Depressive episode is a syndrome that includes depressed mood, anhedonia (loss of interest or pleasure) and fatigue that is present for a period of at least 2 weeks.

Diagnosis is made by a structured interview using the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) (American Psychiatric Association [13]) criteria. Major depressive episode may include: five (or more) of the following symptoms have been present during the same 2-week period and represent a change from previous functioning; at least one of the symptoms is either (1) depressed mood or (2) loss of interest or pleasure.

It takes about 30 min to administer the scale and the gold standard for the diagnosis of bout major and minor depressive symptoms. However, the downside of the scale is time-consuming and requires special training to administer the scale.

The Geriatric Mental State Schedule [14] is a well validated scale used to diagnose clinical depression and anxiety in elderly people including those with chronic diseases. It is one of the most widely used and respected instruments for measuring a wide range of psychopathology in elderly people. It is based on the Present State Examination [15] and the Psychiatric State Schedule [16]. It has been adopted for use on laptop computer via software which generates a diagnosis, the Automated Geriatric Examination for Computer Assisted Taxonomy (AGECAT) [17], and is replicable (when used in hospital and community samples) against psychiatrists’ diagnosis [17]. Its internal validity and reliability have been established [18], and it has been used in a cross-national setting [19]. The GMS delivers a diagnosis using a hierarchy which corresponds to approaches to diagnosis by a trained psychiatrist. There are 5 levels of severity generated by GMS for 5 diagnostic groups and a level of 3 or above corresponds to a “case”. Thus, level 3 and above in the depression group is diagnostic of clinical depression, and level 3 or above in the anxiety group diagnostic of clinical anxiety. It requires a trained member of staff for the interview and takes about 40 min to administer.

The MADRS assesses severity of clinical depression [20], has been validated in medically ill elderly patients and is widely used in the context of other chronic diseases. It has ten items scored compositely which result in a maximum total score of 60. It classifies severity of depression into four categories: normal, i.e. not depressed (0–6); mild (7–18); moderate (19–34); and severe (>35) [21]. Low scores imply mild depression and high scores correspond to severe depression. Subjects rate their responses using a Likert 7-point category scale, for example, 0 = “no sadness”, 6 = “miserable all the time”. The MADRS has performed better in identifying responders and non-responders to antidepressant drug therapy than the Hamilton Depression Rating Scale (HDRS) [20]. A study by Hammond [22] identified the inappropriateness of HDRS (because of very low internal consistency) in determining severity of depression in the physically ill-depressed elderly patients but recommended the MADRS as a preferable choice. A study by Yohannes et al. [23] using the MADRS scale in patients with COPD identified the severity of depression 17 (30%) were mildly depressed (MADRS score 7–19), 39 (68%) were moderately depressed (MADRS score 20–34) and 1 (2%) was severely depressed (MADRS score 35–60). It requires a trained person to administer the scale.

The HDRS measures the severity of depression and change in depressive symptoms [24]. It is a clinician-rated scale and takes about 20–30 min to complete. The 17 items of HRDS, a score of 0–7 is generally accepted within the normal range, (or in clinical remission), whilst a score of 20 or higher (indicating at least moderate severity) is usually is required medical intervention (entry into a clinical trial). Subjects rate their responses using a Likert 4-point category scale, for example, 0 = “absent”, 4 = “attempts at suicide”. This is a widely used clinical assessment tool to assess the responsiveness to intervention, e.g. antidepressant drug therapy.

BASDEC is a valid screening tool for depressive symptomatology in elderly medically ill patients [25]. It consists of a 19-item deck of cards, self-administered as “true”, “false” and “I do not know” responses. Two items are weighted to 2 points; other affirmative responses, 1 point; and “I do not know” 0.5 point with a maximum score of 21. A score of seven or above suggests a “case” of depression [25]. The BASDEC demonstrated a good response when tested against the “gold standard” of the Geriatric Depression Score (GDS) which is recommended as an assessment scale for elderly people by the British Geriatrics Society/Royal College of Physicians [26]. The BASDEC performed well as a screening tool in elderly medically ill inpatients compared with the GDS having a sensitivity of 71%, a specificity of 88%, a positive predictive value of 74% and a negative predictive value of 86% [25]. Studies suggest that BASDEC is user-friendly and can be administered by a non-medical personnel. It takes about 4 min to complete. The BASDEC scale has been validated in patients with COPD. The BASDEC scale performed well against the GMS: having a sensitivity of 100%, a specificity of 93%; a positive predictive value of 91% and a negative predictive value of 100% [23]. The kappa score of BASDEC >7 against GMS >3 was 0.93. However, the BASDEC has not been adequately tested to test the efficacy of clinical intervention in patients with COPD. The minimal clinical importance difference is unknown.

The centre epidemiologic scale for depression (CES-D) is a self-complete questionnaire comprising of 20 items. Each item has a 4-point response choice ranges from 0 to 3. A total score of >16, out of 60 points, is considered to indicate the presence of depression [27]. In addition, the CES-D score can be employed as a continuous measure where higher scores are indicative of elevated depressive symptoms. It measures the presence of depression into three categories: normal, i.e. not depressed (0–15); mild (16–21); and moderate-to-severe depression (>21) [27]. In a recent study, it was found that CES-D has a sensitivity of 80% to identify major depression and a specificity around 70% [28] in COPD patients. However, further work is required to examine the efficacy of CES-D to detect clinically relevant change following an intervention in patients with COPD.

