Establishing the Validity of Using the COPD Assessment Test to Screen for Fatigue in People With Chronic Obstructive Pulmonary Disease Referred to Pulmonary Rehabilitation

Abstract

Objective

Fatigue is the second most prevalent symptom in chronic obstructive pulmonary disease (COPD), yet it is often undetected in pulmonary rehabilitation. The aim of this study was to assess the validity of using a health status questionnaire (COPD Assessment Test [CAT] and CAT-energy score) to detect fatigue in people with COPD referred to a pulmonary rehabilitation program.

Methods

This study was a retrospective audit of people with COPD referred to pulmonary rehabilitation. The validity of the CAT-total score and CAT-energy score for detecting fatigue was analyzed compared to a validated fatigue questionnaire, the Functional Assessment of Chronic Illness Therapy-Fatigue (FACIT-F). Cut-off values defining fatigue included a CAT-total score ≥ 10, a CAT-energy score ≥ 2, and a FACIT-F score ≤ 43. Data were analyzed using 2 × 2 tables from which accuracy, sensitivity, specificity, and likelihood ratios were calculated.

Results

Data from 97 participants with COPD (age in years mean [SD] = 72 [9]; FEV₁% predicted mean [SD] = 46% [18]) were used. The FACIT-F score ≤ 43 classified 84 participants (87%) as fatigued. A CAT-total score ≥ 10 yielded an accuracy of 0.87, sensitivity of 0.95, specificity of 0.31, and positive and negative likelihood ratios of 1.38 and 0.15, respectively. A CAT-energy score ≥ 2 yielded an accuracy of 0.85, sensitivity of 0.93, a specificity of 0.31, and positive and negative likelihood ratios of 1.34 and 0.23, respectively.

Conclusion

The CAT-total score is an accurate and sensitive measure for fatigue, and the CAT could be an appropriate tool to screen for fatigue in people with COPD referred to pulmonary rehabilitation.

Impact

Use of the CAT as a screening tool for fatigue has the potential to improve clinician awareness of fatigue, simplify the pulmonary rehabilitation assessment process by reducing survey burden, and inform fatigue management, which may subsequently reduce the symptomatic burden of fatigue in people with COPD.

Introduction

Chronic obstructive pulmonary disease (COPD) is a chronic lung condition characterized by permanent, progressive airway obstruction resulting from an abnormal inflammatory response to noxious gas exposure.¹ COPD is a significant global health problem, with an estimated prevalence of 12%.² The combination of chronic bronchitis and emphysematous lung tissue changes in COPD manifests in pulmonary and extra-pulmonary symptoms.^1,2 Pulmonary symptoms include dyspnea, cough, and sputum production, whereas extra-pulmonary symptoms include fatigue, pain, muscle weakness, depression, and anxiety.^3,4 The presence of multiple concurrent symptoms is associated with worsening health and functional status.^3,4 Consequently, people with COPD report poor quality of life (QoL), which may be attributed to self-reported limitations in daily activities and social interactions.⁵

Fatigue is a common predictor of morbidity and mortality in chronic disease⁶ and has been described as a “complex phenomenon.”⁷ Ream & Richardson have defined fatigue as a “subjective, unpleasant symptom incorporating total body feelings ranging from tiredness to exhaustion creating an unrelenting overall condition which interferes with individuals’ ability to function to their normal capacity.”⁷ Hence, fatigue is considered an experience and an impact,⁸ which persists in chronic illnesses.^9,10 Acute fatigue follows periods of considerable effort or inadequate rest and can become chronic when it persists after simple activities and rest brings insufficient relief.¹¹ Psychological and physiological ramifications have been associated with the chronicity of fatigue, whereby limitations in an individual’s motivation, concentration, and mobility may reduce participation in work, recreation, and activities of daily living.¹¹ Fatigue, after dyspnea, is the second most prevalent symptom and most common extra-pulmonary symptom in people with stable COPD. A recent systematic review found that the prevalence of fatigue in people with COPD ranged from 17 to 95%.¹² Despite the high prevalence of fatigue, it remains underreported by people with COPD,¹³ delaying identification and implementation of effective clinical interventions.

