Stroke Epidemiology in Australia

Progress Review
Frequency of Depression After Stroke

A Systematic Review of Observational Studies
Maree L. Hackett, MA (Hons); Chaturangi Yapa, BSc; Varsha Parag, MSc (Hons);
Craig S. Anderson, PhD, FRACP, FAFPHM
Background and Purpose—Although depression is an important sequelae of stroke, there is uncertainty regarding its
frequency and outcome.
Methods—We undertook a systematic review of all published nonexperimental studies (to June 2004) with prospective
consecutive patient recruitment and quantification of depressive symptoms/illness after stroke.
Results—Data were available from 51 studies (reported in 96 publications) conducted between 1977 and 2002. Although
frequencies varied considerably across studies, the pooled estimate was 33% (95% confidence interval, 29% to 36%)
of all stroke survivors experiencing depression. Differences in case mix and method of mood assessment could explain
some of the variation in estimates across studies. The data also suggest that depression resolves spontaneously within
Downloaded from http://stroke.ahajournals.org/ by guest on December 25, 2017
several months of onset in the majority of stroke survivors, with few receiving any specific antidepressant therapy or
active management.
Conclusions—Depression is common among stroke patients, with the risks of occurrence being similar for the early,
medium, and late stages of stroke recovery. There is a pressing need for further research to improve clinical practice in
this area of stroke care. (Stroke. 2005;36:1330-1340.)
Key Words: depression 䡲 epidemiology 䡲 meta-analysis 䡲 stroke
S ome form of depression is considered to occur in at least

one-quarter of patients in the first year after acute
stroke,1– 4 with the period of greatest risk being the first few
ical and time-limited boundaries. There were no restrictions on the
basis of language, sample size, or duration of follow-up, but we
excluded studies of mixed populations (such as stroke and head
injury) unless separate results for stroke patients were identified.
months after onset.2,5,6 However, such estimates vary consid- Studies were also excluded if they had any of the following: (1)
erably across studies because of differences in definitions, limited to specific patient characteristics such as sex or location of
study populations and the timing of assessments, as well as stroke lesion; (2) convenience sampling; (3) retrospective recruit-
complexities in the recognition, assessment, and diagnosis of ment; or (4) there was only unstructured assessment of mood.
The primary endpoint was the proportion of patients who met the
abnormal mood in the setting of stroke-related disability.3,4,7 diagnostic category of depression as applied in a study, which
The results of studies may also depend on whether subjects included: (1) depressive disorder, depressive symptoms, or “psycho-
are categorized on the basis of psychiatric interview using logical distress,” as defined by scores above a cut-point for abnor-
standard diagnostic criteria such as the Diagnostic and Sta- mality on a standard mood scale; and (2) major depression or minor
depression (or dysthymia) according to DSM-IIIR,8 DSM-IV,9 or
tistical Manual of Mental Disorders (eg, DSM-IIIR,8 DSM- other standard diagnostic criteria.
IV9), or a psychiatric rating scale such as the Montgomery Studies that reported mood outcomes only as a continuous
Åsberg Depression Rating Scale,10 or on the basis of a variable, without a categorical diagnosis, were excluded.
self-rating mood scale. The goal of this review was to
determine the frequency, outcome, and management of de- Search Strategy and Data Extraction
pression after stroke from all high-quality published studies. Data for this review were identified from the following comput-
erized databases: MEDLINE, EMBASE, CINAHL, PsychINFO,
Applied Science and Technology Plus, Biological Abstracts,
Materials and Methods General Science Plus, Arts and Humanities Index, Science Cita-
This review was restricted to published research articles, abstracts, tion Index, Social Sciences Citation Index, Sociofile, and Digital
and letters of patients with a clinical diagnosis of stroke in whom an Dissertations using terms and strategy based on the Cochrane
assessment of mood was performed at a prespecified time point. The Stroke Group methodology (full details available on request; last
review included incidence studies and case series that had prospec- searched June 2004). In some cases, similarities between study
tive consecutive patient recruitment within clearly defined geograph- reports indicated the possibility of multiple publications from the
Received November 15, 2004; final revision received January 13, 2005; accepted February 15, 2005.
From the George Institute for International Health (M.L.H., C.S.A.), Neurological Diseases and Ageing Division, Royal Prince Alfred Hospital and
the University of Sydney, Australia; and the Clinical Trials Research Unit (C.Y., V.P.), Faculty of Medical and Health Sciences, the University of
Auckland, New Zealand.
Correspondence to Maree Hackett, the George Institute for International Health, PO Box M201, Missenden Road, Sydney, NSW 2050, Australia.
E-mail mhackett@thegeorgeinstitute.org
© 2005 American Heart Association, Inc.
Stroke is available at http://www.strokeaha.org DOI: 10.1161/01.STR.0000165928.19135.35
1330
Hackett et al Frequency of Depression After Stroke 1331
same cohort. In the absence of explicit cross-referencing, we population, data were taken from the first publication that referred
judged articles to be from the same cohort if they met the to each follow-up period. When a variety of methods of mood
following criteria: there was evidence of overlapping recruitment assessment were undertaken, we reported results for each scale/
sites, study dates, and grant funding numbers; and there were method. Because we were unable to obtain enough information on
similar or identical reported patient characteristics in the stud- publications from one group of papers (the Iowa Group), we
ies.11 If several articles reported outcomes from the same study chose to present these studies separately.
TABLE 1. Frequency of Depressed Mood After Stroke: A Summary of the Population-Based Evidence
Study Country No. Assessed Time Since Stroke Depression Diagnosis/Criteria Frequency (95% CI)
Bristol Stroke Study25 UK 379 3 wk WDI (score 19–36, depressed) 22 (18 to 26)
All strokes WDI (score 15–18, probably depressed) 11 (8 to 14)
All strokes 377 6 mo WDI (score 19–36, depressed) 20 (16 to 24)
WDI (score 15–18, probably depressed) 12 (9 to 15)
All strokes 348 12 mo WDI (score 19–36, depressed) 18 (14 to 22)
WDI (score 15–18, probably depressed) 13 (10 to 17)
FINNSTROKE58 Finland 321 3 mo BDI (score ⱖ10) 47 (42 to 53)
BDI (score 30–63 “severe”) 3 (1 to 5)
BDI (score 19–29 “moderate”) 11 (8 to 14)