Beck’s depression inventory (BDI) is a 21-item self-administered rating inventory measuring attitudes and symptoms of depression, with high internal consistency, and good discriminates and convergent validity [29, 30]. It is scored 0–3, with the scores range from 0 to 63. The optimal cut-off score in the BDI >19 distinguishes patients with minimal or mild depressive symptoms from patients with moderate or severe depressive symptoms [30]. This cut-off point was previously used in a recent prospective study that enrolled COPD patients in a randomised controlled clinical trial [30]. It has been recommended as a clinical screening tool for depression in COPD patients by the American College of Chest Physician [31].

The DSM-IV criteria [12] define generalised anxiety Disorder as follows:

Researchers and clinicians, who recognise the need to identify patients with clinically significant anxiety and/or to measure anxiety levels to monitor interventions, have called for a reliable and easily administered screening and measurement tool [36, 45]. However, as Jain and Lolak asserted [46] in 2009, the most appropriate “gold standard” anxiety screening instrument for patients with COPD was yet to be identified. The majority of anxiety screening instruments that are used in clinical practice and within research settings have been developed in and for young healthy populations. Few scales have been specifically developed for use in elderly populations and none have been developed specially for patients with COPD where there is a lack of standardisation of appropriate measures [47]. Clinical guidelines recommend scales such as the Hospital Anxiety and Depression Scale-Anxiety (HADS-A [48]), Beck Anxiety Inventory (BAI [29]) and Depression Anxiety Stress Scales (DASS [49] for measuring and screening anxiety in patients with COPD. However, these scales, although popular within COPD-related research and clinical practice, have a number of documented shortcomings that may make them unsuitable for use in patients with chronic somatic disease, particularly COPD.

Whilst all people experience anxiety to some degree, most do not develop long-term anxiety disorders. Chronic, persistent or severe anxiety is typically classified, in terms of a medical approach, into one of the specific anxiety disorders (PD or GAD, for example), such as those proposed by the DSM-IV-TR criteria [12] or International Classification of Diseases-10 world health organisation [50]. This categorical system allows clinicians to decide whether or not to treat the patient. However, McDowell [51] posits that psychologists, in contrast to medical doctors, typically take a dimensional approach to anxiety, which treats the associated symptoms of anxiety on a continuum of severity. This distinction is characterised by the two styles of measurement: the medical model of dichotomous case or non-case, categorised by a clinical diagnosis, and the psychological model of ordinal measurement of symptom severity, often measured using scales and questionnaires.

GOLD [43] and NICE [52] guidelines recommend that all newly diagnosed COPD patients should undergo a detailed medical assessment, including the assessment of anxiety symptoms. The NICE [52] guidelines for COPD also indicate that clinicians should be alert to the presence of anxiety or depression in their patients. However, these COPD-specific guidelines fail to recommend clear strategies for identifying anxiety in this patient group. Although NICE (2010) guidelines [52] on the management of GAD and PD (with and without agoraphobia) recommend that a formal diagnosis of anxiety should be undertaken using a structured clinical interview, this is not always practical. Therefore, it is recommended that COPD patients seen in clinical settings are screened using self-report screening tools [31]. In clinical settings, a two-step approach is often incorporated in which patients are first screened using brief, inexpensive scales. Those patients who screen positive for anxiety usually undergo a more thorough assessment to confirm diagnosis with a clinical interview [53].

There are a number of barriers to the detection of anxiety and depression in patients with COPD. These typically fall into patient- or clinician-level barriers. Patient-level barriers to anxiety and depression detection include the stigma associated with mental illness which may lead patients with anxiety to exaggerate somatic complaints instead of acknowledging emotional problems, the reluctance to disclose anxiety symptoms and the confusion or masking that may occur in physical symptoms. Clinician-level barriers include the lack of a standardised assessment approach for patients with COPD, the lack of a disease-specific screening tool, the poor utilisation and uptake of existing screening tools, lack of confidence, skills and knowledge of anxiety symptoms and disorders, and the stigma of mental illness [1, 31, 36].

Such barriers may help to explain why in one recent study exploring the prevalence of anxiety disorders in patients with COPD, less than a third (29%) of patients with a clinical anxiety disorder had received a physician’s diagnosis [36].

In clinical practice and research settings, monitoring of anxiety symptoms and screening of anxiety disorders is typically undertaken using self-report anxiety scales. The following section focuses specifically on these scales and critically discusses their use in patients with COPD.

There are number of different scales have been utilised for the measurement and screening of anxiety symptoms and disorders in patients with COPD. Within this section, we critically review five scales that have been either recommended by clinical guidelines for COPD, are widely utilised in COPD-related research and/or are validated for use in patients with COPD (see Table 3.2). A summary of the scales’ psychometric properties is provided, with a focus on reliability and validity. Also, where appropriate, recommended cut-off values will be discussed in order to assess the clinical utility of these scales to screen for anxiety disorders.

Beck et al. [29] inventory is a self-report measure that was specifically designed to minimise confounding symptoms with depression and avoid the non-specific dimension of negative affect. The scale contains 21 items, with 14 items reflecting somatic symptoms of anxiety and panic. The BAI is recommended by the ACCP as a viable screening tool for use in COPD patients [31]. A few studies have utilised the BAI in COPD-related research [37, 54], yet the scale remains one of the most common instruments for measuring anxiety in general medical research [55].

Items are presented as a list of symptoms with respondents asked to rate on a four-point scale how much they have been bothered by each symptom in the preceding week. Scores range from 0 to 63. Beck and Steer’s [56] manual suggests that a cut-off point of ≤9 indicates normal levels of anxiety; 10–18 mild-moderate levels of anxiety; 19–29 moderate-severe levels of anxiety, and 30–63 severe levels of anxiety.