Depression and anxiety in chronic heart failure and chronic obstructive pulmonary disease: prevalence, relevance, clinical implications and management principles

Authors


Abstract

Objective

To review evidence regarding the prevalence, causation, clinical implications, aspects of healthcare utilisation and management of depression and anxiety in chronic heart failure and chronic obstructive pulmonary disease.

Design

A critical review of the literature (1994–2009).

Findings

The prevalence of depression and anxiety is high in both chronic obstructive pulmonary disease (8–80% depression; 6–74% anxiety) and chronic heart failure (10–60% depression; 11–45% anxiety). However, methodological weaknesses and the use of a wide range of diagnostic tools make it difficult to reach a consensus on rates of prevalence. Co-morbid depression and anxiety are associated with increased mortality and healthcare utilisation and impact upon functional disability and quality of life. Despite these negative consequences, the identification and management of co-morbid depression and anxiety in these two diseases is inadequate. There is some evidence for the positive role of pulmonary/cardiac rehabilitation and psychotherapy in the management of co-morbid depression and anxiety, however, this is insufficient to guide recommendations.

Conclusions

The high prevalence and associated increase in morbidity and mortality justifies future research regarding the management of anxiety and depression in both chronic heart failure and chronic obstructive pulmonary disease. Current evidence suggests that multi-faceted interventions such as pulmonary and cardiac rehabilitation may offer the best hope for improving outcomes for depression and anxiety. Copyright © 2009 John Wiley & Sons, Ltd.

Introduction

Chronic disease accounts for a substantial amount of the global health burden (Moussavi et al., 2007). Chronic disease with co-morbid mental disorder, notably depression and anxiety, add considerably to this burden. Data from the World Health Survey indicated that co-morbid depression incrementally worsens health compared with depression alone, any major chronic disease alone, and any combination of chronic diseases without depression (Moussavi et al., 2007).

COPD and CHF are highly prevalent, especially among older people. COPD is responsible for substantial human and economic burden throughout the world (Maurer et al., 2008). Worldwide, COPD was the fifth leading cause of death in 2001 (Pauwels and Rabe, 2004) and is projected to be the third leading cause of death by 2020 (Murray and Lopez, 1997). Recent findings suggest a 5-year mortality rate following an acute exacerbation of COPD as high as 70% (Ai-Ping et al., 2005). CHF has a 5-year life expectancy from diagnosis, comparable to that of many cancers (Davis et al., 2000). In the UK, CHF accounts for approximately 5% of all hospital admissions (MacMahon and Lip, 2002) and COPD for 12.5% (National Collaborating Centre for Chronic Conditions, 2004). In the developed world, as a result of an ageing population the incidence of both COPD and CHF are projected to increase (MacMahon and Lip, 2002).

Recent research on the psychological consequences of chronic illness underlines the substantial adverse impact of disease on patients, their families and society. The two most common and least-treated psychological co-morbidities are anxiety and depression. Despite an increase in awareness of the impact of psychological co-morbidity, anxiety and depression are often undetected and untreated in both COPD (Maurer et al., 2008) and CHF (Rutledge et al., 2006) patients. This paper reviews evidence regarding the prevalence, causation, and aspects of healthcare utilisation and management of these two disorders in patients with COPD and CHF.

Prevalence and causative factors

There are a number of tools validated to screen anxiety and depressive symptoms in patients with COPD and CHF (Tables 1–4). Although these screening tools can recognise the presence of depression, they do not all provide a clinical diagnosis. A clinical diagnosis of major depression is defined by the Diagnostic and Statistical Manual of Mental Disorders, 4th edition (DSM-IV) criteria (American Psychiatric Association, 1994) and is often identified through the use of a structured interview (e.g. SCID-I). This distinction is important as many studies examining the prevalence of anxiety and depression use self-report scales which report rates of distress levels, rather than an actual psychiatric diagnosis.

Table 1. Prevalence of depression in COPD
AuthorNDiagnostic tool% Reporting depression
  • ADIS-IV, Anxiety Disorders Interview Schedule - Version 4; BASDEC, Brief Assessment Schedule for Depression Cards; BDI, Beck Depression Inventory; BDI-II, Beck Depression Inventory-II; CES-D, Centre for Epidemiological Studies Depression Scale; CES-D8, Centre for Epidemiological Studies Depression Scale-8; ESAS, Edmonton Symptom Assessment Scale; GDS, Geriatric Depression Scale; HADS, Hospital Anxiety and Depression Scale; ICD-10, International Statistical Classification of Diseases and Related Health Problems 10th Revision; SCID-I, Structured Clinical Interview; SCL-90-R, Symptom Checklist-90-R; WMH-CIDI, The World Mental Health Composite International Diagnostic Interview.

