European Neuropsychopharmacology (]]]]) ], ]]]–]]]

www.elsevier.com/locate/euroneuro

Cognitive impairment in remitted

and non-remitted depressive patients:
A follow-up comparison between
first and recurrent episodes
Miquel Rocaa,b,n, Emilio López-Navarroa,b, Saray Monzóna,
Margalida Vivesa,b, Mauro García-Toroa,b,
Javier García-Campayob,c, John Harrisond, Margalida Gilia,b

a
Institut Universitari d'Investigació en Ciències de la Salut (IUNICS), University of Balearic Islands, Palma
de Mallorca, Spain
b
Red de Actividades Preventivas y Promoción de la Salud en Atención Primaria (RediAPP),
Institute Carlos III, Spain
c
Department of Psychiatry, Miguel Servet Hospital, University of Zaragoza, Zaragoza, Spain
d
Alzheimer Center, VU University Medical Center, Amsterdam, The Netherlands

Received 17 March 2015; received in revised form 8 June 2015; accepted 24 July 2015

KEYWORDS
Major depressive dis-
order;
Cognition;
Recurrent depression;
Clinical remission;
Executive functions
Abstract
Cognitive impairment is a core symptom of depressive disorders associated with poor social
function. New research is needed to analyze depression-related symptoms in cognitive
impairment and to observe if they are reversible or not during clinical remission in patients
with or without previous episodes. None of the previous studies has analyzed the differences
between first and recurrent episodes in a long-term follow-up study related with remission
state. The aim of our study was to compare cognitive performance and assess the impact of
previous depressive episodes in a sample of patients in acute phase and in remission six month
later. 79 depressive patients were assessed at baseline. The instruments used for clinical and
cognitive assessment were: Hamilton Depression Rating Scale, Mini-Mental State Examination
and the Clinical Global Impression Rating Scales, Trail Making Test parts A and B, Digital Span
subtest of WAIS, Stroop Colour Word Test, Tower of London, Controlled Verbal Fluency Task,
Semantic Verbal Fluency and Finger Tapping Test. A repeated measures MANCOVA with
education as covariate was used. No differences were found at baseline between first episode
and recurrent depressive patients. At six month, remitted patients scored significant better in
TMT-A, TMT-B, Animals and Tower of London total time. Remitted first depressive patients

n
Corresponding author at: Institut Universitari d'Investigació en Ciències de la Salut (IUNICS), University of Balearic Islands, Ctra
Valldemossa km 7,5, Balearic Islands, Spain. Tel.: +34 971173081; fax: +34 971259935.
E-mail address: mroca@uib.es (M. Roca).

http://dx.doi.org/10.1016/j.euroneuro.2015.07.020
0924-977X/& 2015 Elsevier B.V. and ECNP. All rights reserved.

Please cite this article as: Roca, M., et al., Cognitive impairment in remitted and non-remitted depressive patients: A follow-up
comparison between first and.... European Neuropsychopharmacology (2015), http://dx.doi.org/10.1016/j.euroneuro.2015.07.020
2 M. Roca et al.

scored significant worse than remitted recurrent depressive patients. The main finding of the
study is the effect of remission on cognitive function despite previous episodes. However first
episode remitted patients seemed to have poor access to long term memory than recurrent
remitted patients.
& 2015 Elsevier B.V. and ECNP. All rights reserved.