It is poorly understood exactly how fatigue interacts with the disease process in COPD. Fatigue is associated with psychological, physiological, and disease–related components that consequently interact to impair health status, QoL and wellbeing in people with COPD.¹² A recent systematic review found fatigue strongly correlated with psychological factors including depression and anxiety, and moderately correlated with physiological factors such as reduced exercise capacity.¹² Despite the unclear nature of fatigue in COPD, there is some evidence that disease and respiratory–related factors can correlate with fatigue. The evidence remains unclear if the level of airway obstruction contributes to fatigue.¹² However, evidence exists suggesting that fatigue is intensified with dyspnea,^8,14 increasing exacerbation frequency,^3,15 in the presence of more than 3 comorbidities,¹⁶ with coexisting cardiac disease,¹⁴ and with medication use.⁹ A qualitative study explored the ramifications of fatigue on activities of daily living, and became a “heavy burden” that accelerated social isolation.⁵

Fatigue in COPD can be measured with unidimensional, multidimensional, or health–related quality of life (HRQoL) questionnaires.¹⁰ The Functional Assessment of Chronic Illness Therapy-Fatigue (FACIT-F) is a multi-dimensional fatigue questionnaire that has been used in pulmonary rehabilitation.¹⁷ The FACIT-F is a valid and reliable questionnaire to measure fatigue in people with COPD.¹⁸ However, the FACIT-F is rarely used in pulmonary rehabilitation programs, possibly due to its potential to increase the length and complexity of pulmonary rehabilitation assessments and contribute to survey burden.

The COPD Assessment Test (CAT) is a health status measure commonly used in pulmonary rehabilitation to measure the symptomatic burden of COPD on daily life.¹⁹ CAT-total scores correlate with FACIT-F scores in people with stable COPD.¹⁶ The CAT has a fatigue–related item referring to energy levels. Stridsman and colleagues²⁰ analyzed the correlation between fatigue (measured with FACIT-F) and both the CAT-total score and CAT-energy score in a European COPD population. Clinically relevant fatigue correlated with a CAT score ≥ 10 and a CAT-energy score ≥ 2.²⁰ These authors suggest that the CAT and CAT-energy score may identify clinically relevant fatigue. However, the participants in the study had a narrow range of disease severity as only mild to moderate COPD disease severity was represented, with a very small representation of severe COPD.

The question of how to best identify fatigue in people attending pulmonary rehabilitation has not been evaluated in a moderate to severe COPD population. Given that the CAT is already a popular measure used in pulmonary rehabilitation and has been shown to correlate with clinically relevant fatigue in a European population, it may serve as an appropriate fatigue screening tool. Stridsman and colleagues²⁰ provided cut-off values for the CAT-total score and CAT-energy score, which correlates with clinically relevant fatigue identified with the FACIT-F. To our knowledge, these cut-offs and screening tools are yet to be reviewed for their validity (accuracy, sensitivity, and specificity) in detecting fatigue in people with COPD. Validating the CAT-total score and CAT-energy score as appropriate screening tools may increase the identification of those with fatigue attending pulmonary rehabilitation programs.

The primary aim of this study was to assess the validity (accuracy, sensitivity, and specificity) of using a health status questionnaire (CAT-total score and CAT-energy score) to detect fatigue in people with COPD referred to a pulmonary rehabilitation program. We hypothesized that the CAT-energy and CAT-total scores would have high specificity, sensitivity, and accuracy in identifying those with clinically relevant fatigue.

Methods

Study Design

This study was a retrospective audit of a cohort of people with COPD referred to pulmonary rehabilitation at a major tertiary hospital in Sydney, Australia between August 2018 and January 2020. The dataset included results from a standardized pulmonary rehabilitation assessment that was completed by each participant before commencing the pulmonary rehabilitation program.

Study Population

Participants were included in the study from a convenience sample of people referred to a hospital outpatient pulmonary rehabilitation program with stable COPD. Diagnosis of COPD was made by a medical practitioner and confirmed via spirometry according to Global Initiative of Chronic Obstructive Lung Disease (GOLD) criteria, forced expiratory volume in 1 second (FEV₁)/ functional vital capacity (FVC) < 0.70 and severity by FEV₁ percentage predicted; GOLD 1: FEV₁ ≥ 80%; GOLD 2: 50% ≤ FEV₁ < 80%; GOLD 3: 30% ≤ FEV₁ < 50%; GOLD 4: FEV₁ < 30%.¹ Stability was defined as absence of an exacerbation in the previous 4 weeks. Participants were excluded from the study if they had a hospital admission within 4 weeks of assessment and if spirometry assessment did n ot meet the GOLD criteria for COPD (FEV₁ / FVC > 0.70).

