Published online before print December 12, 2008, doi: 10.1161/STROKEAHA.108.527705
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
(Stroke. 2009;40:523.)
© 2009 American Heart Association, Inc.
Original Contributions |
Assessment of Depression After Stroke
A Comparison of Different Screening Instruments
From the Department of Neurology (A.B., H.P., M.K.), Helsinki University Central Hospital, Helsinki, Finland; and the Department of Mental Health and Alcohol Research (J.L.), National Public Health Institute, Helsinki, Finland.
Correspondence to Anu Berg, Lic Psych, South Karelian Central Hospital, Valto Käkelän katu 1, FIN-53130 Lappeenranta, Finland. E-mail anu.berg{at}ekshp.fi
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Background and Purpose— Assessing poststroke depression may be complicated by aphasia, other cognitive deficits, and several somatic stroke-related symptoms. We studied the possible differences in performance of some commonly used instruments in screening depression after stroke.
Methods— We compared the Beck Depression Inventory, Hamilton Rating Scale for Depression, Visual Analogue Mood Scale, proxy assessment, and Clinical Global Impression of the nursing and study personnel, together with Diagnostic and Statistical Manual of Mental Disorders, 3rd Edition, Revised diagnosis, in assessing depression after stroke in a follow-up study of 100 patients. The patients were studied at 2 weeks and at 2, 6, 12, and 18 months after stroke.
Results— The feasibility rates of all assessment instruments studied were fairly similar, but the prevalence rates differed according to the assessment instruments, varying from the lowest rates with a Diagnostic and Statistical Manual of Mental Disorders, 3rd Edition, Revised-based diagnosis up to 3-fold with caregiver ratings. The sensitivity and specificity against the Diagnostic and Statistical Manual of Mental Disorders criteria were acceptable with the Clinical Global Impression, Beck Depression Inventory, and Hamilton Rating Scale for Depression, mostly in the range of 0.70 to 1.00. The caregiver ratings were higher than the patient ratings (P<0.001) and correlated with the caregiver’s own Beck Depression Inventory (0.60 to 0.61, P<0.001). The Visual Analogue Mood Scale was not a sensitive instrument (sensitivity, 0.20 to 0.60) and did not correlate with the Beck Depression Inventory during the first year after stroke.
Conclusions— Beck Depression Inventory, Hamilton Rating Scale for Depression, and Clinical Global Impression assessment by professionals, in addition to the Diagnostic and Statistical Manual of Mental Disorders, 3rd Edition, Revised diagnosis, are useful in assessing depression, but none of these instruments clearly stood apart from the others. Proxy ratings should be used with caution, and the use of the Visual Analogue Mood Scale among patients with aphasia and other cognitive impairments cannot be recommended.
Key Words: assessment of depression • depression • stroke
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Poststroke depression affects approximately one third of ischemic stroke survivors1 and the question of how to reliably screen and diagnose depression is crucial. In addition to diagnosis based on psychiatric interview and the Diagnostic and Statistical Manual of Mental Disorders (DSM-III-R2 or other versions), a variety of self-rating mood scales (eg, Beck Depression Inventory [BDI]3 and interviewer-administered scales (eg, Hamilton Rating Scale for Depression [HRSD])4 are used. These measures are not specifically designed for patients with stroke, who typically have a variety of physical and cognitive impairments that may introduce difficulties in using these instruments. First, vegetative symptoms such as fatigue, psychomotor retardation, or insomnia may be related directly to the medical disorder but are also part of the criteria of depression. Second, aphasia (among other cognitive problems) may even make it impossible to assess depression in a normal way.