1. Introduction associated to the number of previous depressive episodes and

Depression is a disorder associated with impairment of
cognitive function (Bora et al., 2013; McIntyre et al.,
2013). Cognition across a variety of domains is impaired,
including executive functioning. These deficits are associated
with poor social function (Rock et al., 2013). The nature and
specificity of these cognitive impairments remains unclear.
Some authors have argued that difficulties with social func-
tion are best explained in the context of the commonly
observed cognitive deficits (Roca et al., 2015). Others have
characterized these difficulties in terms of ‘social cognition’
deficits (Ladegaard et al., 2014; Weightman et al., 2014),
though it is unclear whether social cognition is a qualitatively
distinct, modular component of cognition (Aboulafia-Brakha
et al., 2011).
Cognitive disturbances appears to be impacted by a variety
of clinical factors, including severity, comorbidity, subtype of
depression and duration of illness (Hasselbalch et al., 2011;
McDermott and Ebmeier, 2009). Attention, visual learning,
verbal memory and executive functions are significantly
impaired in major depressive disorder patients without pre-
vious depressive episodes compared to healthy controls (Lee
et al., 2012). Cognitive deficits leading to poor psychomotor
speed and memory functioning were associated with clinical
state, whereas attention and executive functioning are pro-
posed to be likely trait-markers of the disease (Lee et al.,
2012). Cognitive deficits seems to be more severe in recurrent
episodes (Gorwood and Corruble, 2008; Herrmann et al.,
2007). In a sample of 1140 depressive patients Gorwood
et al. (2014) reported that after 6–8 weeks of treatment, after
controlling for age, educational level and professional activity,
‘psychomotor speed’ was correlated with the number of
previous episodes in patients with clinical remission. However,
a recent meta-analysis did not report significantly worse
performance between samples including both first episode
(FE) and recurrent episodes (RE) compared with studies solely
examining patients in their first episode (Lee et al., 2012).
The relation between cognitive disturbances and the
course of depression is a crucial question to understand the
conceptual models of the disease (Bhardwaj et al., 2010;
Preiss et al., 2009). New research is needed to analyze
depression-related symptoms in cognitive impairment and to
observe if they are reversible or not during clinical remission
in patients with or without previous depressive episodes.
None of the related papers has analyzed the differences
between FE or RE depressive patients in a long-term follow-
up study and related with remission state.
Our study compares cognitive performance in a sample of
patients in acute status and six months later, to determine the
presence of a different clinical pattern of cognitive impairment
how remission could mediate it. We hypothesized that number
of past episodes would affect cognitive performance but this
effect may lessen as a function of clinical remission.
2. Experimental procedures
2.1. Design and setting
An observational longitudinal cohort study was designed to contact
as many patients as possible. Patients were individually recruited
and included consecutively by clinical psychiatrists from four health
care centres. Inclusion criteria were 1) diagnosis of DSM-IV-TR Major
Depressive Disorder (MDD), 2) age between 18 and 55 years, 3) at
least a total score of 17 on Hamilton Depression Rating Scale, 4) be
able to read and understand native language, and 5) have signed
informed consent. Exclusion criteria were: 1) history of medical
conditions that could entail cognitive deterioration, 2) history of
head injury or neurological disorder, 3) current psychotic symp-
toms, 4) current treatment with antipsychotic or mood stabilizer
agents, 5) electroconvulsive therapy in the 6 months prior to the
study, 6) Mini-Mental State Examination score lower or equal to 25,
and 7) diagnosis of intellectual disability or inability to understand
or complete the cognitive assessment. No payment was made due
participation in the study. Informed consent was redacted and
approved by the Ethics and Clinical Research Committee of the
Balearic Islands (Palma de Mallorca, Spain).
Patients were assessed at baseline and six month follow-up. In
the first visit demographic data were collected, while cognitive and
clinical assessment took place at baseline and at six months.
Clinical remission at follow-up was defined as having a Hamilton
Depression Rating Scale total score r7 (Rush et al., 2006). All
neuropsychological assessments were performed by two trained
psychologist (SM & MV), and carried out in the morning in order to
avoid confounding effects like mood and cortisol fluctuation.
2.2. Measurements
2.2.1. Clinical assessment
Socio-demographic and clinical questionnaire designed specifically
for this study. This questionnaire collected information about
gender, date of birth and age, marital status, education level and
occupational status, age of disorder onset, number of previous
episodes and current pharmacological treatment (type of drug,
mean dose and time of administration).
17-item Hamilton Depression Rating Scale (HDRS17) (Hamilton,
1960). The HDRS17 is the most commonly used scale in clinical
practice and research in mood disorders to rate symptom severity in
depressive disorders.
Mini-Mental State Examination (MMSE) (Folstein et al., 1975). It
is a brief 30-point questionnaire of a broad array of cognitive
functions including orientation, memory and language and is often
employed as a screening tool to identify and exclude potential
patients with early onset dementia.

Please cite this article as: Roca, M., et al., Cognitive impairment in remitted and non-remitted depressive patients: A follow-up
comparison between first and.... European Neuropsychopharmacology (2015), http://dx.doi.org/10.1016/j.euroneuro.2015.07.020
Cognitive impairment in remitted and non-remitted depressive patients 3

Table 1 Sociodemographic and clinical characteristics.

Variable Total n =79 First episode n= 26 Recurrent depression n =53 P-value

Gender n (%)
Male 17 (21.52) 5 (19.23) 12 (22.64) 0.729
Female 62 (78.48) 21 (80.77) 41 (77.36)
Age, years mean (SD) 46.11 (8.28) 44.46 (8.77) 47.07(7.93) 0.218
Civil status n (%)
Single 10 (12.66) 3 (11.54) 7 (13.21) 0.641
Married 49 (62.03) 16 (61.54) 33 (62.27)
Widowed 3 (3.80) 2 (7.69) 1 (1.88)
Separated 17 (21.51) 5 (19.23) 12 (22.64)
Education n (%)
Incomplete primary 3 (3.79) 2 (7.69) 1 (1.88) 0.560
Complete primary 46 (58.24) 13 (50) 33 (62.27)
Secondary 18 (22.79) 6 (23.08) 12 (22.64)
University 12 (15.19) 5 (19.23) 7 (13.21)
Employment status n (%)
Employed 49 (62.03) 14 (53.84) 35 (66.05) 0.638
Unemployed 22 (12.66) 9 (34.61) 13 (24.53)
Student 2 (2.53) 1 (3.85) 1 (1.88)
Retired 6 (7.61) 2 (7.69) 4 (7.55)
Lives n (%)
Living alone 13 (16.46) 5 (19.23) 8 (15.1) 0.683
Living accompanied 66 (83.56) 21 (80.77) 45 (84.9)
HRSD mean (SD) 23.43 (4.46) 22.42 (3.16) 23.92 (4.93) 0.162
MMSE mean (SD) 1.21 (0.5) 1.31 (0.68) 1.14 (0.35) 0.478a
CGI-Severity mean (SD) 3.92 (0.86) 3.96 (0.94) 3.89 (0.81) 0.545a
Age at onset mean (SD) 37.06 (11.96) 42.54 (9.32) 34.01 (12.26) 0.004
Number previous episodes – – 3.95 (3.27) –
Pharmacological main treatment n (%) 0.123
No treatment 6 (7.60) 4 (15.38) 2 (3.77)
Antidepressant 71 (89.89) 22 (84.61) 49 (92.46)
SSRIb 32 12 20
SNRIc 1 0 1
TCAd 1 0 1
Other 37 10 27
Benzodiazepines 2 (2.53) 0 (0) 2 (3.77)
a
Mann–Whitney value.
b
Selective Serotonin Reuptake Inhibitors.
c
Serotonin Norepinephrine Reuptake Inhibitors.
d
Tricyclic antidepressants.