BDI (score 10–18 “mild”) 28 (23 to 33)
311 12 mo BDI (score ⱖ10) 47 (42 to 53)
BDI (score 30–63 “severe”) 2 (0 to 4)
BDI (score 19–29 “moderate”) 15 (11 to 19)
BDI (score 10–18 “mild”) 25 (20 to 30)
Örebro Stroke Study 24 Sweden 231 12 mo GDS (score ⬎5) 37 (31 to 43)
251 12 mo DSM-IV depression 27 (22 to 33)
OCSP6 UK 89 1 mo DSM-III: major depression 11 (5 to 18)
DSM-III: dysthymic disorder 1 (0 to 3)
PSE/index of definition 19 (11 to 27)
76 1 mo BDI (score ⱖ10) 32 (22 to 43)
BDI (score ⱖ13) 20 (11 to 29)
BDI (score ⱖ17) 8 (2 to 14)
119 6 mo DSM-III: major depression 9 (4 to 14)
107 6 mo BDI (score ⱖ10) 32 (23 to 41)
BDI (score ⱖ13) 15 (8 to 22)
112 12 mo DSM-III: major depression 5 (1 to 9)
88 12 mo BDI (score ⱖ10) 16 (8 to 24)
Prevalence over 12 mo BDI (score ⱖ10) 39 (30 to 48)
ICD-9 Psychiatric disorder 22 (15 to 29)
60 3–5 y DSM-III-R major depression 18 (8 to 28)
PCSS2 Australia 191 4 mo DSM-III major depression 17 (12 to 22)
DSM-III minor depression 11 (7 to 15)
All strokes 294 4 mo DSM-III major depression 15 (11 to 19)
SELSS63 UK 96 5y HADS (score⬎10) 36 (26 to 46)
BDI indicates Beck Depression Inventory; CI, confidence interval; DSM, Diagnostic and Statistical Manual of Mental Disorders; HADS, Hospital Anxiety and
Depression Scale; OCSP, Oxford Community Stroke Study; PCSS, Perth Community Stroke Study; PSE, Present State Examination; SELSS, South East London Stroke
Study; UK, United Kingdom; WDI, Wakefield Depression Inventory.
1332 Stroke June 2005
TABLE 2. Frequency of Depressed Mood After Stroke: A Summary of the Hospital-Based Evidence
Author, Year Country No. Assessed Time Since Stroke Depression Diagnosis/Criteria Frequency (95% CI)
Aben, 200233 Netherlands 154 1 mo DSM-IV depression 22 (15 to 28)
Prevalence over 1 y DSM-IV major depression 23 (16 to 30)
DSM-IV minor depression 15 (9 to 21)
Andersen, 199432 Denmark 209 4–6 wk HDRS (score ⱖ13) 21 (15 to 26)
Bayer, 200144 Jordan 168 3 mo DSM-IV major depression 25 (19 to 32)
Berg, 200146 Finland 89 2 wk DSM-III-R major depression 6 (1 to 11)
BDI (score ⱖ10) 27 (18 to 36)
Desmond, 200327 USA 421 3 mo HDRS (score ⱖ12) 11 (8 to 14)
Ebrahim, 198764 UK 149 6 mo GHQ (score ⱖ12) 23 (16 to 30)
Eriksson, 2004*17 Sweden 13 999 3 mo Single simple question 14 (14 to 15)
Hayee, 200134 Bangladesh 161 3 mo BDI (score ⱖ10) 41 (33 to 49)
156 1y 42 (34 to 50)
Herrmann, 199526 Germany 47 Within 2 mo DSM-III-R major depression 19 (8 to 30)
DSM-III-R minor depression 17 (6 to 28)