Data Collection

The data were collected by an experienced physical therapist during a comprehensive assessment before pulmonary rehabilitation commencement. All patients provided written consent for their data from this assessment to be used for research studies. This study was approved by the South Eastern Sydney Local District Human Research Ethics Committee.

Primary Outcome Measures

The FACIT-F is a 13-item questionnaire that estimates the severity of fatigue in chronic disease²¹ and has been validated in a COPD population.¹⁸ Each item is scored between 0 and 4, amounting to a total score between 0 and 52. Lower scores indicate greater fatigue severity, whereby clinically relevant fatigue is defined as a score ≤ 43.^13,21 The FACIT-F score ≤ 43 is used as the reference standard for measuring fatigue in this study.

The CAT is a COPD-specific health status questionnaire that measures symptom burden and HRQoL.¹ It consists of 8 items about cough, phlegm, chest tightness, breathlessness, limitations in daily activities, confidence leaving home, sleep quality, and energy levels. Each item can be scored between 0 and 5, equating to a total score between 0 and 40. Higher scores indicate greater symptom burden and poorer HRQoL. A CAT score ≥ 10 represents high symptomatic burden and is the recommended cut-off for management of COPD.¹ A CAT score ≥ 10 and a CAT-energy score ≥ 2 were used to screen for fatigue in this study.

Additional Outcome Measures

Demographic characteristics collected during the initial pulmonary rehabilitation assessment included age, sex, smoking status, body mass index (BMI, kg/m²), and the presence of comorbidities. Lung function [FEV₁ (L) and FVC (L)] was measured during the initial pulmonary rehabilitation assessment using spirometry (Microlab, Carefusion, Lewiston, ME, USA) to confirm a diagnosis of COPD.

HRQoL was measured with the St George Respiratory Questionnaire (SGRQ), which is a respiratory disease-specific questionnaire that has been validated as a reliable assessment tool for people with COPD.²² The SGRQ is comprised of 50 items within 3 domains: symptoms, activity limitations, and impact on wellbeing. A total score, which is a weighted combination of scores from the 3 domains, ranges from no impact on QoL, 0, to maximal impact on QoL, 100.

Functional exercise capacity was measured with the 6-Minute-Walk-Test (6MWT). The 6MWT was performed according to American Thoracic Society/ European Respiratory Society (ATS/ERS) standards.²³ The distance walked (meters) in 6 minutes around a set track was measured and calculated as a percentage predicted to demographic norms.²³

Anxiety and depression were measured with the Hospital Anxiety and Depression Scale (HADS). The HADS is a 14-item scale in which each item is scored between 0 and 3 summating to a score out of 21 for anxiety and 21 for depression. A score ≥ 11 for either anxiety or depression is indicative of clinically relevant symptoms.²⁴

Sample Size

To achieve a sensitivity and specificity of at least 0.80, given a fatigue prevalence of at least 60% with a power of 0.8 and P-value < 0.05, a minimum sample size of 94 participants was required.²⁵

Data Analysis

SPSS version 25.0 (IBM Corp., Armonk, NY, USA) was used for statistical analysis of data.²⁶ A 2 × 2 contingency table was used to evaluate the accuracy, sensitivity, specificity, and positive and negative likelihood ratios (LRs) of the CAT and CAT-energy in detecting clinically relevant fatigue. The dichotomous reference test was the FACIT-F, and the screening tool was the CAT and the CAT-energy item. The criteria defining the presence of fatigue were a FACIT-F score ≤ 43,²¹ whereas the criteria used to screen for fatigue were a CAT-total score ≥ 10 and CAT-energy score ≥ 2.²⁰ Positive LRs will determine how much more likely a fatigued screening test result, for example a CAT-total score ≥ 10, is among fatigued participants, whereas negative LRs determine how much more likely a non-fatigued screening test result, for example a CAT-total score < 10, is among non-fatigued participants. Positive LR between 2 and 5, 5 and 10, and >10 provide weak, moderate, and strong evidence to rule in fatigue, respectively, whereas, negative LR between 0.2 and 0.5, 0.1 and 0.2, and <0.1 provide weak, moderate, and strong evidence to rule out fatigue, respectively.²⁷ Sensitivity, specificity, accuracy, and LR calculations are presented in Table 1.

Table 1.

