Original article
Screening for anxiety and depression after stroke: Comparison of the Hospital Anxiety and Depression Scale and the Montgomery and Åsberg Depression Rating Scale,☆☆

https://doi.org/10.1016/j.jpsychores.2009.03.007Get rights and content

Abstract

Objective

Anxiety and depression after stroke are frequent, but are often overlooked and not assessed. The aims of the study were to (1) assess the prevalence of anxiety and depression and (2) compare the performance of the Hospital Anxiety and Depression Scale (HADS) and Montgomery and Åsberg Depression Rating Scale (MADRS) as screening instruments for anxiety and depression disorders 4 months after stroke.

Methods

Stroke patients, consecutively admitted to a stroke unit, were assessed with HADS and MADRS 4 months after stroke (n=104). Depression and anxiety disorders were diagnosed using the Structured Clinical Interview for DSM-IV (SCID). Measures were compared in terms of correlations, sensitivity, specificity, positive and negative predictive value, overall agreement, kappa, and ROC curves, using DSM-IV diagnoses of “at least one current significant anxiety disorder” (Anxiety) and “any current depression” (Depression), as the clinical criteria.

Results

Anxiety occurred in 23% of patients and Depression in 19% (13% major depression, 3% minor depression, 4% dysthymia). For Anxiety, the optimal screening cut-off was 4 for HADS-A and 6 for HADS-total; for Depression, optimal cut-offs were 4 for HADS-D, 11 for HADS-total, and 8 for MADRS. At cut-offs commonly used in clinical practice for depression screening (HADS-D: 8, MADRS: 12), the MADRS performed marginally better than the HADS.

Conclusion

Anxiety is as prevalent as depression 4 months after stroke. MADRS and HADS-D perform acceptably as screening instruments for depression, and HADS-A for anxiety after stroke. However, lower HADS cut-offs than recommended for the general population should be considered for stroke patients.

Introduction

Anxiety and mood disorders are important consequences of stroke. They affect outcome and recovery and increase caregiver burden [1], [2], [3], [4]. However, even though anxiety and depression are common after stroke, with reported rates of 14–28% for anxiety and 11–61% for depression, they are often not detected nor treated adequately [5], [6], [7], [8], [9]. The variation in prevalence is due to differences in assessment methods, time points, and varying definitions of depression and anxiety across studies [5], [6], [7], [8], [9].

Antidepressant medication has been shown to improve both depression and anxiety after stroke, as well as activities of daily living (ADL) and cognitive impairment, and to decrease poststroke mortality [10]. Significantly greater recovery and improvement in ADL functions have been associated with poststroke depression remission [11]. These findings emphasize the importance of identification and treatment of anxiety and depression after stroke.

Depression after stroke has been extensively studied. There is, however, much less research on anxiety after stroke, despite evidence for its importance. Furthermore, many studies suffer from methodological limitations, such as selection bias (e.g., investigating only samples from persons of low social status), assessment at unstandardized time intervals poststroke, exclusion of patients with physical or cognitive impairments, and lack of diagnostic assessments with structured interviews [12].

The ideal method of diagnosing psychiatric illness is a standardized psychiatric interview. However, this is time consuming, impractical, and mostly unavailable on neurological or medical wards or in general practice. Therefore, there is a need for valid and reliable screening instruments in order to identify vulnerable patients.

Both the Hospital Anxiety and Depression Scale (HADS) [13], as a self-report questionnaire, and the Montgomery and Åsberg Depression Rating Scale (MADRS) [14], as a clinician-rated scale, are frequently used for the assessment of depression and anxiety in stroke patients [15], [16], [17], [18], [19], [20]. HADS was designed for use in nonpsychiatric hospital settings and contains no questions referring to physical complaints to avoid confounding from underlying somatic diseases [13]. In a review of the literature on use of the HADS as an assessment instrument for anxiety and depression, Bjelland et al. [21] found that HADS performed well in the assessment of both symptom severity and diagnosis of anxiety and depression at the recommended diagnostic cut-off of ≥8 for both subscales. However, studies in stroke patients indicate that lower cut-offs (anxiety: 5–7; depression: 4–8) would be more appropriate in this population [18], [19], [20].