Herrmann, 1998 5 Canada 150 3 mo MADRS (score ⱖ7) 27 (20 to 34)
ZDS (score ⱖ50) 22 (15 to 29)
133 1y MADRS (score ⱖ7) 22 (15 to 29)
ZDS (score ⱖ50) 21 (14 to 28)
Hosking, 2000 48 New Zealand 79 3 mo GDS (score ⬎9) 39 (28 to 50)
Hüwel, 199849 Germany 102 5–9 d MADRS (cut-point not stated) 55 (45 to 65)
Jaracz, 200350 Poland 72 6 mo ZDS (score ⱖ50) 46 (34 to 57)
Kotila, 198465 Finland 154 3 mo BDI (cut-point not stated) 44 (36 to 52)
1y 29 (22 to 36)
Pohjasvaara, 199866 Finland 277 3 mo DSM-III-R major depression 26 (21 to 31)
BDI (score ⱖ10) 44 (38 to 50)
276 15 mo BDI (score ⱖ10) 45 (39 to 51)
Williams, 199955 USA 71 1 mo BDI (cut point not stated) 39 (28 to 50)
*Mood assessed using a single standard question.
GDS indicates Geriatric Depression Scale; HDRS: Hamilton Depression Rating Scale; MADRS: Montgomery Åsberg Depression Rating Scale; USA: United States of
America; ZDS: Zung Depression Scale.
One reviewer (M.H.) extracted the data, and a second reviewer that spanned 2 time points (eg, 2 to 6 weeks after stroke) were
(C.Y.) cross-checked a selection of data extractions. Studies were included in the latter time category.
grouped into 3 categories that represented degrees of case selection. We used the depression category that included the most patients to
The first group, “population-based studies,” considered to be of the determine the point estimate, together with 95% confidence intervals
highest (least biased) quality, consisted of studies that attempted to (CIs) for each study. Pooled estimates were calculated using both
recruit all stroke patients, including those who were not admitted to fixed and random effects. When one study contributed more than one
hospital for acute care. Although not all of these studies fulfilled endpoint to the pooled estimate, sensitivity analyses were undertaken
certain “ideal” criteria for population-based stroke incidence stud- using either the earliest or the latest assessment in that study,
ies,12 we considered that these studies included stroke patients with although data were presented herein only for the earliest assessment.
the most representative characteristics. Statistical heterogeneity was assessed using the standard Q statistic,
The other 2 categories were “hospital-based” studies, which with P⬍0.05. When there was evidence of statistical heterogeneity,
included all inpatients from acute care medical wards in general the random effects approach of DerSimonian and Laird15 was used to
hospitals, and “rehabilitation-based” studies, which included patients pool the frequencies.
from rehabilitation wards, or hospitals (including stroke units), with
one study of patients recruited from an outpatient clinic13 and one Results
study of patients from primary care general practices.14 More than 12 000 references were identified, of which 418
were retrieved to assess for inclusion/exclusion criteria, and a
Statistical Analysis total of 96 reports (51 studies) were considered eligible for
Studies were stratified by case selection, and according to the timing inclusion.
of mood assessment from stroke onset, defined as: “acute phase”
(within 1 month and including date of admission to rehabilitation
beds); “medium-term phase” (between 1 and 6 months and including Patient Characteristics
date of discharge from rehabilitation); and “long-term phase” (6 The 6 population-based studies included 2869 patients from a
months or more) after stroke. Studies with follow-up assessments base population of 1 338 981 between 1981 and 2000 (Table
TABLE 3. Frequency of Depressed Mood After Stroke: A Summary of the Rehabilitation-Based Evidence
First Author, Year,
Source of Patients Country No. Assessed Time of Assessment Depression Criteria Frequency (95% CI)
Åstrom, 1993 37 Sweden 76 Discharge from hospital DSM-III-R major depression 25 (15 to 35)
Stroke unit 73 3 mo after stroke 31 (20 to 42)
68 1 y after stroke 16 (7 to 25)
Bacher, 199038 Canada 48 Admission to hospital ZDS (score ⱖ60) 4 (0 to 10)
Rehabilitation hospital ZDS (score 50–59) 25 (13 to 38)
43 6 wk after admission ZDS (score ⱖ60) 9 (1 to 18)
ZDS (score 50–59) 26 (13 to 39)
42 6 mo after admission ZDS (score ⱖ60) 17 (5 to 28)
ZDS (score 50–59) 24 (11 to 37)
39 1 y after admission ZDS (score ⱖ60) 21 (8 to 33)
ZDS (score 50–59) 31 (16 to 45)
Bendsen, 199745 Denmark 128 1.5 mo after stroke DSM-III-R major depression 16 (10 to 22)
Rehabilitation hospital
Carod-Artal, 200067 Spain 90 1 y after stroke HDRS (cut-point not stated) 38 (28 to 48)
Stroke unit
Daily, 198368 USA 32 Admission to unit HDRS & BDI (cut-point not stated) 16 (3 to 29)
Stroke unit 7 d after admission 25 (10 to 40)
Diamond, 199529 USA 14 Admission to unit GDS (score ⬎10) 36 (11 to 61)
Geriatric rehabilitation unit Discharge from unit 29 (5 to 53)
Folstein, 197716 USA 20 30 d after stroke PSE (ⱖ 8 items on the 8th edition) 45 (23 to 67)
Gainotti, 199947 Italy 153 ⬍2 mo after stroke DSM-III-R major depression/ 27 (20 to 34)/
Neurology department & 2–4 mo after stroke HDRS (score ⱖ17 32 (25 to 39)
rehabilitation center ⬎4 mo after stroke DSM-III-R major depression/ 40 (32 to 48)/
HDRS (score ⱖ17) 60 (52 to 68)
Gillen, 200169 USA 243 15 d after stroke GDS (score ⱖ15) 13 (9 to 17)
Inpatient rehabilitation
Gottlieb, 200270 Israel 65 3 mo after stroke GDS (ⱖ50% items endorsed) 63 (51 to 75)
Inpatient rehabilitation 9 mo after stroke 58 (46 to 70)
Jürgensen, 199971 Germany 77 6 mo after discharge DSM-IV major depression 20 (11 to 29)
Geriatric rehabilitation center
Kauhanen, 199951 Finland 101 3 mo after stroke DSM-III-R major depression 9 (3 to 15)
Stroke unit 92 1 y after stroke DSM-III-R minor depression 44 (34 to 54)
DSM-III-R major depression 15 (8 to 22)
Kellermann, 199972 Hungary 82 1 wk after admission BDI (score ⬎10) 20 (11 to 28)
Stroke unit
Kim, 200213 Korea 145 3–12 mo after stroke DSM-IV depression (ⱖ5 items), BDI (score ⬎13) 15 (9 to 21)
Outpatient clinic
King, 200239 USA 53 Discharge from unit CES-D (score ⱖ16) 30 (8 to 42)
Rehabilitation units and 6–10 wk after discharge 26 (14 to 38)
general hospital 1 y after discharge 17 (7 to 27)
2 y after discharge 23 (12 to 34)
Prevalence over 2 y 51 (38 to 65)
1).18 –23 All these studies included patients with standard assessments ranged from 61% (3-week assessment)25 to 99%
clinical criteria for stroke, although one excluded patients (12-month assessment).24
with subarachnoid hemorrhage in the assessment of mood The hospital-based studies included 16 302 patients (see
outcomes,24 and between 38%20 and 92%19 of all stroke Table 2), and the rehabilitation-based studies included 6036
patients were managed in hospital. All studies excluded patients (Table 3) at the initial assessment. One study17