The Clinical Global Impression rating scale (CGI) (Guy et al.,
1976). Is a commonly used measure of symptom severity in
treatment studies of patients with mental disorders. For the aim
of the study we focused on CGI Severity which requires the clinician
to rate the severity of the patient's illness through a 7-point scale.
2.2.2. Cognitive assessment
Trail Making Test-Parts A and B (TMT) (Army Individual Test Battery,
1944). It measures visual attention and task-switching. Participant
has to connect consecutively numbered circles (Trails A) and then
has to connect numbers and letters in alternating sequence (Trails
B) switching from a category to other.
Digit Span subtest of the Wechsler Adult Intelligence Scale, 3rd
edition (WAIS-III) (Wechsler, 1997). Digit span forward is used to measure
attention and working memory span. Digit span backwards also is used to
measure executive function (Lezak, 1995), specifically requires monitor-
ing and replacing old non-relevant information by new-relevant ones.
Stroop Colour Word Test (SCWT) (Golden, 1978). This is a
measure of selective attention, freedom from distractibility and
response inhibition. First, the subject has to read out three colour
names printed in black as fast as possible. Then, is presented with
the same words printed in a colour different from the colour which
it names and the subject has to name the colour of the ink in which
word is printed as fast as possible. The time used to complete the
task increases significantly in the third trial, it is called “inter-
ference effect”. The interference score is calculating by subtract-
ing Trial 1 (naming words) from Trial 3 (naming the colour in which
the word is printed). A higher score indicates greater interference.
Tower of London, 2nd Edition (TOL-DX) (Culberston and Zillmer,
2006). This is an instrument of planning that taxes central executive
function. The subjects have to rearrange a set of spheres to match a
given target arrangement in a specified number of moves. Accuracy
and latency are recorded (Gallagher et al., 2007).
The Controlled Verbal Fluency Task (FAS) (Borkowski et al.,
1967). This task involves development of a strategy to produce

Please cite this article as: Roca, M., et al., Cognitive impairment in remitted and non-remitted depressive patients: A follow-up
comparison between first and.... European Neuropsychopharmacology (2015), http://dx.doi.org/10.1016/j.euroneuro.2015.07.020
4 M. Roca et al.

Table 2 Cognitive assessment at baseline.

SCALE First episode Recurrent ta P Remitted at Non-Remitted at ta P

depression value follow-up follow-up value

TMT-Ab 1.01 (1.26) 2.09 (2.66) 0.581 0.563 0.99 (1.93) 1.56 (1.72) 1.09 0.281
TMT-Bc 3.6 (3.07) 2.57 (3.29) 0.456 0.650 1.73 (2.16) 2.96 (3.05) 1.44d 0.149
WAIS – Digit Span 0.56 (1.17) 0.62 (1.33) 0.562 0.576 1.09 (1.31) 0.49 (1.23) 1.63 0.102
Forwardse
WAIS – Digit Span 0.25 (1.1) 0.07 (1.24) 0.273 0.785 0.04 (1.55) 0.07 (1.02) 0.32d 0.753
Backwardsf
Stroop – Wg 1.43 (0.66) 1.29 (1.07) 1.782 0.079 1.19 (0.95) 1.23 (0.93) 0.13 0.894
Stroop – Ch 1.38 (0.63) 1.22 (1.14) 0.878 0.383 1.1 (0.91) 1.22 (0.78) 0.49 0.621
Stroop – CWi 1.19 (1.13) 1.19 (1.03) 0.587 0.559 0.99 (0.98) 1.18 (1.14) 0.59 0.552
Stroop – Interferencej 0.17 (1.55) 0.16 (1.43) 0.081d 0.935 0.04 (1.19) 0.18 (1.7) 0.07 0.942
DX ToL – Total Movesk 0.12 (0.89) 0.27 (1.03) 0.272 0.787 0.14 (0.88) 0.29 (0.79) 1.78 0.081
DX ToL – Total Timel 0.64 (1.19) 0.66 (1.29) 0.613 0.541 0.37 (1.06) 0.55 (1.1) 0.58 0.564
DX ToL – Initiation 0.25 (0.77) 0.31 (0.75) 0.197 0.845 0.33 (0.91) 0.26 (0.71) 0.32 0.747
Timem
DX ToL – Execution 0.57 (1.29) 0.63 (1.34) 0.306 0.760 0.31 (1.13) 0.46 (1.16) 0.45 0.652
Timen
FASo 1.52 (1.22) 1.31 (0.88) 0.238 0.812 1.24 (0.95) 1.4 (1.14) 0.51 0.614
SVFp 1.03 (1.22) 0.94 (1.18) 1.333 0.187 0.72 (1.37) 0.94 (1.15) 0.62 0.541
a
Statistics have been computed using raw scores.
b
Trail Making Test-Part A (TMT-A).
c
Trail Making Test-Part B (TMT-B).
d
Z Mann–Whitney value.
e
Digit Span subtest of the Weschler Adult Intelligence Scale (WAIS) – Part I (WAIS-I).
f
Digit Span subtest of the Weschler Adult Intelligence Scale (WAIS) – Part II (WAIS-II).
g
Stroop Colour Word Test – Part I (SCWT I).
h
Stroop Colour Word Test – Part II (SCWT II).
i
Stroop Colour Word Test – Part III (SCWT III).
j
Stroop Colour Word Test – Interference.
k
Tower of London Total Move (TOL DX Total Move).
l
Tower of London Total Problem-Solving (TOL DX Total Problem-Solving).
m
Tower of London Total Initiation Time (TOL DX Total Initiation Time).
n
Tower of London (TOL DX Total Execution Time).
o
Controlled Oral Word Association Test (FAS).
p
Semantic Verbal Fluency (Animals).

the words, working memory and language skills. In this instrument
the subject has to generate words starting with letters F, A and S as
fast as possible during 1 min.
Semantic Verbal Fluency (SVF). Participant has to generate words
corresponding to a specific semantic category (animals) (Ardila et al.,
2006). This task mainly measures the executive function of access to long
term memory (Fisk and Sharp, 2004) as well as some aspects of language.
(time  group  remission), repeated-measures multivariate analy-
sis of variance (MANOVA) was conducted. If parametric assumptions
were not met a bootstrapped MANOVA was performed at 2000
iterations. Eta squared was used as effect size estimator. Interac-
tion analysis was calculated using Bonferroni correction to control
Type I error rate across comparisons. Data were analyzed with IBM
SPSS 21 for Windows. Statistical significance was set at 0.05.