  • a

    Inidicates a clinical diagnosis.

Laurin et al. (2007)101ADIS-IV17a
Yohannes et al. (1998)96BASDEC46
Yohannes et al. (2005)100BASDEC56
Carvalho et al. (2007)49BDI29
De Voogd et al. (2009)121BDI20
Fan et al. (2007)610BDI41
Kunik et al. (2005)1334BDI-II80
Van Manen et al. (2002)162CES-D25
Schane et al. (2008)1736CES-D840
Walke et al. (2007)74ESAS16
Kim et al. (2000)43GDS40
Lacasse et al. (2001)109GDS57
Ng et al. (2009)189GDS23
Barr et al. (2009)1003GP Clinical diagnosis37a
Patten and Williams (2007)518GP Clinical diagnosis8a
Cleland et al. (2007)106HADS21
Dowson et al. (2001)79HADS28
Funk et al. (2009)122HADS52
Gudmundsson et al. (2005)406HADS29
Gudmundsson et al. (2006)416HADS8
Lewis et al. (2007)182HADS35
Ng et al. (2007)376HADS44
Withers et al. (1999)95HADS15
Xu et al. (2008)491HADS23
Aghanwa and Erhabors (2001)30ICD-1017
Stage et al. (2003)49ICD-1032
Kunik et al. (2005)204SCID-I23a
Kuhl et al. (2008)143SCL-90-R12
Patten and Williams (2007)518WMH-CIDI8#
Di Marco et al. (2006)202Zung28
Table 2. Prevalence of anxiety in COPD
AuthorNDiagnostic tool% Reporting anxiety
  • ADIS-IV; Anxiety Disorders Interview Schedule for DSM-IV; BAI; Beck Anxiety Inventory; CIDI, Composite International Diagnostic Interview; ESAS, Edmonton Symptom Assessment Scale; F-DIPS, Diagnostic Interview for Psychiatric Disorders-Research Version; GMSS, Geriatric Mental State Schedule; HADS, Hospital Anxiety and Depression Scale; ICD-10, International Statistical Classification of Diseases and Related Health Problems 10th Revision; MHI-15, Mental Health Inventory-15; SCID-I, Structured Clinical Interview; SCL-90-R, Symptom Checklist-90-R; STAI, State-Trait Anxiety Inventory.

  • a

    Inidicates a clinical diagnosis.

Laurin et al. (2007)101ADIS-IV46a
Kim et al. (2000)43BAI33
Aydin and Ulusahin (2001)38CIDI16a
Walke et al. (2007)74ESAS32
Vogele and von Leupoldt (2008)20F-DIPS55a
Yohannes et al. (2000)137GMS18a
Cleland et al. (2007)98HADS33
Funk et al. (2009)122HADS49
Gudmundsson et al. (2005)406HADS41
Gudmundsson et al. (2006)416HADS20
Gurney-Smith et al. (2002)30HADS53
Lewis et al. (2007)182HADS25
Sutton et al. (1999)37HADS57
Withers et al. (1999)95HADS29
Xu et al. (2008)491HADS10
Aghanwa and Erhabors, (2001)30ICD-1010
Stage et al. (2003)49ICD-1035
Blinderman et al. (2009)100MHI-1551
Kunik et al. (2005)204SCID-I19a
Kuhl et al. (2008)143SCL-90-R6
Carvalho et al. (2007)49STAI74
Di Marco et al. (2006)202STAI19
Table 3. Prevalence of depression in CHF
AuthorNDiagnostic tool% Reporting depression
  • BDI, Beck Depression Inventory; BDI-II, Beck Depression Inventory-II; CES-D, Centre for Epidemiological Studies Depression Scale; CES-D-SF, Centre for Epidemiological Studies Depression Scale- Short Form; DSM-IV, Diagnostic and Statistical Manual of Mental Disorders-IV; ESAS, Edmonton Symptom Assessment Scale; GDS, Geriatric Depression Scale; HADS, Hospital Anxiety and Depression Scale; HDS, Hamilton Depression Scale; MHI, Mental Health Index; MOS-D, Medical Outcomes Study-Depression; SCID-I, Structured Clinical Interview; SCID-NP, Structured Clinical Interview-Non Patient Edition.