patients with communication difficulties (eg, aphasia, confu- accounted for the majority (86%) of patients in the hospital-
sion, dementia), so that the proportion who completed mood based studies, with the remaining studies ranging in size from
TABLE 3. Continued
First Author, Year,
Source of Patients Country No. Assessed Time of Assessment Depression Criteria Frequency (95% CI)
Langhorne, 2000* 31 Scotland 311 Duration of hospitalization Single simple question 16 (12 to 20)
Stroke unit & hospital
Lincoln, 199814 UK 84 1 mo after stroke HADS (score ⬎10) 13 (6 to 20)
54 GP practices
Löfgren, 199959 Sweden 47 3 y after stroke DSM-IV ongoing depression 38 (24 to 52)
Geriatric stroke and
rehabilitation ward
Malec, 199052 USA 20 8–30 d after stroke HDRS (score ⬎8) 35 (14 to 56)
Rehabilitation unit
Mast, 200473 USA 195 1 wk after admission GDS (score ⬎10) 36 (30 to 43)
Morris 199040 Australia 99 2 mo after stroke DSM-III major depression 14 (7 to 21)
Hospital, geriatric unit & DSM-III minor depression 18 (10 to 26)
rehabilitation hospital 56 15 mo after stroke DSM-III major depression 7 (2 to 12)
Ng, 199557 Singapore 52 22 da after stroke DSM-III-R depressive illness 55 (42 to 69)
Rehabilitation center 49 Discharge from unit 29 (16 to 42)
Paolucci, 199930 Italy 470 50 d after stroke HDRS (score ⱖ18) 28 (24 to 32)
Rehabilitation unit
Sinyor, 198653 Canada 64 59 d after stroke ZDS (score ⱖ60) 22 (12 to 32)
Rehabilitation hospital ZDS (score 50–59) 25 (14 to 36)
Spalletta, 200256 Italy 153 7–14 d after admission DSM-IV major depression 41 (33 to 49)
Neuropsychiatric rehabilitation DSM-IV minor depression 17 (11 to 23)
Tang, 200274 China 157 1 mo after stroke DSM-III-R depression 17 (11 to 24)
Stroke rehabilitation unit
Van de Weg, 199954 Holland 85 3–6 wk after stroke DSM-III-R depression 35 (25 to 45)
Rehabilitation centres
Verdelho, 200460 France 110 6 mo after stroke MADRS (score ⱖ7) 43 (34 to 52)
Acute stroke unit 96 1 y after stroke 36 (26 to 46)
Weimar, 200228 Germany 2853 1 y after stroke CES- D (score ⱖ17) 33 (31 to 35)
Mixed
*Mood assessed using a single standard question.
CES-D indicates Centre for Epidemiologic Studies-Depression scale; DSM, Diagnostic and Statistical Manual of Mental Disorders; GP, General Practitioner.
4726 to 42127 patients. The largest rehabilitation-based study28 The criteria used to define depression (or depression symp-
contributed 50% of patients, with the remaining studies tom burden) also varied across studies. In the main, DSM
ranging in size from 1429 to 47030 patients. Only 2 separate criteria were used to define depression (in 19 studies) using
cohorts were clearly identified from the Iowa Group, which information from completed mood scales, and occasionally
included up to 463 patients (Table 4). from structured interviews, although the criteria for dysthy-
mia were modified by excluding the requirement for symp-
Diagnosis of Depression toms to be present for at least 2 years. There was also
A variety of methods were used to diagnose depressive illness variation in the cut-points used on the standardized mood
or assess the degree of depressive symptoms, covering 10
scales, whereas there was consistency for the Montgomery
different mood scales and 4 different psychiatric interview
Åsberg Depression Rating Scale (ⱖ7) and the Zung Depres-
schedules. Mood scales were either self-completed by pa-
sion Scale (ⱖ50), these scales were used in only a minority
tients (in 4 studies) or interviewer administered and scored
(20 studies), and the psychiatric interviews were conducted (6) of studies, whereas multiple cut-points were used for the
either alone (in 4 studies) or in combination with a self- Beck Depression Inventory (ⱖ10, ⬎10, ⬎13, and ⱖ17), the
administered or interviewer-administered mood scale (10 Geriatric Depression Scale (⬎5, ⬎10, ⱖ15, ⬎50% items
studies). The remaining studies used either varying combina- positively endorsed), and the Hamilton Depression Rating
tions of these methods, or the method was not explicitly Scale (⬎8, ⱖ12, ⱖ13, ⱖ17, ⱖ18).
stated in the report. Two studies used a single simple question
to diagnose depression,17,31 but these data were not included Frequency of Depression
in the pooled estimates because of the nonstandard method of Although there was considerable variation in the reported
assessment. frequency of depression after stroke across individual studies,
TABLE 4. Selection of Publications by the Iowa Group on the Frequency of Depression After Stroke
First Author, Year Assessed/Alive Time Since Stroke Topic Depression Criteria Frequency
Cohort 1*
Robinson, 198275† 103/154 Acute assessment Prevalence and severity of depression GHQ-28 ⱖ5 29
GHQ-28 ⱖ6 23
GHQ-28 ⱖ8 17
Robinson, 198376 103 2 wk Predictors of depression DSM-III Major D 27
DSM-III Minor D 20
Robinson, 198477 40/154 3 mo Prevalence and duration of depression DSM-III Major D 22
DSM-III Minor D 27
50/154 6 mo DSM-III Major D 34
DSM-III Minor D 26
Robinson, 198778 37/154 12 mo Prevalence of depression DSM-III Major D 14
DSM-III Minor D 19
48/154 2y DSM-III Major D 21
DSM-III Minor D 21
Parikh, 199079 63 2y Recovery over 2 y DSM-III Major D 24
DSM-III Minor D 6
Cohort 2‡
Castillo, 199380 288/309 2 wk Generalized anxiety disorder and depression HDRS (no cutpoint), PSE (no cutpoint) 38
Downhill, 199481 140/309 Baseline Depression and cognitive impairment DSM-IV Major D 40
DSM-IV Minor D 35
Castillo, 199582 142/215 Baseline Correlates of early and late onset GAD ‘Depression’ 51
78 3 mo DSM-III-R GAD 27
80 6 mo 27
70 12 mo 36
66 2y 17
27
Kishi, 199683 301 Baseline Validity of observed depression DSM-IV Major D 17
DSM-IV Minor D 18
Paradiso, 199784 142 Baseline Psychological symptoms associated with depression HDRS (no cut-point), PSE (no cut-point) 42
76 3 mo 38
79 6 mo 39
69 12 mo 29
66 2y 38
Schultz, 199785 142 Baseline GAD and depression and recovery DSM-IV GAD 19
77 3 mo 22
79 6 mo 25
70 12 mo 11
66 2y 18
Shimoda, 199886 142 Baseline GAD and depression on recovery DSM-IV Major D 19
DSM-IV GAD 23
Shimoda, 199887 142 Baseline Social functioning and depression on recovery DSM-IV Major D 19
DSM-IV Minor D 25
Paradiso, 199888 301 2 wk Gender differences in depression DSM-IV Major D 17
DSM-IV Minor D 18
GAD indicates generalized anxiety disorder; GHQ, General Health Questionnaire.
*Funding numbers: MH00163, NS15178, NS16332, NS18622, NS9 –2032; recruitment period 1.5 years, or between January 1980 and July/August 1981;
populations had similar demographic characteristics.
†Did not report funding number NS9-2032.
‡Funding numbers: MH00163, MH40355; populations had similar demographic characteristics.
Observational studies of the proportional