Several instruments commonly used in the assessment of poststroke depression are described and their measurement properties critically reviewed by Salter et al.5 In validation studies in patients with stroke, the BDI6–8 and the HRSD7,9 proved to be acceptable screening instruments, but their specificity is too low to provide a basis for diagnosis.7,8 Some authors support the use of the DSM criteria despite the nature or origin of the symptoms, and see both the psychological and somatic symptoms as associated with poststroke depression.10–12 Others stress differences in symptom profiles between poststroke and endogenous depression13,14 or suggest that diagnosis should rely more heavily on nonsomatic symptoms15 or consider different symptoms individually according to their diagnostic sensitivity.9 More information on the responsiveness of the measures to change over time and appropriateness of cutoff values are needed. In addition, overall validation studies do not reveal whether there are stroke-related or patient-related factors influencing the sensitivity of the assessment tool. Studies on the validity and reliability of common verbal tools among patients with aphasia are very rare.16 Use of the Visual Analogue Mood Scale (VAMS)17,18 in patients with stroke with impaired language function has only vague support in some studies.18,19 Few studies have used nurses’ or rehabilitation personnel observation and proxy ratings in assessing poststroke depression.6,20–22
We compared several depression assessment instruments in the detection of depression in patients with stroke. We studied the feasibility of using these instruments, their overall accuracy in detection of depression, sensitivity to various symptoms, and the factors related to the differences found. A follow-up study enabled us to examine possible differences occurring between the acute and chronic poststroke periods.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
A total of 100 consecutive patients with their first ischemic stroke, admitted to the Department of Neurology, Helsinki University Central Hospital, were included in the study. Patients >70 years of age and those with known histories of alcohol abuse, dementia, psychosis, current antidepressant treatment, or severe concomitant disease were not included. The mean age of the patients was 55.2 years; 68 of the patients were male. In all, 38 patients had a right hemispheric lesion, 42 a left hemispheric lesion, and 20 a lesion in the brainstem region. Stroke severity was measured with the Scandinavian Stroke Scale (SSS)23 (mean, 43.7; SD, 11.6 at the acute phase) and cognitive functions were assessed with a comprehensive neuropsychological test battery (for details, see Berg et al24), including an Albert test25 and finger tapping followed by aphasia with the Western Aphasia Battery.26 According to the Western Aphasia Battery, 31% of all patients had aphasia (aphasia quotient
93.7). The patient group was described in detail earlier.24,27 The patients were studied at 2 weeks and at 2, 6, 12, and 18 months after stroke. The depression ratings and SSS assessments were repeated at each time point; neuropsychological assessment was done in whole at the acute phase and at 12 months and, in part, at 6 months. A person providing the closest contact with the patient was identified in 98 cases. The present study is part of a medical treatment trial for poststroke depression.27
Depression was diagnosed with the DSM-III-R2 by the neurologist (H.P.). The patients completed a self-rating instrument BDI,3 assisted when needed, and the HRSD4 was used as an observer-rated instrument by the neuropsychologist (A.B.). A single-item VAMS,17,18 a 100-mm vertical line connecting 2 schematic faces, a happy face positioned at the top pole and a sad face at the bottom, with corresponding words, was presented to the patients, who were asked to assess their mood by placing a mark on the line. When dichotomy was needed, a patient was rated as depressive if he or she marked the line closer to the negative end point (
50 mm); otherwise, the VAMS was used as a continuous variable. The neurologist, neuropsychologist, and study nurse (R.L.) rated the patient with a Clinical Global Impression (CGI)28; in further analysis, we used the mean of these CGI ratings. At the acute phase, a nurse in the ward caring for the patient rated the CGI during routine clinical work. At 6 months and at 18 months, the caregivers were asked to assess patients’ moods by completing the BDI. Caregivers’ mood was assessed with the BDI at the acute phase, at 6 months, and at 18 months.
Statistical Methods
We first calculated the response rates of each instrument and compared the percentages of depressive patients using different criteria. The discriminatory power, sensitivity, and specificity of these instruments were calculated with the DSM-III-R as the reference. To assess different cutoff points, receiver operating characteristic curves were obtained, and the area under the curve was calculated. The internal consistencies of the BDI and HRSD were measured with Cronbach’s alpha. One-way analysis of variance was used to compare different symptoms between various patient groups. The sensitivity of individual depressive symptoms and their contribution to the diagnosis of poststroke depression were also assessed. Based on the scores for the individual items of the BDI, a discriminant model was calculated that would predict whether patients would be classified as depressed or nondepressed according to the various assessment methods used, and correlation coefficients with the discriminant function for each of the individual items were obtained. Spearman correlations, multiple linear regression analysis, and logistic regression analysis were also used to model some differences. A significance level of 0.05 was used in all analyses. The data were analyzed with the SPSS statistical package (SPSS, Inc, Chicago, Ill).29
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Response Rates
Of the 100 patients included in the study, 100 continued at 2 months, 96 at 6 months, 93 at 12 months, and 92 at 18 months. Of all patients in the acute phase, approximately 90% could be assessed with the DSM-III-R, BDI, and HRSD (Table 1). The CGI achieved the highest feasibility and the VAMS the lowest. In the follow-up, the rates varied from 92% to 96% with the DSM-III-R, BDI, and HRSD to 99% to 100% with the CGI.
