A Systematic Evaluation of the Adaptation of Depression Diagnostic Methods for Stroke Survivors Who Have Aphasia
Background and Purpose— One in 3 stroke survivors has aphasia (impaired language comprehension and expressive abilities). Conventionally, depression diagnosis uses language-based methods. We aimed to systematically review methods that have been used to diagnose depression and adaptations to these methods intended for people with aphasia.
Methods— We systematically reviewed stroke studies (to January 2006) that included a depression diagnosis and individuals with aphasia. We extracted data related to depression diagnostic methods used for individuals with/without aphasia. We sought clarification from authors when required.
Results— A total of 60 studies included people with aphasia. Almost half the studies (29/60; 48%) adapted their main depression diagnostic method (most typically a clinical interview and published criteria) for individuals with aphasia. Adaptive methods included: using informants (relatives or staff), clinical observation, modifying questions and visual analogue scales. Evidence of the validity or reliability of these adaptations was rarely reported. However, use of informants or clinical observation did achieve the inclusion of most people with aphasia in the diagnosis of depression. Remaining studies, that did not report adaptive methods, suggested that conventional language-based methods are suitable for individuals with only ‘mild’ aphasia.
Conclusions— People with aphasia can be and have been included in depression diagnostic assessments. However, we suggest that depression and language experts collaborate to develop a more valid method of depression diagnosis for patients with aphasia that has good reliability.
Aphasia affects ≈20%1 to 38%2 of stroke survivors. People with aphasia can have problems with understanding and producing spoken language (eg, conversations), reading, writing and numerical skills. Accordingly, aphasia may negatively influence their personal relationships, employment and social re-engagement.3,4 Aphasia is typically associated with left hemisphere damage.5 Compared with people with right hemisphere damage, individuals with left hemisphere damage may be more likely to retain emotional awareness6,7 and demonstrate observable extreme emotional reactions.6 Therefore, studies of emotion conducted on participants with mainly right hemisphere damage cannot be generalized to individuals with aphasia. The prevalence and associated bio-psycho-social characteristics of aphasia highlight the importance of being able to assess and study depression in this population.
Depression is an important complication poststroke that affects patients’ well-being, recovery and survival.8,9 Clinical guidelines recommend routine screening of all patients followed by diagnostic assessment of all those who have been identified as possibly depressed by clinicians with specialist training.10,11 A dependence on language-based assessment (eg, questionnaire or verbal interview) to either screen or diagnose depression is clearly going to present barriers for people with aphasia. Much recent research has aimed to develop nonlanguage-based depression screening tools,12–16 but has been limited by a failure to actually include patients with aphasia (whose emotional presentation may differ qualitatively from samples of patients with mainly right hemisphere damage).6,7 This may be due to the absence of an agreed ‘gold standard’ method for depression diagnosis in aphasia to test screening tools against. It is therefore very important to both clinical practice and research to understand how communication difficulties may be overcome to enable depression diagnosis in aphasia.
We aimed to systematically search, identify, appraise, synthesize and present the methods that have been used to diagnose depression in individuals with aphasia after stroke.
Criteria for Selecting Studies for This Review
We included studies that diagnosed depression after stroke in individuals with aphasia. We excluded studies published in languages other than English but did not exclude studies on the basis of sample size, study design or recruitment setting. Our definition of ‘depression diagnosis’ was the categorical division of participants into ‘depressed’ and ‘not depressed’, by consideration of prespecified symptoms including low mood. We classified as ‘aphasia’ any description of aphasia, dysphasia, language impairment or related synonyms (such as language comprehension or expressive problems) but not descriptions of cognitive impairment, dementia or decreased consciousness. We classified individuals with aphasia as having participated if this was an explicitly reported inclusion criteria, or if specific results for participants with aphasia were reported. We did not assume that individuals with aphasia had participated based on descriptions of qualified exclusion criteria such as ‘significant’ aphasia.
Search Strategy, Data Extraction and Synthesis
We modified a previously published Cochrane Stroke Group search strategy17 by using those terms and synonyms relating to ‘stroke’ and ‘depression’ but not ‘clinical trials’. MEDLINE, CINAHL and PsychINFO were searched from inception up to January 2006 and we also scanned reference sections of obtained publications.
Titles and abstracts were screened and rescreened to identify studies that diagnosed depression after stroke. Reviewers (E.T. and K.M.) then independently extracted data to determine the aphasia screening, exclusion and inclusion criteria that had been used. Studies that had both diagnosed depression after stroke and had included people with aphasia were then targeted for closer examination.
We identified whether studies had administered depression diagnostic assessments to all available individuals with aphasia (‘unlimited’) or only to a subgroup of available individuals with aphasia (‘limited’) on the basis of lesser severity or types (as defined by study authors). We took into account whether depression diagnostic assessments had been attempted with all available patients with aphasia during the recruitment stage of studies as well as whether they were finally included in studies.