frequency of depression after stroke.
the pooled estimate indicates that depressive symptoms are estimates again varied when studies were grouped by type of
present in 33% (95% CI, 29% to 36%) of all stroke survivors mood scale and timing of assessment.
at any time during follow-up (Figure). The pooled estimate The Oxfordshire Community Stroke Project,6 conducted
from the population-based studies was 33% in the acute and nearly 20 years ago, is the only population-based study to
medium-term phases, with a slight increase to 34% in the date that recruited controls to allow estimates of the relative
long-term phase of recovery after stroke. There was modest risks of depression after stroke. Using Beck Depression
variation in the pooled frequencies in the hospital-based Inventory (scores ⱖ10) and the Present State Examination
(acute 36%, medium-term 32%, and long-term 34%) and psychiatric interview schedule (scores ⱖ5) to determine
rehabilitation-based studies (acute 30%, medium 36%, and “caseness,” the study showed that the frequency of depression
long-term 34%) over time, but the 95% CIs around all pooled in stroke survivors (20% Beck Depression Inventory; 11%
frequencies overlapped. Present State Examination) was twice that in controls, al-
There was, however, more variation in the pooled frequen- though this difference only reached statistical significance at
cies when studies were grouped by method of mood assess- the 6-month follow-up assessment. In addition, 4 hospital-
ment (data not presented). The 2 studies that used a single based studies have compared mood data in cases with
simple question to determine depression status17,31 had a controls, although only 1 study had controls selected from the
pooled frequency of 14% (95% CI, 14% to 15%). The community.32 However, this latter study only assessed pa-
smallest pooled frequency with standardized questionnaires tients with new episodes of depression within the 6 months
was from studies that used the Hamilton Depression Rating after stroke, yet still included patients with depression that
Scale (26%; 95% CI, 11% to 42%), whereas the highest had been present for ⬎1 year. Despite this anomaly, the 5%
frequency was in studies that used the Montgomery Åsberg proportional frequency of “new” depression was similar in
Depression Rating Scale (41%; 95% CI, 23% to 60%) or the stroke patients and controls. Another study showed no dif-
Zung Depression Scale (41%; 95% CI, 34% to 48%). The ference in the cumulative 1-year incidence of depression after
stroke compared with a randomly selected control group of available on the proportion of stroke patients who had
patients with first-ever myocardial infarction (in contrast to received psychotherapy.
consecutive recruitment of stroke patients).33 Different results
were found in 2 other studies: one study in which controls Discussion
were randomly selected from the neighboring hospital com- We report that one third of all people experience significant
munity and including spouses of stroke patients27 showed depressive symptoms at some time after the onset of stroke.
more depression in stroke survivors (11%) than controls (5%) We recognize, however, that this is likely to be a conservative
at 3 months after stroke (odds ratio [OR], 2.52; 95% CI, 1.35 estimate because of potential under-reporting (or under-
to 4.80); and the other study that did not provide any details recognition) of abnormal mood, and the difficulties inherent
on the selection of controls, reported that depression was also to the assessment of mood in patients with neurological
more common in stroke patients than controls (OR, 3.16; disability, particularly when there are communication prob-
95% CI, 1.48 to 4.12).34 lems caused by dysphasia and/or dementia. Given the impor-
Another robust examination of the relative frequency of tance of mood, which along with cognition, motivation, and
depression in stroke survivors was undertaken in The Fra- social support is a key factor influencing recovery from
mingham Study, a prospective, observational, community- stroke, it is surprising that there is much misconception over
based study that enrolled middle-aged subjects who have the epidemiology of stroke-associated depression, although
been followed-up biennially since the middle of the past the generally poor quality of studies has obviously contrib-
century.35 When data from 74 of the 251 subjects who uted to this situation.
experienced a stroke between 1982 and 1994 were isolated Whereas previous reports have acknowledged wide varia-
and compared with data from 74 control subjects matched for tion in the frequency of depression after stroke across studies
age and sex, significantly more stroke survivors (38%) were largely because of differences in patient characteristics and
depressed at 6 months after stroke than controls (10%). In a study designs, they have also suggested that the lowest and
separate analysis of ⱖ20-year outcomes from 148 subjects highest frequency of depression is found among patients in
with a stroke between 1972 and 1974,36 only 1 (11%) of 9 population-based and rehabilitation-based studies, respec-
stroke survivors met the criteria for depression compared tively, potentially reflecting selection bias toward the inclu-
with 3 (15%) of 20 nonstroke controls. sion of more disabled stroke survivors in the latter studies.2,5,6
Moreover, the time period of greatest risk of depression has
History of Depression, Antidepressants, traditionally been considered to be the first few months of
and Psychotherapy stroke onset. Our review, conversely, showed consistency in
Only 2 population-based studies have assessed the natural the overall frequency of depression across the 3 different
history of depression within individual stroke patients.6,25 types of studies and in relation to the time periods from stroke
Both found that only a small proportion of patients had onset, thus raising doubts about specific biological theories
depressive symptoms that persisted for most of the first year related to an acute stroke lesion as the major cause of
after stroke, despite the relatively constant overall frequency depression in this condition. In addition, we found that few
of depression among the total study population during this stroke patients receive effective management (antidepressants
time. Although some patients recovered spontaneously from or psychotherapy) for their depression, although the limited
depression within a few months, others had depressive data would suggest that these symptoms are self-limited in
symptoms for the first time later after stroke. Similar results most after several months.
were found in 1 hospital-based,5 and 4 rehabilitation-based Because the design of observational studies, by their very
studies,37– 40 and are further confirmed by this review. nature, may differ in a number of important ways,61 it is
It is well-recognized that personal or family history of useful to explore and quantify the reasons for such heteroge-
depression may place an individual at increased risk of neity. We identified variation in the cut-points used to
depression.41– 43 Although no population-based studies ex- determine “caseness” in the standardized mood scales as one
cluded patients with a history of depression from analyses, 17 key source of variability in the data. It would appear that