  • a

    Inidicates a clinical diagnosis.

Berkman et al. (2003)2481BDI39
Freedland et al. (2003)612BDI51
Friedmann et al. (2006)153BDI36
Gottlieb et al. (2004)155BDI48
Jiang et al. (2001)374BDI35
Jiang et al. (2004)291BDI30
Schiffer et al. (2008)149BDI32
Schweitzer et al. (2007)102BDI33
Westlake et al. (2005)200BDI17
Tsuchihashi-Makaya et al. (2009)139BDI-II37
Havranek et al. (1999)45CES-D24
Koenig, (1998)107CES-D17
Lesman-Leegte et al. (2006)572CES-D41
Lesman-Leegte et al. ((2009))781CES-D39
Skotzko et al. (2000)33CES-D42
Sullivan et al. (2009)208CES-D20
Turvey et al. (2002)199CES-D11
Friedman and Griffin, (2001)170CES-D-SF30
Freedland et al. (2003)682DSM-IV20a
Ola et al. (2006)100DSM-IV28a
Walke et al. (2007)74ESAS12
Fulop et al. (2003)203GDS36
Vaccarino et al. (2001)391GDS22
Faris et al. (2002)396GP Clinical diagnosis21a
Junger et al. (2005)209HADS30
Yu et al. (2009)95HADS10
Sullivan et al. (2004)142HDS20
Sauve et al. (2009)50MHI26
Havranek et al. (2004)245MOS-D21
Rumsfeld et al. (2003)460MOS-D30
Haworth et al. (2005)100SCID-I29a
Fulop et al. (2003)203SCID-NP22a
Pihl et al. (2005)47Zung60
Table 4. Prevalence of anxiety in CHF
AuthorNDiagnostic tool% Reporting anxiety
  • ESAS, Edmonton Symptom Assessment Scale; HADS, Hospital Anxiety and Depression Scale; MHI, Mental Health Index; POMS, Profile Of Mood States; SCID-I, Structured Clinical Interview; STAI, State-Trait Anxiety Inventory.

  • a

    Inidicates a clinical diagnosis.

Walke et al. (2007)59ESAS12
Yu et al. (2009)95HADS11
Sauve et al. (2009)50MHI34
Sullivan et al. (2009)208POMS11
Haworth et al. (2005)100SCID-I18a
Tsuchihashi-Makaya et al. (2009)139STAI37
Schweitzer et al. (2007)102STAI31
Friedmann et al. (2006)153STAI45

COPD

Estimates of the prevalence of depression in COPD range widely. In a systematic review of 17 studies (Hynninen et al., 2005) the prevalence of depression ranged from 7 to 79%. Putman-Casdorph and McCrone (2009) report a prevalence of depression of 16–88%. Similar data was found in this review where thirty studies (since 1998) were identified with a prevalence of depression of 8–80% (Table 1). In addition, studies which utilised a validated clinical interview demonstrate a prevalence of clinical depression as high as 37%. The large variance in this data may be attributed to a number of factors including the use of different diagnostic tools, different degrees of illness severity, and settings of the studies (e.g. hospital or community based patients). The large range complicates diagnosis as does the overlap of depressive symptoms with ones caused by severe COPD including fatigue and sleep disturbance (Maurer et al., 2008) (Figure 1). Yohannes et al. (2000) conducted a meta-analysis of 13 studies (n = 900) and the pooled analysis found that the prevalence of depression in patients with COPD was 40% (95% CI: 36–44%). This may be considered a benchmark figure for clinicians.

Figure 1.

Overlap of depressive symptoms in relation to COPD and CHF.

Hynninen et al. (2005) identified 19 studies that investigated the prevalence of anxiety in COPD patients. The prevalence of anxiety varied from 10 to 100%. This review found 22 studies since 1999 which report the prevalence of anxiety (Table 2). The prevalence of anxious symptomology was 6–74% whilst clinical anxiety was found in up to 55% of COPD patients. In a similar vein to depression, the wide variety of screening tools and classifications make it difficult to reach a consensus regarding the prevalence of anxiety in patients with COPD. An earlier meta-analysis found a prevalence of anxiety in patients with moderate to severe COPD of 36% (95% CI: 31–46%) (Yohannes et al., 2000). In COPD, anxiety is often secondary to depressive disorder. In the study by Yohannes et al. (2000), the prevalence of anxiety was found to be 37% in those with COPD-related depression compared to 5% in the non-depressed patient group (Yohannes et al., 2000).

