You are here

Article Text

Article menu

Download PDFPDF

Cerebrovascular disease
Research paper
The long-term outcomes of depression up to 10 years after stroke; the South London Stroke Register
  1. L Ayerbe1,2,
  2. S Ayis1,
  3. S Crichton1,
  4. C D A Wolfe1,3,
  5. A G Rudd1,4
  1. 1Division of Health and Social Care Research, King's College London, London, UK
  2. 2Blizard Institute, Centre for Primary Care and Public Health, Barts and The London School of Medicine and Dentistry, Queen Mary University of London, London, UK
  3. 3National Institute for Health Research (NIHR) Biomedical Research Centre, Guy's and St Thomas’ NHS Foundation Trust, London, UK
  4. 4Stroke Unit, Guy's and St. Thomas’ NHS Foundation Trust, St. Thomas’ Hospital London, London, UK
  1. Correspondence to Dr Luis Ayerbe, Blizard Institute, Centre for Primary Care and Public Health, Barts and The London School of Medicine and Dentistry, Queen Mary University of London, Yvonne Carter Building, 58 Turner Street, London E1 2AB, UK; l.garcia-morzon{at}qmul.ac.uk

Abstract

Background Post-stroke depression is a frequent chronic and recurrent problem that starts shortly after stroke and affects patients in the long term. The health outcomes of depression after stroke are unclear.

Aims (1) To investigate the associations between depression at 3 months and mortality, stroke recurrence, disability, cognitive impairment, anxiety and quality of life (QoL), up to 5 years post-stroke. (2) To investigate these associations in patients recovering from depression by year 1. (3) To investigate associations between depression at 5 years and these outcomes up to 10 years.

Methods Data from the South London Stroke Register (1997–2010) were used. Patients (n at registration=3240) were assessed at stroke onset, 3 months after stroke and annually thereafter. Baseline data included sociodemographics and stroke severity measures. Follow-up assessments included anxiety and depression (Hospital Anxiety and Depression scale), disability, QoL and stroke recurrence. Multivariable regression models adjusted for age, gender, ethnicity, stroke severity and disability were used to investigate the association between depression and outcomes at follow-up.

Results Depression at 3 months was associated with: increased mortality (HR: 1.27 (1.04 to 1.55)), disability (RRs up to 4.71 (2.96 to 7.48)), anxiety (ORs up to 3.49 (1.71 to 7.12)) and lower QoL (coefficients up to −8.16 (−10.23−6.15)) up to year 5. Recovery from depression by 1 year did not alter these risks to 5 years. Depression in year 5 was associated with anxiety (ORs up to 4.06 (1.92 to 8.58)) and QoL (coefficients up to −11.36 (−14.86 to −7.85)) up to year 10.

Conclusions Depression is independently associated with poor health outcomes.

  • Stroke
  • Depression

https://doi.org/10.1136/jnnp-2013-306448

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Introduction

    Depression after stroke is a common, chronic and recurrent problem affecting patients in the long term.1–5 However, in order to understand the burden of depression in stroke patients, it is also necessary to estimate its potential association with mortality and morbidity over time.

    A systematic review6 identified studies investigating the association between depression and other health outcomes.2 ,7–10 Most of these studies had limitations, including small sample size, short follow-up, and poorly reported methods. Although the evidence suggests that depression is associated with higher risk of stroke,11 cognitive impairment12 and anxiety,13 to the best of our knowledge, no studies have investigated the association between depression and these outcomes at follow-up.6 The association of depression with other health outcomes could change in patients who recover from depression. However, the effect of recovery has not been investigated in any of the studies.6 Finally, while the long-term outcomes of depression may be affected by episodes of depression occurring more than 1 year after stroke, no studies investigated depression as an exposure more than 3 months after stroke.6 Therefore, the prognosis and consequences of depression in the long term after stroke remain unclear and this limits its clinical management.

