Prevalence, Contribution to Disease Burden and Management of Comorbid Depression and Anxiety in Chronic Obstructive Pulmonary Disease: A Narrative Review

Abstract

Mental health issues such as depression and anxiety are common comorbidities of individuals suffering from chronic obstructive pulmonary disease (COPD), though they are frequently underdiagnosed and consequently undertreated. To that end we sought to identify the impact of co-morbid anxiety and depression in COPD patients and evaluate recent evidence regarding the management of such cases. A Pubmed search was conducted for relevant original articles with emphasis on the past 5 years. The studies we identified indicate that psychiatric comorbidity negatively impacts the prognosis of COPD, as it is associated with reduced adherence to treatment, reduced physical activity and a general reduction in quality of life, in turn leading to more frequent exacerbations and increased severity of exacerbations (which are more likely to require hospitalization), resulting in increased mortality. Despite the evidence showing a high prevalence and exceedingly negative impact of depression and anxiety in patients with COPD, very few clinical trials (for both pharmacological and psychological treatments) with small sample sizes have been conducted in this population. As treatment for co-morbid mental health conditions may reduce mortality, interventions to ensure prompt identification of mental health issues and subsequent initiation of treatment are warranted.

Keywords:

• Antidepressant
• anxiety
• COPD
• depression
• mental health
• quality of life
• psychotherapy
• smoking
• smoking cessation

Introduction

As it is true of most chronic conditions, chronic obstructive pulmonary disease (COPD) negatively affects quality of life and the disease burden is further increased by multiple comorbidities, including coronary artery disease and cerebrovascular disease. Furthermore, patients with COPD are more likely to develop lung and other cancers due to the effects of smoking [1]. Their mental health is also severely affected by the condition. Depression and anxiety disorders may be more common in patients suffering from COPD compared to the general population and they may be associated with a detrimental effect of prognosis. This interplay between mental health and COPD has become a subject of intense study in recent years.

The chief symptoms of COPD are productive cough and dyspnea, both of which may cause significant emotional distress and interfere with daily functioning. Dyspnea in particular is intimately linked with anxiety disorders and anxious individuals may complain of dyspnea in the absence of any organic cause, usually in the context of panic disorder and agoraphobia [2]. A link between depression and chronic organic conditions has also been established, as the diminished quality of life due to a physical illness may predispose individuals to the development of depression, which in turn may reduce their motivation to adhere to treatment, leading to further quality of life deterioration [3]. On the other hand, mental illness has been associated with a higher likelihood of smoking and may also lead to the adoption of a high-risk lifestyle (poor eating habits, minimal physical activity, obesity, poor sleep quality and avoidance of indicated visits to the physician), thus pre-disposing to the development of physical illness. In the case of established comorbidities, mental health issues may be associated with a worse prognosis due to treatment non-adherence, particularly for interventions which require active participation on behalf of the patient (complex medication regimens, rehabilitation problems, and, in the case of COPD, home oxygen therapy) [4]. Depression and anxiety are frequently examined in parallel as they are frequently co-morbid, the diagnostic tools employed (psychiatric assessment, HADS scale) are similar and the treatment of choice is also similar (antidepressant drugs either alone or in combination with psychotherapy, most commonly cognitive behavioural therapy – CBT) [5, 6] – Questions which invariably arise concerning the interplay between anxiety, depression and COPD are as follows: just how common depression and anxiety in COPD patients is, which COPD patients are at increased risk for mental health issues, how to properly conduct mental health screening, how mental illness affects outcomes, how does it affect quality of life and how to properly treat it. Studies attempting to address these questions have been conducted for the past two decades along with reviews highlighting key findings, though many issues remain as of yet unresolved.

For the purposes of this review, we will use the term depression to refer to major depressive disorder and the term anxiety to refer to generalized anxiety disorder. Major depressive disorder is characterized by the presence of depressed mood throughout the day or anhedonia along with at least four of the following seven symptoms, for a period of at least 2 weeks: hypersomnia or insomnia, subjective fatigue, psychomotor agitation or retardation, inappropriate feelings of worthlessness or guilt, unintentional 5% change in body weight, impaired concentration and recurrent thoughts of death (suicidal ideation) [7]. Generalized anxiety disorder is characterized by excessive anxiety and worry present for most days of a period of at least 6 months which is difficult to control and is related with three or more of the following symptoms: restlessness, excessive fatigue, impaired concentration, irritability, muscle tension and sleep disturbances [7]. The conditions can only be diagnosed definitively by an interview with a mental health professional, though a variety of screening tools can be used to identify those at risk. In epidemiological studies these rating scales are often used in the place of a clinical interview, which may lead to an overestimation of the aforementioned mental health comorbidities. It should be noted that generalized anxiety disorder is frequently co-morbid with other anxiety disorders (phobias, panic disorder, social anxiety disorder) but in the context of this review the term anxiety refers to the symptoms of generalized anxiety disorder unless specified otherwise. Due to the increasing emphasis placed on mental health on a global scale, many studies have been conducted recently, during the past 5 years. In this review we attempt to examine the findings of the most recent studies and their implications both for the clinical practice and for the direction of future research on the association between COPD, major depression and anxiety disorders.