CHF

The symptoms of CHF overlap with those of depression, often making the latter difficult to diagnose (Figure 1). Both CHF and depression are characterised by fatigue, loss of energy, poor appetite, sleep disturbances, psychomotor retardation and concentration deficits (Simon and von Korff, 2006).

A Cochrane review (Lane et al., 2005) evaluated eight studies which examined the prevalence of depressive symptoms in CHF. Reported prevalence varied considerably with data suggesting that depressive symptoms may be present in up to 85% of CHF patients. Additionally, major clinical depression was found in up to 26%. This is significantly higher than levels of depression found in the general population (2.1%) (NICE, 2004a,b; Scherer et al., 2007). A search of CHF studies of the past decade found 33 studies which reported the prevalence of depression (Table 3). Prevalence ranged widely between 10 and 60%, again the variation in prevalence are likely to be similar to the reasons outlined earlier for COPD. Lesman-Leegte et al. (2009) examined the prevalence of depression in a community sample of patients with CHF (n = 781) compared to an age- and gender matched control. Depressive symptoms (measured by the CES-D) in the CHF patients were almost double that of the healthy matched sample (39% vs. 21%). A recent meta-analysis by Rutledge et al. (2006) indicated that the prevalence of depression in in-patients with CHF was 35–37%, three times that seen in out-patients (O'Connor and Joynt, 2004).

Few studies have examined co-morbid anxiety in CHF patients. This review identified just eight studies which report on the prevalence of anxiety (Table 4). Of 100 out-patients with CHF, 18.4% had clinical anxiety (measured using SCID-I) (Haworth et al., 2005) and it is likely that anxiety disorders are under-diagnosed (MacMahon and Lip, 2002).

Mechanisms

The exact causes of or mechanisms that lead to anxiety and depression in patients with COPD and CHF are poorly understood. Maurer et al. (2008) propose a number of risk factors associated with depression and anxiety in COPD, including physical disability, chronic hypoxia, low body mass index, severe dyspnoea, lower social class and poor social support. A study by Koenig (2006) investigated the aetiology and management differences between depressed COPD patients (n = 527) and CHF patients (n = 174) using the DSM-IV criteria. Using a multivariate analysis, depression in COPD patients was found to be associated with the presence of stressful life events, psychiatric history and co-morbid psychiatric illness. In contrast, depression in CHF patients was related to greater medical co-morbidity and severe medical illness. Haworth et al. (2005) investigated the predictors of anxiety and depression in patients with CHF (n = 100) using the SCID-I. In a logistic regression, depression and anxiety in patients with CHF was related to previous psychiatric history and a higher NYHA (New York Heart Association) class. Additionally, anxiety in CHF patients was related to co-morbid physical illness (diabetes and angina).

Clinical implications

Functional disability

Depression is associated with a decrease in exercise capacity and poor health perception in patients with chronic illness (Putman-Casdorph and McCrone, 2009). A recent large study of chronic diseases which included patients with COPD (n = 1681) and CHF (n = 391) found that co-morbid depressive symptoms (measured with the CIDI-SF) correlated significantly with higher levels of functional disability. Patients with chronic illness and depression were 2.5 times more likely to have a functional disability than non-depressed patients with chronic illness (Egede, 2007).

In patients with COPD, functional capacity was associated more with depression and anxiety than with traditional physiological markers such as lung function (Graydon and Ross, 1995; Yohannes et al., 2005). In a cross-sectional analysis of 1252 patients with COPD, Felker et al. (2001) found that patients with COPD and depressive symptoms reported significantly more impaired functioning and worse health status when compared to those patients without depressive symptoms. In addition, Kim et al. (2000) reported that anxiety and depression had a greater impact on functional status than COPD severity. Depression in patients with COPD and CHF has also been found to be an independent risk factor for increased physical disability (Yohannes et al., 1998; Ola et al., 2006) and elevated mortality (Yohannes et al., 2005; Albert et al., 2009; de Voogd et al., 2009).

Quality of life (QoL)

Psychological well being is a powerful predictor of QoL in patients with COPD (Yohannes et al., 1998). Evidence indicates that both depression (Yohannes et al., 1998) and anxiety (Brenes, 2003) are significantly correlated with impaired QoL in patients with COPD rather than lung function. A single-blinded, match-controlled study by Yohannes et al., (1998) examined the predictive factors for QoL in elderly patients (n = 96) with COPD compared to an elderly, healthy matched sample with normal lung function (n = 55). Multiple regression analysis revealed that depression (measured with the BASDEC) along with activities of daily living (ADL) function were the most important determinants of QoL. Interestingly, this study also found that lung-function was not a significant predictor of QoL.