    This study estimates the association between depression 3 months after stroke and mortality, stroke recurrence, disability, cognitive impairment, anxiety and quality of life (QoL) up to 5 years after stroke. Second, the association between depression and these health outcomes is also estimated in patients who recover from depression within the first year of stroke (depressed at 3 months and not depressed at 1 year). Finally, the association between depression 5 years after stroke (depressed or not in year 5) and the same outcomes up to 10 years is estimated.

    Methods

    First in a lifetime stroke patients were recruited from the South London Stroke Register (SLSR), a prospective population-based cohort study covering an inner-city population of 271 817.14 Data from patients, registered in the SLSR between June 1997 and December 2009, and followed-up until August 2010, were used. The methodology has been described in detail previously and is summarised below.14 WHO definition of stroke was used.15 Sixteen overlapping referral sources were used to increase completeness of notification. All patients with a suspected diagnosis of stroke were identified by one of the sources of notification and investigated for eligibility of study inclusion. Patients were registered during the acute phase of stroke and then they were followed-up 3 months after stroke, 1 year after stroke and annually, thereafter, for up to 10 years. A study clinician verified the diagnosis of all patients being registered. Data collected at baseline included sociodemographics and stroke severity measures. Demographic data included age, gender and ethnicity (white, black and other ethnicity). Stroke severity measures included Glasgow Coma Scale (GCS), categorised as severe (3–8), moderate (9–12) and mild (13–15), levels of impairment, urinary incontinence and paresis. Disability was also assessed 7–10 days after stroke using the Barthel Index (BI)16: scores of 0–14 were categorised as severe disability, 15–19 moderate disability and 20 independent. The BI was chosen to measure disability as it has been widely used in clinical settings and research, and it has excellent test-retest (κ w=0.98) and inter-rater reliability (κ w=0.88).16 ,17

    Follow-up was by postal questionnaire or interview, depending on the capacity of the patient to fill in the questionnaire. Patients who could not be followed-up at one time point remained registered and were contacted again for the following assessment. At follow-up, patients were assessed for depression and anxiety, using the Hospital Anxiety and Depression scale (HADS).18 HADS has been validated in stroke patients showing a good performance when it is used in a face-to-face interview, and when it is self-administered19 (optimum performance when HADS subscales scores above 7 are used to identify anxiety and depression: sensitivity 0.82 and specificity 0.74 for depression, and sensitivity 0.78 and specificity 0.74 for anxiety).18 Despite its good performance, HADS is not a diagnostic scale but a screening tool that detects only symptoms of depression and anxiety. However, the terms ‘depression’ and ‘anxiety’ will be used in this paper for succinctness in patients with scores above 7. Patients with HADS scores above 7 at 3 months, presenting scores below 7 at 1 year, were categorised as patients recovering from depression. HADS cannot be answered by proxy, so all information was collected directly from patients. No data on anxiety or depression could be collected from patients with cognitive or communication impairment if they were considered unable to give valid responses. The exclusion of these patients was based on the clinical judgement of the SLSR field workers. Other data collected at follow-up included measures of disability, cognitive impairment and QoL. Cognitive impairment was assessed with the Mini Mental State examination (MMSE)20 between 1997 and 2001, and the Abbreviated Memory Test (AMT)21 between 2002 and 2010. Patients with MMSE score <24 or AMT scores 0–7 were considered cognitively impaired.20 ,21 QoL was assessed with the SF-36 and SF-12.22 ,23 The physical and mental components summary scores of these scales were used. Scores collated from the scales ranged from 0 to 100 with high scores representing better QoL. Mortality data were collected by the SLSR follow-up team and verified by the Office of National Statistics (ONS). The definition of stroke recurrence was the same as for the index stroke.15 Only recurrences 21 days after the initial event, or if earlier, clearly in a different vascular territory were included. The same overlapping sources used by the SLSR to identify first ever strokes14 were used to identify recurrent strokes.