2.3. Statistical analysis

Before proceeding to conduct any analysis over the outcomes
variables, raw scores were converted to standard scores using mean
and standard deviation of normative samples according to age and
gender. Parametric requirements – homoscedasticity and normality
were tested, and ‘number of previous episodes’ was dichotomized.
Participants were assigned to FE group if they had no previous
history of MDD, or to RE group if they reported previous history of
MDD. Both groups of participants were compared on baseline
variables using Chi square and independent sample t tests, or
Mann–Whitney U-test if parametric assumptions were not adequate.
To identify possible differences between the course of cognitive
functions in FE and RE patients as a function of remission
3. Results
The sample was composed of seventy nine depressive patients,
26 met first episode and 53 recurrent depression criteria. The
sample was mainly composed of middle-aged (X =46.11,
SD=8.28) women (78.48%), married (62.03%) and living accom-
panied (83.56%), employed (62.03%), with at least primary
education studies (58.24%). At baseline mean HRSD total score
for the whole sample was 23.43 (SD=4.46), while MMSE mean
score was 1.21(SD=0.5), and CGI Severity mean score was 3.92
(SD=0.86). Most of the participants were treated with anti-
depressant (89.89%). The comparison between FE and RE

Please cite this article as: Roca, M., et al., Cognitive impairment in remitted and non-remitted depressive patients: A follow-up
comparison between first and.... European Neuropsychopharmacology (2015), http://dx.doi.org/10.1016/j.euroneuro.2015.07.020
Cognitive impairment in remitted and non-remitted depressive patients 5

Table 3 Cognitive assessment at 6 month follow-up. Comparison between remitted and non-remitted patients.

SCALE Remitted (n =27) Non-Remitted (n =40) F P value Eta squared

a
TMT-A 0.41 (1.4) 2.23 (3.31) 7.13 0.010 0.121
TMT-Bb 1.41 (1.52) 4.03 (4) 13.08 0.001 0.211
WAIS – Digit Span Forwardsc 1.55 (1.52) 0.84 (1.16) 5.01 0.029 0.085
WAIS – Digit Span Backwardsd 0.41 (1.46) 0.11 (1.16) 1.29 0.262 0.024
Stroop – We 0.87 (0.95) 1.53 (1.25) 5.01 0.029 0.088
Stroop – Cf 0.86 (0.87) 1.33 (1.28) 8.48 0.005 0.14
Stroop – CWg 0.94 (0.97) 1.35 (0.98) 0.02 0.887 o0.001
Stroop – Interferenceh 0.13 (0.87) 0.28 (0.88) 1.41 0.241 0.027
DX ToL – Total Movesi 0.25 (0.68) 0.41 (1.39) 8.74 0.005 0.130
DX ToL – Total Timej 0.13 (0.59) 0.65 (1.51) 8.37 0.005 0.134
DX ToL – Initiation Timek 0.22 (1.03) 0.78 (1.9) 2.18 0.145 0.039
DX ToL – Execution Timel 0.26 (0.59) 0.46 (1.52) 6.93 0.011 0.114
FASm 1.11 (0.98) 1.46 (1.05) 2.82 0.099 0.050
SVFn 0.67 (0.87) 1.37 (0.84) 6.99 0.011 0.115
a
Trail Making Test-Part A (TMT-A).
b
Trail Making Test-Part B (TMT-B).
c
Digit Span subtest of the Weschler Adult Intelligence Scale (WAIS) – Part I (WAIS-I).
d
Digit Span subtest of the Weschler Adult Intelligence Scale (WAIS) – Part II (WAIS-II).
e
Stroop Colour Word Test – Part I (SCWT I).
f
Stroop Colour Word Test – Part II (SCWT II).
g
Stroop Colour Word Test – Part III (SCWT III).
h
Stroop Colour Word Test – Interference.
i
Tower of London Total Move (TOL DX Total Move).
j
Tower of London Total Problem-Solving (TOL DX Total Problem-Solving).
k
Tower of London Total Initiation Time (TOL DX Total Initiation Time).
l
Tower of London (TOL DX Total Execution Time).
m
Controlled Oral Word Association Test (FAS).
n
Semantic Verbal Fluency (Animals).