Similar findings have also been reported in patients with CHF. Depressive symptoms were associated with impaired QoL more than severity of cardiac function or functional impairment in patients with CHF (Havranek et al., 2004; Lee et al., 2005).

Depression and anxiety have a major impact on psychological function and social interaction in patients with COPD and CHF. Recent findings suggest that depression in patients with COPD and CHF is often marked by feelings of hopelessness and pessimism, reduced sleep, decreased appetite, increased lethargy, difficulty with concentration and social withdrawal (Ola et al., 2006; Emery et al., 2008; Maurer et al., 2008). Anxiety in patients with COPD and CHF is associated with ‘chronic worry’, nausea, fear of loosing control and physiological symptoms such as tachycardia, sweating and dyspnoea (Emery et al., 2008; Tsuchihashi-Makaya et al., 2009).

In summary, psychological well being plays a powerful role in the QoL of patients with COPD and CHF. Specifically, anxiety and depression may have a greater role in QoL than disease severity and physical disability.

Healthcare utilisation

It is estimated that patients with COPD with co-morbid depression have 50–100% higher medical costs than those without depression, even when controlling for demographic factors and severity of medical illness (Maurer et al., 2008). COPD patients with co-morbid anxiety and depressive symptoms exhibited elevated annual rate of exacerbations (relative risk = 1.56) compared with COPD patients without these symptoms in an out-patient setting (Laurin et al., 2009).

A community-based study of healthcare utilisation in a sample controlled for age, chronic disease and functional limitation, found that anxiety in COPD patients was associated with increased general practitioner and emergency care visits (Kim et al., 2000). Additionally, a large, prospective, multi-centre study (n = 416) found that in COPD patients with low health status who were discharged after an acute exacerbation, the risk of rehospitalisation was far higher in patients with anxiety compared to patients without anxiety (Gudmundsson et al., 2005). More recently, a study on 491 stable COPD patients with case-level depression (HADS depression score ≥ 11) was associated with an increased risk of exacerbations and hospitalisations (Xu et al., 2008).

Similar findings have been reported in patients with CHF. For example, Sullivan et al. (2002) found that depression in CHF is associated with significantly increased medical costs. After adjusting for age, sex, medical co-morbidity and length of stay at hospital, costs were found to be 26% higher in patients with CHF who had been prescribed anti-depressants (but did not necessarily have clinically diagnosed depression). Furthermore, CHF patients who had a clinical diagnosis of depression had 29% higher medical costs than non-depressed patients with CHF. Depression may also play a significant role in increased hospital re-admissions and longer hospital stays in patients with CHF (Albert et al., 2009). Patients with CHF and major depression (measured by DSM-IV) were twice as likely to be re-admitted in comparison with non-depressed patients up to 1 year after discharge (Jiang et al., 2001).

Interactions between depression and CHF/COPD

The relationship between chronic disease and depression is multidimensional. An earlier review (Yohannes, 2008) outlined the multiple interactions and pathways through which depression can manifest in patients with COPD. Similar complex interactions have been proposed for the development of co-morbid depression in patients with CHF (Figure 2). The physical, psychological and social consequences of depression negatively impact on CHF and CHF symptoms generate depression, especially in those with risk factors. Important risk factors are lower social class, worse NHYA class and a previous history of mental ill-health (Haworth et al., 2005). Depression itself may also lead to negative health outcomes similar to those directly caused by CHF. Co-morbid CHF and depression have an additive effect on poor health outcomes and this may at least partly explain the increased mortality found in patients with CHF (Rutledge et al., 2006).

Figure 2.

Interactions and consequences of depression in CHF patients.

Patients with CHF often have cerebrovascular disease and where this affects small vessel function vascular depression may ensure. This features apathy, poor executive cognitive function and psychomotor impairment (Alexopoulos et al., 1997). Whether chronic hypoxia from COPD may produce a similar picture is unclear but there is evidence that brain hypoperfusion may predispose to late-life depression (Baldwin, 2005). Lastly, depression and vascular disease are in a two way relationship, each making the other worse (Teper and O'Brien, 2007).

Management

It is likely that depression and anxiety in COPD and CHF are under-diagnosed. This may be in part due to the diagnostic complexities as discussed, but there are also barriers (from both patients and healthcare professionals) to the recognition and treatment of these two co-morbidities at several levels. These include patient perception of depression and lack of confidence in detection and management of depression by healthcare professionals (Table 5).