    The associations between depression 3 months after stroke (explanatory variable) and mortality, stroke recurrence, disability, cognitive impairment and QoL (dependent variables) up to 5 years of follow-up were investigated. Second, the associations between depression at 3 months and these outcomes at follow-up were compared in patients who recovered from depression in year 1 and in those who did not have depression. These associations were also compared in patients who recovered from depression in year 1 and those who did not recover. Finally, the associations between depression in year 5 and mortality, disability, anxiety, cognitive impairment and Qol up to 10 years were estimated.

    Statistical methods

    χ2 And t tests were used to compare sociodemographic and clinical characteristics, at baseline, at 3 months and at 5 years, of patients assessed and not assessed for depression. Kaplan Meir curves, Log rank tests and multivariate Cox regression models were used to analyse the association between depression at 3 months and mortality up to 5 years after stroke. The same analyses, with Kaplan Meir curves, Log rank tests and Cox regression models, were used to investigate the association between depression 3 months after stroke and stroke recurrence among survivors within 5 years of the first event. Multinomial logistic regression models were used to analyse the association between depression 3 months after stroke, and disability at follow-up. The associations between depression at 3 months and anxiety, and between depression and cognitive impairment at follow-up, were analysed with logistic regression. Linear regression models were used to analyse the association between depression 3 months after stroke, and QoL at follow-up. All regression models were adjusted for age, gender, ethnicity, disability at baseline, depression at 3 months and stroke severity measures (GCS, incontinence and paresis). These case-mix variables were chosen because of their clinical relevance and also because of their reported prognostic value.24 ,25 Results from the multivariate analysis are reported. Only patients with complete outcome data were included in the analysis. Most variables analysed as potential predictors or confounders had some missing data. A separate category was assigned to it, for example, Paresis 0 (No), 1 (Yes), and 2 (Missing). Sensitivity analysis was conducted to compare estimates obtained in multivariate analysis when the category for missing data was included and when it was not included. All estimates and SEs were stable and similar to those based on complete data, so the results obtained using separate categories for missing data were reported.

    Similar statistical analyses were used to compare outcomes up to 5 years in patients recovering from depression in year 1. Outcomes of those depressed at 3 months and not depressed in year 1 were compared, first with outcomes of patients not having depression at 3 months and neither at year 1, and second with outcomes of patients showing depression at 3 months and 1 year.

    Similar regression models, adjusted for age gender ethnicity, stroke severity (GCS, paresis and incontinence) and disability in year 5, where used to observe the association between depression in year 5 and mortality, disability anxiety and QoL up to 10 years after stroke.

    Results

    The SLSR recruited 3240 patients between June 1997 and December 2009. Eight hundred and fourteen patients died before the 3-month assessment. Among the 1458 survivors who were followed-up, 1101 were able to complete the HADS scale at 3 months, of which 361 were depressed (32.8%). The baseline and follow-up characteristics of these patients are presented in tables 1 and 2.

    View this table:
    Table 1

    Baseline characteristics of patients assessed for depression at 3 months

    View this table:
    Table 2

    Characteristics of patients, assessed for depression at 3 months, in the 10 years of follow up

    Patients completing the HADS scale had less severe strokes and lower disability rate at baseline. There were no significant differences in age and gender between survivors assessed for depression and those not assessed.

    Outcomes of depression at 3 months

    Mortality rate up to year 5 was higher for patients who were depressed at 3 months (dead n=292), Log rank test p=0.0015 (figure 1). This result was confirmed in the multivariate analysis HR: 1.34 (1.06 to 1.70) p=0.014. Depression at 3 months, was not associated with higher risk of recurrence up to year 5 (recurrences n=72), Log rank test p=0.9519 (figure 2). This result was confirmed in the multivariate analysis, HR: 0.98 (0.60 to 1.62) p=0.951. Patients with depression at 3 months had higher disability rate (BI: 0–14) at all time points until year 5 (table 3). Depression at 3 months showed no association with cognitive impairment at any time point during the 5 years of follow-up. A significant association between depression at 3 months and anxiety in the first 5 years after stroke was identified (table 3). Depression 3 months after stroke was associated with lower scores in the mental and physical domains of QoL in the 5 years of follow-up (table 3).