|
Prevalence of Depression
To diagnose depression with the DSM-III-R, at least 5 different symptoms from the list of 9 depression symptoms are needed, at least one of which is either depressed mood or loss of interest or pleasure.2 Several questionnaires and other measures, typically developed to assess the severity of depression, are also used to screen and identify depression. However, the cutoff points are not independent of the sample and sensitivity and specificity needed. The variable rates for prevalence of depression during the follow-up according to the various assessment instruments and with different cutoff points are presented in Table 2. The DSM-III-R-based diagnoses produced the lowest prevalence rates: 6% to 16% of patients had major depression, as reported earlier.27 When a cutoff point of 107,30 in the BDI was used as the criterion, depression was rated in 23% to 29% of patients, and when the same cutoff point was used in the HRSD, the percentages were 10% to 14%. In all, 22% to 27% were at least mildly depressive using the CGI administered by the study personnel. The CGIs of the 3 raters were significantly correlated with each other (Spearman r=0.6 to 0.87, P<0.01) and with the mean CGI (Spearman r=0.83 to 0.95, P<0.01) at every assessment point during the follow-up. The highest rates were by the nurse on the ward and patient’s proxy; approximately half of the patients were rated as depressive either with the CGI (at least mildly by the nurse) or BDI (at least 10 points in proxy assessment).
|
Discriminatory Power of the Methods With DSM-III-R as the Criterion
The classification ability (the number of correct classification in percent), sensitivity, specificity, and area under the curve of each assessment method are presented in Table 3
. The CGI ratings of the study personnel had a sensitivity of 0.80 and specificity of 0.79 at the acute phase and even higher (0.82 to 1.00) during the follow-up. Although the BDI (with a cutoff point of 10) appeared to be more sensitive (0.71 to 1.00) than the HRSD, the HRSD (with a cutoff point of 10) showed higher specificity (0.92 to 0.94). The BDI showed highest sensitivity and specificity at 12 months, missing no patients with a major depression diagnosis. The HRSD showed highest sensitivity and specificity at 2 months, after which the sensitivity became poor.
|
|
The sensitivity and specificity of caregiver assessment were poor. The nurse on the ward was able to detect 80% of depressive patients, but the specificity was low. The receiver operating characteristic of the VAMS was not analyzed in the acute phase because the VAMS was recorded only by 42 patients for technical reasons. The VAMS proved not to be a satisfactory measure of depression, and the area under the curve of the VAMS was significant only at 18 months.
Sensitivity of the Symptoms
The internal consistency of the BDI and HRSD was good at every time point with Cronbach alpha values of 0.82 to 0.86 and 0.70 to 0.84, respectively. The BDI and HRSD correlated significantly (Spearman r=0.63 to 0.71, P<0.001) during the follow-up.
When the BDI was divided into cognitive–affective items (1 to 14) and somatic items (15 to 21),31 both subscales correlated with the BDI (Spearman r=0.78 to 0.91 and 0.75 to 0.89, respectively; P<0.001). The internal consistency remained favorable in the affective subscale (0.85 to 0.91), but not in the somatic subscale (0.37 to 0.56). The somatic symptoms were more common among patients older than 55 years than in younger patients at the acute phase (4.9 versus 3.0, P<0.01) and at 2 months (4.2 versus 2.9, P<0.01). The scores in the cognitive–affective and somatic subscales tended to be slightly higher among patients with lower SSS than among those with higher SSS, but none of these differences was significant. The scores in these subscales were also similar in patients with 3 location groups (left hemisphere, right hemisphere, brainstem). At 18 months, men had higher scores than women both in the cognitive–affective and somatic subscales.
