Original Paper

Cerebrovasc Dis 2009;28:157165

DOI: 10.1159/000226114

Received: October 31, 2008

Accepted: March 3, 2009
Published online: June 25, 2009

Poststroke Depression: Prevalence and

Determinants in Brazilian Stroke Patients
Francisco Javier Carod-Artal Luciane Ferreira Coral Daniele Stieven Trizotto
Clarissa Menezes Moreira
Department of Neurology, Sarah Network of Rehabilitation Hospitals, Sarah Hospital, Brasilia, Brazil

Key Words
Disability HADS Mood disorders Poststroke
depression Stroke, prognosis

Abstract
Background: Poststroke depression (PSD) is one of the most
important long-term adverse psychosocial consequences in
stroke survivors. Our objective was to assess the prevalence
of PSD in Brazilian stroke patients and identify significant associated factors. Methods: A cross-sectional study of stroke
patients consecutively admitted for rehabilitation was conducted. The patients were evaluated by means of the NIH
Stroke Scale, Mini-Mental State Examination, Barthel Index,
Lawton Scale, modified Rankin Scale, Hospital Anxiety and
Depression Scale (HADS), Geriatric Depression Scale (GDS)
and MOS-Short Form 36. Patients with a HADS-depression
subscale score 611 and/or GDS score 68 were classified as
depressed. Results: Three hundred stroke survivors were assessed (mean age: 56.3 years; 51.7% males). Half (46.7%) of
the stroke patients had an m-RS score ^2. The proportion
of stroke patients who scored 611 points on the HADS-depression and HADS-anxiety subscales were 19.2 and 23.7%,
respectively. One third (29.7%) had a GDS mean score 68.
The GDS scores significantly correlated (p ! 0.0001) with the
HADS-depression (r = 0.51) and HADS-anxiety subscales (r =
0.54). The prevalence of mood disorders was significantly
higher in females than in males (24.8 vs. 14.2%; 2, p = 0.03).

2009 S. Karger AG, Basel

10159770/09/02820157$26.00/0
Fax +41 61 306 12 34
E-Mail karger@karger.ch
www.karger.com

Accessible online at:

www.karger.com/ced

PSD was significantly associated (p ! 0.0001) with work status (housewife), education level, lower social and cognitive
functioning, dependence in the instrumental activities of
daily living and presence of diabetes in the multivariable regression analysis (R adjusted = 0.32). Conclusions: PSD was
highly prevalent in the chronic phase of stroke. Early detection and recognition of associated risk factors is important
to treat and prevent PSD in a rehabilitation setting.
Copyright 2009 S. Karger AG, Basel

Introduction

Poststroke depression (PSD) is one of the most important long-term adverse psychosocial consequences in
stroke survivors. Depression commonly occurs after a
stroke, with an estimated prevalence as high as 30% in the
first year after the event [1]. PSD is associated with poor
functional and psychosocial outcome in chronic stroke
patients [2, 3]. It affects cognitive function, functional recovery, health-related quality of life and health care use
in stroke survivors [35]. Emotional factors also affect
the stroke survivors considered fully recovered. Patients
who are perceived functionally independent still experience social participation difficulties and depressive
symptoms [6]. This empirical evidence has provided support for a biopsychosocial model of PSD in chronic stroke
patients.
Francisco Javier Carod-Artal, MD, PhD
Neurology Department, Sarah Hospital
SMHS quadra 501 conjunto A
Brasilia DF 70330-150 (Brazil)
Tel. +55 61 3 319 1555, Fax +55 61 3 319 1538, E-Mail fjcarod-artal@hotmail.com

The screening of patients with stroke for rehabilitation

needs has been proposed [7]. PSD is often underrecognized and undertreated, and it may be associated with
increased morbidity and mortality [3]. The diagnosis of
PSD may be hindered by problems with self-report in patients with cognitive and communicative deficits following stroke. So, the identification of PSD determinants
that can be targeted early in rehabilitation is needed.
Social and cultural influences may have a different impact on stroke survivors in developed and developing
countries. The global burden of stroke has been rising in
Latin America, with the increase in life expectancy and
the aging of the population, and specifically in Brazil,
where stroke is the number one cause of death [8]. Data
regarding the long-term consequences of stroke and its
psychosocial impact in stroke survivors are scarce in
Brazil.
The primary objective of this study was to assess the
prevalence and determinants of PSD in a cohort of Brazilian stroke patients consecutively admitted for rehabilitation.