    View this table:
    Table 3

    Comparison of anxiety, disability and quality of life (QoL) at follow-up in patients with and without depression at 3 months. Multivariate analysis

    Figure 1
    Figure 1

    Mortality up to 5 years after stroke by depression status at 3 months.

    Figure 2
    Figure 2

    Stroke recurrence up to 5 years by depression status at 3 months.

    Outcomes of patients depressed at 3 months not depressed in year 1

    A total of 116 patients depressed at 3 months recovered from depression in year 1, while 433 patients did not have depression at 3 months or at 1 year. Patients who recovered from depression showed higher mortality than those not having depression at 5 years of follow-up, HR: 1.69 (1.09 to 2.62) p=0.019. No differences in recurrence of stroke at follow-up were observed between patients who recovered from depression and those not having depression. Patients recovering showed higher anxiety rate in years 1–4 (table 4). Disability rate (BI: 0–14) was also higher up to year 5 among patients recovering from depression than in those not having depression (table 4). No significant differences in cognitive impairment at follow-up were observed. The mental domain of QoL was significantly lower in years 1 to three, and five, in those recovering from depression. Physical domain of QoL was also lower for those recovering in years one, two, three and five than in those not having depression (table 4).

    View this table:
    Table 4

    Comparison of anxiety, disability and quality of life (QoL) at follow-up in patients recovering from depression and those not having depression. Multivariate analysis

    The 116 patients with depression at 3 months who recovered from depression in year 1 were compared with the 116 patients who had depression at 3 months and 1 year, and no differences in mortality, stroke recurrence, anxiety, disability, cognitive impairment, or physical domain of QoL up to year 5 were observed. The mental health domain of QoL was higher among those recovering at 1 year than among those having depression at 3 months and 1 year, only in year 4 coef.: 5.66 (1.46 to 9.87) p=0.09.

    Outcomes of depression in year 5

    Patients numbering 2358 had died before year 5. Among the 624 patients who were followed-up, 532 could complete the HADS questionnaire of which 156 (29.3%) were depressed. Patients assessed in year 5 were younger and had had less severe strokes and lower rate of disability than those not assessed.

    There was no association between depression in year 5 and mortality up to 10 years after stroke (dead n=111). HR: 1.25 (0.83 to 1.89) p=0.277. Depression in year 5 was not associated with disability or cognitive impairment up to year 10. However, depression in year 5 was associated with higher rates of anxiety in the following 5 years except in year 9 (table 5). Depression in year 5 was also associated with lower mental domain of QoL up to 9 years after stroke and with lower physical QoL in years 6 and 7.

    View this table:
    Table 5

    Comparison of anxiety, and quality of life (QoL) at follow-up in patients with and without depression in year 5. Multivariate analysis

    Discussion

    Outcomes of depression up to 10 years after stroke

    The association between depression and higher mortality after stroke is not consistent in previous studies.2 ,8 ,9 Apart from suicide, a number of biobehavioral mechanisms have been hypothesised to underlie the relationship between depression and mortality. First, depression is associated with poor health behaviour (eg, smoking, physical inactivity, lack of medication compliance). Second, depression is correlated with other major comorbidities, such as diabetes and hypertension, that are cardiovascular risk factors. Third, depression is associated with biological changes, which include neuroendocrine (eg, sympathetic nervous system activation, dysregulation of the hypothalamic-pituitary-adrenocortical axis) and immunological/inflammation changes, that could influence mortality.11 ,26–28 The short duration but high recurrence rate of depression5 may explain its persistent association with higher mortality rate up to 10 years after stroke. Patients surviving one episode of depression remain at risk of dying during a recurrent one.