patients showed no statistically significant differences for
demographic or clinical characteristics (Table 1).
At baseline no statistical differences were found between
FE patients and RE patients. Homoscedasticity assumption
was not met in Stroop Interference (Levene test – F=11.829,
p=0.001) so Mann–Whitney test was used to compare FE and
RE patients in this outcome. There were no significant
differences between remitted and no remitted patients in
any cognitive domain assessed at baseline. Homoscedasticity
assumption was not met in TMT Part B (Levene test –
F=10.01, p=0.003) WAIS – Digit Span Backwards (Levene
test – F=6.55, p=0.013), so groups were compared using
Mann–Whitney test. Detailed information about cognitive
function at baseline is shown in Table 2.
At six-months 12 patients were lost to follow-up. Further
analyses were conducted in 67 patients, 27 were in clinical
remission. When compared, remitted patients scored signifi-
cant better than non-remitted patients in TMT Part A
(F=7.136, p=0.010, η2 =0.121), TMT Part B (F=13.08,
p=0.001, η2 =0.211), WAIS Digit Span Forwards (F=5.01,
p=0.029, η2 =0.085), Stroop – Words (F=5.01, p=0.029,
η2 =0.088), Stroop – Colours (F=8.48, p=0.05, η2 =0.140), DX
ToL – Total Moves (F=8.74, p=0.005, η2 =0.130), DX ToL – Total
Time (F=8.37, p=0.0.005, η2 =0.07134), DX ToL – Execution
Time (F=6.93, p=0.011, η2 =0.114), and Semantic Verbal
Fluency (F=6.99, p=0.011, η2 =0.115). Detailed data about
cognitive performance of both groups are shown in Table 3.
When FE and RE patients were compared within-remission
factor, significant differences were found on the SVF scale, RE
remitted patients score better than FE remitted patients
(F=6.65, p=0.013, η2 =0.110), this means there is a 69%
chance that a patient selected at random from the RE remitted
group will have a higher score than a patient picked at random
from the FE remitted group. In non-remitted condition there
were no statically significant differences, nevertheless large
effect size was registered at TMT Part A (F=2.09, p=0.164,
η2 =0.083). Detailed information about comparisons between
groups within remission factor is shown in Table 4.
4. Discussion
The main result of our study is the positive impact of
remission on cognitive impairment in depressive patients
regardless of the number of previous episodes after six
months of treatment. Remitted patients scored significantly
better than non-remitted in nine of the cognitive tasks used
in our study. When remitted FE depressive patients were
compared with remitted RE depressive patients no statisti-
cally differences were found except for SVF scale.
The impact of clinical remission was remarkable on TMT Part
B scores which accounted for 11% of variance. TMT Part A, WAIS
Digit Span Forwards, Stroop Test Words and Colours, ToL Total
Moves, ToL Total Time, ToL Total Execution Time and SVF

Please cite this article as: Roca, M., et al., Cognitive impairment in remitted and non-remitted depressive patients: A follow-up
comparison between first and.... European Neuropsychopharmacology (2015), http://dx.doi.org/10.1016/j.euroneuro.2015.07.020
6 M. Roca et al.

Table 4 Interaction between type of depression and remission at 6 month follow-up.

SCALE Remitted Non-Remitted

FEP (n =7) RD (n =20) F P η 2

FEP (n =14) RD (n =26) F P η2
value value

TMT-Aa 0.31 (0.68) 0.29 (1.09) 0.315 0.577 0.007 1.24 (2.11) 3.35 (4.22) 2.09 0.164 0.083
TMT-Bb 1.45 (1.77) 1.04 (1.69) 0.01 0.982 o0.001 3.87 (2.6) 4.66 (5) 0.01 0.934 o0.001
WAIS – Digit Span 1.39 (1.75) 1.72 (1.6) 0.04 0.845 0.001 0.88 (1.58) 0.45 (1.34) 1.07 0.305 0.019
Forwards c
WAIS – Digit Span 0.06 (1.87) 0.61 (1.35) 0.73 0.397 0.014 0.22 (1.26) 0.08 (0.95) 0.57 0.458 0.010
Backwardsd
Stroop – We -0.98 (0.81) 0.64 (0.95) 0.63 0.431 0.012 1.61 (0.92) 1.64 (1.45) 0.58 0.449 0.011
Stroop – Cf 0.71 (0.49) 0.69 (0.94) 0.03 0.854 o0.001 1.42 (0.63) 1.34 (1.61) 0.06 0.798 0.001
Stroop – CWg 0.77 (0.87) 0.72 (0.92) 0.02 0.887 o0.001 1.29 (0.95) 1.47 (1.03) 0.02 0.899 o0.001
Stroop – 0.35 (1.11) 0.22 (0.54) 0.88 0.352 0.017 0.39 (0.75) 0.25 (0.97) 0.09 0.755 0.002
Interferenceh
DX ToL – Total 0.53 (0.48) 0.24 (0.64) 0.42 0.518 0.008 0.42 (1.03) 0.69 (1.69) 0.54 0.467 0.010
Movesi
DX ToL – Total 0.08 (0.77) 0.17 (0.56) 0.01 0.938 o0.001 0.69 (1.01) 1.04 (1.88) 0.81 0.373 0.015
Timej
DX ToL – Initiation 0.37 (1.14) 0.29 (1.08) 0.02 0.889 o0.001 1.08 (2.67) 1.06 (1.76) o0.01 0.980 o0.001
Timek
DX ToL – Execution 0.29 (0.58) 0.33 (0.59) o0.01 0.962 o0.001 0.43 (1.02) 0.8 (1.93) 0.74 0.394 0.014
Timel
FASm 0.98 (1.07) 1.06 (1.03) 0.04 0.848 0.001 1.58 (1.55) 1.44 (0.77) o0.01 0.998 o0.001
SVFn 1.32 (0.89) 0.37 (0.81) 6.65 0.013 0.110 1.43 (0.93) 1.48 (0.77) 0.05 0.823 0.001
a
Trail Making Test-Part A (TMT-A).
b
Trail Making Test-Part B (TMT-B).
c
Digit Span subtest of the Weschler Adult Intelligence Scale (WAIS) – Part I (WAIS-I).
d
Digit Span subtest of the Weschler Adult Intelligence Scale (WAIS) – Part II (WAIS-II).
e
Stroop Colour Word Test – Part I (SCWT I).
f
Stroop Colour Word Test – Part II (SCWT II).
g
Stroop Colour Word Test – Part III (SCWT III).
h
Stroop Colour Word Test – Interference.
i
Tower of London Total Move (TOL DX Total Move).
j
Tower of London Total Problem-Solving (TOL DX Total Problem-Solving).
k
Tower of London Total Initiation Time (TOL DX Total Initiation Time).
l
Tower of London (TOL DX Total Execution Time).
m
Controlled Oral Word Association Test (FAS).
n
Semantic Verbal Fluency (Animals).