Subjects and Methods

Patients
Patients with a diagnosis of stroke who were consecutively admitted to the outpatient Stroke Rehabilitation Clinics of the Sarah
Hospital in Brasilia DF, between July 2007 and June 2008, were
included in the study. The Sarah Network of Rehabilitation Hospitals is a Brazilian public institution dedicated to rehabilitation
of neurological disorders. All subjects had been previously hospitalized in a general hospital before admission to the rehabilitation
center.
Stroke was defined as a focal deficit of sudden onset that lasted
at least 24 h with no known alternative to a vascular cause [9] and
was confirmed by clinical examination and radiological findings.
Patients with either ischemic or hemorrhagic stroke as well as
subjects with first ever and recurrent stroke were included in the
study. The Trial of ORG 10172 in Acute Stroke Treatment criteria
were used to define ischemic stroke subtypes [10].
Exclusion criteria were the following: (1) patients !18 years of
age; (2) severe disability caused by a previous neurological disorder; (3) patients with lesions related to brain injury or tumor; (4)
concomitant severe systemic illness; (5) impaired ability to understand the functional and depression scales (e.g. due to severe
aphasia, dementia), and (6) refusal to participate in the study. The
institutional board reviewed the design and methods section of
the study. All patients included in the study gave their informed
consent.
Assessments
The study protocol included the National Institute of Health
Stroke Scale (NIHSS) [11], the modified version of the Rankin
scale (m-RS) [12], the Barthel Index (BI) [13], the Lawton and Bro-

158

Cerebrovasc Dis 2009;28:157165

dy Scale [14] and the Folstein Mini-Mental State Examination

(MMSE) [15]. The patients answered the Hospital Anxiety and
Depression Scale (HADS) [16], the Geriatric Depression Scale
(GDS) [17] and the Short-Form 36 (SF-36) [18].
Functional and neurological assessments were done by stroketrained neurologists. The patients were examined for depression
diagnosis in the chronic phase of stroke when they were admitted
to the rehabilitation hospital. The scales were administered by the
interviewers to avoid missing data and answered by the patients
during their visit to the clinic before starting rehabilitation. The
selected scales had been previously adapted and validated to the
Brazilian population [1922].
Neurological impairment was measured using the NIHSS.
The NIHSS is a 15-item stroke severity scale [11] that assesses
level of consciousness, extrinsic ocular movements, visual fields,
facial muscle function, arm and leg strength, coordination (limb
ataxia), sensory function, language (aphasia), speech (dysarthria),
extinction and inattention. The maximum possible score is 42,
and higher scores indicate greater impairment.
The m-RS was used to measure global functional independence [12]. The scale is defined categorically with 7 different
grades: 0 (no symptoms); 1 (no significant disability, despite
symptoms); 2 (unable to perform all previous activities but able to
look after own affairs without assistance); 3 (patient requires
some help but is able to walk without assistance); 4 (unable to walk
without assistance and unable to attend to own bodily needs without assistance); 5 (severe disability), and 6 (dead).
Activities of daily living (ADL) were evaluated by means of the
100-point BI [13]. The BI measures 10 basic aspects of ADL related to self-care and mobility: control of bowels and bladder,
grooming, toilet use, feeding, transfer, mobility, dressing, stairs
and bathing. The BI score ranges from 0 to 100 and lower scores
indicate greater dependency. Three categories of functional disability were defined using the following cutoff values: severe (BI
^60), moderate (BI 6590) and mild dependence/independence
in the ADL (BI 95 and 100).
Instrumental ADL were assessed with the Lawton and Brody
Scale [14]. The scale assesses the following functional abilities: using the telephone; getting to places beyond walking distance; grocery shopping; preparing meals; doing housework or handyman
work; doing laundry; taking medications, and managing money.
The score ranges from 8 (completely unable to handle instrumental activities) to 24 (without help).
Cognitive function was assessed by means of the MMSE [15].
The MMSE measures certain areas of cognitive functioning, including memory, orientation in place and time, naming, reading,
copying (visuospatial orientation), writing and the ability to follow a 3-stage command. The MMSE has 19 items and is scored
from 0 to 30 points. The Brazilian version of the MMSE was used
[19].
Depression was measured by means of the HADS and the
GDS. The HADS is a bidimensional scale developed specifically
to identify cases of depression and anxiety disorders among physically ill patients [16]. The HADS consists of 7 items for the assessment of anxiety and 7 for depression, with each item scored from
0 (no problem) to 3 (severe problem). The scores on individual
items can be summed to calculate a score for anxiety (HADSanxiety) and for depression (HADS-depression). Scores 611
points for a subscale are indicative of mood disturbance. The
HADS assessment of depression focuses on the core symptoms of

Carod-Artal/Ferreira Coral/Trizotto/
Menezes Moreira

Table 1. Demographic characteristics of the stroke population

(n = 300)
Age, years
Education, years
Mean time since stroke, months
Male gender
Marital status
Married
Single
Widower
Divorced
Occupation
Retired
Still working
Housewife
Unemployed
Previous stroke
Stroke subtype
Ischemic stroke
Hemorrhage
Vascular risk factors
Hypertension
Smoking
Hyperlipidemia
Diabetes
Atrial fibrillation

56.3814.3
8.285.5
22.5828.7
155 (51.7)
163 (54.3)
47 (15.7)
43 (14.3)
47 (15.7)
139 (46.3)
86 (28.7)
48 (16)
27 (9)
32 (10.7)
258 (86)
42 (14)
227 (75.6)
162 (54)
103 (34.3)
62 (20.7)
18 (6)

Spearmans correlation coefficient (r) was used to evaluate the

correlation between HADS and GDS mean scores and other
scales. Depression scores were also dichotomized into nondepressed (HADS !11; GDS !7) and depressed (HADS-depression 611; GDS score 68). The significance level was established
at 0.05.
A multivariable regression analysis was used to assess the
main predictors of PSD. PSD, as measured by the HADS-depression subscale and/or the GDS, was considered the dependent variable. The independent variables included in the regression model
were age, gender, years of education, time since stroke, dependence in the ADL (BI) and instrumental ADL (Lawton), comorbidities, social functioning and work status. The significant determinants were tested for collinearity to prevent overparametrization of the prediction model.
Statistical Package for the Social Science 13.0 for Windows
(SPSS, Chicago, Ill., USA) was used for the data analysis.