    Two previous systematic reviews reported an increased cardiovascular risk among patients with depression.11 ,29 This study did not find evidence of an association between depression and stroke recurrence. Depression seems to be a less relevant cardiovascular risk factor after stroke than before. Another possibility is that the number of recurrences registered in our study is not enough to observe the association between depression and recurrence of stroke. The association between depression and disability at follow-up may be explained by the poor involvement in rehabilitation30 or by the low level of activity that these patients have.12 Previous literature had reported an association between depression and lower QoL only in the first year after stroke.7 ,10 This study shows a very consistent association between depression and lower QoL, and the mental and physical domains in the longer term. The possible overlapping between the HADS, and the mental health domain of the SF-12 and SF-36 may explain, in part, the association between depression and the lower mental health QoL. The high comorbidity rate of anxiety and depression may explain the association between depression in the short term and anxiety at follow-up.31

    Recovering from depression does not reduce the risk of negative health outcomes in the long term. The high recurrence rate of depression can explain the independent association between depression shortly after stroke and poor health outcomes in the long term.5

    Depression in year 5 was not associated with higher mortality up to 10 years suggesting that the link between depression and mortality may be different in the long term after stroke. It also shows that depression shortly after stroke may be a more relevant clinical predictor of stroke prognosis in the long term. However, depression at 5 years still predicts lower QoL and anxiety at follow-up, and therefore, requires clinical attention.

    Strengths and limitations

    The SLSR is a possibly the largest population-based cohort of stroke patients, and it has data collected in the very long term after the acute event. It provides the least biased sampling frame, and good statistical power in the analyses, by contrast with many previous studies.6 It also has data on numerous variables allowing for a comprehensive description of the clinical situation of the patient up to 10 years after stroke. Most outcomes of depression reported before in the literature had been observed shortly after the acute event. The analyses could only be conducted in patients with complete outcome data. The missing data were due to the difficulty following patients for so long and to the difficulty of some patients with cognitive or communication impairment to respond to the HADS. While the patients assessed for depression had less severe strokes than those unable to respond to the HADS, the proportion of patients responding to the HADS showing clinical features of severe stroke was significant (urinary incontinence 26%, paresis 62% and disability 56%). Furthermore, the sensitivity analysis suggests that, while some sociodemographic groups are more likely to be missing than others, this had little impact when analysing the association between depression and health outcomes in the long term. The inability of some patients to be assessed for depression is a problem affecting most studies of depression after stroke.32 The estimation of clinical outcomes using scales introduces some limitations. Nonetheless, all scales have good performance16–19 ,22 ,23 and they allowed assessing a very large number of patients during a long follow-up which would have been unfeasible otherwise. The National Institute of Health Stroke Score (NIHSS), would have been a suitable tool to measure stroke severity.33 However, the SLSR collected NIHSS data from 2001, and this study presents estimates from 1997 to 2010. Nonetheless, the stroke severity variables included in the statistical models of this study were chosen for their clinical relevance, and their prognostic value, which was observed in two previous studies.24 ,25 The assessment of depression with a screening scale instead of a diagnostic tool, such as the DSM-IV,34 represents a limitation. However, the HAD scale was designed to assess depression in patients with medical problems, it has been widely used and it has good performance.18 Additionally, it would have been unfeasible to assess with DSM-IV, so many patients for such a long period of time.

    Clinical and research implications

    Depression after stroke deserves more clinical and research attention as it is an important outcome of stroke and an independent predictor of higher mortality, disability, anxiety and lower QoL in the long term.

    Further studies are needed to understand the mechanisms underlying the elevated morbidity and mortality in depressed patients. This would strengthen the evidence for causality and provide important prognostic information. It would also suggest innovative approaches for improving outcomes in patients with depression after stroke. Future clinical trials should investigate whether interventions for depression after stroke improve depression symptoms and also have a positive effect on mortality, disability, anxiety and QoL in the long term. The evidence on the beneficial or harmful effect of antidepressants, which are already being prescribed to many patients, is poor and contradicting.11 ,35 Observational studies of stroke patients with depression may help to clarify the safety of antidepressants and their effect on mood, functionality and mortality after stroke. Since recovering from depression does not improve prognosis up to 5 years after stroke, interventions for depression in the longer term should also be considered. If effective interventions on depression after stroke actually improve the long-term prognosis of stroke, a substantial change in clinical practice could be suggested. The results presented in this paper could inform studies of burden of depression in the future improving the estimation of loss of DALYs attributed to mental and behavioural disorders.36 The potential association between stroke location, size and depression, has been discussed in the literature.6 ,37 These variables are not routinely collected by the SLSR and could not be investigated in this study. The analysis of these associations remains a matter of further research. Finally, this study has tested some of the links between mental and physical health providing evidence that may be applicable in the context of other long-term conditions.