Methods and literature search strategy

This study is a narrative review examining the association between COPD and symptoms of depression and anxiety with emphasis placed on the advances of the last 5 years. A literature search was conducted on the Pubmed database using the search string (COPD OR COPD) AND (Depression OR Major Depressive Disorder OR Dysthymia OR MDD) to identify studies examining the association between COPD and depression and the string (COPD OR COPD) AND (Anxiety OR Generalized Anxiety Disorder OR Generalized Anxiety Disorder OR GAD) to identify studies examining the association between COPD and Anxiety. The first search string revealed 2,081 non-duplicate studies as of March 2019, whereas the second revealed 1,317. Initial screening for relevance was based on the abstract and title. The full text of articles which were deemed relevant by the initial screening were retrieved and used for the determination of the type of study, sample size, main outcomes and possible limitations. Our intention was to include all recent (published within the last 5 years) studies with data on the following subjects: use of screening instruments for depression and anxiety in COPD patients, prevalence and epidemiological associations of depression and anxiety in COPD patients, prevalence and epidemiological associations of depression and anxiety in the carers of individuals suffering from COPD, effect of co-morbid depression and anxiety on COPD outcomes (including hospitalizations, incidence of exacerbations and mortality) and management of co-morbid depression and anxiety in individuals suffering from COPD. The initial screening was performed by two of the authors independently and a consensus among all authors was reached regarding which studies to include in the review. Other reviews, older studies and articles describing rating instruments were cited where appropriate and are present in the reference list.

Screening for depression and anxiety in COPD patients

As the association between COPD and mental illness has become more widely recognized in recent years, the need arises to efficiently screen COPD patients for anxiety and depression. The gold standard for the diagnosis of these conditions is a clinical interview by a psychiatrist, but the use of screening instruments is beneficial in determining which patients should receive such an interview. The Severe Respiratory Insufficiency questionnaire is a screening tool used to assess disease burden in patients suffering from COPD and other respiratory conditions, and anxiety is one of the central parameters it evaluates. A recent Chinese study [8] of 149 patients on non-invasive positive pressure ventilation validated the Chinese version of the rating scale for use in this population. A simple screening test for depression is the Patient Health Questionnaire (PHQ) PHQ9 [9] and its abbreviated version, the PHQ2. In a study [10] of 561 COPD patients who completed a rehabilitation program and followed up over 1 year, it was determined that the long version is more sensitive than the abbreviated one, as was expected. However, the participants did not receive a clinical interview, so the study could not present a clear picture of the sensitivity and specificity of the tests. Another study [11] on 56 COPD chronic patients was designed to determine the best screening tool for depression and anxiety in this population. It evaluated the Hospital Anxiety and Depression – Depression Subscale (HADS-D) [12] as well as the hospital anxiety and depression – anxiety subscale (HADS-A) scale, the Beck Depression Inventory [13] (BDI-I) and the Beck Anxiety Inventory (BAI) [14], whereas all patients also underwent a structured clinical interview. All but the HADS-D scale had adequate specificity and sensitivity unmodified, whereas the depression scales’ sensitivity reached 100% with the removal of items which may also be complications of COPD, specifically those related to lack of energy and sex drive. An Italian study [15] on 65 inpatients (GOLD stages III and IV) using the aforementioned screening tools for depression reported similar findings, while also assessing screening tools for mild cognitive impairment, another issue which commonly affects COPD patients. A few years ago a group of researchers [16] designed a simple screening test for anxiety in individuals suffering from respiratory conditions and found its validity to be similar to the aforementioned anxiety rating scales in a two-phase study. The same screening test was evaluated again in a more recent study [17], alongside the GAD-7 [18] and the HADS-A against the clinical interview gold standard. A total of 220 patients were enrolled, with 11% meeting the DSM-5 criteria for the diagnosis of generalized anxiety disorder based on the clinical interview, whereas the screening tools found anxiety symptoms in a significantly larger percentage of the study population (ranging from 20% for the HADS-A to 38% for the AIR). Differences in sensitivity between the tests did not reach statistical significance, whereas the HADS-A was significantly more specific than the other two. Another rating scale which was recently introduced is the Assessment of Burden in COPD (ABC) scale, which evaluates depression and anxiety among other symptoms in patients suffering from COPD. It was evaluated alongside the Saint George Respiratory Questionnaire [19] (SGRQ) in a Dutch study [20] of 133 patients. Reliability (convergence between test and retest at 2 weeks) was similar between the two screening tools, as was convergent validity. These advances indicate a shift towards a more holistic approach to COPD treatment, which aims to address the psychiatric comorbidities associated with the underlying disease process. Appropriate screening tools are required to identify which patients would be more likely to benefit from a psychological assessment, as it would not be practical to request a clinical interview by a mental health professional with frequent follow-up visits for all patients diagnosed with COPD. The screening tests available for anxiety and depression, the most common psychiatric comorbidities associated with COPD have not yet been conclusively evaluated and it remains unclear whether their sensitivity is equivalent, so more research on this matter is expected in the near future [21].