The sensitivity of individual depressive symptoms in the BDI was assessed with discriminant models with major depression (DSM-III-R), BDI (cutoff point of 10), CGI (at least mild), nurse CGI, caregiver BDI, and VAMS (cutoff point of 50 mm) as classification criteria. Models with DSM, BDI, CGI, and nurse CGI were significant, and the results are presented in Table 4; the models were not significant with the VAMS and caregiver BDI. At the acute phase, discouraged about the future, feeling like a failure, feeling guilty, and looking unattractive were items that were important discriminators in major depression, BDI, and CGI. At 18 months, the best discriminators were sadness, dissatisfaction, discouraged about the future, feeling disappointed, loss of interest in people, and difficulty with decisions.
|
Caregiver Ratings
The caregivers rated patient depressive symptoms with the BDI approximately 4 points higher than the patients themselves (11.2 [SD, 8.5] versus 6.6 [SD, 5.5] at 6 months; 10.8 [SD, 8.0] versus 6.9 [SD, 6.6] at 18 months; P<0.001). At 6 months, disagreement was significantly higher among patients with more severe stroke (SSS at the acute phase <48; 6.7 [SD, 6.9] versus 3.0 [SD, 8.6], P<0.05), but not significantly among nonspouse caregivers than among spouses (6.8 [SD, 8.8] versus 3.8 [SD, 7.7], P=0.16); neither were there differences in disagreement between male and female patients or patients with different lesion locations. In addition, there was a significant correlation between caregivers’ ratings of patients and caregivers’ own BDI (0.60 to 0.61, P<0.001), which was even higher than that between caregivers’ ratings of patients and patient ratings (0.37, P<0.005 to 0.43, P<0.001). With linear regression analysis at 6 months (F[3,75]=22.83, P<0.001, adjusted r2=0.46), we found that higher levels of disagreement were independently associated with higher values for the caregiver’s own BDI score (beta 0.60, P<0.001) and lower values for the patient’s own BDI score (beta –0.41, P<0.001) when the effect of the SSS was not significant (beta –0.17, P=0.05). The results were similar at 18 months (F[3,70]=17.84, P<0.001, adjusted r2=0.41; caregiver BDI beta 0.58, P<0.001; patient BDI beta –0.51, P<0.001; SSS beta 0.02, not significant).
We further divided patients with their caregivers into 2 groups and compared the patient and caregiver variables between these groups with t tests. In the first group, the patients were assessed as depressed by caregivers (BDI 10) but not by researchers (mean CGI no more than borderline), whereas in the other group, the researchers, but not the caregivers, rated the patients as depressed. In the first group, the caregivers had higher BDIs than in the latter group both at 6 months (12.1 versus 3.1, P<0.001) and at 18 months (10.1 versus 3.4). This could be confirmed with logistic regression analysis; models with patient BDI, caregiver BDI, and SSS as covariates were significant (
2[3]=16.2, P<0.001, Nagelkerke R2=0.52 at 6 months; 2[3]=9.3, P<0.05, R2=0.48 at 18 months), and caregiver BDI was the only independent significant predictor (beta=–0.28, P<0.01, OR, 0.76 at 6 months; beta –0.35, P=0.05, OR, 0.70 at 18 months).
Visual Analogue Mood Scale
We further studied the associations between the VAMS and BDI among all patients and among subgroups with or without aphasia, inattention disorder, impaired hand motor function, or overall stroke severity to determine whether there were different types of associations in the various subgroups. The VAMS did not correlate significantly with the BDI until at 18 months, when the correlation was significant (0.52, P<0.001). Among patients with aphasia (Western Aphasia Battery aphasia quotient <93.8), the VAMS was not significantly correlated with the BDI during the follow-up; among patients without aphasia, the correlation was significant at 18 months (0.60, P<0.001). When the Albert test score was <40, indicating neglect or other inattention disorder, the correlations between the VAMS and the BDI were also nonsignificant during the follow-up. Among patients with normal test results, significant correlations were attained at the acute phase (0.45, P<0.01), at 6 months (0.35, P<0.05), and at 18 months (0.49, P<0.001). Based on left and right finger-tapping scores, indicating motor disorder, the correlation between the VAMS and BDI was significant more often in the poorer group (<30). The correlation was significant among patients with poor right hand tapping at the acute phase (0.75, P<0.01) and at 18 months (0.57, P<0.05) and among patients with poor left hand tapping at 6 months (0.51, P<0.05) and at 18 months (0.50, P<0.01). Among the better groups, the correlation was not significant until at 18 months (right 0.50, left 0.53; P<0.001). Differences of a similar type were found when the patients were divided based on the SSS, indicating overall stroke severity. In the poorer group (SSS <44), a significant correlation was found at the acute phase (0.60, P<0.05), at 12 months (0.47, P<0.05), and at 18 months (0.47, P<0.05), whereas in the better group, the correlation was not significant until 18 months (0.54, P<0.001).