hospital-based and rehabilitation-based studies excluded pa- many studies failed to consider that older people, and those
tients with recent or severe depression,13,26,30,32,38,44 –55 includ- with physical illnesses, might require higher cut-points on
ing 6 studies in which patients using antidepressants at entry these scales. Two other problems were that multiple methods
were also excluded.30,46,48,52–54 were often used to diagnose depression, with few investiga-
Use of antidepressants at entry or during follow-up was tors clearly identifying an a priori primary endpoint in their
assessed in less than half of the included studies, and ranged study, so that the endpoints reported varied between “any
from 0% in the first few weeks56 to 31% at 2 years after depression,” “first-ever depression,” and “severity” of de-
stroke.39 The highest proportion of “adequately treated” pression. Without greater uniformity or standardization of
patients (ie, those using antidepressants who did not fulfil such methodological issues, it will remain difficult to deter-
criteria for depression) at the time of assessment in any one mine whether heterogeneity in study findings represent true
study was 84%.27 The highest proportion of participants differences in characteristics of populations or simply artifact
receiving antidepressants who were “inadequately treated” caused by measurement bias and other error.
(ie, those on antidepressants who still fulfilled criteria for Of course, heterogeneity across studies can also be attrib-
depression) was 71%,57 but this frequency ranged between uted to differences in case mix, including variation in stroke
17% and 36% in most studies.5,24,34,51,54,58 – 60 No data were features, clinical characteristics, source of patient recruit-
ment, and the timing of assessment. In this review, we have 5. Herrmann N, Black SE, Lawrence J, Szekely C, Szalai JP. The Sun-
attempted to minimize heterogeneity by grouping studies by nybrook Stroke Study: a prospective study of depressive symptoms and
functional outcome. Stroke. 1998;29:618 – 624.
source of case selection. It is particularly noteworthy that a 6. House A, Dennis M, Mogridge L, Warlow C, Hawton K, Jones L. Mood
number of studies excluded patients at the greatest risk for disorders in the year after first stroke. Br J Psychiatry. 1991;158:83–92.
depression, that is those with a history of depression. Al- 7. House A, Dennis M, Hawton K, Warlow C. Methods of identifying mood
disorders in stroke patients: experience in the Oxfordshire Community
though a systematic review with meta-analysis can overcome
Stroke Project. Age Ageing. 1989;18:371–379.
some of the shortcomings of unstructured examination of 8. American Psychiatric Association. Diagnostic and Statistical Manual of
individual studies, the gold standard review would include an Mental Disorders: DSM-III-R. Washington, DC: American Psychiatric
individual patient data analysis with adjustment for potential Association; 1987.
9. American Psychiatric Association. Diagnostic and Statistical Manual of
confounding factors to calculate frequency estimates, but this Mental Disorders: DSM-IV. Washington DC: American Psychiatric
is complex and challenging to undertake. Association; 1994.
There are other problems of study design that limit the 10. Montgomery SA, Asberg M. A new depression scale designed to be
reliability of the comparisons of frequency estimates between sensitive to change. Br J Psychiatry. 1979;134:382–389.
11. Carson AJ, MacHale S, Allen K, Lawrie SM, Dennis M, House A, Sharpe
stroke patients and controls. It is important to note that stroke M. Depression after stroke and lesion location: a systematic review.
patients are generally older, female, and more likely to have Lancet. 2000;356(9224):122–126.
concomitant illnesses and to have experienced some bereave- 12. Sudlow CLM, Warlow CP. Comparing stroke incidence worldwide. What
makes studies comparable? Stroke. 1996;27:550 –558.
ment or other major life event than that of the general 13. Kim JS, Choi S, Kwon SU, Seo YS. Inability to control anger or
population.62 In this respect, controls need to be matched by aggression after stroke. Neurology. 2002;58:1106 –1108.
age and sex, because these factors are associated with 14. Lincoln NB, Gladman JRF, Berman P, Luther A, Challen K. Rehabili-
depression at any age. Yet the selection of controls in 3 of the tation needs of community stroke patients. Disabil Rehabil. 1998;20:
457– 463.
5 case-control studies did not account for such factors. 15. Egger M, Davey Smith G, Altman DG. Systematic Reviews in Health
Although it may appear likely that stroke-associated depres- Care: Meta-Analysis in Context, 2nd Ed. London: BMJ Publishing
sion is no more common than other types of depression in the Group; 2001.
16. Folstein MF, Maiberger R, McHugh PR. Mood disorder as a specific
general population, there have been no direct comparisons
complication of stroke. J Neurol Neurosurg Psychiatry. 1977;40:
between stroke patients and people with other disabling and 1018 –1020.
potentially life-threatening illnesses. 17. Eriksson M, Asplund K, Glader E-L, Norrving B, Stegmayr B, Terent A,
Although we attempted to adhere to the guidelines for Asberg KH, Wester PO; Riks-Stroke Collaboration. Self-reported
depression and use of antidepressants after stroke: a national survey.
reporting meta-analyses of observational studies,61 this re- Stroke. 2004;35:936 –941.
view does have several limitations. First, we did not hand- 18. Anderson CS. The Epidemiology of Stroke-Rrelated Disability: Fre-
search journals and made no attempt to identify unpublished quency, Patterns and Determinants of Care. Doctor of Philosophy: Uni-
studies, raising the possibility that some studies have been versity of Western Australia; 1995.
19. Appelros P, Nydevik I, Seiger A, A. T. High incidence rates of stroke in
missed. Second, only one person extracted most of the data. Orebro, Sweden: further support for regional incidence differences within
Ideally, 2 people should complete data extraction to reduce Scandinavia. Cerebrovasc Dis. 2002;14:161–168.
the possibility of errors when transcribing study results. Even 20. Bamford J, Sandercock P, Dennis M, Warlow C, Jones L, McPherson K,
Vessey M, Fowler G, Molyneux A, Hughes T. A prospective study of
so, all the data were checked for accuracy on multiple
acute cerebrovascular disease in the community: the Oxfordshire Com-
occasions and all analyses were conducted twice. Finally, it is munity Stroke Project 1981–1986. J Neurol Neurosurg Psychiatry. 1988;
possible that some “multiple publications” have been mis- 51:1373–1380.
coded or missed altogether. We have paid particular attention 21. Numminen H, Kotila M, Waltimo O, Aho K, Kaste M. Declining
incidence and mortality rates of stroke in Finland from 1972–1991.
to addressing this source of publication bias, because the lack Stroke. 1996;27:1487–1491.
of cross-referencing of data from some cohorts has only 22. Wade DT, Langton-Hewer R, Skilbeck CE, Bainton D, Burns-Cox C.
served to mislead the research community, specifically in Controlled trail of a home-care service for acute stroke patients. Lancet.
reference to the amount of information available on the 1985;1:323–326.
23. Wolfe CD, Taub NA, Woodrow J, Richardson E, Warburton FG, Burney