Table 5. Potential barriers to the treatment of anxiety and depression for COPD and CHF (Yohannes et al., 2001; Sirey et al., 2007; Maurer et al., 2008)
PatientsHealth professionals
(1) Fear of side effects of medications(1) Inadequate awareness of symptoms of anxiety and depression
(2) Stigma of psychological diseases(2) Stigma of psychological disease
(3) Guilt and embarrassment(3) Inadequate knowledge of management
(4) Unwillingness to discuss issues of anxiety and depression with friends, family and health professionals(4) Uncertainty about the correct information to provide patient
(5) Reluctance to take additional medication(5) Lack of clinical guidelines regarding effective management
(6) Lack of knowledge about the symptoms and management of depression and anxiety(6) Unwillingness to discuss potentially difficult issues
(7) Masking of psychological symptoms with physical symptoms e.g. excess response dyspnoea(7) Insufficient time in which to assess patients
(8) Patient or family reluctance to accept diagnosis of depression(8) Difficulty distinguishing depressive symptoms with the physical problems
 (9) Lack of access to mental health specialists
 (10) Incomplete knowledge of diagnostic criteria

Despite high levels of anxiety and depression, there is a paucity of evidence regarding their effective management. This is reflected in the inadequate management that many patients receive. For example, in one study, less than 25% of COPD patients with moderate-to-severe anxiety or depressive symptoms received relevant interventions (Kim et al., 2000).

Medication

There is little data regarding the efficacy of antidepressants in the treatment of anxiety and depression in patients with COPD or CHF (Glassman et al., 2002; Maurer et al., 2008). Furthermore, such patients are usually already taking multiple medications and may be reluctant to accept antidepressant drugs. Thus, a feasibility trial of antidepressant therapy in patients with COPD found that the majority (72%) of depressed patients refused antidepressant therapy (Yohannes et al., 2001). Other treatments and modalities are therefore important. Traditional management of COPD and CHF often incorporates a multi-faceted approach in the form of pulmonary and cardiac rehabilitation and these will be discussed.

Pulmonary rehabilitation

A recent review and meta-analysis by Coventry and Hind (2007) suggest that a pulmonary rehabilitation (PR) programme that incorporates exercise, education and social support significantly reduced anxiety and depression compared with standard care in patients with COPD. It is unclear which aspects of the PR programme have the greatest impact upon anxiety and depression symptoms.

There is evidence that exercise is beneficial in the treatment of co-morbid anxiety and depression in chronic illness. Exercise may facilitate the release of endogenous opiates and increase production of serotonin and dopamine which are often found to be low in depressed individuals. Exercise may also have additional psychosocial benefits such as improved self-efficacy, increased social support via group networks and quality of life (Wielenga et al., 1998; Emery et al., 2008). Furthermore, regular exercise may have a desensitising effect on anxiety symptoms such as dyspnoea and hyperventilation (Emery et al., 2008).

Exercise is a ‘hall-mark’ of an effective PR programme. Guell et al., (2006) found that a 16 week exercise intervention (twice weekly 30-min cycling and breathing training) in COPD patients can decrease anxiety and depression when compared to standard treatment. Emery et al. (2008) compared an intervention of exercise, stress management and education with a control group (education and stress management only) and a waiting list group. The randomised, controlled trial (n = 79) found that anxiety symptoms were reduced (measured by SCL-90-R and STAI) in the exercise intervention group but not in the control groups.

The impact of education within a PR intervention is less clear. A review by Coventry and Hind (2007) concluded that education alone is insufficient to bring about reductions in anxiety and depression in patients with COPD. Nevertheless, the authors state that education is a key part of a successful PR programme. It is possible that this educational component may help patients to understand the benefits of PR and subsequently enable them to take control of their symptoms. This is supported by a recent study which compared COPD education and cognitive behavioural therapy (CBT) in COPD patients with co-morbid anxiety and/or depression (Kunik et al., 2008). The study enrolled 238 COPD patients with clinical anxiety (BAI) and depression (BDI-II) and both were significantly reduced after the 8-week intervention (education and group CBT). Kunik et al. (2008) suggest that the social interaction and support that takes place during educational sessions may help to explain the positive changes in anxiety and depression in patients with COPD.

The exact mechanisms and the long-term benefits of PR are still poorly understood. Further research is needed to examine the effects of specific components of PR on mood.