Results

Data Analysis
Mean scores of the HADS subscales, GDS and functional
stroke scales were calculated for the patients. The unpaired t test,
2 test and ANOVA were used for comparison between groups.

Three hundred and ninety stroke patients were consecutively admitted to the rehabilitation hospital during
the 1-year period of recruitment. On the basis of clinical
examination, 50 patients with severe aphasia and 32 with
vascular dementia were excluded; 5 patients refused to
participate in the study, and 3 were !18 years old. Three
hundred patients (mean age: 56.3 years; 51.7% males)
were included in the study. The demographic characteristics of the study population and vascular risk factors are
shown in table 1. One third (29.7%) of the patients were
665 years old. Females were slightly younger than males
(54.7 vs. 57.8 years; unpaired t test, p = 0.06), whereas the
mean time since stroke and level of education were similar. Most patients (87%) were living in their own home
and 13% with family members. A delay in stroke rehabilitation was observed and approximately two thirds
(57%) of the stroke patients started stroke rehabilitation
in the first 12 months following stroke.
Two hundred and fifty-eight subjects (86%) had an
ischemic stroke. The most common ischemic stroke subtypes were: cryptogenic stroke (25.6%), small-vessel infarction (25.2%), large artery occlusion (19.8%), cardioembolism (18.6%) and other causes (10.8%). Hypertension was significantly more common in males (81.3 vs.
69.7%; unpaired t test, p = 0.03).
Functional evaluation of stroke survivors is shown in
table 2. Forty-three percent of the stroke patients had an
NIHSS score ^5, 51% scored between 6 and 13, and 5.7%
had an NIHSS score 614. The median of BI score was 80
(interquartile range: 40), and the median of the m-RS was
3 (interquartile range: 2). Approximately half (46.7%) of

Poststroke Depression

Cerebrovasc Dis 2009;28:157165

Figures are means 8 standard deviation or numbers of cases

with percentages in parentheses.

mood and anhedonia [23]. The Brazilian version of the HADS was
used for the purpose of this study [20].
The GDS was developed as a screening test to detect depression in elderly individuals [17]. Subjects answer yes/no to questions about feelings, interests, activities and hopes. The focus of
the GDS items on affective rather than somatic components of
depression may increase its suitability for use among individuals
with stroke. The shortened 15-item Brazilian version was used.
The scores range from 0 to 15. A score of 66 indicates possible
depression and higher scores (cutoff value of 68) are associated
with clinically important depressive symptoms [21].
The SF-36 is a generic health-related quality of life measure
[18] that includes the following 8 domains: physical functioning,
role limitations due to physical problems, bodily pain, general
health perception, vitality, social functioning, role limitations
due to emotional problems and emotional well-being. The scores
for each domain range from 0 to 100, and higher values indicate
better function. Two summary scores of SF-36, the Physical Component Summary and the Mental Component Summary, can be
obtained.

159

Table 2. Functional evaluation of stroke patients broken down by sex

NIHSS2
MMSE1
BI1
Lawton instrumental ADL scale1
HADS-depression subscale2
HADS-anxiety subscale2
HADS-total score2
GDS2
SF361
Physical functioning
Physical role
Bodily pain
Social functioning
Mental health
Emotional role
Vitality
General health
Physical component summary
Mental component summary

Sample
(n = 300)

Male
(n = 155)

Female
(n = 145)

6.883.8
25.983.4
72.7826.5
16.984.7
7.184.1
7.684.2
14.687.3
5.983.4

6.683.9
26.483.6
74.4825.9
17.084.8
6.583.6
7.284.1
13.786.7
5.783.5

7.083.7
25.483.8
70.8827.1
16.684.6
7.784.4
7.984.4
15.687.7
6.283.3

NS
0.04
NS
NS
<0.01
NS
0.02
NS

24.4826.8
15.1830.2
59.8829.3
51.0827.9
59.2823.8
37.9844.1
60.6821.8
63.6820.7
33.388.8
50.1811.2

27.9828.5
15.7831.2
66.5827.8
53.5827.3
62.5823.3
39.7845.2
63.4823.4
66.5820.7
33.989.4
50.5811.7

24.4826.1
14.6829.1
54.3829.7
48.6828.4
55.8824.0
35.8842.7
57.7821.9
60.5820.3
33.288.2
49.6811.2

NS
NS
<0.001
NS
0.01
NS
0.03
0.01
NS
NS

Unpaired t test; means 8 standard deviation.