Prevalence and risk factors for depression and anxiety disorders in patients suffering from COPD

The association between COPD, depression and anxiety has been established in the literature for almost a decade, with a systematic review of eight studies [3, 22–29] (39,587 patients and 39,431 controls in total) finding COPD patients to be almost three times as likely to experience depression compared to matched controls [OR = 2.81 (1.69–4.66)]. Another review [30] examined the association between COPD and clinically significant anxiety symptoms, with the authors reporting a greater prevalence of anxiety in individuals suffering from COPD, while also pointing out the weak quality of the evidence and the heterogeneity of the previous studies. A higher prevalence of depression was also reported in another review which included data from the clinical trials of tiotropium [31]. Given the similar symptoms of chronic respiratory conditions, it would be worth investigating whether the association between respiratory and affective disorders in exclusive to COPD and if it is evident for other respiratory conditions as well.

In the last 5 years, eight recent studies which examined the prevalence and risk factors for both anxiety and depression in COPD patients were published [32–39]. The studies were extremely heterogenous, differing in sample size, methodological design and in the means of depression and anxiety ascertainment. However, all concluded that anxiety and depression are highly prevalent in COPD. The heterogeneity of the studies does not enable safe conclusions to be drawn regarding the contribution of additional risk factors for the development of depression or anxiety in individuals suffering from COPD. Eight additional studies were identified which focused on prevalence and risk factors of depression alone in individuals suffering from COPD [40–46]. There was a consensus among the researchers that depression is highly prevalent in individuals suffering from COPD, though the methodological heterogeneity of the studies does not facilitate a synthesis of their results (Table 1). We noted that authors are likely to consider anxiety as common comorbidity or even a complication of depression, and while most studies opted to identify patients as depressed via a formal diagnosis of MDD via a clinical interview, only two studies [32, 35] utilized the same approach for the diagnosis of GAD. Use of standardized rating scales to estimate the prevalence of a mental illness may overestimate its prevalence (which is the reason why these scales can also be used as screening tools to identify the patients who would most benefit from a mental health assessment), whereas registry-based medical records may underestimate the prevalence of the mental illness in question [47].

Table 1. Summary of studies examining the prevalence and risk factors of anxiety and depression in individuals suffering from COPD.