2 x 2 Table Sensitivity, Specificity, Accuracy, and Likelihood Ratio Calculations of CAT-total Score ≥ 10 and CAT-energy Score ≥ 2^a

Screening Tool		Reference Test (FACIT-F Score ≤ 43)
		Yes	No
(CAT-total score ≥ 10) or (CAT-energy score ≥ 2)	Yes	(a) “true” fatigue	(b) “false” fatigue
	No	(c) “false” non-fatigue	(d) “true” non-fatigue

^a Calculations: accuracy = (a + d) / (a + b + c + d); positive likelihood ratio = sensitivity/(1 − specificity); prevalence = (a + c)/(a + b + c + d); negative likelihood ratio = (1 − sensitivity)/specificity; sensitivity = a/(a + c); specificity = d/(b + d). CAT-energy = COPD Assessment Test energy score; CAT-total = COPD Assessment Test total score; FACIT-F = Functional Assessment of Chronic Illness Therapy-Fatigue questionnaire.

Results

Characteristics of the Study Population

Data from 97 participants were used for analysis. The characteristics of participants are presented in Table 2. Over half of the participants (57%) were classified with severe or very severe airway obstruction (GOLD stages 3 and 4).¹ The mean number of comorbidities per participant was 2, with the majority (70%) of participants having 2 or more comorbidities. The proportion of participants having 0, 1, 2, and 3 or more comorbidities was 6, 24, 35, and 35%, respectively.

Table 2.

Characteristics of Participants^a

Characteristic	COPD (n = 97)
Sex, n (%)  Female  Male	55 (57) 42 (43)
Age y, mean (SD)	72 (9)
BMI, mean (SD)	27 (7)
Smoking status, n (%)  Former  Current  Never	81 (84) 12 (12) 4 (4)
Lung function, mean (SD)  FEV1 (L)  FEV1 (% predicted)  FVC (L)  FVC (% predicted)  FEV1 /FVC Disease severity, n (%)  GOLD stage  1  2  3  4	1.15 (0.50) 46 (18) 2.43 (0.72) 78 (21) 0.47 (0.13) 6 (6) 36 (37) 35 (36) 20 (21)
Comorbidities, n (%)  Cancer  Cardiac  Musculoskeletal  Metabolic  Neurological  Mental health	20 (21) 67 (69) 57 (59) 31 (32) 14 (14) 17 (18)
mMRC, mean (SD)	3 (1)
6MWD, mean (SD)  Distance, m  Distance (% predicted)	337 (114) 57 (18)
SGRQ total score, mean (SD)	48 (16)
HADS, mean (SD)  Anxiety  Depression	6 (4) 6 (4)

^a 6MWD = 6-Minute Walk Test distance; BMI = body mass index; COPD = chronic obstructive pulmonary disease; FEV₁ = forced expiratory volume in 1 second; FVC = functional vital capacity; GOLD = Global Initiative of Chronic Obstructive Lung Disease; HADS = Hospital Anxiety and Depression Scale; mMRC = Modified Medical Research Council Dyspnea Scale; SGRQ = St George’s Respiratory Questionnaire total score.

FACIT-F, CAT-Total, and CAT-Energy Scores

The FACIT-F, CAT-total, and CAT-energy scores are reported in Table 3. The prevalence of clinically relevant fatigue, being the proportion of participants scoring ≤ 43 on the FACIT-F, was 87%. Nearly all participants, 92%, had a high symptomatic burden, scoring ≥ 10 on the CAT. Similarly, 90% of participants had a CAT-energy score of ≥2.

Table 3.

FACIT-F and CAT Mean Scores and Cut-Offs Among Participants^a

Outcome Measure	Value
FACIT-F  Total score, mean (SD)  Fatigued (FACIT-F ≤ 43), n (%)  Not fatigued (FACIT-F > 43), n (%)	32 (10) 84 (87) 13 (13)
CAT  Total, mean (SD)  High symptomatic burden (CAT ≥ 10), n (%)	20 (8) 89 (92)
CAT-energy  Energy, mean (SD)  CAT-energy ≥ 2, n (%)	3 (1.2) 87 (90)

^a CAT = COPD Assessment Test; CAT-energy = individual item of the CAT-total questionnaire relating to energy levels; FACIT-F = Functional Assessment of Chronic Illness Therapy-Fatigue questionnaire.