remission effect was large sized. The data suggest specific
impairment in shifting attention and access to long-term
memory. Compared with remitted patients poor execution of
non-remitted patients in Tower of London task might be
explained by a combination of slow psychomotor processing,
poor attention shifting, and rigid access to long-term memory.
Our findings are also consistent with previous data (Biringer
et al., 2007) reporting that clinical remission is associated with
improvement in verbal memory. Improvement in executive
functions of shifting and access to information might be
underlying remitted patient's performance in verbal memory
task and working memory. Our results did not show an
improvement in cognitive inhibition in remitted patients,
contrary to the findings of (Schmid et al., 2013). However, we
observed improved semantic fluency, problem solving skills, and
attentional shifting. Our results may differ from Schmid (2013)
due sample characteristics: sample in our study is older and at
baseline there are no differences between remitted and
non-remitted patients. In contrast with Reppermund et al.
(2009), remission is associated with cognitive improvement in
our study. The difference in cognitive outcome may be due to
the remitted patients' rate at follow-up. In our study remission
was found in 40% of sample, whereas Reppermund report a
level of 81%. This difference is likely explained by a combina-
tion of statistical comparison biases and the use of different
clinical remission criteria. With respect to the later issue, we
define it as a score lower than 7, which is in line with ACNP Task
Force recommendations (Rush et al., 2006), while Reppermund
et al. (2009) set a score lower than 9 as remission criteria.
It is important to point out that in our study FE and RE
patients were not different at baseline. This finding is
convergent with prior studies about the influence of pre-
vious episodes in cognitive performance (Kaygusuz et al.,
2013; Preiss et al., 2009) but does not concur with other
data raising the possibility of neurotoxic role of previous
depressive episodes in psychomotor speed assessed through