Cardiac rehabilitation

Until the last decade, patients with CHF were not encouraged to participate in exercise for the fear of decompensation (Kulcu et al., 2007). Today, the benefits of a cardiac rehabilitation (CR) programme extend to many clinical populations who experience limited exercise tolerance including patients with CHF. The benefits of CR are well documented and may include increased exercise tolerance, improved functional capacity and better QoL in patients with CHF (Shen et al., 2005).

A review by Shen et al. (2005) found 24 studies that examined the efficacy of CR on depression. Their findings were inconclusive. All but three studies demonstrated reduced depression as a result of a CR intervention. However, many of these studies did not have control groups and the samples were predominately male. Despite this, some studies were of large cohorts with positive findings in diverse groups of participants including the elderly and diabetics. A study which compared community and hospital CR programmes demonstrated that depression was amenable to CR interventions in both hospital and community samples (Benzer et al., 2007). Two hundred and sixteen patients were randomly allocated to receive either a 4-week in-patient CR programme, a 3-month out-patient CR programme or usual care. Results indicated that depression was significantly reduced in both in-patient and out-patient CR programmes. However, the long term effects (>3-months) were only present in the out-patient intervention group. Further studies of large cohorts and controlled interventions are needed to confirm the possible benefits of CR for depression.

CR may also reduce symptoms of anxiety in patients with CHF. In a systematic review, Shen et al. (2005) identified 27 studies that examined the effects of CR on anxiety. Twenty five of the studies found that CR reduced anxiety symptoms. However, the generalisability of these findings was compromised by relatively small sample sizes and lack of control groups. In a study that examined the effects of CR in CHF patients (n = 1077), Turner et al. (2003) demonstrated that anxiety symptoms (measured with HADS) were significantly decreased in a CR intervention which included exercise, education, relaxation and stress management. However again, there was no control group, raising the possibility that any reductions in anxiety may have been the result of the passage of time rather than the CR intervention itself. Nevertheless, data from both uncontrolled studies and studies with control groups suggest that CR incorporating exercise, education and relaxation may help to decrease anxiety symptoms in patients with CHF (Shen et al., 2005).

At present there is little data available addressing possible long-term benefits of CR on anxiety symptoms (>12-months). A previous study by Oldridge et al. (1995) found that patients who completed a CR intervention had greater reductions in anxiety than a control group at eight weeks, but by 12 months improvements were similar in both groups.

Similarly to PR, it is unclear which elements of the CR intervention are responsible for the potential decrease in depression and anxiety symptoms. Few studies have examined the benefits of individual components of CR with the exception of exercise. Exercise alone was found to increase self-efficacy and decrease depression in a 12-week CR programme (Evon and Burns, 2004). Furthermore, Kulcu et al. (2007) found that an 8-week aerobic exercise programme (thrice weekly) decreased anxiety (measured with STAI) and depression (measured with BDI) in patients with CHF. Conversely, a recent in-hospital combined endurance and muscle strength training programme found no impact on measures of anxiety and depression (Miche et al., 2008).

Psychological interventions

There is a growing interest in the role of psychotherapy (including CBT) in the management of anxiety and depression in patients with COPD and CHF. The cognitive model proposes that patients become anxious in response to physiological events such as dyspnoea and misinterpret or catastrophise them, leading to extreme negative cognitions, increases in physiological arousal and poor behavioural coping (Rose et al., 2002). A few studies have examined the effects of CBT compared to education in patients with COPD. In a single-blinded RCT (n = 56), Kunik et al., (2001) examined the effects of 2 h of group CBT with a 6-week follow-up including weekly calls and homework versus a group which received COPD education followed by weekly calls. Their findings showed that those who received CBT reported a clinically significant reduction in depressive (measured with the GDS) and anxiety (measured with the BAI) symptoms compared to the education group.

The additional benefits of psychotherapy combined with a PR programme have also been examined in patients with COPD. In a small, RCT (n = 30), de Goody and de Goody (2003) examined the effects of a comprehensive PR programme (including 24 sessions of physical exercise, 24 sessions of physiotherapy, 12 psychotherapy sessions and 3 educational sessions) versus a control group which did not have the psychotherapy component. They showed that patients who received additional psychotherapy with PR had a clinically significant change in depression (measured with the BDI) and anxiety (measured with the BAI) compared to the PR group alone.