The 2 × 2 table of the CAT-total score (screening tool) and FACIT-F (reference test) is shown in Table 4. The CAT-total score had a very high accuracy of 87%. A CAT-total score ≥ 10 had high sensitivity (0.95), correctly identifying 95% of those who were fatigued. The CAT-total score correctly identified 31% of those who were not fatigued, yielding a specificity of 0.31. A CAT-total score of ≥10 is 1.38 times more likely among participants with fatigue than among those without fatigue (positive LR = 1.38). A CAT-total score of <10 is less than one-sixth as likely among participants with fatigue than those without fatigue (negative LR = 0.15). Thus, the CAT-total score provides weak evidence to rule in clinically relevant fatigue and moderate evidence to rule out clinically relevant fatigue.²⁷

Table 4.

2 × 2 Table CAT-Total Score ≥ 10 Identifying Clinically Relevant Fatigue^a

Screening Tool		Reference Test (FACIT-F score ≤ 43)
		Yes	No	Sensitivity	Specificity	+LR	−LR
(CAT-total score ≥ 10)	Yes	80	9	0.95	0.31	1.38	0.15
	No	4	4

^a CAT-total = COPD Assessment Test total score; FACIT-F = Functional Assessment of Chronic Illness Therapy-Fatigue questionnaire.

The 2 × 2 table of the CAT-energy score (screening tool) and FACIT-F (reference test) is shown in Table 5. The CAT-energy score had a high accuracy of 85%. A CAT-energy score of ≥2 correctly identified 93% of those who were fatigued, yielding a sensitivity of 0.93. A score of <2 yielded the same specificity of a CAT-total score (0.31) and only identified 31% of those who were not fatigued. A CAT-energy score of ≥2 is 1.34 times more likely among participants with fatigue than those without fatigue (positive LR =1.34). A CAT-energy score of <2 is less than one-quarter as likely among participants with fatigue than those without fatigue (negative LR = 0.23). Thus, the CAT-energy score provides weak evidence to rule clinically relevant fatigue in and out.²⁷

Table 5.

2 × 2 Table CAT-Energy Score ≥ 2 Identifying Clinically Relevant Fatigue^a

Screening Tool		Reference Test (FACIT-F score ≤ 43)
		Yes	No	Sensitivity	Specificity	+LR	−LR
(CAT-energy score ≥ 2)	Yes	78	9	0.93	0.31	1.34	0.23
	No	6	4

^a CAT-energy = individual item of the CAT-total questionnaire relating to energy levels; FACIT-F = Functional Assessment of Chronic Illness Therapy-Fatigue questionnaire.

Discussion

This is the first study to identify that the CAT-total score is an accurate and sensitive measure for detecting fatigue in people with COPD referred to pulmonary rehabilitation. To our knowledge, no other study has assessed the validity of the CAT-total score and CAT-energy score in detecting participants with COPD who are fatigued or not fatigued. The results showed that a CAT-total score of ≥10 yielded the same specificity and a higher accuracy and sensitivity than a CAT-energy score of ≥2. The positive LR was similar for both a CAT-total score of ≥10 and CAT-energy score of ≥2 in ruling in fatigue, while the negative LR for the CAT-total score was superior to that of the CAT-energy score in ruling out fatigue. This study also identified a high prevalence of fatigue.

The prevalence of fatigue in people with COPD has been reported in numerous studies that used different fatigue measurement tools. The COPD cohorts within these studies have been varied and included hospital inpatients,³ pulmonary rehabilitation participants,^17,28,29 and population–based studies.^16,20 The prevalence of fatigue is higher in a pulmonary rehabilitation cohort of people with COPD (77%–95%)^17,28,29 in comparison to a general population cohort of people with COPD (37%).²⁰ The prevalence of fatigue in this study of pulmonary rehabilitation participants with moderate to severe COPD resembles Chen and colleagues’²⁸ pulmonary rehabilitation cohort, which had a similar severity of COPD.

Previous studies that used the FACIT-F to measure fatigue were population–based and cross-sectional epidemiological studies,^16,20 which found a lower prevalence of fatigue compared to our pulmonary rehabilitation cohort. The participants in epidemiological cohorts studied by Jones and colleagues¹⁶ and Stridsman and colleagues²⁰ presented with less severe airway obstruction and a very small proportion of GOLD 3 and 4 participants compared with our pulmonary rehabilitation cohort.^16,20 Despite this, conflicting evidence exists as to whether severity of airway obstruction contributes to fatigue.¹² Most participants in this current study presented with 2 or more coexisting comorbidities, which is similar to the pulmonary rehabilitation cohort of Wong and colleagues¹⁷ but greater than the population–based study of Jones and colleagues.¹⁶ The presence of comorbidities has correlated with the severity of fatigue.^8,12 Comorbidities may contribute to fatigue symptom severity, although more evidence is required to support this.¹² Those referred to pulmonary rehabilitation are seeking assistance for symptom management, including fatigue, and they may be more likely to report symptoms, consequential impacts on HRQoL, and comorbidities. These findings promote the inherent need to screen for, and subsequently appropriately manage fatigue, in pulmonary rehabilitation programs.