The findings in previous studies that compared HADS with other screening instruments for the detection of anxiety and depression in stroke patients are contradictory. O'Rourke et al. [19] and Aben et al. [18] did not find significant differences between HADS and other instruments, such as the General Health Questionnaire (GHQ), Beck Depression Inventory, Hamilton Depression Rating Scale, and Symptom Check List, while Johnson et al. [20] found that the performance of HADS was inferior to instruments such as Geriatric Depression Scale and GHQ. The contradictory findings may be attributed to differences in diagnostic methods, definitions of anxiety and depression, and differences in time frames poststroke. Obviously, there is a need for further studies to elucidate the suitability of HADS as a screening instrument for anxiety and depression in a stroke population.

MADRS has been recommended as an assessment tool in stroke patients. It has demonstrated reliability and validity comparable to other observer-rated scales, but places greater emphasis on the psychological symptoms of depression, which partially limits the possible confounding impact of symptoms of physical illness being mistaken as indicative of depression [14], [22]. However, a consensus on the diagnostic cut-off score for MADRS does not exist [23]. Snaith et al. [24] established score ranges for MADRS, defining four grades of severity of depression. Diagnosis was defined as MADRS score >6. This cut-off has been used in studies on stroke patients. In psychiatric practice, a diagnostic threshold of >12 is used [25], [26]. To our knowledge, it has not been previously assessed how well these thresholds agree with a DSM-IV diagnosis of depression. We do not know of any study that has investigated the accuracy of MADRS regarding the identification of clinical depression in poststroke patients and compared it to other screening instruments.

According to rehabilitation guidelines and clinical practice, patients are assessed 3 to 4 months after stroke [27]. Diagnostic assessment of both affective and anxiety disorders using structured clinical psychiatric interviews for DSM-IV criteria together with self-rated and clinician-rated scales has not been conducted 4 months after stroke. Therefore the relationship between diagnostic assessment and the scores on self-report and clinician-rated instruments in the poststroke phase is not known.

With this background, the aims of this study were to (1) assess the prevalence of anxiety and depression and (2) evaluate the performance of HADS and MADRS as screening instruments for the detection of anxiety and depression 4 months after stroke.

Section snippets

Method

The research protocol was accepted by the local ethics committee and the data board in May 2001. Informed written consent was obtained from all participants.

Sample characteristics

The participants at baseline (n=150) were significantly younger than the nonparticipants (n=34) (mean age=66.4 years, S.D.=13.3 vs. mean age=73.5 years, S.D.=9.2; P<.001). There was no significant difference in sex as regards participation (male: 64% vs. 62%; P=.807). One hundred and twenty-two patients (82%) suffered from their first stroke; 27 (18%) had a recurrent stroke. One patient had a CT scan abroad; that scan was not available for evaluation.

The participants at follow-up (n=104) were

Prevalence of anxiety and depression

Anxiety and depression were frequent 4 months after stroke. The prevalence was in line with findings in other studies 3–4 months after stroke [5], [6], [8], [10], [34]. The same applies for the comorbidity rate, which was in the range of 11–18% for comorbid anxiety and depression [5], [6], [8]. The rates of both anxiety disorders and depression are considerably higher than those reported in the general elderly population (anxiety: 3%, depression: 12%, comorbid anxiety and depression: 2%) [35].

Acknowledgments

We want to thank the staff of the stroke unit at the Department of Neurology, Telemark Hospital, Norway, for their engagement and assistance.

References (36)

  • AstromM

    Generalized anxiety disorder in stroke patients. A 3-year longitudinal study

    Stroke

    (1996)
  • HackettML et al.

    Frequency of depression after stroke: a systematic review of observational studies

    Stroke

    (2005)
  • ChemerinskiE et al.

    The effect of remission of poststroke depression on activities of daily living in a double-blind randomized treatment study

    J Nerv Ment Dis

    (2001)
  • JohnsonJL et al.

    Poststroke depression incidence and risk factors: an integrative literature review

    J Neurosci Nurs

    (2006)
  • ZigmondAS et al.

    The Hospital Anxiety and Depression Scale

    Acta Psychiatr Scand

    (1983)
  • MontgomeryS et al.

    A new depression scale designed to be sensitive to change

    Br J Psychiatry

    (1979)
  • FureB et al.

    Emotional symptoms in acute ischemic stroke

    Int J Geriatr Psychiatry

    (2006)
  • HerrmannM et al.

    Poststroke depression. Is there a pathoanatomic correlate for depression in the postacute stage of stroke?

    Stroke

    (1995)
  • Cited by (0)

    The stroke study was funded by the Norwegian Foundation for Health and Rehabilitation and the Council for Mental Health, Norway, and the Thora and Sverre Lofthus Foundation.

    ☆☆

    Declaration of interest: none.

    View full text