Details of sample source and size, participant characteristics, time since stroke at first full depression diagnosis, the main depression diagnostic method used and adaptations to this method for aphasia were extracted. Main depression diagnostic methods were identified as were any adaptive methods used for people with aphasia. We noted tools used to assess depressive symptoms (interviews based on published syndromal criteria, interviews based on observer-rating scales, or questionnaires) and diagnostic criteria used (eg, Diagnostic and Statistical Manual of Mental Disorders [DSM]18 or a cut-off point on a scale). Information regarding success or failure in completing both main and adaptive depression diagnostic methods in participants with aphasia was extracted, as were comments and evidence related to the validity and reliability of these methods. We contacted authors to clarify any ambiguities and collated the different types of main depression diagnostic methods and adaptive methods for participants with aphasia using simple content analysis.19
Sixty studies, involving a total of 8242 participants (range 18 to 1074), met our inclusion criteria and were reported across 138 publications (for further details, see supplemental Figure I, available online at http://stroke.ahajournals.org). Of the 60 studies, 3 sampled from the community, 24 from acute hospitals and 33 from rehabilitation hospitals or other sources such as outpatient clinics. Time elapsed since stroke ranged from less than a week to 5 years across studies (for details of individual studies see supplemental Table I⇓⇓⇓, available online at http://stroke.ahajournals.org).
Characteristics of Participants With Aphasia Included in Depression Diagnostic Assessments
Only 37 of the 60 studies specified the number of participants with aphasia (n=829, range=5 to 60, median proportion=22% with a range of 1% to 100%). Altogether we classified 22 (22/60; 37%) studies as having conducted depression diagnostic assessments in individuals with ‘unlimited’ aphasia and 38 (38/60; 63%) studies as having done so in individuals with ‘limited’ aphasia.
Main Depression Diagnostic Methods
The majority of studies conducted clinical interviews (48/60; 80%) or a clinical interview and questionnaire (6/60; 10%) whereas a minority used only a questionnaire (6/60; 10%). Of those conducting interviews alone, all but 9 used an observer-rating scale or structured interview schedule to structure the interview (see Table 1). Once information on depressive symptoms was obtained, the diagnosis of depression in most studies (43/60; 72%) was based on published syndromal criteria (usually a version of the DSM), whereas other studies used a cut-off point on an observer-rating scale or questionnaire.
The main depression methods used in 3 studies20–22 (Cornell Depression Scale,20,21 Structured Assessment of Depression in Brain Damage22) were specifically chosen for their suitability for participants with aphasia, so these were counted as both main and adaptive methods.
Reporting of Adaptive Depression Diagnostic Methods for Individuals With Aphasia
Overall just less than half of studies (29/60; 48%) indicated having adapted their depression diagnostic method for participants with aphasia. Studies that involved only individuals whose aphasia was ‘limited’ (11/38; 29%) were far less likely to report an adaptive method than studies that involved individuals with ‘unlimited’ aphasia (18/22; 82%).
Types of Adaptive Methods
The adaptive methods reported by the 29 studies included using informants, clinical observation, modifying questions or responses required, modifying the timing of interviews or using visual analogue scales (see Table 2).
Informants to Supplement Interview Findings
The most common adaptation used informants to supplement information obtained from patients (18 studies: 12 with ‘unlimited’ and 6 with ‘limited’ aphasia). Whereas most supplemented information obtained from patient interviews, 2 studies used informants in a more extensive and structured manner.22,23 The SADBD interview schedule22 could become entirely informant-based if a patient was deemed to lack sufficient awareness in all domains of depressive symptoms (physical, cognitive, affective) in a preliminary assessment. In this case the use of informants was sometimes ‘alternative’ rather than ‘supplementary’. Nurses in the Robinson and Szetela study23 completed a structured Nurses Rating Scale for Depression (NRSD) that covered a wide range of depressive symptoms with a focus on nonverbal ones. This formed 1 of 4 assessments that contributed equally to the diagnosis of depression.
Some studies specified whether they used staff (7 studies), relatives/friends (4 studies), or both (3 studies) as the informant(s). The manner of engaging with informants and the nature of the information obtained was rarely described. However, 4 studies specified gauging patient’s mood and motivation from staff reports on day-to-day behavior,24–27 whereas Finklestein et al specified using nurses’ observations to assess nonverbal depressive symptoms (appetite and sleep).28
One study5 stated that using relatives and staff to supplement clinical interviews had been established as a ‘reliable’ method; however, although the supporting references21,29 did use informants they did not report reliability data. Robinson et al30 state that their NRSD had established reliability and validity, but the supporting reference is a duplicate of the same study. In this publication23 they report correlations of the NRSD with the Hamilton Depression Rating Scale (HDRS), Zung depression Scale (SDS) and Folstein’s Visual Analogue Mood Scale (VAMS) of r=0.55, r=0.79 and r=0.60 respectively, but it is unclear how the HDRS itself was completed.