|
Discussion |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Our study provides longitudinal data on several depression assessment instruments among patients with stroke to an extent not reported earlier. The studies of House et al,6 Paradiso et al,12 and Laska et al16 are the few follow-up studies dealing with this question, whereas studies on proxy assessment in poststroke depression are largely unavailable. In addition to overall sensitivity and specificity of the instruments studied, we also provided data on the sensitivity of different symptoms during the follow-up. The size of our patient group, however, limited the use of very refined analyses with different subgroups. Another possible limitation is that this investigation was part of a treatment study. We did not include patients older than 70 years and patients with severe health problems that might interfere with the medication. This may have reduced the symptoms of depression, and so the results may be more conservative. In addition to the advantages of the very favorable follow-up rate and broadly examined patients in a study like this, by including only patients without other severe health problems, we may be better able to study the impact of stroke without a very large sample size. With the inclusion criteria described earlier, the patients are, however, consecutive patients with their first ever ischemic stroke. Another question concerns a learning effect that cannot be avoided in follow-up studies. The feasibility of the measures may possibly increase in the course of study, due to learning effect, but we believe it to be of minor importance compared with cognitive recovery and the demands of the measures. In addition, the intervals between follow-up studies are quite long, from 2 to 6 months, and the same for every instrument.
There was little variation in feasibility of the different measures of depression. The VAMS proved not to be much easier for patients with aphasia to complete than conventional verbal measures; the predominant cause of not being able to complete the measures was aphasia. Only the CGI stood out from the other measures in this aspect. Wider differences were found in the prevalence rates for depression using different instruments. Although 6% to 16% of patients had major depression when the DSM-III-R criteria, similar to those of the DSM-IV-TR, were used, caregivers assessed approximately half of the patients as depressive. The sensitivity and specificity of different rating scales varied during the follow-up. Few studies have reported the sensitivity and specificity of the rating scales used in assessing poststroke depression; we found only the study of House et al,6 which reported the results of a follow-up of up to 12 months. The sensitivity of the BDI with cutoff point 9 out of 10 found in our study is comparable to that of House et al6 (0.71 to 1.00; 0.70 to 0.85) and similar to that of Aben et al7 at the acute phase (0.80 in both studies). Although in our study there was some fall in sensitivity during rehabilitation, both the reliability and sensitivity of the BDI were better in late follow-up than in the acute phase. In our study, lower cutoff points would be needed at 6 months and in the study of House et al6 at 12 months. The results of Lincoln et al8 are not comparable; they did not report the time after stroke and included a sample with opportunity cases and patients recruited in behavioral therapy. For a screening instrument, high sensitivity is more important than high specificity so that it does not miss true cases. In line with previous investigators,6–8 we can state that the BDI is an acceptable screening instrument but is not specific enough to be used as a diagnostic tool. In contrast to Andersen et al,32 we cannot recommend higher cutoff points with patients with stroke.
In our study, the sensitivity of the HRSD was lower than that of the BDI, but it showed higher specificity. Unfortunately, there are no previous follow-up studies with the HRSD to compare the results after the acute phase.