Author	Year	Country	Study design	Characteristics of study population	Groups compared	Number of participants	Ascertainment of COPD	Affective disorder ascertainment	Study findings and conclusions
Studies assessing both depression and anxiety
Pothirat	2015	Thaland	Cross-sectional	Individuals over 40 years of age	Individuals with chronic respiratory conditions and those without	571	History and spirometry	Clinical interview	Individuals suffering from COPD were more likely to suffer from anxiety disorders (GAD and panic disorder) compared to controls without respiratory disorders. Individuals suffering from COPD were less likely to suffer from any affective disorder [(RR = 0.2 (0.1, 0.5)] and depression [(RR = 0.3 (0.1, 0.8)] compared to individuals suffering from asthma
Bania	2016	Greece	Cross-sectional	Individuals in Greek correctional institutes (91.3% male)	Individuals with COPD and those without	552	Medical history and spirometry	PSS-14, EQ-5D-3L	Individuals suffering from COPD were more likely to exhibit signs of depression and anxiety as assessed by the PSS-14 and EQ-5D-3L rating scales
Gonzalez-Gutierrez	2016	Spain	Cross-sectional	Stable patients with COPD	Those who had anxiety or depression and those who did not	204	History and spirometry	Clinical interview, ICD-10 criteria for GAD and MDD	19% of the Study patients had GAD, 9.8% had isolated depression and 7.3% had a mixed anxiety–depression disorder
Biswas	2017	India	Cross-sectional	Individuals being treated for COPD in a hospital in India	Patients with COPD and depression and those without Patients with COPD and anxiety and those without	75	Spirometry and medical history	HAM-D for depression, STAI for anxiety	Patients with COPD were more likely to exhibit symptoms of both depression and anxiety simultaneously. HAM-D and STAI scores did not correlate with 6-min walk test performance
Thapa	2017	Nepal	Case–control	Individuals being treated for COPD in Nepalese hospitals and matched controls (visitors to the hospital)	COPD patients and matched controls	198	Medical history and spirometry	BDI, BAI	Individuals suffering from COPD had significantly higher higher BDI and BAI scores compared to controls
Raherison	2018	France	Cohort study	Patients suffering from COPD in southern France	Comparison based on COPD severity as assessed by the GOLD score	1,584	History and spirometry	Medical records	A much higher (p < .01) prevalence of depression and anxiety was observed in patients with symptomatic disease (GOLD scores B and D) compared to those without severe symptoms (GOLD scores A and C). Anxiety was more prevalent than depression
Stallberg	2018	Sweden	Cohort study	General population	Individuals with COPD and those without	101,993 (17,479 COPD patients and 84,514 individuals without COPD)	Registry records	Registry records	The prevalence of depression or anxiety rose from 5 to 21.3% in the 2 years following COPD diagnosis, while the prevalence of depression rose only slightly in the control population during the same time period (from 2.8 to 4.1%). Another distressing finding is the association of depression with increased mortality in the COPD patients
Xiao	2018	China	Cross-sectional	Inidviduals with mild COPD	COPD patients with anxiety and depression and those without	275	History and spirometry	HADS	8.1% of Patients had anxiety and 13.4% had depression. Female patients were more likely to suffer from anxiety than male patients (OR = 6.41, 95% CI: 1.73–23.80)
Studies assessing only depression
Silva Junior	2014	Brazil	Case–control	Individuals over 40 years of age suffering from COPD with mild daytime hypoxemia and PaO2 < 60 mm Hg	COPD patients with depression and those without	60	History and spirometry	Clinical interview, DSM-IV criteria	A significant association was found between higher CAT scores (>20) and the presence of depression
Kamour	2015	USA	Cross-sectional	General population	Individuals with COPD and those without	9,585	Telephone interview	Telephone interview	Self-reported comorbidity of COPD and depression was estimated at 4.32% (CI = 4.27, 4.36)
Albrecht	2016	USA	Cohort study	Medicare recipients with a first diagnosis of COPD in the 2006–2012 period	Patients newly diagnosed with COPD who developed depression and those who did not	125,348	Medicare records	Medicare records	A significant association was found between depression and a recent diagnosis of COPD [risk ratio 1.76 (1.73, 1.79)], with an even stronger association between depression and COPD-related hospitalizations [risk ratio 4.59 (4.09, 5.15)]
Garin	2016	Multinational registry	Cross-sectional	Individuals over 50 years of age	Multiple comparisons to detect comorbidity patterns	41,909	Medical records – registry-based study	Medical records (depression only) – Registry-based study	The prevalence of COPD and depression comorbidity in the multinational sample was 18%
Rosrita	2016	Indonesia	Cross-sectional	Outpatients being treated for COPD in a hospital in Jakarta	COPD patients with depression and those without	141	Medical history and spirometry	Clinical interview	Patients with higher CAT scores were more likely to suffer from depression, whereas there was no correlation between the risk of depression and the degree of airflow limitation as assessed by spirometry
Himani	2018	Pakistan	Cross-Sectional	COPD outpatients monitored at the pulmonology clinic of Agha Khan uiversity hospital	COPD patients with depression and those without	556	Medical records	HADS scale for depression	57.2% of the Sample were depressed according to the HADS scale, with age >59 years, being unemployed or retired and using inhaled steroids being associated with a higher prevalence of depression
Kayhan	2018	Turkey	Cross-sectional	Patients with COPD on long term oxygen therapy	Those who were depressed depression and those who were not	54	History, use of long-term oxygen therapy	Clinical interview, DSM-IV criteria for depression	63% of Patients met the criteria for the diagnosis of MDD, and MDD was more prevalent in non-compliant individuals
Lee	2018	South Korea	Cross-sectional	Patients with early COPD diagnosis	Those who were depressed according to the PHQ-9 score and those who were not	211	Medical records	PHQ-9	A high prevalence (14.2%) of depression assessed via the PHQ-9 was found in this population with significant associations between depression and low income [OR 2.17 (1.55, 3.04)], whereas women and those who lived alone were also at higher risks [OR = 1.79 (1.38, 2.31) and 1.86 (1.37, 2.51), respectively]

Abbreviations: BAI, Beck Anxiety Inventory; BDI, Beck Depression Inventory; COPD, chronic obstructive pulmonary disease; DSM, diagnostic and statistic manual; EQ-5D-3L, EuroQol five-dimension three-level questionnaire; GAD, generalized anxiety disorder; GOLD, global initiative for chronic obstructive lung disease; HADS, Hospital Anxiety and Depression Scale; HAM-D, Hamilton Rating Scale for depression; ICD, International Classification of Diseases; MDD, major depressive disorder; PHQ-9, Patient Health Questionnaire-9; PSS-14, Perceived Stress Scale-14.

There has been little investigation thus far into the way COPD symptoms may directly be related to the development of depression. A cross-sectional study of 115 patients in Spain evaluated the association between depression and COPD symptoms and found that 24.3% of the patients suffered from depression. The severity of COPD as assessed by the BODE index, quality of life and dyspnea as assessed by the MRC were significantly correlated with depression [48]. In a study of 590 patients [2] (the same population which was used to evaluate the MHQ2 and MHQ9 questionnaires), an association was found between characteristics of dyspnea on the BORG scale and certain items of the MHQ9 scale. Specifically, the authors report that duration and intensity of dyspnea correlated with more severe somatic symptoms of depression, whereas the patient’s emotional response to dyspnea was more strongly correlated with the psychological features of depression, hopelessness and anhedonia.