development of depression after stroke. PGJ. Does the incidence, severity, or case fatality of stroke vary in
southern England? J Epidemiol Commun Health. 1993;47:139 –143.
Acknowledgments 24. Appelros P, Viitanen M. Prevalence and predictors of depression at one
Maree Hackett holds a Senior Health Research Scholarship, and year in a Swedish population-based cohort with first-ever stroke. J Stroke
Chaturangi Yapa was in receipt of a Faculty of Medical and Health Cerebrovasc Dis. 2004;13:52–57.
25. Wade DT, Legh-Smith J, Hewer RA. Depressed mood after stroke. A
Sciences Summer Studentship, of The University of Auckland. The
community study of its frequency. Br J Psychiatry. 1987;151:200 –205.
funding body had no role in the conduct or reporting of the review.
26. Herrmann M, Bartels C, Schumacher M, Wallesch CW. Poststroke
depression. Is there a pathoanatomic correlate for depression in the
References postacute stage of stroke? Stroke. 1995;26:850 – 856.
1. Burvill PW, Johnson GA, Jamrozik KD, Anderson CS, Stewart-Wynne 27. Desmond DW, Remien RH, Moroney JT, Stern Y, Sano M, Williams JB.
EG, Chakera TM. Anxiety disorders after stroke: results from the Perth Ischemic stroke and depression. J Int Neuropsychol Soc. 2003;9:
Community Stroke Study. Br J Psychiatry. 1995;166:328 –332. 429 – 439.
2. Burvill PW, Johnson GA, Jamrozik KD, Anderson CS, Stewart-Wynne 28. Weimar C, Kurth T, Kraywinkel K, Wagner M, Busse O, Haberl RL,
EG, Chakera TMH. Prevalence of depression after stroke: the Perth Diener HC; German Stroke Data Bank Collaborators. Assessment of
Community Stroke Study. Br J Psychiatry. 1995;166:320 –327. functioning and disability after ischemic stroke. Stroke. 2002;33:
3. House A. Mood disorders after stroke: a review of the evidence. Int J 2053–2059.
Geriatr Psychiatry. 1987;2:211–221. 29. Diamond PT, Holroyd S, Macciocchi SN, Felsenthal G. Prevalence of
4. Johnson GA. Research into psychiatric disorder after stroke: the need for depression and outcome on the geriatric rehabilitation unit. Am J Phys
further studies. Aust N Z J Psychiatry. 1991;25:358 –370. Med Rehab. 1995;74:214 –217.
30. Paolucci S, Antonucci G, Pratesi L, Traballesi M, Grasso MG, Lubich S. 53. Sinyor D, Amato P, Kaloupek DG, Becker R, Goldenberg M,
Poststroke depression and its role in rehabilitation of inpatients. Arch Coopersmith H. Post-stroke depression: relationships to functional
Phys Med Rehab. 1999;80:985–990. impairment, coping strategies, and rehabilitation outcome. Stroke. 1986;
31. Langhorne P, Stott DJ, Robertson L, MacDonald J, Jones L, McAlpine C, 17:1102–1107.
Dick F, Taylor GS, Murray G. Medical complications after stroke: a 54. van de Weg FB, Kuik DJ, Lankhorst GJ. Post-stroke depression and
multicenter study. Stroke. 2000;31:1223–1229. functional outcome: a cohort study investigating the influence of
32. Andersen G, Vestergaard K, Riis JO, Lauritzen L. Incidence of post- depression on functional recovery from stroke. Clin Rehab. 1999;13:
stroke depression during the first year in a large unselected stroke pop- 268 –272.
ulation determined using a valid standardized rating scale. Acta Psychiatr 55. Williams LS, Weingerger M, Harris LE, Biller J. Measuring quality of
Scand. 1994;90:190 –195. life in a way that is meaningful to stroke patients. Neurology. 1999;53:
33. Aben I, Verhey F, Strik J, Lousberg R, Lodder J, Honig A. A comparative 1839 –1943.
study into the one year cumulative incidence of depression after stroke 56. Spalletta G, Guida G, De Angelis D, Caltagirone C. Predictors of cog-
and myocardial infarction. J Neurol Neurosurg Psychiatry. 2003;74: nitive level and depression severity are different in patients with left and
581–585. right hemispheric stroke within the first year of illness. J Neurol. 2002;
34. Hayee MA, Akhtar N, Haque A, Rabbani MG. Depression after stroke- 249:1541–1551.
analysis of 297 stroke patients. Bangladesh Medical Research Council 57. Ng KC, Chan KL, Straughan PT. A study of post-stroke depression in a
Bulletin. 2001;27:96 –102. rehabilitative center. Acta Psych Scand. 1995;92:75–79.
35. Kase CS, Wolf PA, Kelly-Hayes M, Kannel WB, Beiser A, D’Agostino 58. Kotila M, Numminen H, Waltimo O, Kaste M. Depression after stroke:
RB. Intellectual decline after stroke: the Framingham Study. Stroke. results of the FINNSTROKE study. Stroke. 1998;29:368 –372.
1998;29:805– 812. 59. Löfgren B, Gustafson Y, Nyberg L. Psychological well-being 3 years
36. Gresham GE, Kelly-Hayes M, Wolf PA, Beiser AS, Kase CS, D’Agostino after severe stroke. Stroke. 1999;30:567–572.
RB. Survival and functional status 20 or more years after first stroke: the 60. Verdelho A, Henon H, Lebert F, Pasquier F, Leys D. Depressive
Framingham Study. Stroke. 1998;29:793–797. symptoms after stroke and relationship with dementia: a three-year
37. Åström M, Adolfsson R, Asplund K. Major depression in stroke patients: follow-up study. Neurology. 2004;62:905–911.
a 3-year longitudinal study. Stroke. 1993;24:976 –982. 61. Stroup DF, Berlin JA, Morton SC, Olkin I, Williamson GD, Rennie D,
38. Bacher Y, Korner-Bitensky N, Mayo N, Becker R, Coopersmith H. A Meta-analysis of observational studies in epidemiology: a proposal for
longitudinal study of depression among stroke patients participating in a reporting. Meta-analysis of observational studies in epidemiology: a
rehabilitation program. Can J Rehab. 1990;4:27–37. proposal for reporting. Meta-analysis Of Observational Studies in Epide-
39. King RB, Shade-Zeldow Y, Carlson CE, Feldman JL, Philip M. Adap- miology (MOOSE) group. JAMA. 2000;283:2008 –2012.
tation to stroke: a longitudinal study of depressive symptoms, physical 62. Beekman ATF, Copeland JRM, Prince MJ. Review of community prev-
health, and coping process. Topics Stroke Rehab. 2002;9:46 – 66. alence of depression in later life. Br J Psychiatry. 1999;174:307–311.
63. Wilkinson PR, Wolfe CD, Warburton FG, Rudd AG, Howard RS, Ross-
40. Morris PLP, Robinson RG, Raphael B. Prevalence and course of
Russell RW, Beech R. Longer term quality of life and outcome in stroke
depressive disorders in hospitalized stroke patients. Int J Psychiatry Med.
patients: is the Barthel index alone an adequate measure of outcome?
1990;20:349 –364.
Qual Health Care. 1997;6:125–130.
41. Charney DS, Reynolds CF 3rd, Lewis L, Lebowitz BD, Sunderland T,
64. Ebrahim S, Barer D, Nouri F. Affective illness after stroke. Br J Psy-
Alexopoulos GS, Blazer DG, Katz IR, Meyers BS, Arean PA, Borson S,
chiatry. 1987;151:52–56.
Brown C, Bruce ML, Callahan CM, Charlson ME, Conwell Y, Cuthbert
65. Kotila M, Waltimo O, Niemi ML, Laaksonen R, Lempinen M. The profile
BN, Devanand DP, Gibson MJ, Gottlieb GL, Krishnan KR, Laden SK,
of recovery from stroke and factors influencing outcome. Stroke. 1984;
Lyketsos CG, Mulsant BH, Niederehe G, Olin JT, Oslin DW, Pearson J,
15:1039 –1044.
Persky T, Pollock BG, Raetzman S, Reynolds M, Salzman C, Schulz R,
66. Pohjasvaara T, Leppavuori A, Siira I, Vataja R, Kaste M, Erkinjuntti T.
Schwenk TL, Scolnick E, Unutzer J, Weissman MM, Young RC;
Frequency and clinical determinants of poststroke depression. Stroke.
Depression and Bipolar Support Alliance. Depression and Bipolar
1998;29:2311–2317.
Support Alliance consensus statement on the unmet needs in diagnosis
67. Carod-Artal J, Egido JA, Gonzalez JL, Varela de Seijas E. Quality of life