    Acknowledgments

    We thank all patients, and healthcare professionals involved. Particular thanks to field workers working since 1995 for the SLSR. We thank Toby Prevost for his input as medical statistician and Ruoling Chen for his contribution as clinical epidemiologist.

    References

    1. ↵
      1. Astrom M,
      2. Adolfsson R,
      3. Asplund K
      . Major depression in stroke patients. A 3-year longitudinal study. Stroke 1993;24:976–82.
      OpenUrlAbstract/FREE Full Text
    2. ↵
      1. Farner L,
      2. Wagle J,
      3. Engedal K,
      4. et al
      . Depressive symptoms in stroke patients: a 13 month follow-up study of patients referred to a rehabilitation unit. J Affect Disord 2010;127:211–18.
      OpenUrlCrossRefPubMed
    3. ↵
      1. House A,
      2. Dennis M,
      3. Mogridge L,
      4. et al
      . Mood disorders in the year after first stroke. Br J Psychiatry 1991;158:83–92.
      OpenUrlAbstract/FREE Full Text
    4. ↵
      1. Wade DT,
      2. Legh-Smith J,
      3. Hewer RA
      . Depressed mood after stroke. A community study of its frequency. Br J Psychiatry 1987;151:200–5.
      OpenUrlAbstract/FREE Full Text
    5. ↵
      1. Ayerbe L,
      2. Ayis S,
      3. Crichton S,
      4. et al
      . The natural history of depression up to 15 years after stroke: the South London Stroke Register. Stroke 2013;44:1105–10.
      OpenUrlAbstract/FREE Full Text
    6. ↵
      1. Ayerbe L,
      2. Ayis S,
      3. Wolfe CD,
      4. et al
      . Natural history, predictors and outcomes of depression after stroke: systematic review and meta-analysis. Br J Psychiatry 2013;202:14–21.
      OpenUrlAbstract/FREE Full Text
    7. ↵
      1. Kwok T,
      2. Lo RS,
      3. Wong E,
      4. et al
      . Quality of life of stroke survivors: a 1-year follow-up study. Arch Phys Med Rehabil 2006;87:1177–82.
      OpenUrlCrossRefPubMedWeb of Science
    8. ↵
      1. Morris PL,
      2. Robinson RG,
      3. Samuels J
      . Depression, introversion and mortality following stroke. Aust N Z J Psychiatry 1993;27:443–9.
      OpenUrlAbstract/FREE Full Text
    9. ↵
      1. Morris PLP,
      2. Robinson RG,
      3. Andrzejewski P,
      4. et al
      . Association of depression with 10-year poststroke mortality. Am J Psychiatry 1993;150:124–9.
      OpenUrlPubMedWeb of Science
    10. ↵
      1. Wulsin L,
      2. Alwell K,
      3. Moomaw CJ,
      4. et al
      . Lifetime depression and post-stroke depressive symptoms as predictors of stroke outcomes at 12 months. Stroke 2008;39:281.
      OpenUrl
    11. ↵
      1. Pan A,
      2. Sun Q,
      3. Okereke OI
      ,et al. Depression and risk of stroke morbidity and mortality: a meta-analysis and systematic review. JAMA 2011;306:1241–9.
      OpenUrlCrossRefPubMedWeb of Science
    12. ↵
      1. Ayerbe L,
      2. Ayis S,
      3. Rudd AG,
      4. et al
      . Natural history, predictors, and associations of depression 5 years after stroke: the South London Stroke Register. Stroke 2011;42:1907–11.
      OpenUrlAbstract/FREE Full Text
    13. ↵
      1. Sagen U,
      2. Finset A,
      3. Moum T,
      4. et al
      . Early detection of patients at risk for anxiety, depression and apathy after stroke. Gen Hosp Psychiatry 2010;32:80–5.
      OpenUrlCrossRefPubMed
    14. ↵
      1. Heuschmann PU,
      2. Grieve AP,
      3. Toschke AM,
      4. et al
      . Ethnic group disparities in 10-year trends in stroke incidence and vascular risk factors: the South London Stroke Register (SLSR). Stroke 2008;39:2204–10.
      OpenUrlAbstract/FREE Full Text
    15. ↵
      1. Hatano S
      . Experience from a multicentre stroke register: a preliminary report. Bull World Health Organ 1976;54:541–53.
      OpenUrlPubMedWeb of Science
    16. ↵
      1. Wolfe CD,
      2. Taub NA,
      3. Woodrow EJ,
      4. et al
      . Assessment of scales of disability and handicap for stroke patients. Stroke 1991;22:1242–4.
      OpenUrlAbstract/FREE Full Text
    17. ↵
      1. Quinn TJ,
      2. Dawson J,
      3. Walters MR,
      4. et al
      . Functional outcome measures in contemporary stroke trials. Int J Stroke 2009;4:200–5.
      OpenUrlCrossRefPubMedWeb of Science
    18. ↵
      1. Brennan C,
      2. Worrall-Davies A,
      3. McMillan D,
      4. et al
      . The Hospital Anxiety and Depression Scale: a diagnostic meta-analysis of case-finding ability. J Psychosom Res 2010;69:371–8.