A noncompletion rate of 19 from at least 252 attempts was achieved (based on 12 studies that reported numbers of individuals with aphasia that did not complete5,20,21,23,24,27–29,31–34 and 10 studies that reported the number that did.5,20,21,23,24,28,29,31–33
Diagnosticians, most of whom were experienced psychiatrists, used what was described as observations of ‘nonverbal’ behavior (presumably body language, delayed time to respond and other markers of emotion) to infer emotional states in 6 studies with ‘unlimited’ aphasia. In 3 studies, observations of such ‘nonverbal behavior’ was used to provide information to supplement that which had been obtained verbally from interviews. Hermann and colleagues20,21 combined patients and informants reports with his observations using the CDS. Finkelstein et al28 combined patients and informants reports assessed with a modified HDRS and 3 ‘experienced observers’ observations and specified that “Ratings of aphasic patients (N=8) were based heavily on evidence of weeping, anger, agitation, or retardation during the interview.”28
In 3 studies, psychiatrist observations were used as an alternative method after noncompletion of interviews with the HDRS. The psychiatrist in Benaim et al’s study35 observed patients generally during their rehabilitation then rated them globally for depressive severity (between 0 to 100) and finally made a categorical judgment as to whether they were depressed. Palomaki et al36 judged whether patients were depressed using the Clinical Global Index that requires the clinician to be an experienced diagnostician (and is generally used to assess improvement after treatment). Andersen et al37 report making clinical observation of whether patients had depression or uncontrollable crying.
Hermann et al21 reported that scores on the CDS were well matched to categorical Research Diagnostic Criteria diagnoses of not depressed, minor depression, probable and definite major depression. However, RDC-based diagnoses appear to have been extracted from the information obtained with the CDS.
All 6 studies using psychiatrist observation reported numbers of individuals with aphasia assessed and only 1 of these36 reported not being able to assess everybody. Overall, a low noncompletion rate of only 4 of 169 individuals was achieved with this method.
Modified Questions, or Responses Required
Four studies (1 with ‘unlimited’ and 3 with ‘limited’ aphasia) reported that interviews were supplemented with the use of simple questions designed to require only ‘Yes/No’ or categorical responses.22,26,38,39 Reding et al26 also indicated using participants’ gestures as responses. A noncompletion rate of 14 from at least 46 attempts was achieved (based on 3 studies that reported numbers of individuals with aphasia that did not complete26,38,39 and 2 studies that reported the number that did.38,39
Two studies22,40 (both with ‘limited’ aphasia) reported using response cards as a supplementary method: as part of the interview assessment with the SADBD22; and to aid administration of the SDS questionnaire.40 Neither study reported the number of individuals with aphasia assessed but neither reported any noncompletion.
Four studies (2 with ‘limited’23,41 and 2 with ‘unlimited’ aphasia25,26) reported a supplementary method of verbally administering either the SDS questionnaire23,25,26 or the Geriatric Depression Scale (GDS) questionnaire.41 Robinson et al23 report that 10 carers also completed the SDS and that their responses were highly correlated with patients (r=0.83). A noncompletion rate of 14 from at least 92 attempts was achieved (based on 3 studies that reported numbers of individuals with aphasia that did not complete23,26,41 and 2 studies that reported the number that did.23,41
Modified Timing of Interviews
Two studies (1 with ‘limited’42 and 1 with ‘unlimited’ aphasia33) supplemented interviews by delaying their timing with individuals with more severe aphasia. For example, Morris et al33 suggest that the ‘delay of weeks’ to 2 months poststroke was what accounted for ‘reliable’ communication in all but 5 of their participants, which was evidenced by consistent responses to the GDS questionnaire. Both studies that used delayed interviews reported numbers of individuals with aphasia assessed (n=100) and neither reported any failure to complete depression diagnosis.
Visual Analogue Mood Scales
Four studies (1 with ‘limited’23 and 3 with ‘unlimited’ aphasia43–45) reported using visual analogue scales. In Robinson and colleague’s study,23 1 of 4 assessments contributed equally to the diagnosis of depression, but in the other studies visual analogues were used as an alternative method for individuals with aphasia.
Three slightly different visual analogue scales were used. All were based on schematic emotional face pictures joined with a 100-mm line on which respondents were asked to indicate their mood. Robinson et al used an adapted version of Folstein’s horizontal scale46 (with a ‘Frowning’ and ‘Happy’ face as well as the words ‘worst’ and ‘best’ mood at opposite poles). Gainotti et al43 and Paolucci et al44 used Stern and Bachman’s47 vertical scale that joins a ‘Happy’ and ‘Sad’ face. The DESTRO study45 used the ‘Sad’ item from Arruda et al’s48 set of visual analogue scales that is a vertical scale with a ‘Neutral’ and ‘Sad’ face.
Arruda et al’s45 scale is referred to as a ‘validated diagnostic instrument’; however, as the instrument’s authors, they acknowledge their validation study does not generalize to people with aphasia who were excluded. Similarly, Stern and Bachman’s44 scale is stated as ‘valid in aphasic patients’ but people with aphasia were excluded from correlative analyses comparing it to other measures in the supporting references43,47 because they were unable to complete the other measures. Robinson et al23 report administering their visual analogue scale at the start and end of assessment interviews and obtaining a very high test-retest reliability (r=0.98, P<0.0001).