Another recent development is the increasing recognition of cases of COPD which cannot be directly attributable to smoking. This is a matter of considerable epidemiological interest, as these patients may have different comorbidity patterns than chronic smokers who develop COPD. A recent Korean population-based cross-sectional study [49] found that non-smokers (399 in total, defined as individuals diagnosed with COPD who had smoked <100 cigarettes in their lifetime) were almost twice as likely to suffer from depression compared to smokers who developed COPD (n = 823). The association was statistically significant (p < .001). However, as this was a cross-sectional study the significance of this association remains unclear. Although depression is more common in women than men, this study found an association between male sex and depression. Another approach is to examine whether current smoking status can be associated with mental illness in individuals suffering from COPD. Due to the psychotropic properties and addictive nature of nicotine, one would expect current smoking to be associated with a decreased risk for concurrent depression or anxiety disorders and a higher risk would be expected in recent quitters. A small study of 60 patients, however, did not find any association between smoking status and depression, anxiety or panic disorder, whereas COPD severity and the number of exacerbations were the most potent predictive factors. The generalizability of these findings is, however, limited by the small sample size [50]. Only one study [51] examined the contribution of possible risk factors to the severity of depressive symptoms in COPD. A total of 836 patients with stable COPD were recruited, and the most notable finding of the study was that the risk factors which have been identified in the literature for the development of depression in COPD patients (female gender, living alone, increased severity COPD symptoms such as dyspnea, frequent exacerbations) are also associated with an increased likelihood for the presence of severe depressive symptoms as assessed by the BDI.

In recent years the role of the immune system in the pathogenesis of chronic disease has been extensively studied. Aberrant immune system activation may also be implicated in the association between COPD and periodontitis which was reviewed recently [52]. This trend is also prominent in psychiatry and systemic inflammation as well as localized modulation of microglial cell activity in the CNS have been associated with most psychiatric conditions. Depression in particular may be associated with systemic inflammation and higher serum concentrations of inflammatory biomarkers such as IL-1 and TNF-α, though the nature of the association has yet to be clearly elucidated. A recent study [53] found elevated levels of IL-2, IL-6 and IFN-γ in patients suffering from co-morbid COPD and depression compared to healthy controls, though the implications of these findings remain uncertain. However, results obtained from the cohort of the ECLIPSE study [54] (data available for 481 patients) indicate that there is no significant association between inflammatory biomarkers and symptoms of depression in individuals suffering from COPD. Further research may be required to establish an association, if one exists the clinical implications of such knowledge would be limited in the short term, as immune-modulating agents are not indicated for the treatment of depression.

Chronic obstructive pulmonary disease may further be categorized into clinical phenotypes and the prevalence of mental health issues may differ between them. These distinct phenotypes include non-exacerbators, asthma–COPD overlap, frequent exacerbators and frequent exacerbators with chronic bronchitis [55]. Some researchers prefer to simplify the distinction into frequent exacerbators and none exacerbators, as exacerbations are by far the most significant clinical manifestation of COPD [56]. Controlled studies based on this categorization are more challenging as they require a large patient base to ensure an adequate number of individuals of all phenotypes are recruited and to avoid selection bias favouring recruitment of the more severely ill (and thus more likely to be hospitalized) patients [57]. In the Polish sub-cohort of the POPE study (430 patients), it was reported that frequent exacerbators with chronic bronchitis were at higher risk for both depression and anxiety compared with the other three phenotypes [58]. In a recent cross-sectional study of 161 patients conducted in Hong Kong [59], a significant association was found between depressive symptomatology and frequency of exacerbations, recent exacerbations and also severity of dyspnea rated by the MMRC.

Prevalence and risk factors for depression and anxiety disorders in carers of COPD patients

Individuals suffering from COPD exhibit variable degrees of disability and their autonomy may be markedly reduced, especially in later stages. This leads them to become dependent on carers (typically spouses and family members) who may in turn be affected by the disease burden of COPD. This phenomenon is not exclusive to COPD, as it has been studied in most chronic conditions which significantly limit the patient’s capacity to carry out daily tasks autonomously. A significant correlation was found between depression assessed by the BDI scale in the partners of the patients and COPD severity, number of exacerbations and depression in the patients themselves in a Greek study of 230 people hospitalized for a COPD exacerbation and their spouses [60]. The assessment occurred a year after the first admission. Two additional cross-sectional studies were recently published on this issue. One of these examined characteristics associated with mental health issues in carers (n = 203) as assessed by the HADS-A and HADS-D scales [61] – Female carers were more likely to experience mental health issues compared to males, as were older individuals and those who cared for patients with greater activity limitations. A similar study [62] of 119 patient–carer pairs found using the same rating scales that depression or anxiety in patients was significantly correlated with the same findings in carers. Also noteworthy was the observation that while both carers and patients exhibited similar high levels of anxiety, the prevalence of depression was almost twice as high in the patients suffering from COPD. A large Spanish registry based study on 461,884 COPD patients and the 220,892 carers that could be identified found a significant burden in caregivers encompassing all aspects of their life, including social/leisure activities, employment and personal health. Carers of more severely disabled COPD patients were affected to a greater extent than carers of individuals with milder COPD. The authors concluded that the burden on caregivers was similar to that experienced by caregivers of patients with other chronic medical conditions such as ischemic heart disease, cancer and stroke [63].