Both affective and somatic items were associated with depression in our study. However, when the BDI was divided into cognitive–affective and somatic subscales, the somatic subscale could not retain good internal consistency; somatic symptoms may be based on different causes. Of the BDI items, "discouraged about the future" was among the best discriminators both at the acute phase and at 18 months and for major depression, BDI, and CGI. In addition, "feeling like a failure," "feeling guilty," and "looking unattractive" were also important discriminators at the acute phase for all 3 instruments and "sadness," "dissatisfaction," "feeling disappointed," "loss of interest in people," and "difficulty with decisions" at 18 months. Weight loss was not associated with depression, which is in line with the result of Paradiso et al.12 Crying was not among the items best discriminating depression when major depression and BDI were used as criteria, whereas with the CGI assessment at the acute phase, it was the best discriminator. The symptoms that characterize depression also appeared to change between the subacute and chronic poststroke periods in the study of Paradiso et al.12 The ranking of depressive items by DeCoster et al,9 in which depressed mood showed the highest sensitivity followed by loss of appetite, is thus difficult to compare with our results; the researchers used scores from different time points according to the first diagnosis of depression. Stein et al15 studied whether the frequency of depressive symptoms differed as a function of laterality, but they found the pattern of depressive symptomatology very similar in both hemispheric groups. The distribution for each symptom in the HRSD was similar in right-sided lesions and left-sided lesions also in the study of Andersen et al.32
The VAMS was less successful than the BDI. If a patient’s mood cannot be assessed with the BDI or DSM ratings, it likewise cannot be assessed with the VAMS, as was also found by House et al.6 In addition, the VAMS scores appeared not to be satisfactorily comparable with other measures during the first year after stroke, which is in line with House et al6 and with Tang et al.33 In our study, the correlation between the VAMS and other measures of depression appeared for the first time at 18 months among all patients and not at all among patients with aphasia or inattention disorder. The more positive conclusions of Stern et al19 were based on normal volunteers and those of Arruda et al18 on a very small sample of patients with no severe aphasia. Price et al34 also concluded that many patients after a stroke are unable to successfully complete the Visual Analogue Scales of any format. Based on our data, the VAMS seems not to be a reliable way to assess depression after stroke among patients with aphasia or other cognitive impairments.
Caregiver ratings with the BDI were neither sensitive nor specific against the DSM criteria. Our results differ from those of House et al6; in their study, caregivers reported all cases with major depression. They used a special caregiver’s depression rating scale; however, only 49% of identified caregivers completed the rating scale. In our study, caregivers rated patients as more depressive than did the patients themselves, although there was some correlation between the caregiver ratings and patient ratings. This result is in line with studies on the quality of life among patients with stroke,35,36 in which proxies tended to score the patients as more severely affected than the patients scored themselves. In our study, this disagreement 6 and 18 months after stroke was higher when the caregivers themselves had high BDI scores and patients low BDI scores. In the study of Williams et al,35 caregiver depression 1 to 2 months after stroke was associated with proxy health-related quality of life ratings, but not with agreement between the patient and proxy health-related quality of life 1 to 2 months after stroke. We also found disagreement between the caregiver ratings with BDI and CGI ratings of the study personnel, a discrepancy associated with caregiver depression. The caregiver’s own mood may be important when he or she assesses the mood of a patient with whom he or she is in a close relationship. Laska et al16 found in their recent study that the assistance of relatives and staff increases feasibility and decreases validity. The use of informants, however, is the most common adaptive method used in assessing patients with aphasia.37 We conclude that caregivers may be valuable sources of information, but their assessment should be used with caution.
We did not study the interrater reliability of the nurses’ observational formula, and we can see these results only as preliminary. The problem in rating by nurse with no specific training was poor specificity; when half of the patients were rated as depressive, there were many false-positive ratings. In the study of House et al,6 the sensitivity was also fairly poor, whereas in our study, it was 80%. Trained study personnel with special interests in depression were better in assessment accuracy. Special training of nurses and instruction on eliciting symptoms would be needed.
In conclusion, self-rating instruments such as the BDI are useful in screening depression in patients with stroke. The BDI needs no psychiatric professionals or specially trained personnel and is satisfactory in its sensitivity. The CGI of the professionals provided a fairly satisfactory screening instrument for further assessment. Nurses identified most patients with depression, but their assessment was poor in specificity. More research on observational and CGI-based assessment is needed instead of relying VAMS and proxy ratings. The VAMS was not more often ratable than other assessment methods and it showed insufficient sensitivity and/or specificity during the first year after stroke and especially among patients with aphasia and cognitive impairment. Proxies often rate patients’ depression as more severe than the patients themselves, and their ratings seem to be influenced by their own depression. Various assessment methods may be based on somewhat different symptoms, which may also be an important point to consider, because the symptoms characterizing depression may change during the transition from the acute phase to a chronic poststroke period.
|
Acknowledgments |
|---|
We thank Riitta Lönnqvist for her valuable work with patients in data collection, Matti Lehtihalmes for aphasia ratings, Jari Lipsanen for statistical assistance, and the Orion Corporation, Orion Pharmaceutica for its support for data collection.