Please cite this article as: Roca, M., et al., Cognitive impairment in remitted and non-remitted depressive patients: A follow-up
comparison between first and.... European Neuropsychopharmacology (2015), http://dx.doi.org/10.1016/j.euroneuro.2015.07.020
Cognitive impairment in remitted and non-remitted depressive patients
TMT-A and TMT-B scores (Gorwood et al., 2014). Talarowska
et al. (2015) also reported significant differences between
FE and RE patients in their second episode in verbal memory
learning, but did not find a relationship between cognitive
impairment and the number of previous episodes in a
sample of medication-free hospitalized patients. Our results
could differ from Gorwood et al. (2014) due to time
between assessments, i.e. 6–8 weeks versus 6 months.
The present study also assesses the effect of previous episodes
on the impact of clinical remission in cognitive performance. Our
data suggests that FE remitted patients perform worse than RE
remitted patients in semantic fluency task. Similar results have
also been found by Lee et al. (2012) who reported that studies
including samples composed only of FE patients showed poorer
neuropsychological functioning than studies with samples com-
posed of FE and RE patients. As stated by Schmid et al. (2011)
this impairment combined with poor inhibition may lead depres-
sive patients failing to inhibit negative thoughts and reduced
attention to positive thoughts. This phenomenon might explain
the increased vulnerability of recurrent depressive episodes.
Moreover the blunted effect of remission in FE patient's semantic
fluency in contrast with RE patients, might support the hypoth-
esis of greater relevance of cognitive bias in onset of FE patients
than RE patients (Vrijsen et al., 2014).
The main strength of our study is its follow-up design.
This allows us to assess the impact of remission on cognitive
performance within two types of depressive patients over
time. There are some limitations of the study. First, the
small sample size, especially in the remitted condition, may
affect the power of statistical tests and increase the like-
lihood of a type II error. A second limitation is the inclusion
of depressive patients with melancholic features who could
be more severe and less willing to accept treatment (Roca
et al., 2014). Finally, not all patients were under pharma-
cological treatment, although at baseline there were no
differences between FE and RE patients in HDRS17 scores.
In summary, the main finding of our study is that remission
and not number of previous episodes has a high impact on
cognitive performance of MDD patients. Future research should
address if improvement in cognitive function as a consequence
of clinical remission is related with functional recovery in daily
life (Romera et al., 2013). To facilitate comparison between
results across the same cognitive tests researchers should use a
common model of executive functions. This would help to a
better understanding of executive functioning in MDD, which
may extend to other cognitive domains as well. Executive
function is currently employed as an umbrella term for a variety
of cognitive skills related to planning, organizing, dealing with
novelty, flexible problem solving strategies, inhibition and
elements of working memory. Differentiation of these various
functions would likely facilitate a better understanding of the
cognitive difficulties experienced by patients with various
psychiatric disorders, including MDD. As a first step to find
new treatments targeting the cognitive impairment in MDD
(Solé et al., 2015), further studies should incorporate samples
composed of treatment-naïve groups into study design to
control the effects of drug interventions.
Role of funding source
None.
Please cite this article as: Roca, M., et al., Cognitive impairment in remitted and non-remitted depressive patients: A follow-up
comparison between first and.... European Neuropsychopharmacology (2015), http://dx.doi.org/10.1016/j.euroneuro.2015.07.020
7
Contributors
MR, SM and MG designed the study and wrote the protocol. MR, ELN,
MV managed the literature searches. SM and MV assess the patients
included in the study. MR and ELN undertook the statistical analysis.
MR, ELN, MGT, JGC and JH wrote the first draft of the manuscript.
All authors contributed to and have approved the final manuscript.
Conflict of interest
Dr. Roca, Ms. Monzón, Ms. Vives, Dr. García-Toro and Dr. Gili report
grants from the Instituto de Salud Carlos III (Institute of Health
Carlos III) of the Ministry of Economy and Competitiveness (Spain)
and the European Union ERDF funds, during the conduct of the
study; Dr. Roca reports personal fees from Eli Lilly, Servier and
Pfizer; and grants and personal fees from Lundbeck and Jannsen,
outside the submitted work. Dr. García-Campayo reports personal
fees from Pfizer, Eli Lilly, Rovi and GSK, outside the submitted work.
Dr. Harrison reports personal fees from Abbvie, Anavex, AstraZe-
neca, Avraham, Axon, Boehringer Ingelheim, Catenion, CRF Health,
DeNDRoN, EnVivo Pharma, ePharmaSolutions, Eisai, Eli Lilly, Hep-
tares, Janssen AI, Kyowa Hakko Kirin, Lundbeck, MedAvante, Merck,
Mind Agilis, MyCognition, Neurocog, Neuroscios, Novartis, Nutricia,
Orion Pharma, Pharmanet/i3, Pfizer, Prana Biotech, PriceSpective,
Prophase, Prostrakan, Roche, Reviva, Servier, Shire, TCG, TransTech
Pharma, Velacor, outside the submitted work. Mr. López-Navarro
reports no conflict of interest.
Acknowledgements
This project has been funded by a Grant from the Instituto de Salud
Carlos III (Institute of Health Carlos III) (FIS no. PI08 1270) of the
Ministry of Economy and Competitiveness (Spain), and co-financed
with European Union ERDF Funds.
References
Aboulafia-Brakha, T., Christe, B., Martory, M.-D., Annoni, J.-M.,
2011. Theory of mind tasks and executive functions: a systema-
tic review of group studies in neurology. J. Neuropsychol. 5,
39–55. http://dx.doi.org/10.1348/174866410X533660.
Ardila, A., Ostrosky-Solís, F., Bernal, B., 2006. Cognitive testing
toward the future: The example of Semantic Verbal Fluency
(ANIMALS). Int. J. Psychol. 41, 324–332. http://dx.doi.org/
10.1080/00207590500345542.
Army Individual Test Battery, 1944. Manual of Directions and
Scoring. War Department, Adjutant General's Office, Washing-
ton, DC.
Bhardwaj, A., Wilkinson, P., Srivastava, C., Sharma, M., 2010.
Cognitive deficits in euthymic patients with recurrent depres-
sion. J. Nerv. Ment. Dis. 198, 513–515. http://dx.doi.org/
10.1097/NMD.0b013e3181e4c5ba.
Biringer, E., Mykletun, A., Sundet, K., Kroken, R., Stordal, K.I.,
Lund, A., 2007. A longitudinal analysis of neurocognitive func-
tion in unipolar depression. J. Clin. Exp. Neuropsychol. 29,
879–891. http://dx.doi.org/10.1080/13803390601147686.
Bora, E., Harrison, B.J., Yücel, M., Pantelis, C., 2013. Cognitive
impairment in euthymic major depressive disorder: a meta-
analysis. Psychol. Med. 43, 2017–2026. http://dx.doi.org/