      OpenUrlCrossRefPubMed
    19. ↵
      1. Aben I,
      2. Verhey F,
      3. Lousberg R,
      4. et al
      . 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;43:386–93.
      OpenUrlCrossRefPubMedWeb of Science
    20. ↵
      1. Jitapunkul S,
      2. Pillay I,
      3. Ebrahim S
      . The abbreviated mental test: its use and validity. Age ageing 1991;20:332–6.
      OpenUrlAbstract/FREE Full Text
    21. ↵
      1. Tombaugh TN,
      2. McIntyre NJ
      . The mini-mental state examination: a comprehensive review. J Am Geriatr Soc 1992;40:922–35.
      OpenUrlPubMedWeb of Science
    22. ↵
      1. Ware JE,
      2. Kosinski M
      . SF-36 physical and mental health summary scales: a user's manual. Boston: The Health Assessment Laboratory, 1994.
    23. ↵
      1. Ware JE,
      2. Kosinski M,
      3. Keller SD
      . SF-12: How to score the SF-12 Physical and Mental Health Summary Scales. Boston: The Health Institute, New England Medical Center, 1995.
    24. ↵
      1. Ayis SA,
      2. Coker B,
      3. Rudd AG,
      4. et al
      . Predicting independent survival after stroke: a European study for the development and validation of standardised stroke scales and prediction models of outcome. J Neurol Neurosurg Psychiatry 2013;84:288–96.
      OpenUrlAbstract/FREE Full Text
    25. ↵
      1. Wolfe CD,
      2. Smeeton NC,
      3. Coshall C,
      4. et al
      . Survival differences after stroke in a multiethnic population: follow-up study with the South London stroke register. BMJ 2005;331:431.
      OpenUrlAbstract/FREE Full Text
    26. ↵
      1. Blazer DG,
      2. Hybels CF,
      3. Pieper CF
      . The association of depression and mortality in elderly persons: a case for multiple, independent pathways. J Gerontol A Biol Sci Med Sci 2001;56:505–9.
      OpenUrlCrossRef
    27. ↵
      1. Mykletun A,
      2. Bjerkeset O,
      3. Dewey M,
      4. et al
      . Anxiety, depression, and cause-specific mortality: the HUNT study. Psychosom Med 2007;69:323–31.
      OpenUrlAbstract/FREE Full Text
    28. ↵
      1. Roy T,
      2. Lloyd CE
      . Epidemiology of depression and diabetes: a systematic review. J Affect Disorders 2012;142(Suppl):S8–21.
      OpenUrlCrossRefPubMed
    29. ↵
      1. Van der Kooy K,
      2. Van Hout H,
      3. Marwijk H,
      4. et al
      . Depression and the risk for cardiovascular diseases: systematic review and meta analysis. Int J Geriatr Psychiatry 2007;22:613–26.
      OpenUrlCrossRefPubMedWeb of Science
    30. ↵
      1. Turner-Stokes L,
      2. Hassan N
      . Depression after stroke: a review of the evidence base to inform the development of an integrated care pathway. Part 1: Diagnosis, frequency and impact. Clin Rehabil 2002;16:231–47.
      OpenUrlAbstract/FREE Full Text
    31. ↵
      1. Raju RS,
      2. Sarma PS,
      3. Pandian JD
      . Psychosocial Problems, Quality of Life, and Functional Independence Among Indian Stroke Survivors. Stroke 2010;41:2932–7.
      OpenUrlAbstract/FREE Full Text
    32. ↵
      1. Hackett ML,
      2. Anderson CS,
      3. House AO,
      4. et al
      . Interventions for Treating Depression After Stroke. Stroke 2009;40:487–8.
      OpenUrlAbstract/FREE Full Text
    33. ↵
      1. Goldstein LB,
      2. Samsa GP
      . Reliability of the National Institutes of Health Stroke Scale. Extension to non-neurologists in the context of a clinical trial. Stroke 1997;28:307–10.
      OpenUrlAbstract/FREE Full Text
    34. ↵
      American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders: DSM-IV. Washington, DC: American Psychiatric Association, 1994.
    35. ↵
      1. Mead GE,
      2. Hsieh CF,
      3. Lee R,
      4. et al
      . Selective serotonin reuptake inhibitors (SSRIs) for stroke recovery. Cochrane Database Syst Rev 2012;(11):CD009286. http://dx.doi.org/10.1002/14651858.CD009286.pub2 (accessed 15 Jul 2013).
    36. ↵
      1. Murray CJ,
      2. Vos T,
      3. Lozano R,
      4. et al
      . Disability-adjusted life years (DALYs) for 291 diseases and injuries in 21 regions, 1990–2010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet 2012;380:2197–223.
      OpenUrlCrossRefPubMed
    37. ↵
      1. Carson AJ,
      2. MacHale S,
      3. Allen K,
      4. et al
      . Depression after stroke and lesion location: a systematic review. Lancet 2000;356:122–6.
      OpenUrlCrossRefPubMedWeb of Science