A noncompletion rate of 437 of at least 521 attempts was achieved (based on 4 studies that reported numbers of individuals with aphasia that did not complete23,43–45 and 3 studies that reported the number that did complete.23,43,44 Gainotti and colleagues43 discuss why using a visual analogue scale only enabled them to assess 9 of the 23 participants with aphasia (with whom they had been unable to complete the HDRS interview). They highlight that although participants are only required to respond nonverbally, the instructions about how to respond are verbal. They tried using nonverbal gestures as instructions but found participants were also often unable to understand these. As Gainotti et al state, “The majority of severely aphasic patients are unable to understand not only the verbal questions but also the nonverbal requests of the examiner.”43
Confidence Expressed in Adaptive Depression Diagnostic Methods for Aphasia
Three studies that used adaptive methods to enable depression diagnosis in individuals with ‘unlimited’ aphasia expressed lacking confidence in their diagnoses. Andersen and colleagues37 used clinical observation when patients could not complete the HDRS. They expressed doubt that they had been able to distinguish between depression and uncontrolled crying in these individuals and actually decided to then exclude them from their study. Palomaki et al,36 who also used clinical observation when participants could complete neither the HDRS or BDI, excluded these individuals from the analysis. Although Reding et al used adapted depression diagnoses based on informants,25,26 modified questions25,26 and clinical observation26 to include individuals with aphasia, they also expressed uncertainty: “The more neurologically impaired a patient is, the more subjective are our assessments of his mood state.”26 “The clinical diagnosis of depression in the stroke population is difficult.”25
In contrast, 1 study with ‘unlimited’ aphasia severity33 that delayed interviewing participants with more severe aphasia indicated that aphasia was not generally a hindrance to conventional depression diagnostic interviewing: “Although 49% of the sample had some degree of aphasia following their stroke, only two had receptive aphasia and by the time of evaluation [two months but delayed this ‘by weeks’ if required] most aphasic patients had mild or minimal deficits not severe enough to interfere with the assessment process [Composite International Diagnostic Interview].”33
Completion of Main Depression Diagnostic Methods in Studies That Did Not Report Adaptations
Of studies that did not report an adaptive method, only 4 included individuals with ‘unlimited’ aphasia.49–52 Three of these49–51 reported being unable to complete depression diagnostic assessments. It is noteworthy that the study by Kotila et al51 was the only study with ‘unlimited’ aphasia that had used a questionnaire alone as the main depression diagnostic tool. Therefore, only 1 study52 that included individuals with ‘unlimited’ aphasia appeared able to complete a conventional method of diagnosing depression (the HDRS) without the use of adaptations. However, in contrast, they reported using informants to be able to complete a quality-of-life measure.
None of the 27 studies that did not report an adaptive method and had only individuals with ‘limited’ aphasia reported any failures in completing main depression diagnostic methods. However, 4 of these commented that the level of impairment among their participants with aphasia was so mild that completion of their main depression diagnostic method was still possible53,54 or that communication was still considered reliable.55,56 For example, “None of the patients had aphasia severe enough to disrupt communication.”56 “In all aphasic subjects, however, sufficient speech comprehension and production ability to complete a structured clinical interview and the Beck Depression Inventory could be secured.”53
Summary of Results
This review systematically identified 60 studies that diagnosed depression and included at least some individuals with aphasia. Of these, 29 provided information on how conventional language-based methods of depression diagnosis (most typically clinical interviews) were adapted in aphasia. The most common adaptive method was to use informants, whereas others used observation of nonverbal behavior in clinical interviews, simplified questions or a visual analogue scale.
Both the use of informants and clinical observation achieved very high combined completion rates, despite most studies using them for individuals with ‘unlimited’ aphasia. Studies that reported modifying interview questions (to only require ‘Yes/No’ responses, or with cue cards) or questionnaires (with verbal administration) also achieved good completion rates. However, all but 1 only included individuals with ‘limited’ aphasia, so it is unclear how successful they would be for individuals with more severe language-impairments. One small study23 used a visual analogue scale successfully with all their participants with ‘limited’ aphasia; however, the combined completion rate for 3 studies43–45 that used a visual analogue scale with individuals with ‘unlimited’ aphasia was very poor.
Nearly all studies used a standardized, structured tool and published criteria or cut-off points to diagnose depression in participants without aphasia. However, adaptive depression diagnostic methods for individuals with aphasia tended to lack structure and be so briefly described that exact replication would not be possible. Also, evidence of the validity or reliability of these adaptive methods was generally not reported or not substantiated by an independent study.
A few researchers, who did not report an adaptive depression diagnostic method, suggested that their participants’ aphasia was too mild to affect communication.53–56 Logically, this seems to be a contradiction in terms. However, 27 studies, that were restricted to people with ‘limited’ aphasia, neither reported having adapted their depression diagnostic method, nor reported failing to complete assessments. Therefore, the idea that some people with aphasia can communicate well enough to undergo a language-based depression diagnostic interview appeared to be a common one.
Strengths and Limitations
We thoroughly reviewed the adaptation of depression diagnostic methods for people with aphasia. Authors were contacted for clarification about ambiguous reporting, but others may have also used adaptive depression diagnostic techniques for aphasia but simply failed to report their use. Completion rates for the different adaptive methods we have calculated should only be taken as rough guides because only a minority of studies provided this information.