10.1017/S0033291712002085.
Borkowski, J.G., Benton, A.L., Spreen, O., 1967. Word fluency and
brain damage. Neuropsychologia 5, 135–140.
Culberston, W.C., Zillmer, E.A., 2006. Multi-Health Systems Inx.
Tower of London-Drexel University (TOLDX), Canada.
8 M. Roca et al.
Fisk, J.E., Sharp, C.A., 2004. Age-related impairment in executive
functioning: updating, inhibition, shifting, and access. J. Clin.
Exp. Neuropsychol. 26, 874–890.
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. J. Psychiatr. Res. 12, 189–198.
Gallagher, P., Robinson, L.J., Gray, J.M., Porter, R.J., Young, A.H.,
2007. Neurocognitive function following remission in major
depressive disorder: potential objective marker of response?
Aust. N. Z. J. Psychiatry 41, 54–61. http://dx.doi.org/10.1080/
00048670601057734.
Golden, M.D., 1978. Stroop Color and Word Test. Stoetling Com-
pany, Chicago, IL.
Gorwood, P., Corruble, E., 2008. Toxic effects of depression on
brain function: impairment of delayed recall and the cumulative
length of depressive disorder in a large sample of depressed.
Am. J. Psychiatry, 731–739.
Gorwood, P., Richard-Devantoy, S., Baylé, F., Cléry-Melun, M.L., 2014.
Psychomotor retardation is a scar of past depressive episodes,
revealed by simple cognitive tests. Eur. Neuropsychopharmacol. 24,
1630–1640. http://dx.doi.org/10.1016/j.euroneuro.2014.07.013.
Guy, W., (U.S.)., N.I. of M.H., Branch., P.R., Program., E.C.D.E.,
1976. ECDEU Assessment Manual for Psychopharmacology. U.S.
Dept. of Health, Education, and Welfare, Public Health Service,
Alcohol, Drug Abuse, and Mental Health Administration, National
Institute of Mental Health, Psychopharmacology Research Branch,
Division of Extramural Research Programs, Rockville, MD.
Hamilton, M., 1960. A rating scale for depression. J. Neurol.
Neurosurg. Psychiatry 23, 56–62.
Hasselbalch, B.J., Knorr, U., Kessing, L.V., 2011. Cognitive impair-
ment in the remitted state of unipolar depressive disorder: a
systematic review. J. Affect. Disord. 134, 20–31. http://dx.doi.
org/10.1016/j.jad.2010.11.011.
Herrmann, L.L., Goodwin, G.M., Ebmeier, K.P., 2007. The cognitive
neuropsychology of depression in the elderly. Psychol. Med. 37,
1693–1702. http://dx.doi.org/10.1017/S0033291707001134.
Kaygusuz, C.C., Arisoy, O., Boztas, M.H., Sercan, M., 2013.
Comparison of first episode and recurrent major depression
patients in terms of cognitive function. Dusunen Adam: J.
Psychiatry Neurol. Sci. 26, 320–332. http://dx.doi.org/
10.5350/DAJPN2013260401.
Ladegaard, N., Lysaker, P.H., Larsen, E.R., Videbech, P., 2014. A
comparison of capacities for social cognition and metacognition
in first episode and prolonged depression. Psychiatry Res. 220,
883–889.
Lee, R.S.C., Hermens, D.F., Porter, M.A., Redoblado-Hodge, M.A.,
2012. A meta-analysis of cognitive deficits in first-episode Major
Depressive Disorder. J. Affect. Disord. 140, 113–124.
Lezak, M.D., 1995. Neuropsychological Assessment. Oxford Univer-
sity Press, USA.
McDermott, L.M., Ebmeier, K.P., 2009. A meta-analysis of depres-
sion severity and cognitive function. J. Affect. Disord. 119, 1–8.
http://dx.doi.org/10.1016/j.jad.2009.04.022.
McIntyre, R.S., Cha, D.S., Soczynska, J.K., Woldeyohannes, H.O.,
Gallaugher, L.A., Kudlow, P., Alsuwaidan, M., Baskaran, A.,
2013. Cognitive deficits and functional outcomes in major
depressive disorder: determinants, substrates, and treatment
interventions. Depress. Anxiety 30, 515–527. http://dx.doi.org/
10.1002/da.22063.
Please cite this article as: Roca, M., et al., Cognitive impairment in remitted and non-remitted depressive patients: A follow-up
comparison between first and.... European Neuropsychopharmacology (2015), http://dx.doi.org/10.1016/j.euroneuro.2015.07.020
Preiss, M., Kucerova, H., Lukavsky, J., Stepankova, H., Sos, P.,
Kawaciukova, R., 2009. Cognitive deficits in the euthymic phase
of unipolar depression. Psychiatry Res. 169, 235–239. http://dx.
doi.org/10.1016/j.psychres.2008.06.042.
Reppermund, S., Ising, M., Lucae, S., Zihl, J., 2009. Cognitive
impairment in unipolar depression is persistent and non-specific:
further evidence for the final common pathway disorder hypoth-
esis. Psychol. Med. 39, 603–614.
Roca, M., Monzón, S., Vives, M., López-Navarro, E., Garcia-Toro,
M., Vicens, C., Garcia-Campayo, J., Harrison, J., Gili, M., 2014.
Cognitive function after clinical remission in patients with
melancholic and non-melancholic depression: a 6 month
follow-up study. J. Affect. Disord. 171C, 85–92. http://dx.doi.
org/10.1016/j.jad.2014.09.018.
Roca, M., Vives, M., Lopez-Navarro, E., Garcia-Campayo, J., Gili,
M., 2015. Cognitive impairments and depression: a critical
review. Actas Esp. Psiquiatr. 43, 187–193.
Rock, P.L., Roiser, J.P., Riedel, W.J., Blackwell, A.D., 2013.
Cognitive impairment in depression: a systematic review and
meta-analysis. Psychol. Med. 44, 1–12. http://dx.doi.org/
10.1017/S0033291713002535.
Romera, I., Pérez, V., Ciudad, A., Caballero, L., Roca, M., Polavieja,
P., Gilaberte, I., 2013. Residual symptoms and functioning in
depression, does the type of residual symptom matter? A post-
hoc analysis. BMC Psychiatry 13, 51. http://dx.doi.org/10.1186/
1471-244X-13-51.
Rush, A.J., Kraemer, H.C., Sackeim, H.A., Fava, M., Trivedi, M.H.,
Frank, E., Ninan, P.T., Thase, M.E., Gelenberg, A.J., Kupfer, D.
J., Regier, D.A., Rosenbaum, J.F., Ray, O., Schatzberg, A.F.,
2006. Report by the ACNP Task Force on response and remission
in major depressive disorder. Neuropsychopharmacology 31,
1841–1853.
Schmid, M., Hammar, A., Hammar, Å., 2013. Cognitive function in
first episode major depressive disorder: poor inhibition and
semantic fluency performance. Cogn. Neuropsychiatry 18,
515–530. http://dx.doi.org/10.1080/13546805.2012.754748.
Schmid, M., Strand, M., Ardal, G., Lund, A., Hammar, A., 2011.
Prolonged impairment in inhibition and semantic fluency in a
follow-up study of recurrent major depression. Arch. Clin.
Neuropsychol. 26, 677–686. http://dx.doi.org/10.1093/arclin/
acr048.
Solé, B., Jiménez, E., Martinez-Aran, A., Vieta, E., 2015. Cognition
as a target in major depression: new developments. Eur.
Neuropsychopharmacol. 25, 231–247. http://dx.doi.org/
10.1016/j.euroneuro.2014.12.004.
Talarowska, M., Zajàczkowska, M., Gałecki, P., 2015. Cognitive
functions in first-episode depression and recurrent depressive
disorder. Psychiatr. Danub. 27, 38–43.
Vrijsen, J.N., Becker, E.S., Arias-Vásquez, A., van Dijk, M.K.,
Speckens, A., Oostrom, I. Van, 2014. What is the contribution
of different cognitive biases and stressful childhood events to
the presence and number of previous depressive episodes?
Psychiatry Res. 217, 134–142, http://dx.doi.org/10.1016/j.
psychres.2014.02.033.
Wechsler, D., 1997. Wechsler Adult Intelligence Scale (WAIS-III).
Pearson, San Antonio.
Weightman, M.J., Air, T.M., Baune, B.T., 2014. A review of the role
of social cognition in major depressive disorder. Front. Psychia-
try 5, http://dx.doi.org/10.3389/fpsyt.2014.00179.