There is also little data on the effects of psychological interventions for co-morbid depression and anxiety in CHF. A Cochrane review by Lane et al. (2005) found no completed RCTs of psychological interventions for depression in patients with CHF. One small, non-randomised, controlled pilot study (n = 33) examined the effects of psychological training (Freeze-Frame stress management programme) on depression in elderly patients with mild to moderate heart failure (Luskin et al., 2002). The intervention consisted of eight 75-min sessions from a psychotherapist compared to a waiting-list control group. Their findings showed that those who received psychological training had clinically significant improvement of depression. More recently, a large RCT by Sullivan et al. (2009) investigated the benefits of an 8-week psycho-educational intervention on depression in patients with CHF (n = 208). Those who received the psycho-educational intervention (consisting of mindfulness meditation, coping skills and support group discussion) were reported to have a clinically significant reduction in depression (measured with the CES-D) and anxiety (measured with the POMS) compared to a control group that received standard care.

The effect of CBT in CHF patients with depression and anxiety has received even less attention. A meta-analysis (Rutledge et al., 2006) and a recent systematic review by Dekker (2008) both identified just one study which examined the effects of CBT on depression and anxiety in patients with CHF (Kostis et al., 1994). This randomised controlled study investigated the effects of a non-pharmacological intervention which incorporated exercise training (1 h, thrice weekly), CBT (90 min, twice weekly) and dietary control (reduced salt, fat and alcohol intake) versus digoxin and versus a placebo for 12-weeks. The non-pharmacological group displayed clinically significant reductions in depression (measured with the BDI) and anxiety (measured with the HADS) compared with the digoxin and placebo groups. However, it is not possible to ascertain which elements of the non-pharmacological intervention were the most effective.

Thus, as in COPD, there is a paucity of robust evidence demonstrating the effectiveness of psychological interventions for patients with CHF. Generally, studies had poor methodologies, often with small sample sizes and inadequate randomisation. The limited available evidence suggests that patients with CHF may benefit from CBT or some other similar type of psychological intervention aimed at reducing their symptoms of depression (Lane et al., 2005).

Further research is required to establish the cost-benefit of psychological treatments. Furthermore, future studies should endeavour to select patients who have a clinical diagnosis of depression or anxiety so that sufficient comparisons can be made between pre and post intervention.

Other multi-faceted interventions: collaborative care

In COPD and CHF, multi-faceted interventions probably offer the best hope for improving outcomes for depression and anxiety. Pulmonary and cardiac rehabilitation are one means but are only modestly successful. Another approach that is promising is that of collaborative care which has been extensively investigated in primary care setting for patients with depression, including older people as a result of the large Improving Mood-Promoting Access to Collaborative Treatment (IMPACT) study (Unűtzer et al., 2002). This involves the use of a depression care manager (usually a nurse, psychologist or social worker) who coordinates the care, including medication concordance, a particular problem with depressed COPD and cardiac patients. In the IMPACT study Problem-Solving Treatment was the psychological modality. Although harder to standardise than CBT, it is easier to train personnel and likely to be more accessible. The World Health Organisation (WHO) concluded that collaborative care improves patient outcomes in diabetes, depression, chronic heart failure and cardiovascular diseases (although not necessarily mortality in the latter case) (Velasco-Garrido et al., 2003). In the IMPACT study, physical co-morbidity did not adversely affect outcomes (Callaghan et al., 2005). Bringing together the skills of this approach and that of rehabilitation in a research programme may offer a viable way forward to lessen the burden of affective morbidity in COPD and CHF.

Key Points

  • Depression and anxiety are common in patients with COPD and CHF.

  • Untreated and undiagnosed anxiety and depression in COPD and CHF have serious negative effects on quality of life.

  • Combining the principles of chronic disease management and those from the field of rehabilitation may benefit in the treatment of depression and anxiety in patients with CHF and COPD.

Conclusions

Despite ranging estimates, it is evident that co-morbid anxiety and depression are common in both COPD and CHF. This has major clinical implications as the additive effects of anxiety and depression in COPD and CHF lead to increased mortality, morbidity and healthcare burden, and have serious negative effects on quality of life. Despite this, there is still a paucity of evidence for the identification and management of co-morbid anxiety and depression. Combining the principles of chronic disease management with those from the field of rehabilitation may offer the optimum way forward although further research is warranted. There is some evidence to suggest that both PR and CR are beneficial in ameliorating anxiety and depressive symptoms in patients with COPD and CHF in the short-term. However, the long-term benefits in reducing healthcare utilisation and improving psychological well-being, physical disability and quality of life are unknown.

Conflict of interest

None known.

Ancillary