We found studies that assessed all available individuals with aphasia regardless of severity or type (‘unlimited’) to be more likely to report adaptive depression diagnostic methods and to report difficulty completing assessments in aphasia than studies that only included a subgroup of available individuals with aphasia (‘limited’). We acknowledge that gross descriptions of aphasia severity (‘limited’, ‘unlimited’) fail to communicate the complexity of retained communicative skills across comprehension, expression and the verbal and written modalities. Ideally, we would have been able to compare studies by participants’ aphasia severity level and type less crudely. However, this was not possible because there is no internationally agreed standard for defining aphasia severity and because many studies failed to even report how aphasia itself was assessed.57
Commentary and Critique of Adaptive Methods Used
Informants to Supplement Interview Findings
Despite high completion rates achieved using informants, certain reservations should be noted. Firstly, only 2 studies provided a structured, replicable format for informant use and these present different problems. The SADBD used by Gordon et al22 is a very extensive assessment containing over 70 questions. Their study actually included only a tiny fraction of people available with aphasia. Informants may struggle to estimate the detailed symptoms involved in the SADBD for individuals with more marked communication impairment. Whereas Robinson et al23 report no difficulties with their NRSD, Sinyor and colleagues40 report having to omit questions about verbal communication which their nurses could not rate and no significant association between the NRSD and the SDS. Similarly, House et al58 report that nurses in their study often failed to complete a similar screening tool. An interesting explanation may be that nurses in Robinson et al’s study were more confident at observing behavior in aphasia because they worked within a specialist aphasia institute.
The use of informants to estimate subjective cognitive symptoms of depression (eg, worthlessness) is of doubtful face-validity. A recent quality-of-life after stroke study found that carers’ proxy responses for patient mood significantly differed from patients and were affected by their own perception of carer burden.59 However, informants’ estimation of somatic symptoms of depression (eg, insomnia) appears much more reasonable. The Structured Aphasia Depression Questionnaire, which uses informant ratings of somatic symptoms and nonverbal behavior to screen for depression, may provide a useful measure of these symptoms to partially aid the diagnosis of depression in aphasia (although it remains to be tested in a study that has diagnosed depression in aphasia).60
Studies that used this method all achieved high completion rates. However, authors of studies were divided by whether they were confident in their diagnoses, or whether they disregarded them as too subjective or too influenced by emotionalism. An interesting explanation may be that the more confident diagnosticians were more experienced in working with patients with aphasia and may have ‘intuitively’ developed a greater range of communication skills and proficiency for enabling their communication.
Modified Questions or Responses Required
The use of simplified questions, supplementary key written phrases or verbal rather than written material are recommended techniques for maximizing communication with people with aphasia.61,62 However, other techniques such as repetition and use of pictorial material are also recommended and the techniques used should be tailored to an individuals needs. As the studies reviewed here only tested such techniques in people with ‘limited’ aphasia, they do not support their isolated use as sufficient to fully adapt depression diagnostic methods for all individuals with aphasia.
Delaying Timing of Interviews
Prevalence rates for aphasia do decrease in the months after stroke, with most spontaneous recovery taking place in the first month.1,2,5 However, like other poststroke impairments, aphasia does become a chronic condition in many cases.5 Therefore, delayed timing of depression diagnostic assessment is only likely to increase completion rates for some patients with aphasia, and this may not always be a clinically appropriate strategy.
Visual Analogue Scales
The visual analogue scales used in reviewed studies do not claim to provide sufficient assessment for the diagnosis of depression.46–48 Only 1 symptom of depression (low mood) is addressed, and although the scales provide a nonverbal response format they provide only verbal instructions. Gainotti and colleagues43 discussed having attempted to provide alternative nonverbal instructions but with a low success rate. The concept of expressing one’s mood metaphorically as a point on a line between highly stylized face pictures is very abstract. House and colleagues,58 who tested the performance of Folstein’s visual analogue scale as a screening tool, report finding participants ‘bewildered’ by it and to be more likely to respond to a depression questionnaire.
Of course, Robinson and colleagues23,30 did not report problems with Folstein’s scale and achieved a high test-retest result (that suggested participants’ responses at different times differed by only minimal amounts on the 100-mm line). However, this was a very small study (n=1823 or n=2530) and this visual analogue scale that uses a horizontal not a vertical line is unsuitable for individuals with haemianopia. On balance, the visual analogue scales reviewed provide neither a suitably comprehensive nor a particularly successful method of diagnosing depression in aphasia. Elsewhere, the suitability of visual analogue scales with stroke patients has been more generally criticized.63 However, the use of realistic looking pictures (with which people with stroke may more easily identify) may usefully support communication about mood with people with aphasia.62,64
Implications and Recommendations
The ability to diagnose depression in people with aphasia is essential to clinical practice. It is also important to further investigations into emotional disorder in this population and the development of depression screening measures. We would recommend future research to describe aphasia exclusion and inclusion criteria,57 whether and how depression diagnostic methods are adapted for individuals with aphasia along with completion and failure numbers. This would allow readers to judge the use of methods for people with different severity levels or types of aphasia and enable replication.
