D5. Treat anxiety and depression

Anxious and depressive symptoms and disorders are common comorbidities in people with COPD (Yellowlees 1987, Kunik 2005, Ng 2007, Xu 2008, Eisner 2010b) and have a range of negative impacts [evidence level III]

Anxiety symptoms in COPD are associated with worse quality of life (Giardino 2010, Blakemore 2014), self-management (Dowson 2004) and exercise performance (Eisner 2010b) [evidence level III], and with increased medical symptom reporting (Katon 2007), exacerbations (Laurin 2012), hospitalisations (Yellowlees 1987, Gudmundsson 2005, Livermore 2010), length of hospitalisations (Xu 2008), medical costs (Katon 2007, Livermore 2010), and mortality (Celli 2008) [evidence level III]. The prevalence of one anxiety disorder in particular, panic disorder, is approximately 10 times greater in COPD than the population prevalence of 1.5 to 3.5%, and panic attacks are commonly experienced (American Psychiatric Association 2004, Smoller 1996).

Cognitive behaviour therapy has been shown to be an effective treatment for panic disorder in the physically healthy (Mitte 2005) [evidence level I]. There is promising evidence from a number of small randomised controlled trials that cognitive behaviour therapy can treat anxiety symptoms in COPD (de Godoy 2003, Hynninen 2010, Livermore 2010), prevent the development of panic attacks, panic disorders (Livermore 2010 Yohannes 2017) and reduce ratings of dyspnoea (Livermore 2015). The health service burden associated with anxiety in COPD is well established.  A nurse-delivered minimalist version of Cognitive Behaviour Therapy facilitated by the use of laminated cards and delivered in 1-2 home visits of 20-60 minute duration, provided clinically and statistically significant improvements in the Hospital Anxiety Depression Scale (HADS) and also the Chronic Respiratory Disease Questionnaire (CRQ) Mastery scale, in the intervention arm (n=22) compared to the control arm (n=22), when followed up at 3 months (Bove 2016).  In a trial of 28 patients undergoing pulmonary rehabilitation with a three month follow up, cognitive behaviour therapy showed a short term improvement in fatigue, stress and depression and anxiety scores, thus demonstrating a non-pharmacological intervention with important positive outcomes, although the follow up of 3 months is short and unknown if the benefits are sustained (Luk 2017).

A record linkage study in Canada found that elderly COPD patients prescribed benzodiazepines were at increased risk of an outpatient exacerbation (NNH 66, 95% CI 57–79) or an emergency department visit for COPD or pneumonia (NNH 147, 95% CI 123–181).  There was also a slightly elevated albeit not significant risk of hospital admission (Vozoris 2014) [evidence level III-2]. Caution is warranted in using these medications, due to their potential depressive effects on respiratory drive (Shanmugam 2007), and their inherent risks in the elderly of dependence, cognitive impairment, and falls (Uchida 2009).

This retrospective cohort study of 80,088 U.S. Medicare recipients demonstrated 34% higher 30-day readmission rate in COPD patients who had depression, and 43% higher with anxiety (Singh 2016). These and other co-existing psychological disorders were associated with being less likely to have follow up appointments, and more 30 day readmissions (23.8% versus 16.25%). Although this study design has the potential for confounding by severity of disease, it is noteworthy that these relationships with readmissions were much higher than index admission ICU length of stay, or need for the use of mechanical ventilation, which supports the case that depression and anxiety are important independent predictors of readmission.

Selective serotonin reuptake inhibitors (SSRIs) such as sertraline, have been recommended as first line pharmacological therapies for anxiety in COPD. However, in a 2018 Cochrane systematic review which was conducted to assess the effectiveness and safety of pharmacological interventions for the treatment of depression in patients with COPD, there was not enough evidence relating to efficacy and safety to make recommendations on the use of SSRIs for treating patients with COPD and depression (Pollok 2018) [evidence level I]. In a meta-analysis involving two RCTs of 148 participants there was no difference in the primary outcome of change in depressive symptoms post-intervention (SMD 0.75, 95% CI -1.14 to 2.64; I2 = 95%).  Due to the risk of bias and high level of heterogeneity in depression levels, as well as in the types of medication and doses used, these results should be interpreted with caution (Pollok 2018).  Psychiatrists can advise on the most appropriate medications for particular patients (Shanmugam 2007). Case management to support adherence to antidepressant medication in conjunction with attending pulmonary rehabilitation has been associated with improvements in both depression and dyspnoea-related disability (Alexopoulos 2014, Alexopoulos 2016).

People with COPD are not only at high risk of depressive symptoms and mood disorders, but are at higher risk than people with other chronic conditions (Ng 2007, Omachi 2009) [evidence level III]. When depressive symptoms are comorbid with COPD they are associated with worse health-related quality of life (HRQoL) (Ng 2007, Omachi 2009, Hanania 2011, Blakemore 2014) and difficulty with smoking cessation (Ng 2007) [evidence level III], and with increased exacerbations (Laurin 2012) hospitalisations (Bula 2001, Xu 2008, Hanania 2011, Iyer 2016), length of hospitalisations (Ng 2007) [evidence level III], medical costs (Bula 2001), and mortality (Bula 2001, Ng 2007) [evidence level III]. Depression may also influence decisions about end of life issues (Stapleton 2005). As is the case for anxiety symptoms in COPD, there is evidence from small, randomised controlled trials that depressive symptoms can be decreased by cognitive behaviour therapy (de Godoy 2003, Hynninen 2010). Mindfulness-based cognitive therapy in conjunction to pulmonary rehabilitation also improved depressive symptoms compared to pulmonary rehabilitation alone (Farver-Vestergaard 2018).  A 2019 Cochrane review concluded that, while cognitive behaviour therapy may be an effective treatment for depression in COPD, the quality of the evidence is currently limited (Pollok 2019).  Evidence for the effectiveness of particular antidepressant medications for mood disorders in COPD is still limited, with a few small, randomised controlled trials conducted (Argyropoulou 1993, Lacasse 2004, Eiser 2005, He 2016). Treatment with antidepressants can be complicated by poor tolerance of side effects such as sedation, which may cause respiratory depression (Evans 1997). As with anxiety symptoms, psychiatrists can advise on which pharmacological treatments may be most appropriate for patients.

However, the existing evidence still warrants the referral of anxious and depressed people with COPD to clinical psychologists and psychiatrists for assessment and treatment. Depressed COPD patients referred to mental health specialists have lower odds of two year mortality than those treated in primary care settings (Jordan 2009). Screening for clinically significant anxiety and depression, given their serious impacts, should therefore be part of routine care (including during admissions for exacerbations) (Lecheler 2017).  The Hospital Anxiety Depression Scale (HADS) is an example of an easily administered, widely used screening questionnaire, developed for use with medical patients (Zigmond 1983), and utilised in numerous studies of people with COPD (Gudmundsson 2005, Ng 2007, Xu 2008, Livermore 2010, Eisner 2010b, Bock 2017) [evidence level III].  Another screening option is the Patient Health Questionnaire (PHQ), which screens for symptoms of the most commonly seen mental disorders in medical patients – depression, generalised anxiety, panic attacks, somatoform and eating disorders. The full scale, or the depression and anxiety subscales, may be administered (Spitzer 1994). The PHQ has the advantages of high statistical reliability and validity, while being an easily administered measure that is available on the internet at no cost (Kroenke 2010).