1
Higher scores indicate better function.
2
Higher scores indicate worse function.

the stroke patients (140) had an m-RS score ^2; 71 patients (23.7%) were independent in the ADL and achieved
a BI score of 100.
No significant differences in neurological impairment
(as measured by NIHSS), disability (as measured by BI
and/or Lawton scale) or functional status (as measured by
m-RS) were observed between males and females. The
Lawton instrumental ADL scale significantly correlated
(p ! 0.0001) with BI (r = 0.82), m-RS (r = 0.78), NIHSS
(r = 0.68) and MMSE (r = 0.35). In addition, no significant differences were found in the mean scores of the
scales that evaluated stroke severity (NIHSS), cognition
(MMSE) and depression (HADS, GDS) between young
(!65 years) and elderly stroke patients. Young stroke survivors had higher levels of anxiety (p = 0.05) as measured
by the HADS-anxiety subscale and were more independent in the instrumental ADL (p = 0.03). The correlation
between age and mean number of comorbidities was
moderate (r = 0.31).
The proportions of stroke patients who scored 611
points on the HADS-depression and HADS-anxiety subscales were 19.2 and 23.7%, respectively. A GDS score 66
160

Cerebrovasc Dis 2009;28:157165

was observed in 41.7% of the stroke patients; 89 patients

(29.7%) had a GDS mean score 68, indicating clinical
depression. The GDS mean scores significantly correlated (p ! 0.0001) with the HADS total score (r = 0.60),
HADS-depression (r = 0.51) and HADS-anxiety subscales (r = 0.54). Low correlations were observed between
age and cognition (as measured by the MMSE; r = 0.17);
and between age and depression (as measured by the
HADS-depression subscale, r = 0.07; or the GDS, r =
0.04). The prevalence of mood disorders was significantly higher in females than in males (24.8 vs. 14.2%; 2,
p = 0.03). The HADS-depression mean score was 7.1, and
females scored significantly worse than males (7.7 vs. 6.5;
unpaired t test, p ! 0.01). Sixty-two percent of the stroke
patients who scored 611 on the HADS-depression subscale were females.
Persons with diabetes scored significantly worse on
the HADS-depression subscale (mean score: 8.6 vs. 6.7;
unpaired t test, p = 0.001). Neither stroke laterality nor
anatomic stroke location was associated with PSD. No
significant differences in the HADS-depression and GDS
mean scores were observed by stroke subtype according
Carod-Artal/Ferreira Coral/Trizotto/
Menezes Moreira

Table 3. Depression and cognition scores of stroke survivors according to disability and functional status

Disability (BI)

HADS-depression
HADS-anxiety
HADS total score
GDS
MMSE

Functional status (m-RS)

BI 60
(n = 99)

BI 6590
(n = 103)

BI 95
(n = 98)

8.183.9
8.284.1
16.286.9
6.783.4
24.984.1

7.483.9
7.684.6
15.287.5
5.982.9
25.784.4

5.283.8
6.983.9
12.486.9
5.283.7
27.182.7

p
<0.0001
NS
0.0005
0.01
<0.0001

Rankin 1
(n = 59)

Rankin 2
(n = 81)

Rankin 3
(n = 69)

Rankin 4
(n = 91)

5.583.7
6.983.9
13.386.9
4.782.8
27.282.5

6.783.9
7.184.2
13.887.2
6.083.7
26.883.0

7.883.9
8.484.8
16.387.5
6.683.3
24.684.7

7.884.1
7.683.8
15.686.9
6.283.3
25.083.9

0.001
0.1
0.006
0.01
0.001

Means 8 standard deviation. p value: ANOVA test. NS = Nonsignificant.

to Trial of ORG 10172 in Acute Stroke Treatment classification. The HADS-depression mean scores (7.4 and 8.5;
ANOVA, p = 0.04) were significantly increased in moderate (NIHSS score: 613) and severely impaired stroke patients (NIHSS score 614), compared to patients with
mild impairment (NIHSS ^5), who scored 6.5.
Education, measured as the number of years of formal education, inversely correlated with the HADS-depression subscale (r = 0.25; p ! 0.0001). Patients who
were living alone scored significantly lower (worse) on
the HADS-depression subscale (5.6 8 3.9 vs. 7.3 8 3.9;
p = 0.01) and in the HADS total score (12.3 8 6.9 vs.
15.0 8 7.1; p = 0.02). No significant differences were observed in the GDS mean scores (5.3 8 3.6 vs. 6.2 8 3.3).
Depressed patients scored significantly worse (lower) in
the social function domain of the SF-36 (35.5 8 23.7 vs.
54.7 8 27.6, p ! 0.0001). Depressive symptoms were
more common in housewives (HADS-depression mean
score: 9.1) compared to retired and active workers
(HADS-depression mean score: 6.8 and 6.3, respectively; ANOVA, p = 0.001). Housewives who were married
scored significantly worse on the HADS-depression as
compared to those who were single (8.5 8 4.1 vs. 6.4 8
4.1; p = 0.04).
The depression and cognition scores broken down by
disability and functional status are shown in table 3. Disability and dependence in instrumental ADL were associated with PSD. Depressed patients had significantly
lower BI (65.9 vs. 74.2; unpaired t test, p = 0.03) and Lawton scores (14.8 vs. 17.2; p = 0.0003). Patients independent
in the ADL scored 5.2 on the HADS-depression subscale,
whereas severely disabled patients (BI ^60) scored 8.1
(p ! 0.0001). The Lawton mean score significantly (p !
0.001) correlated with the GDS mean score (r = 0.51),