Future research to assess the reliability and validity of adaptive methods for diagnosing depression in aphasia is clearly required. Of course, the validity of a diagnostic gold standard can only be estimated through consensus expert opinion. We would recommend depression experts consider involvement and collaboration with language experts, such as speech and language therapists (whose absence within reviewed studies was apparent). This would allow research to develop adaptive methods that are accurately matched to individual’s specific aphasic profiles—to maximize retained skills, avoid severely impaired language components or modalities, or provide maximum support for these deficits. Ideally, because certain symptoms of depression are largely subjective (notably low mood, anhedonia, worthlessness and suicidal ideation), depression diagnosis in aphasia should involve patient self-report as it does for patients without aphasia.
Speech and language therapists are skilled in using a set of techniques for supporting communication in aphasia, including: observation of gestures (as questions and responses) with written key phrases and pictures (both preprepared and created with pen and paper during a conversation) along with simplified verbal communication (eg, Yes/No questions) that can be tailored to suit the person with aphasia’s individual needs and language abilities.61,62 Although aspects of these techniques were described in reviewed studies, we would recommend that formal training and use of supportive communication in combination with depression diagnostic interviewing skills could be developed into a semistructured clinical interview schedule for use with patients with aphasia.
Individuals with aphasia can and have been included in studies that have diagnosed depression using a range of adaptive, minimally language-dependent methods. However, the validity and reliability of these methods has not been established. Only a few studies reported using >1 adaptive technique and none described the depression diagnostician having been trained in, or using, the full range of skills that are available to support communication in people with aphasia. We would suggest that depression and language experts collaborate to develop and test communication training and materials to support the comprehensive assessment of depressive symptoms in people with aphasia, particularly more severe aphasia.
We thank Sarah Hinks for contributing to the screening of articles.
Source of Funding
This study was partially funded by Chest Heart and Stroke Scotland (CHSS).
- Received February 2, 2007.
- Revision received April 13, 2007.
- Accepted April 23, 2007.
Warlow CP, Sandercock P, Dennis M, Wardlaw J. Stroke: a Practical Guide to Management. Oxford: Blackwell Science; 2000.
Parr S, Byng S, Gilpin S. Talking About Aphasia. Buckingham: Open University Press; 1997.
Wade DT, Hewer RL, David RM, Enderby P. Aphasia after stroke: natural history and associated deficits. J Neurol Neurosurg Psychiatry. 1986; 49: 11–16.
Spalletta G, Pasini A, Costa A, De Angelis D, Ramundo N, Paolucci S, Caltagirone C. Alexithymic features in stroke: effects of laterality and gender. Psychosom Med. 2001; 63: 944–950.
Herrmann 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: 618–624.
SIGN: Scottish Intercollegiate Guidelines Network. Management of Patients With Stroke: Rehabilitation Prevention and Management of Complications and Discharge Planning. Edinburgh: SIGN; 2002: 64.
Royal College of Physicians. National Clinical Guidelines for Stroke. London: RCP; 2004.
Sutcliffe LM, Lincoln NB. The assessment of depression in aphasic stroke patients: the development of the Stroke Aphasic Depression Questionnaire. Clinical Rehabilitation. 1998; 12: 506–513.
Leeds L, Meara RJ, Hobson JP. The utility of the Stroke Aphasia Depression Questionnaire (SADQ) in a stroke rehabilitation unit. Clinical Rehabilitation. 2004; 18: 229–231.
Hammond M, O’Keefe ST, Barer DH. Development and validation of a brief observer-rated screening scale for depression elderly medical patients. Age and Ageing. 2000; 29: 511–515.
Watkins C, Leathley M, Daniels L, Dickinson H, Lightbody CE, van der Broek M, Jack CIA. The signs of depression scale in stroke: how useful are nurses’ observations? Clinical Rehabilitation. 2001; 15: 447.
The dexamethasone suppression test for diagnosing depression in stroke patients. [Abstract 973035] Database of Abstracts of Reviews of Effects. Available at: http://www.york.ac.uk/inst/crd/crddatabases.htm. Accessed on August 7, 2006. Abstract of: The dexamethasone suppression test for diagnosing depression in stroke patients. Harvey SA, Black KJ. Ann Clin Psychiatry. 1996; 8: 35–39.
Anderson CS, Hackett ML, House AO. Interventions for preventing depression after stroke. Cochrane Database of Systematic Reviews. 2004; Issue 2. Art. No.: CD003689. DOI: 10.1002/14651858.CD003689.pub2.
American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders: DSM-IV. Washington, DC: American Psychiatric Association; 1994.
Robson C. Real World Research: A Resource for Social Scientists and Practitioner-Researchers. Oxford: Blackwell Science; 1993.
Herrmann M, Bartels C, Schumacher M, Wallesch CW. Poststroke depression: is there a pathoanatomic correlate for depression in the postacute stage of stroke? Stroke. 1995; 26: 850–856.
Herrmann M, Bartels C, Wallesch CW. Depression in acute and chronic aphasia: symptoms, pathoanatomical-clinical correlations and functional implications. J Neurol Neurosurg Psychiatry. 1993; 56: 672–678.