The impact of co-morbid depression and anxiety in COPD patients

Due to the high rates of co-morbid depression and anxiety in individuals suffering from COPD, it would be important to evaluate how these mental health conditions contribute to the overall disease burden and whether they significantly affect the prognosis of COPD. A recent European study [64] of 408 patients attempted to evaluate the effect of a number of common comorbidities on quality of life in COPD patients. Depression and anxiety were assessed by a clinical interview, and the patients’ overall quality of life was assessed via the subjective feeling thermometer, a visual analog scale. Co-morbid depression and anxiety had, by far, the largest impact on the feeling thermometer score, leading to a reduction of approximately 11/100 points for depression and 9/100 for anxiety. Based on this finding the researchers propose the introduction of a weighted index of comorbidities that significantly reduce quality of life in COPD patients, the COMCOLD index. It is used to examine five common comorbidities: depression (given six points), anxiety – four points and peripheral artery disease, cerebrovascular disease and symptomatic cardiac disease (either CHF or CAD), given three points each for a maximum score of 19. This is the first clinical index designed to assess quality of life in COPD, in contrast to other existing indices used to predict mortality. It seems that even in severely ill patients, even minor mental health issues may affect quality of life more than severe life-threatening physical illnesses. A similar effect was found in a small Korean study of 80 patients, where depression and anxiety had the highest correlation with lower SGRQ and SF-36 scores [65].

Aside from the self-evident detrimental effect of mental illness on quality of life, recent studies show that its impact on the COPD patient population may be even greater, affecting a variety of health-related behaviours and even leading to an increase of overall mortality and morbidity. Daily physical activity is generally recommended for all COPD patients and a multicenter observational study [66] was undertaken in order to assess the determinants of physical activity in this population. In total 4,574 patients were included, and they were examined for the indicators of COPD severity and general quality-of-life measures. A specific assessment for depression and anxiety was also performed via the HADS scales. Physical activity was evaluated via the mean daily walking time. The presence of depression as defined by a score >8 was associated with a low (<30 min) walking time [OR = 1.58 (1.25–2.01)]. A smaller (220 patients) prospective cohort study [67] found a similar association between depression and reduced physical activity, which was assessed by an accelerometer. Incremental one-point increases in the HADS-D scale score were associated with 81 (12,149) fewer steps taken per day. Measures of anxiety did not correlate with the levels of physical activity. A Danish cohort study [68] for the determinants of smoking cessation with 5 years of follow-up evaluated depression as a possible barrier to smoking cessation. In a total of 3,233 patients lower 5-year quit rates were observed for depressed compared to not depressed individuals, but the difference was not statistically significant (p = .189).

Only one study [69] was identified which examined the association between illness acceptance in individuals suffering from COPD and co-morbid depression or anxiety. In a small sample of 102 individuals, a low degree of illness acceptance was associated with higher score on the HADS rating scale for both depression and anxiety. The authors hypothesized that early initiation of antidepressant treatment in patients at risk may contribute to increased adherence to treatment for COPD via increased illness acceptance.

An increased risk of suicide is a hallmark of mental illness, but it is also observed in chronic physical illness, particularly in younger individuals. A Korean population-based cohort study of individuals with COPD (n = 2,506) and controls (n = 13,212) attempted to assess whether COPD in independently associated with suicidal ideation, attempts and completed suicide [70]. Participants were also assessed for co-morbid depression and other mental health issues, but the researchers used the data only for the multivariate analysis to determine whether COPD is an independent risk factor. They reported no significant association between mild COPD (GOLD stages I and II) and suicide, whereas the association for severe COPD (GOLD stages III and IV) was significant both for suicide attempts and for suicidal ideation. It is, however, unclear whether these findings can be generalized, as suicidal behaviour is greatly influenced by cultural standards and South Korea is the country with the highest suicide rate in the world. A smaller (n = 202) cross-sectional analysis conducted in the United States identified 22 patients with suicidal ideation and conducted a qualitative discussion of the themes underlying their suicidal thoughts [71]. Adverse life events, recent loss and a history of untreated depression or other mental health issues were particularly prominent.

An alarming finding reported independently in a Chinese and an American cohort study was an association between depressive or anxiety symptoms and increased mortality in COPD patients. The Chinese study [72] included 7,787 COPD patients, who were evaluated for smoking status, history of smoking, depression and anxiety (via the HADS scales). Both depression and anxiety were associated with higher mortality (p < .001) as was current smoking and greater lifetime exposure to smoking. Concurrent smoking and mental illness had an additive effect on mortality, whereas the association between increased mortality and mental illness was also present even for subjects who never smoked. In the American cohort [73] of 17,320 patients, among which 6,455 had depression. Most of the depressed participants received treatment with antidepressants, with 852 patients (13.2% of the sample diagnosed with depression) receiving no treatment. Depression was associated with increased mortality [HR = 1.21 (1.07, 1.37)], whereas depressed patients on antidepressant treatment had a significantly lower mortality compared with unmedicated depressed COPD patients [HR = 0.55 (0.44, 0.68)]. Additionally, bipolar disorder was also associated with a significantly decreased risk of mortality [(HR = 0.78 (0.65, 0.94)], though the significance of this observation is unclear. These findings constitute evidence in favour of the initiation of antidepressant treatment in depressed COPD patients, as it may reduce overall mortality.