HADS-depression subscale (r = 0.34) and HADS-anxiety subscale (r = 0.18). The HADS-depression subscale
scores significantly increased (ANOVA; p = 0.001) across
the m-RS stages. The HADS-depression mean score
ranged from 5.5 (m-RS 1) to 7.8 (m-RS 4).
Depressed stroke patients also scored worse in the
MMSE (25.4 vs. 26.4; p ! 0.0001). A significant association between cognition and dependence in the ADL was
also detected. Cognitive function, as measured by the
MMSE, significantly decreased as disability (measured
by the BI and/or Lawton scales) and functional status
(m-RS) worsened (ANOVA, p ! 0.0001).
Low scores were also observed in the Mental Component Summary (50.1 8 11.2) and the emotional role domain of the SF-36 (37.9 8 44.1). The HADS-depression
subscale significantly correlated with most domains of
the SF-36: physical functioning (r = 0.28), role limitations due to physical problems (r = 0.22), bodily pain
(r = 0.25), general health perception (r = 0.44), vitality (r = 0.49), social functioning (r = 0.40), role limitations due to emotional problems (r = 0.38) and emotional well being (r = 0.57). Significant correlations
(p ! 0.0001) were also observed between GDS scores and
SF-36 domains and ranged from 0.29 (physical functioning) to 0.59 (mental health). The SF-36 Mental
Component Summary significantly correlated (p !
0.0001) with HADS-depression (r = 0.37), HADS-anxiety (r = 0.40), HADS total score (r = 0.44) and GDS
(r = 0.34).
Several models of multivariate regression analysis
were tested for HADS-depression subscale and GDS (table 4), due to the alternative introduction of disability
(BI vs. Lawton scale) and functional impairment (m-RS
vs. NIHSS) scales. Dependence in instrumental ADL

Poststroke Depression

Cerebrovasc Dis 2009;28:157165

161

Table 4. Determinants for poststroke depression

Measure
HADS-depression
Diabetes
Housewife
SF-36 social
Education
Lawton scale
MMSE
GDS
SF-36 social
Lawton

Coefficient

Standard error

95% CI

R2 adjusted
0.32

1.76
1.80
0.05
0.11
0.16
0.21

0.54
0.71
0.01
0.05
0.07
0.06

0.001
0.01
<0.0001
0.01
0.02
0.002

2.800.69
2.711.12
0.06 to 0.03
0.20 to 0.02
0.29 to 0.02
0.33 to 0.08

0.05
0.19

0.01
0.61

<0.0001
0.002

0.06 to 0.03
0.31 to 0.07

0.23

Multivariable regression analysis. CI = Confidence interval.

and lower social functioning showed to have an independent effect in chronic stroke patients in both models.
In addition, PSD was significantly associated with
housewives, lower education level, decreased cognitive
functioning (as measured by the MMSE) and presence
of diabetes in the HADS-depression model (R adjusted = 0.32).

Discussion

The prevalence of PSD and their determinants varies

in the literature due to differences in diagnostic criteria,
selection of patients and the time elapsed since the stroke
[24, 25]. Although the frequency of PSD varies with time,
it probably affects between a third and a half of the patients at some stage. The rate of depressive symptoms
found in the Brazilian stroke population by means of 2
self-reported scales (HADS-depression: 19.2%; 15-item
GDS: 29.7%) is similar to those reported in developed
countries [1].
Epidemiological studies have used different measures to determine PSD, so the validation of prediction
models may be very difficult in different cultural settings. Most of the scales used to assess PSD suffer from
the limitation of having been structured for patients affected for functional depressive disorders. In addition,
the influence of stroke lesions on PSD may be moderated by psychosocial risk factors, and predictors of mood
may change over time. Multivariate analysis showed
that disability, social functioning, lower cognitive function, education, being a housewife and diabetes were
162

Cerebrovasc Dis 2009;28:157165

significant factors associated with PSD (as measured by

the HADS and GDS) in Brazilian stroke patients. Depression was mainly related to restriction in social activities and dependence in instrumental ADL in the
chronic phase of stroke.
Neither stroke subtypes nor location of vascular brain
lesion were predictors of PSD in our study. Some researchers postulated that left hemisphere lesions [26]
and/or basal ganglion infarctions [27] would be anatomical indicators for PSD. Nevertheless, the results for location of stroke and PSD are conflicting and no clear neuroanatomical factor underlying PSD has been found in
other studies [28, 29].
The HADS and GDS were used for the diagnosis of
PSD instead of the DSM-IV criteria. Many of the somatic symptoms of depression may arise from stroke itself
[30]. Somatic items such as fatigue, weight loss or headache that could be attributable to physical illness rather
than psychological states are not included in the HADS
[31]. Instead, the evaluation of depression is based largely
on items that reflect the concept of anhedonia. Five of the
9 criteria for depression included in the DSM-IV reflect
somatic symptoms, and none of them are represented on
the HADS [32]. The Hamilton Depression Rating Scale
also includes several physical items (weight loss, lack of
energy, lack of appetite, fatigue, insomnia, psychomotor
changes) that can overlap with physical symptoms observed in stroke survivors.
We chose the HADS as a primary measure to detect
PSD because most of our patients were !65 years old.
Nevertheless, we also decided to use the short version of
the GDS to assess mood in the group of elderly stroke paCarod-Artal/Ferreira Coral/Trizotto/
Menezes Moreira