Astrom M, Adolfsson R, Asplund K. Major depression in stroke patients: a 3-year longitudinal study. Stroke. 1993; 24: 976–982.
Verdelho A, Henon H, Lebert F, Pasquier F, Leys D. Depressive symptoms after stroke and relationship with dementia: a three-year follow-up study. Neurology. 2004; 62: 905–911.
Benaim C, Cailly B, Perennou D, Pelissier J. Validation of the aphasic depression rating scale. Stroke. 2004; 35: 1692–1696.
Palomaki H, Kaste M, Berg A, Lonnqvist R, Lonnqvist 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.
Starkstein SE, Fedoroff JP, Price TR, Leiguarda R, Robinson RG. Anosognosia in patients with cerebrovascular lesions: a study of causative factors. Stroke. 1992; 23: 1446–1453.
Sinyor D, Amato P, Kaloupek DG, Becker R, Goldenberg M, Coopersmith H. Post-stroke depression: relationships to functional impairment, coping strategies, and rehabilitation outcome. Stroke. 1986; 17: 1102–1107.
Hosking SG, Marsh NV, Friedman PJ. Depression at 3 months poststroke in the elderly: predictors and indicators of prevalence. Aging Neuropsychol Cognition. 2000; 7: 205–216.
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.
Gainotti G, Azzoni A, Gasparini F, Marra C, Razzano C. Relation of lesion location to verbal and nonverbal mood measures in stroke patients. Stroke. 1997; 28: 2145–2149.
House A, Dennis M, Warlow C, Hawton K, Molyneux A. Mood disorders after stroke and their relation to lesion location: a CT scan study. Brain. 1990; 113: 1113–1129.
Kotila M, Numminen H, Waltimo O, Kaste M. Depression after stroke: results of the FINNSTROKE Study. Stroke. 1998; 29: 368–372.
Carod-Artal J, Egido JA, Gonzalez 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.
Beblo T, Wallesch C-W, Herrmann M. The crucial role of frontostriatal circuits for depressive disorders in the postacute stage after stroke. Neuropsychiatry Neuropsychol Behavl Neurol. 1999; 12: 36–46.
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.
Williams L, Bakas T, Brizendine E, Plue L, Wanzhu T, Hendrie H, Kroenke K. How valid are family proxy assessments of stroke patients’ health-related quality of life? Stroke. 2006; 37: 2081–2085.
Price CIM, Curless RH, Rodgers H. Can stroke patients use visual analogue scales? Stroke. 1999; 30: 1357–1361.
Swinburn K, Byng S. The Communication Disability Profile. London: Connect Press; 2006.
Aben I, Verhey F, Strik J, Lousberg R, Lodder J, Honig A. A comparative study into the one year cumulative incidence of depression after stroke and myocardial infarction. J Neurol Neurosurg Psychiatry. 2003; 74: 581–585.
Aben I, Denollet J, Lousberg R, Verhey F, Wojciechowski F, Honig A. Personality and vulnerability to depression in stroke patients: a 1-year prospective follow-up study. Stroke. 2002; 33: 2391–2395.
Andersen G, Vestergaard K, Ingeman-Nielsen M. Post-stroke pathological crying: Frequency and correlation to depression. Eur J Neurol. 1995; 2: 45–50.
Andersen G, Vestergaard K, Lauritzen L. Effective treatment of poststroke depression with the selective serotonin reuptake inhibitor citalopram. Stroke. 1994; 25: 1099–1104.
Astrom M, Olsson T, Asplund K. Different linkage of depression to hypercortisolism early versus late after stroke. A 3-year longitudinal study. Stroke. 1993; 24: 52–57.
Astrom M, Asplund K, Astrom T. Psychosocial function and life satisfaction after stroke. Stroke. 1992; 23: 527–531.
Olsson T, Astrom M, Eriksson S, Forssell A. Hypercortisolism revealed by the dexamethasone suppression test in patients [corrected] with acute ischemic stroke.[erratum appears in Stroke. 1990;21:681]. Stroke. 1989; 20: 1685–1690.
Beblo T, Driessen M. No melancholia in poststroke depression? A phenomenologic comparison of primary and poststroke depression. J Geriatr Psychiatry Neurol. 2002; 15: 44–49.
Eastwood MR, Rifat SL, Nobbs H, Ruderman J. Mood disorder following cerebrovascular accident. Brit J Psychiatry. 1989; 154: 195–200.
Gainotti G, Antonucci G, Marra C, Paolucci S. Relation between depression after stroke, antidepressant therapy, and functional recovery. J Neurol Neurosurg Psychiatry. 2001; 71: 258–261.
Gainotti G, Azzoni A, Marra C. Frequency, phenomenology and anatomical-clinical correlates of major post-stroke depression. Brit J Psychiatry. 1999; 175: 163–167.
Kauhanen M, Korpelainen JT, Hiltunen P, Brusin E, Mononen H, Maatta R, Nieminen P, Sotaniemi KA, Myllyla VV. Poststroke depression correlates with cognitive impairment and neurological deficits. Stroke. 1999; 30: 1875–1880.