    Footnotes

    • Contributors CDAW and AGR secured the funding. LA, SA, SC, CDAW and AGR designed the study. LA collected part of the data. LA, SA and SC conducted the analysis. LA, SA, SC, CDAW and AGR co-wrote the paper.

    • Funding The study was funded by Guy's and St Thomas’ Hospital Charity, The Stroke Association, Department of Health HQIP Grant, UK, National Institute for Health Research Programme Grant (RP-PG-0407-10184). The authors (CW) acknowledge financial support from the Department of Health via the National Institute for Health Research (NIHR) Biomedical Research Centre award to Guy's & St Thomas’ NHS Foundation Trust in partnership with King’s College London. CDAW is an NIHR Senior Investigator. This article presents independent research commissioned by the National Institute for Health Research (NIHR) under its Programme Grants for Applied Research funding scheme (RP-PG-0407-10184). The views expressed in this (article/presentation/poster etc) are those of the author(s) and not necessarily those of the NHS, the NIHR or the Department of Health.

    • Competing interests None.

    • Ethics approval The ethics committees of the following institutions, all based in London (UK), approved the study: Guy’s and St Thomas’ Hospital National Health Service Foundation Trust, King’s College Hospital Foundation, National Hospital for Nervous Diseases, Queen’s Square Hospital, St George’s Hospital and Westminster Hospital.

    • Patient consent Obtained.

    • Provenance and peer review Not commissioned; externally peer reviewed.