Recent evidence also suggests that aside from mortality, mental illness may also independently increase the risk for COPD exacerbations and hospitalizations in this population, with a reduction in treatment adherence being a possible mechanism. In a British cohort [74] of 58,589 individuals diagnosed with COPD, patients were categorized by number of exacerbations (none, one, two or more) during the first 12 months of follow-up. Depression and anxiety were associated with more frequent exacerbation and the affect, although small, reached statistical significance [OR for two or more exacerbations vs. none for depression 1.25 (1.16, 1.35) and for anxiety 1.16 (1.08–1.25)]. In a sample of 2,059 patients from the Eclipse cohort [75], an association was found between depression as assessed by the CES-D scale and both mild (requiring use of antibiotics or systemic steroids) and severe (requiring hospitalization) exacerbations. The OR for depression and mild exacerbations were 1.18 (1.07, 1.30) and 1.36 (1.09, 1.69) for severe ones. A prospective cohort study of 512 individuals with COPD was conducted in Spain with 2 years of follow-up (2012–2014) with the assessment of both anxiety and depression via the HADS scale at baseline. The presence of anxiety and depression was associated with an increased risk of hospitalization (OR = 1.94). The effect was, however, much more pronounced during the first year of follow-up [76]. Two studies were identified which focused on the sub-population of COPD who were hospitalized for the treatment of an exacerbation. The first of these [77] sampled Medicare beneficiaries hospitalized for COPD in the 2001–2011 time period, including 135,498 patients in total, of which 22.3% had some psychiatric diagnosis. The outcome assessed was the readmission rate during the first 30 days after hospital discharge, and the researchers found an association between depression and anxiety with higher readmission rates, with an OR of 1.34 (1.29, 1.39) for depression and 1.43 (1.37, 1.50) for anxiety. A smaller study conducted in a single US hospital [78] included 422 patients hospitalized for a COPD exacerbation, and the presence of depression was associated with an increased readmission rate at 30 days [OR = 3.83 (1.84, 7.96)] 90 days [OR = 2.47 (1.34, 4.55)] and 1 year after discharge [OR = 2.67, (1.59–4.47)]. These findings suggest an association between depression and a worse prognosis, but it is unclear whether a causal relationship exists. These patients could exhibit depression due to their severe illness, but depression could also negatively impact their outcomes by reducing treatment adherence which during hospitalization entails effective communication with providers. The association between depression and reduced adherence to COPD medication was found in another study of Medicare beneficiaries [79], which included 31,033 patients diagnosed with COPD between 2006 and 2010. In this sample both a new (after the diagnosis of COPD) episode of depression and a history of depression (at least one episode leading the diagnosis prior to being diagnosed with COPD) were significantly associated with reduced adherence [OR = 0.93 (0.89, 0.98) and 0.81 (0.78, 0.85), respectively]. It is notable that a stronger association with reduced adherence was found in individuals with a history of depression, compared to those who first exhibited signs of mental illness following COPD diagnosis. Another study by the same principal investigator [80] on Medicare beneficiaries (n = 16,075) found that a subset of the population had zero adherence to medication (never filled prescriptions), with 39.2% of the cohort having 0 adherence to COPD maintenance medication and 19.2% of those for whom antidepressants were indicated exhibiting 0 adherence. For both categories of drugs, both high (>80%) and low adherence were associated with a decreased risk of both hospitalization and emergency department visits, though there was no significant difference between high and low adherence. Good adherence to antidepressants led to a hazard ratio of 0.74 (0.70, 0.78) for emergency room visits compared to 0 adherence and 0.77 (0.73, 0.81) for hospitalizations. This finding constitutes evidence in favour of prescribing antidepressants for depressed individuals suffering from COPD as they may reduce overall morbidity, regardless of their efficacy in the treatment of depression.

Treatment of co-morbid depression and anxiety in COPD patients

Individuals with co-morbid mental and physical illness are usually treated empirically with the pharmacological and psychotherapeutic intervention indicated, without taking the co-morbid conditions into account unless they constitute a contraindication to usual treatment. There is a relative lack of data from randomized controlled trials regarding the use of antidepressant drugs in patient populations with a specific physical comorbidity, due to the great costs associated with such studies. A single RCT was identified for the use of sertraline in patients suffering from co-morbid COPD and depression, which found that antidepressant treatment was associated with an increased likelihood for depression improvement (absolute score reduction on the HADS-D scale), as well as a reduced CAT (COPD Assessment Test score, a rating scale introduced in 2006 [81]) score and increased distance traversed on the 6-min walk test. Both groups received standard COPD treatment, whereas the active treatment group was treated with sertraline 50 mg/day for 6 weeks, whereas the control group received placebo. The clinical significance of these results is, however, uncertain, due to the small sample size (n = 60 for each group, 120 total) and also because only one dose of sertraline was evaluated, the minimum recommended therapeutic dose [82].