Source of Funding
The study was supported by the Finnish Cultural Foundation.
Disclosure
None.
Received June 3, 2008; accepted June 24, 2008.
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
1. Hackett ML, Yapa C, Parag V, Anderson CS. Frequency of depression after stroke: a systematic review of observational studies. Stroke. 2005; 36: 1330–1340.
2. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, Third Edition, Revised. Washington, DC: APA; 1987.
3. Beck AT, Ward CH, Mendelson M, Mock J, Erbaugh J. An inventory for measuring depression. Arch Gen Psychiatry. 1961; 4: 561–571.
4. Hamilton M. A rating scale for depression. J Neurol Neurosurg Psychiatry. 1960; 23: 56–62.
5. Salter K, Bhogal SK, Foley N, Jutai J, Teasell R. The assessment of poststroke depression. Top Stroke Rehabil. 2007; 14: 1–24.[Medline] [Order article via Infotrieve]
6. House A, Dennis M, Hawton K, Warlow C. Methods of identifying mood disorders in stroke patients: experience in the Oxfordshire Community Stroke Project. Age Ageing. 1989; 18: 371–379.
7. Aben I, Verhey F, Lousberg R, Lodder 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: 386–393.[CrossRef][Medline] [Order article via Infotrieve]
8. Lincoln NB, Nicholl CR, Flannaghan T, Leonard M, Van der Gucht E. The validity of questionnaire measures for assessing depression after stroke. Clin Rehabil. 2003; 17: 840–846.
9. DeCoster L, Leentjens AFG, Lodder J, Verhey FRJ. The sensitivity of somatic symptoms in post-stroke depression: a discriminant analytic approach. Int J Geriatr Psychiatry. 2005; 20: 358–362.[CrossRef][Medline] [Order article via Infotrieve]
10. Spalletta G, Ripa A, Caltagirone C. Symptom profile of DSM-IV major and minor depressive disorders in first-ever stroke patients. Am J Geriatr Psychiatry. 2005; 13: 108–115.[CrossRef][Medline] [Order article via Infotrieve]
11. Fedoroff JP, Strakstein SE, Parikh RM, Price TR, Robinson RG. Are depressive symptoms nonspecific in patients with acute stroke? Am J Psychiatry. 1991; 148: 1172–1176.
12. Paradiso S, Ohkubo T, Robinson RG. Vegetative and psychological symptoms associated with depressed mood over the first two years after stroke. Int J Psychiatry Med. 1997; 27: 137–157.[Medline] [Order article via Infotrieve]
13. Gainotti G, Azzoni A, Razzano C, Lanzillotta M, Marra C, Gasparini F. The Post-Stroke Depression Rating Scale: a test specifically devised to investigate affective disorders of stroke patients. J Clin Exp Neuropsychol. 1997; 19: 340–356.[Medline] [Order article via Infotrieve]
14. Gainotti G, Azzoni A, Marra C. Frequency, phenomenology and anatomical–clinical correlates of major post-stroke depression. Br J Psychiatry. 1999; 175: 163–167.
15. Stein PN, Sliwinski MJ, Gordon WA, Hibbard MR. Discriminative properties of somatic and nonsomatic symptoms for post stroke depression. Clin Neuropsychol. 1996; 10: 141–148.[CrossRef]
16. Laska AC, Mårtensson B, Kahan T, vonArbin M, Murray V. Recognition of depression in aphasic stroke patients. Cerebrovasc Dis. 2007; 24: 74–79.[CrossRef][Medline] [Order article via Infotrieve]
17. Stern RA, Bachman DL. Depressive symptoms following stroke. Am J Psychiatry. 1991; 148: 351–356.
18. Arruda JE, Stern RA, Somerville JA. Measurement of mood states in stroke patients: validation of the Visual Analog Mood Scales. Arch Phys Med Rehabil. 1999; 80: 676–680.[CrossRef][Medline] [Order article via Infotrieve]
19. Stern RA, Arruda JE, Hooper CR. Wolfner GD, Morey CE. Visual analogue mood scales to measure internal mood state in neurologically impaired patients: description and initial validity evidence. Aphasiology. 1997; 11: 59–71.[CrossRef]
20. 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: 428–433.