Kim JS, Choi-Kwon S. Poststroke depression and emotional incontinence: correlation with lesion location. Neurology. 2004; 54: 1805–1810.
Ghika-Schmid F, van Melle G, Guex P, Bogousslavsky J. Subjective experience and behavior in acute stroke: the Lausanne Emotion in Acute Stroke Study. Neurology. 1999; 52: 22–28.
Carota A, Rossetti AO, Karapanayiotides T, Bogousslavsky J. Catastrophic reaction in acute stroke: a reflex behavior in aphasic patients. Neurology. 2001; 57: 1902–1905.
House A, Dennis M, Molyneux A, Warlow C, Hawton K. Emotionalism after stroke. BMJ. 1989; 298: 991–994.
Sharpe M, Hawton K, Seagroatt V, Bamford J, House A, Molyneux A, Sandercock P, Warlow C. Depressive disorders in long-term survivors of stroke. Associations with demographic and social factors, functional status, and brain lesion volume. Brit J Psychiatry. 1994; 164: 380–386.
Berg A, Palomaki H, Lehtihalmes M, Lonnqvist J, Kaste M. Poststroke depression: an 8-month follow-up. Stroke. 2003; 34: 138–143.
Pohjasvaara T, Leppavuori A, Siira I, Vataja R, Kaste M, Erkinjuntti T. Frequency and clinical determinants of poststroke depression. Stroke. 1998; 29: 2311–2317.
Robinson RG, Price TR. Post-stroke depressive disorders: a follow-up study of 103 patients. Stroke. 1982; 13: 635–641.
Robinson RG, Schultz SK, Castillo C, Kopel T, Kosier JT, Newman RM, Curdue K, Petracca G, Starkstein SE. Nortriptyline versus fluoxetine in the treatment of depression and in short-term recovery after stroke: a placebo-controlled, double-blind study. Am J Psychiatry. 2000; 157: 351–359.
Robinson RG, Starr LB, Kubos KL, Price TR. A two-year longitudinal study of post-stroke mood disorders: findings during the initial evaluation. Stroke. 1983; 14: 736–741.
Robinson RG, Starr LB, Lipsey JR, Rao K, Price TR. A two-year longitudinal study of post-stroke mood disorders: dynamic changes in associated variables over the first six months of follow-up. Stroke. 1984; 15: 510–517.
Robinson RG, Starr LB, Lipsey JR, Rao K, Price TR. A two-year longitudinal study of poststroke mood disorders. In-hospital prognostic factors associated with six-month outcome. Journal of Nervous & Mental Disease. 1985; 173: 221–226.
Parikh RM, Lipsey JR, Robinson RG, Price TR. Two-year longitudinal study of post-stroke mood disorders: dynamic changes in correlates of depression at one and two years. Stroke. 1987; 18: 579–584.
Robinson RG, Starr LB, Price TR. A two year longitudinal study of mood disorders following stroke. Prevalence and duration at six months follow-up. Brit J Psychiatry. 1984; 144: 256–262.
Robinson RG, Bolduc PL, Price TR. Two-year longitudinal study of poststroke mood disorders: diagnosis and outcome at one and two years. Stroke. 1987; 18: 837–843.
Robinson RG, Bolla-Wilson K, Kaplan E, Lipsey JR, Price TR. Depression influences intellectual impairment in stroke patients. Brit J Psychiatry. 1986; 148: 541–547.
Robinson RG, Kubos KL, Starr LB, Rao K, Price TR. Mood disorders in stroke patients. Importance of location of lesion. Brain. 1984; 107: 81–93.
Schramke CJ, Stowe RM, Ratcliff G, Goldstein G, Condray R. Poststroke depression and anxiety: different assessment methods result in variations in incidence and severity estimates. Journal of Clinical & Experimental Neuropsychology. 1998; 20: 723–737.
Sinyor D, Jacques P, Kaloupek DG, Becker R, Goldenberg M, Coopersmith H. Poststroke depression and lesion location. An attempted replication. Brain. 1986; 109: 537–546.
Starkstein SE, Robinson RG, Honig MA, Parikh RM, Joselyn J, Price TR. Mood changes after right-hemisphere lesions. Brit J Psychiatry. 1989; 155: 79–85.
Starkstein SE, Robinson RG, Price TR. Comparison of cortical and subcortical lesions in the production of poststroke mood disorders. Brain. 1987; 110: 1045–1059.
Starkstein SE, Robinson RG, Price TR. Comparison of spontaneously recovered versus nonrecovered patients with poststroke depression. Stroke. 1988; 19: 1491–1496.
Starkstein SE, Fedoroff JP, Price TR, Leiguarda R, Robinson RG. Apathy following cerebrovascular lesions. Stroke. 1993; 24: 1625–1630.
Wiart L, Petit H, Joseph PA, Mazaux JM, Barat M. Fluoxetine in early poststroke depression: a double-blind placebo-controlled study. Stroke. 2000; 31: 1829–1832.
Williams LS, Brizendine EJ, Plue L, Bakas T, Tu W, Hendrie H, Kroenke K. Performance of the PHQ-9 as a screening tool for depression after stroke. Stroke. 2005; 36: 635–638.