A single study was also identified which attempted to evaluate any correlation between the use of the inhaled drugs indicated for COPD treatment with depression [83]. In all 84 patients were evaluated for 12 weeks after the initiation of COPD maintenance treatment, and no significant association was found between COPD drug use and the development (or the improvement) of depressive symptoms. Theoretically if the inhaled drugs had any systematic effect, an improvement in depression would be expected with anticholinergic, a worsening of anxiety would be associated with beta adrenergic agonists, whereas corticosteroids might have increased psychopathology in a more unpredictable manner. It would be difficult, however, to accurately assess such an effect, as it would be necessary to include a control group for which treatment would be withheld, which raises significant ethical concerns. To our knowledge, no studies examining the association between drug use during exacerbations and anxiety or depression have been published. Such research would provide valuable insight into the established epidemiological association between symptoms of anxiety and depression and exacerbations, as there is a risk of treatment interactions. Systemic steroids are characterized by unpredictable psychological effects ranging from worsening of depression to an induction of euphoric states, and in some cases they can even precipitate mania or toxic psychosis. Antibiotics on the other hand can exert psychotropic effects in their own right, while they may also exhibit pharmacokinetic interactions with psychotropic drugs [84].

Another cause of concern is the use of psychotropic drugs to aid with smoking cessation in COPD patients. Aside from the gold standard of nicotine replacement therapy, smokers with COPD may also be prescribed varenicline (a partial agonist of nicotinic Ach receptors) or bupropion (an activating antidepressant with an unclear monoaminergic mechanism of action, which also happens to be a nicotinic Ach receptor antagonist) to assist quitting. Among these options, nicotine replacement therapy would be expected to exert no significant psychological effects, as it merely mimics the established effects of smoking, whereas bupropion would be expected to confer additional benefit in depressed patients. Varenicline is more problematic, as it has been associated with psychological adverse effects, including worsening of depression, anxiety and aggressive behavior. A cohort study of COPD patients who were prescribed smoking cessation aids in the United Kingdom found no increased risk of cardiovascular or neuropsychiatric side effects of bupropion (n = 350) or varenicline (n = 3,574) compared to nicotine replacement therapy (n = 10,426), but the study was not designed to evaluate possible beneficial effects of the drugs on co-morbid depression and anxiety [85].

Studies evaluating psychological treatment approaches for co-morbid mental illness in specific patient populations are more common than clinical trials for psychotropic drugs, due to reduced financial and regulatory barriers. One trial of telephone-based CBT versus a controlled friendly conversation [86], one trial of two specific behavioral interventions designed for COPD [87, 88] and a recent review of three other trials [89–92] were recently published regarding the use of psychological therapies for COPD, but the results are not promising, as little difference was observed in the outcomes among comparison groups and the sample sizes were small. A cross-sectional study of 527 patients which attempted to identify how co-morbid depression was actually managed in patients with COPD was recently conducted [93] but not enough data was collected to justify clinical recommendations. Notable findings included a strong association between anxiety and receiving combined treatment with psychotherapy and antidepressants [OR = 6.01 (3.11, 11.61)], and an increased risk of receiving no treatment at all for the uninsured [OR = 0.21 (0.05, 0.86)]. It would be beneficial to conduct cohort studies with a similar design, prospective or retrospective (provided sufficient data is available) to examine association between psychiatric treatment (specific drugs or standardized psychotherapy methods) and outcomes of both mental (improvement in rating scales for depression and anxiety, relapse rates) and physical health (mortality, exacerbations, GOLD score progression and number of hospitalizations).

Conclusion

Individuals suffering from COPD are more likely to suffer from depression and anxiety compared to the general population, and mental illness is a significant contributor of the overall disease burden of COPD. Symptom severity and exacerbation frequency may be the strongest predictors of co-morbid mental health issues as they greatly diminish quality of life. End-stage COPD patients may also be affected by the despair commonly associated with terminal illness. Depression and anxiety may initiate a vicious cycle consisting of reduced physical activity, social isolation, treatment non-adherence and avoidance of physician visits, leading to increased COPD severity and more rapid progression, which in turn may cause further mental health deterioration as well. This is the most likely explanation for the increased rate of exacerbations, hospital admissions and overall mortality affecting patients suffering from co-morbid COPD and mental illness compared to those suffering from COPD alone. These observations are a great cause for concern and in recent years the body of literature examining the association between COPD, depression and anxiety has been growing rapidly [94]. Further research is, however, required to determine how to effectively screen for common mental health complaints in this population and the optimal treatment strategies. In practice, this patient group may receive empirical antidepressant treatment, but there is no evidence from randomized controlled trials to support specific drug choice, and the likelihood of receiving combined psychotherapy and pharmacotherapy (which is considered optimal) is also low. Given the high prevalence of anxiety and depression comorbidity in COPD, the formulation of consensus evidence-based guidelines on their identification and treatment would be a first step towards ensuring adequate quality of mental health care in this population.

Declaration of Interest

The authors have no competing interests to declare.