21. Benaim C, Cailly B, Perennou D, Pelissier J. Validation of the Aphasic Depression Rating Scale. Stroke. 2004; 35: 1692–1696.
22. Bennett HE, Thomas SA, Austen R, Morris AMS, Lincoln NB. Validation of screening measures for assessing mood in stroke patients. Br J Clin Psychol. 2006; 45: 367–376.[CrossRef][Medline] [Order article via Infotrieve]
23. Scandinavian Stroke Study Group. Multicenter trial of hemodilution in ischemic stroke: background study and study protocol. Stroke. 1985; 16: 885–890.
24. Berg A, Palomäki H, Lehtihalmes M, Lönnqvist J, Kaste M. Poststroke depression. An 18-month follow-up. Stroke. 2003; 34: 138–143.
25. Albert ML. A simple test of visual neglect. Neurology. 1973; 23: 658–664.
26. Kertesz A. The Western Aphasia Battery. New York: Grune & Stratton; 1982.
27. Palomäki H, Kaste M, Berg A, Lönnqvist R, Lönnqvist J, Lehtihalmes M, Hares J. Prevention of poststroke depression: 1 year randomised placebo controlled double blind trial of mianserin with 6 month follow up after therapy. J Neurol Neurosurg Psychiatry. 1999; 66: 490–494.
28. Guy W. Clinical Global Impressions. In: New Clinical Drug Evaluation Unit (ECDEU) Assessment Manual for Psychopharmacology. Rockville, MD: National Institute of Mental Health, Psychopharmacology Research Brand, Division of Extramural Research Programs; 1976: 218–222.
29. SPSS Inc. SPSS Advanced Models 10.0. Englewood Cliffs, NJ: Prentice Hall; 2000.
30. Kendall P, Hollon SD, Beck AT, Hammen CL, Ingram RE. Issues and recommendations regarding use of the Beck Depression Inventory. Cogn Ther Res. 1987; 11: 289–299.[CrossRef]
31. Cavanaugh S, Clark DC, Gibbons RD. Diagnosing depression in the hospitalised medically ill. Psychosomatics. 1983; 24: 809–815.[Medline] [Order article via Infotrieve]
32. Andersen G, Vestergaard K, Riis JØ, Lauritzen L. Incidence of post-stroke depression during the first year in a large unselected stroke population determined using a valid standardized rating scale. Acta Psychiatr Scand. 1994; 90: 190–195.[Medline] [Order article via Infotrieve]
33. Tang WK, Ungvari GS, Chiu HFK, Chiu HF, Sze KH. Detecting depression in Chinese stroke patients: a pilot study comparing four screening instruments. Int J Psychiatry Med. 2004; 34: 155–163.[CrossRef][Medline] [Order article via Infotrieve]
34. Price CIM, Curless RH, Rodgers H. Can stroke patients use visual analogue scales? Stroke. 1999; 30: 1357–1361.
35. Williams LS, Bakas T, Brizendine E, Plue L, Tu W, Hendrie H, Kroenke K. How valid are family proxy assessments of stroke patients’ health-related quality of life? Stroke. 2006; 37: 2081–2085.
36. Hilari K, Owen S, Farrelly SJ. Proxy and self report agreement on the Stroke and Aphasia Quality of Life Scale–39. J Neurol Neurosurg Psychiatry. 2007; 78: 1072–1075.
37. Townend E, Brady M, McLaughlan K. A systematic evaluation of the adaptation of depression diagnostic methods for stroke survivors who have aphasia. Stroke. 2007; 38: 3076–3083.
This article has been cited by other articles:
 |
|
 |
|
  L. De Wit, M. Molas, E. Dejaeger, W. De Weerdt, H. Feys, W. Jenni, N. Lincoln, K. Putman, W. Schupp, and E. Lesaffre The Use of a Biplot in Studying Outcomes After Stroke Neurorehabil Neural Repair, October 1, 2009; 23(8): 825 - 830. [Abstract] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||