and treatment of mood disorders in late life. Arch Gen Psychiatry. among stroke survivors evaluated 1 year after stroke: experience of a
2003;60:664 – 672. stroke unit. Stroke. 2000;31:2995–3000.
42. Klein DN, Santiago NJ. Dysthymia and chronic depression: introduction, 68. Daily RD, Sinaki M, Richardson JW. The assessment of depressed mood
classification, risk factors, and course. J Clin Psychol. 2003;59:807– 816. following stroke. Arch Phys Med Rehab. 1983;64:519.
43. Reinherz HZ, Paradis AD, Giaconia RM, Stashwick CK, Fitzmaurice G. 69. Gillen R, Tennen H, McKee TE, Gernert-Dott P, Affleck G. Depressive
Childhood and adolescent predictors of major depression in the transition symptoms and history of depression predict rehabilitation efficiency in
to adulthood. Am J Psychiatry. 2003;160:2141–2147. stroke patients. Archives of Physical Medicine and Rehabilitation. 2001;
44. Bayer R, Marji Z, Shunaigat W. Frequency and clinical determinants of 82:1645–1649.
major post stroke depression in Jordan. Qatar Med J. 2001;10:47–50. 70. Gottlieb D, Salagnik I, Kipnis M, Brill S. Post stroke depression, first year
45. Bendsen BB, Bendsen EB, Lauritzen L, Vilmar T, Bech P. Post-stroke post stroke, in middle band patients. Int J Geriatr Psychiatry. 2002;17:
patients in rehabilitation: the relationship between biological impairment 486 – 487.
(CT scanning), physical disability and clinical depression. Eur Psy- 71. Jürgensen F, Meins W, Meier HP. Depression after stroke: prevalance,
chiatry. 1997;12:399 – 404. functional outcome and recovery six month after discharge from a
46. Berg A, Palomaki H, Lehtihalmes M, Lonnqvist J, Kaste M. Poststroke geriatric rehabilitation center. Zeitschrift Gerontol Geriatr. 1999;32:251.
depression in acute phase after stroke. Cerebrovasc Dis. 2001;12:14 –20. 72. Kellermann M, Fekete I, Gesztelyi R, Csiba L, Kollar J, Sikula J,
47. Gainotti G, Azzoni A, Marra C. Frequency, phenomenology and anatom- Bereczki D. Screening for depressive symptoms in the acute phase of
ical— clinical correlates of major post-stroke depression. Br J Psychiatry. stroke. Gen Hospital Psychiatry. 1999;21:116 –121.
1999;175:163–167. 73. Mast BT, MacNeill SE, Lichtenberg PA. Post-stroke and clinically-
48. Hosking SG, Marsh NV, Friedman PJ. Depression at 3 months poststroke defined vascular depression in geriatric rehabilitation patients. Am J
in the elderly: predictors and indicators of prevalence. Aging Neuro- Geriatc Psychiatry. 2004;12:84 –92.
psychol Cognition. 2000;7:205–216. 74. Tang W-k, Ungvari GS, Chiu HFK, Sze K-h, Woo J, Kay R. Psychiatric
49. Hüwel J, Weisner B, Kemmer H, Heyder J. Depression in the acute stage morbidity in first time stroke patients in Hong Kong: A pilot study in a
after the first ischemic stroke. Nervenarzt. 1998;69:330 –334. rehabilitation unit. Aust N Z J Psychiatry. 2002;36:544 –549.
50. Jaracz K, Kozubski W. Quality of life in stroke patients. Acta Neurol 75. Robinson RG, Price TR. Post-stroke depressive disorders: a follow-up
Scand. 2003;107:324 –329. study of 103 patients. Stroke. 1982;13:635– 641.
51. Kauhanen M, Korpelainen JT, Hiltunen P, Brusin E, Mononen H, Maatta 76. Robinson RG, Starr LB, Kubos KL, Price TR. A two-year longitudinal
R, Nieminen P, Sotaniemi KA, Myllyla VV. Poststroke depression cor- study of post-stroke mood disorders: findings during the initial eval-
relates with cognitive impairment and neurological deficits. Stroke. 1999; uation. Stroke. 1983;14:736 –741.
30:1875–1880. 77. Robinson RG, Starr LB, Price TR. A two year longitudinal study of mood
52. Malec JF, Richardson JW, Sinaki M, O’Brien MW. Types of affective disorders following stroke. Prevalence and duration at six months
response to stroke. Arch Phys Med Rehab. 1990;71:279 –284. follow-up. Br J Psychiatry. 1984;144:256 –262.
78. Robinson RG, Bolduc PL, Price TR. Two-year longitudinal study of 83. Kishi Y, Robinson RG, Kosier JT. The validity of observed depression as
poststroke mood disorders: diagnosis and outcome at one and two years. a criteria for mood disorders in patients with acute stroke. J Affect Dis.
Stroke. 1987;18:837– 843. 1996;40:53– 60.
79. Parikh RM, Robinson RG, Lipsey JR, Starkstein SE, Fedoroff JP, 84. Paradiso S, Ohkubo T, Robinson RG. Vegetative and psychological
Price TR. The impact of poststroke depression on recovery in symptoms associated with depressed mood over the first two years after
activities of daily living over a 2-year follow-up. Arch Neurol. 1990; stroke. Int J Psychiatry Med. 1997;27:137–157.
47(7):785–789.
85. Schultz SK, Castillo CS, Kosier JT, Robinson RG. Generalized anxiety
80. Castillo CS, Starkstein SE, Fedoroff JP, Price TR, Robinson RG. Gen-
and depression: assessment over 2 years after stroke. Am J Geriatr
eralized anxiety disorder after stroke. J Nervous Mental Dis 1993;181:
100 –106. Psychiatry. 1997;5:229 –237.
81. Downhill Jr JE, Robinson RG. Longitudinal assessment of depression and 86. Shimoda K, Robinson RG. Effects of anxiety disorder on impairment and
cognitive impairment following stroke. J Nervous Mental Dis. 1994;182: recovery from stroke. J Neuropsych Clin Neurosci. 1998;10:34 – 40.
425– 431. 87. Shimoda K, Robinson RG. The relationship between social impairment
82. Castillo CS, Schultz SK, Robinson RG. Clinical correlates of early-onset and recovery from stroke. Psychiatry. 1998;61:101–111.
and late-onset poststroke generalized anxiety. Am J Psychiatry. 1995;152: 88. Paradiso S, Robinson RG. Gender differences in poststroke depression.
1174 –1179. J Neuropsych Clin Neurosci. 1998;10:41– 47.
Frequency of Depression After Stroke: A Systematic Review of Observational Studies
Maree L. Hackett, Chaturangi Yapa, Varsha Parag and Craig S. Anderson
Stroke. 2005;36:1330-1340; originally published online May 5, 2005;

doi: 10.1161/01.STR.0000165928.19135.35
Stroke is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 2005 American Heart Association, Inc. All rights reserved.
Print ISSN: 0039-2499. Online ISSN: 1524-4628
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://stroke.ahajournals.org/content/36/6/1330
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published
in Stroke can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office.
Once the online version of the published article for which permission is being requested is located, click
Request Permissions in the middle column of the Web page under Services. Further information about this
process is available in the Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:

http://www.lww.com/reprints
Subscriptions: Information about subscribing to Stroke is online at:

http://stroke.ahajournals.org//subscriptions/

Stroke Epidemiology in Australia

Uploaded by

Copyright:

Available Formats

Stroke Epidemiology in Australia

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Stroke Epidemiology in Australia

Uploaded by

Copyright:

Available Formats

Progress Review

Frequency of Depression After Stroke

S ome form of depression is considered to occur in at least

BDI (score 19–29 “moderate”) 11 (8 to 14)

DSM-III-R minor depression 17 (6 to 28)

Observational studies of the proportional

Stroke. 2005;36:1330-1340; originally published online May 5, 2005;

Reprints: Information about reprints can be found online at:

Subscriptions: Information about subscribing to Stroke is online at:

You might also like