). Assessment of older people: Self-maintaining and with major depression. Journal of Clinical and Experimental Neuropsychology, 19,
instrumental activities of daily living. Gerontologist, 9, 179–186. 587–603.
Macdonald, A. J. D. (1997). Mental health in old age. British Medical Journal, 315, Wechsler D. Test de inteligencia para adultos (WAIS), 1988, Editorial Paidos; Buenos
413–417. Aires.
Martinez-Aran, A., Vieta, E., Colom, F., Reinares, M., Benabarre, A., Torrent, C., et al. Wing, J. K., Babor, T., Brugha, T., Burke, J., Cooper, J. E., Giel, R., et al. (1990). SCAN.
(2002). Neuropsychological performance in depressed and euthymic bipolar Schedules for Clinical Assessment in Neuropsychiatry. Archives of General Psychia-
patients. Neuropsychobiology, 46(Suppl. 1), 16–21. try, 47, 6.
Migliorelli, R., Teson, A., Sabe, L., Petracchi, M., Leiguarda, R., & Starkstein, S. E. (1995). World Health Organization. (2008). The Global Burden of Disease 2004 update. http://
Prevalence and correlates of dysthymia and major depression among patients with www.who.int/healthinfo/global_burden_disease/GBD_report_2004update_full.pdf
Alzheimer’s disease. American Journal of Psychiatry, 152, 37–44. (Accessed 12.5.2014).
Milian, M., Leiherr, A. M., Straten, G., Müller, S., Leyhe, T., & Eschweiler, G. W. (2013). Zhao, K. X., Huang, C. Q., Xiao, Q., Gao, Y., Liu, Q. X., Wang, Z. R., et al. (2012). Age and risk
The Mini-Cog, Clock Drawing Test, and the Mini-Mental State Examination in a for depression among the elderly: A meta-analysis of the published literature. CN
German Memory Clinic: Specificity of separation dementia from depression. Spectrums, 17, 142–154.
International Psychogeriatrics, 25, 96–104. Zubenko, G. S., Zubenko, W. N., McPherson, S., Spoor, E., Marin, D. B., Farlow, M. R., et al.
Petersen, R., Doody, C., Kurz, R., Mohs, A., Morris, R. C., Rabins, J. C., et al. (2001). Current (2003). A collaborative study of the emergence and clinical features of the major
concepts in mild cognitive impairment. Archives of Neurology, 58, 985–1992. depressive syndrome of Alzheimer’s disease. American Journal of Psychiatry, 160,
Reitan, R. M. (1958). Validity of the Trail Making Test as an indication of organic brain 857–866.
damage. Perceptual and Motor Skills, 8, 271.