tients. The GDS is most useful in the identification of

depression among individuals who are higher functioning, community dwelling and have no more than mild
cognitive impairment. The neurologist examiner orally
administered the GDS in order to include a wider range
of individuals [33].
Functional impairment, disability and restriction in
participation are important factors associated with PSD
[34]. Loss of employment, financial difficulties, social
isolation, poor self-esteem, relationship and sexual difficulties are increasingly recognized psychosocial problems in stroke survivors [35]. Reduced social activity, failure to return to work and poor participation in rehabilitation are associated with PSD. Living alone, having few
social contacts and lack of social support have also been
found to contribute to depression [36]. Other possible risk
factors include age, sex, and both personal and family
psychiatric history [37]. There is conflicting evidence regarding gender and PSD. In some trials, women have
been more depressive than men [36, 38]. In our study, female sex and status as a housewife were significantly associated with PSD. Although housewives were analyzed
as a working status category, the relationship between depression and housewife category might also be related to
the altered family roles caused by a stroke.
Cognitive impairment and depression occur commonly after stroke and adversely affect functional recovery. In addition, depressed stroke patients may be more
cognitively impaired [39]. The possibility of cognitive
deficits associated with PSD or coexisting cognitive impairment as confounding factors should be considered in
epidemiological studies about PSD. Cognitive dysfunction, as measured by the MMSE, was significantly associated with PSD and other functional measures (BI, Lawton, m-RS) in Brazilian stroke patients. Severe functional
dependence was less likely in subjects who were neither
cognitively impaired nor depressed. Although the relationship between depression and cognitive impairment
increases with age, the interaction of age was not found
in this study.
Diabetes was a significant factor associated with PSD.
An interaction of depression and diabetes has been proposed. Recognizing depression in diabetic individuals is
critical because it may play a role in worse control of diabetes and worse diabetes outcomes [40]. The relationship
between PSD and diabetes might be partially explained
by the presence of diffuse white matter disease or multiple ischemic lesions. Although this fact was not specifically assessed, no association between ischemic stroke
subtypes and diabetes was found in this study.

The assessment of mood in rehabilitation settings has

been recommended because PSD has a known adverse
impact on rehabilitation outcomes [41]. As there is some
evidence that the treatment of PSD improves the functional outcome [42], early diagnosis and treatment are
needed. The recognition of PSD risk factors may help
clinicians apply interventions aimed at preventing depression in the chronic phase of stroke. Psychosocial
and behavioral intervention adjunctive to antidepressant treatment may have efficacy in reducing PSD and
improving functional outcomes [43, 44]. Improvement
in depressive symptoms may be associated with recovery in the ADL, although cognitive factors may also determine the level of physical functioning achieved after
stroke [45]. Intensive community-based multidisciplinary rehabilitation has been shown to improve social
participation and instrumental ADL in stroke survivors
[46]. A positive effect of integrated care on depressive
symptoms in stroke survivors has also been reported
[47].
As patients with PSD were identified at the beginning
of the rehabilitation process, pharmacological and therapeutic interventions were performed early. In addition,
the identification of patients with PSD-associated risk
factors helped us elaborate a neuropsychological and psychosocial program to prevent PSD during stroke rehabilitation.
Crosscultural studies of PSD may differ because of
heterogeneity in study settings and patient selection, and
the range of measurement variables. A strength of our
study is the inclusion of a wide range of age, stroke
subtypes and degrees of disability and functional status
among Brazilian stroke patients. However, this study has
some limitations. The regression model was not controlled by the effect of potential cofounders, such as antidepressive treatment or the impact of prestroke depression. As these patients were assessed in a rehabilitation
hospital in the chronic phase of stroke, early or mild depression symptoms may have resolved within several
months after stroke in some stroke patients in their community. Aphasic stroke patients were also excluded from
the study, since they were unable to answer the depression
measures. PSD may be common in aphasic stroke patients [48]. Emotional indifference and other emotional
states may be confused with depression, mainly in the
acute phase of stroke [49].
Potential confounders of mood assessment (anosodiaphoria, anosognosia) and personality issues (emotionality, responsivity and empathy) should be specifically addressed in the future. Recently, a diagnostic tool specifi-

Poststroke Depression

Cerebrovasc Dis 2009;28:157165

163

cally developed to assess PSD (the Poststroke Depression

Rating Scale) has been developed. Crosscultural studies
are needed to adapt and compare the psychometric properties of this new scale [50].

In conclusion, a high prevalence of depressed mood

was observed in chronic stroke survivors. Functional, social and cognitive factors were independently associated
with PSD in Brazilian stroke patients. Pharmacologic
and rehabilitation strategies are needed to treat this
chronic condition early on.

References
1 Hackett ML, Yapa C, Parag V, Anderson CS:
Frequency of depression after stroke: a systematic review of observational studies.
Stroke 2005; 36:13301340.
2 Berg A, Palomki H, Lehtihalmes M, Lnnqvist J, Kaste M: Poststroke depression: an
18-month follow-up. Stroke 2003; 34: 138
143.
3 Pohjasvaara T, Vataja R, Leppavuori A, Kaste
M, Erkinjuntti T: Depression is an independent predictor of poor long-term functional
outcome post-stroke. Eur J Neurol 2001; 8:
315319.
4 Carod-Artal J, Egido JA, Gonzlez JL, Varela
de Seijas E: Quality of life among stroke survivors evaluated 1 year after stroke: experience of a stroke unit. Stroke 2000; 31: 2995
3000.
5 Carota A, Berney A, Aybek S, Iaria G, Staub
F, Ghika-Schmid F, Annable L, Guex P, Bogousslavsky J: A prospective study of predictors of poststroke depression. Neurology
2005;64:428433.
6 Lai SM, Studenski S, Duncan P, Perera S: Persisting consequences of stroke measured by
the Stroke Impact Scale. Stroke 2002; 33:
18401844.
7 Edwards DF, Hahn G, Baum CM, Perlmutter
MS, Sheedy C, Dromerick AW: Screening patients with stroke for rehabilitation needs:
validation of the post-stroke rehabilitation
guidelines. Neurorehabil Neural Repair
2006;20:4248.
8 De Padua Mansur A, de Ftima Marinho do
Souza M, Favarato D, Avakian SD, Machado
Csar LA, Mendes Aldrigui J, Franchini
Ramires JA: Stroke and ischemic heart disease mortality trends in Brazil from 1979 to
1996. Neuroepidemiology 2003;22:179183.
9 Foulkes MA, Wolf PA, Price TR, Mohr JP,
Hier DB: The Stroke Data Bank: design,
methods, and baseline characteristics. Stroke
1988;19:547554.
10 Adams HP Jr, Bendixen BH, Kappelle LJ,
Biller J, Love BB, Gordon DL, Marsh EE 3rd:
Classification of subtype of acute ischemic
stroke: definitions for use in a multicenter
clinical trial. Stroke 1993; 24:3541.
11 Brott T, Adams HP Jr, Olinger CP, Marler JR,
Barsan WG, Biller J, Spilker J, Holleran J,
Eberle R, Hertzberg V: Measurements of
acute cerebral infarction: a clinical examination scale. Stroke 1989; 20:864870.

164

12 Bonita R, Beaglehole R: Modification of

Rankin scale: recovery of motor function after stroke. Stroke 1988; 19:14971500.
13 Mahoney PL, Barthel DW: Functional evaluation: the Barthel Index. Md State Med J
1965;14:6165.
14 Lawton MP, Brody EM: Assessment of older
people: self-maintaining and instrumental
activities of daily living. Gerontologist 1969;
9:179186.
15 Folstein MF, Folstein SE, McHugh PR: MiniMental State: a practical method for grading
the cognitive state of patients for the clinician. J Psychiatr Res 1975;12:189198.
16 Zigmond AS, Snaith RP: The hospital anxiety and depression scale. Acta Psychiatr
Scand 1983;67:361370.
17 Sheikh JI, Yesavage JA: Geriatric Depression
Scale (GDS): recent evidence and development of a shorter version. Clin Gerontol
1986;5:165172.
18 Ware JE Jr, Sherbourne CD: The MOS 36Item Short-Form Health Survey (SF-36). I.
conceptual framework and item selection.
Med Care 1992;30:473483.
19 Almeida OP: Mini mental state examination
and the diagnosis of dementia in Brazil. Arq
Neuropsiquiatr 1998;56:605612.
20 Carod-Artal FJ, Vargas AP, Martinez-Martin P: Determinants of quality of life in Brazilian patients with Parkinsons disease. Mov
Disord 2007;22:14081415.
21 Almeida OP, Almeida SA: Short versions of
the geriatric depression scale: a study of their
validity for the diagnosis of a major depressive episode according to ICD-10 and DSMIV. Int J Geriatr Psychiatry 1999; 14: 858
865.
22 Martinez JE, Barauna Filho IS, Kubokawa K,
Pedreira IS, Machado LAM, Cevasco G:
Evaluation of the quality of life in Brazilian
women with fibromyalgia, through the medical outcome survey 36 item short-form
study. Disabil Rehabil 2001;23:6468.
23 Flint AJ, Rifat Sl: Factor structure of the Hospital Anxiety and Depression Scale in older
patients with major depression. Int J Geriatr
Psychiatry 2002;17:117123.
24 Van de Port IG, Kwakkel G, Bruin M, Lindeman E: Determinants of depression in chronic stroke: a prospective cohort study. Disabil
Rehabil 2007;29:353358.

Cerebrovasc Dis 2009;28:157165

25 Turner-Stokes L, Hassan N: Depression after

stroke: a review of the evidence base to inform the development of an integrated care
pathway. 1 Diagnosis, frequency and impact.
Clin Rehabil 2002;16:231247.
26 Robinson RG, Bolduc PL, Price TR: Twoyear longitudinal study of poststroke mood
disorders: diagnosis and outcome at one and
two years. Stroke 1987; 18:837843.
27 Vataja R, Leppvuori A, Pohjasvaara T, Mntyl R, Aronen HJ, Salonen O, Kaste M,
Erkinjuntti T: Poststroke depression and lesion location revisited. J Neuropsychiatry
Clin Neurosci 2004;16:156162.
28 Carson AJ, MacHale S, Allen K, Lawrie SM,
Dennis M, House A, Sharpe M: Depression
after stroke and lesion location: a systematic
review. Lancet 2000;356:122126.
29 Provinciali L, Paolucci S, Torta R, Toso V,
Gobbi B, Gandolfo C: Depression after firstever ischemic stroke: the prognosis role of
neuroanatomic subtypes in clinical practice.
Cerebrovasc Dis 2008;26:592599.
30 Salter K, Bhogal SK, Foley N, Jutai J, Teasell
R: The assessment of poststroke depression.
Top Stroke Rehabil 2007; 14:124.
31 ORourke S, MacHale S, Signorini D, Dennis
M: Detecting psychiatric morbidity after
stroke: comparison of the GHQ and the
HAD scale. Stroke 1998; 29:980985.
32 Aben I, Verhey F, Lousberg R, Rodder J, Honig A: Validity of the Beck Depression Inventory, Hospital Anxiety and Depression Scale,
SCL-90 and Hamilton Depression Rating
Scale as screening instruments for depression in stroke patients. Psychosomatics 2002;
43:386393.
33 Incalzi RA, Cesari M, Pedone C, Carbonin
PU: Construct validity of the 15-item geriatric depression scale in older medical inpatients. J Geriatr Psychiatry Neurol 2003; 16:
2328.
34 Lo RS, Cheng JO, Wong EM, Tang WK,
Wong LK, Woo J, Kwok T: Handicap and its
determinants of change in stroke survivors:
one-year follow-up study. Stroke 2008; 39:
148153.
35 Robinson RG, Murata Y, Shimoda K: Dimensions of social impairment and their effect on depression and recovery following
stroke. Int Psychogeriatr 1999; 11:375384.
36 Andersen G, Vestergaard K, IngemannNielsen M, Lauritzen L: Risk factors for poststroke depression. Acta Psychiatr Scand
1995;92:193198.

Carod-Artal/Ferreira Coral/Trizotto/
Menezes Moreira

37 Carod-Artal FJ: Post-stroke depression. I.

Epidemiology, diagnostic criteria and risk
factors. Rev Neurol 2006;42:169175.
38 Hermann N, Black SE, Lawrence J, Szekely C,
Szalai JP: The Sunnybrook Stroke Study: a
prospective study of depressive symptoms
and functional outcome. Stroke 1998; 29:
618624.
39 MacHAle SM, ORourke SJ, Wardlaw JM,
Dennis MS: Depression and its relation to lesion location after stroke. J Neurol Neurosurg Psychiatry 1998;64:371374.
40 Pirraglia PA, Gupta S: The interaction of depression and diabetes: a review. Curr Diabetes Rev 2007;3:249251.
41 Donovan NJ, Kendall DL, Heaton SC, Kwon
S, Velozo CA, Duncan PW: Conceptualizing
functional cognition in stroke. Neurorehabil
Neural Repair 2008;22:122135.

Poststroke Depression

42 Hackett ML, Anderson CS, House A, Xia J:

Interventions for treating depression after
stroke. Cochrane Database Syst Rev
2008;CD003437.
43 Williams LS, Kroenke K, Bakas T, Plue LD,
Brizendine E, Tu W, Hendrie H: Care management of poststroke depression: a randomized, controlled trial. Stroke 2007; 38:
9981003.
44 Robinson RG, Jorge RE, Moser DJ, Acion L,
Solodkin A, Small SL, Fonzetti P, Hegel M,
Arndt S: Escitalopram and problem-solving
therapy for prevention of poststroke depression: a randomized controlled trial. JAMA
2008;299:23912400.
45 Saxena SK, Ng TP, Koh G, Yong D, Fong NP:
Is improvement in impaired cognition and
depressive symptoms in post-stroke patients
associated with recovery in activities of daily
living? Acta Neurol Scand 2007; 115: 339
346.
46 Ryan T, Enderby P, Rigby AS: A randomized
controlled trial to evaluate intensity of community-based rehabilitation provision following stroke or hip fracture in old age. Clin
Rehabil 2006;20:123131.

47 Joubert J, Joubert L, Reid C, Barton D, Cumming T, Mitchell P, House M, Heng R, Meadows G, Walterfang M, Pantelis C, Ames D,
Davis S: The positive effect of integrated care
on depressive symptoms in stroke survivors.
Cerebrovasc Dis 2008;26:199205.
48 Laska AC, Martensson B, Kahan T, von Arbin M, Murray V: Recognition of depression
in aphasic stroke patients. Cerebrovasc Dis
2007;24:7479.
49 Starkstein SE, Fedoroff JP, Price TR, Leiguarda R, Robinson RG: Apathy following
cerebrovascular lesions. Stroke 1993; 24:
16251630.
50 Quaranta D, Marra C, Gainotti G: Mood disorders after stroke: diagnostic validation of
the poststroke depression rating scale. Cerebrovasc Dis 2008;26:237243.

Cerebrovasc Dis 2009;28:157165

165

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.