The validity results for the CAT-total score and CAT-energy are somewhat similar. Houben-Wilke and colleagues¹⁹ found that the CAT-energy item contributes the most to the CAT-total score compared to the other 7 items in the CAT, which could be the reason for accuracy agreement.¹⁹ Slight differences in validity scores may be due to other CAT items contributing to fatigue, as well as CAT-energy. Some elements of COPD assessed in the CAT are known to be associated with fatigue. For example, dyspnea is assessed in item 4 and is known to intensify fatigue severity.^8,14 Both the CAT-total score and CAT-energy score had high sensitivity, therefore both screening tools accurately identified those with clinically relevant fatigue. The specificity of the CAT-total and CAT-energy was less than 0.50, thus less accurately detected participants who were not fatigued. A very small proportion of non-fatigued participants in the sample, 13%, compared to a large proportion of fatigued participants, 87%, may describe the low specificities, as only 13 participants were included in the specificity ratios. Negative LR indicated that the CAT-energy score had small evidence to rule out fatigue, whereas the CAT-total score had moderate evidence to rule out fatigue. Positive LR for both screening tools indicated weak evidence to rule in clinically relevant fatigue. LRs are dependent on sensitivity and specificity scores, and, therefore, a low specificity score directly contributed to low LRs.

Despite the LR for the CAT-total score generating small and moderate changes in the probability of a participant being fatigued or non-fatigued, these changes can be important,²⁷ especially as these cut-off scores will determine the need, provision, and improvement of fatigue management in pulmonary rehabilitation. The CAT-total score produced a higher sensitivity than the CAT-energy score at identifying fatigue. The high sensitivity indicated that the CAT-total score will rarely miss detection of clinically relevant fatigue in someone with COPD referred to pulmonary rehabilitation. Although the CAT-total score of <10 only identified 31% of participants who were not fatigued, fatigue intervention and education would still be beneficial for those inaccurately identified as fatigued. This will promote fatigue self-management which may be required at a later stage of the disease, as fatigue intensifies in an exacerbation state,^3,14 during hospitalization, and as the disease progresses.¹⁵

We have found that the CAT-total score has been shown to be an appropriate tool to screen for fatigue in pulmonary rehabilitation participants, particularly with moderate to severe COPD. Research indicates that fatigue can improve with pulmonary rehabilitation intervention for people with COPD.^15,30,31 Only 2 systematic reviews have determined the effect of pulmonary rehabilitation on fatigue as a primary outcome.^30,31 These reviews provide low-quality evidence that pulmonary rehabilitation has a positive impact on subjective fatigue,³¹ and that no specific form of training (resistance, endurance, or combination exercise) is superior in managing and improving fatigue.³⁰ Paneroni and colleagues³¹ suggested that further evidence is required to clarify whether fatigue is more responsive to specific fatigue intervention or standardized pulmonary rehabilitation programs. Employing the CAT-total score as a fatigue screening tool will improve clinician awareness of fatigue, simplify the pulmonary rehabilitation assessment process, reduce patient survey burden, inform fatigue management, and subsequently may reduce the symptomatic burden of fatigue. A patient with a CAT-total score of ≥10 may present with clinically relevant fatigue, and this may subsequently require a change in intervention to assist in the management of fatigue.

Limitations

The retrospective design of this study allowed existing data to be analyzed, thereby using minimal resources. However, a limitation of the retrospective design is that the data are restricted to the measures taken in the pulmonary rehabilitation assessment. An additional limitation that reduces the generalizability of the results is that the data used in this retrospective audit were collected from only 1 institution. Furthermore, the fatigue measures taken were subjective in nature, as there are difficulties associated with measuring fatigue objectively, and therefore this introduced some element of recollection bias into the results. Despite this, the instruments used for primary outcome measures and participant characteristics were disease-specific, valid, and reliable.

Conclusion

This is the first study to identify that the CAT-total score is an accurate, sensitive, and an acceptable tool to screen for fatigue in those with COPD referred to pulmonary rehabilitation. This study found a high prevalence of fatigue in people with moderate to severe COPD, indicating the importance of screening for fatigue in pulmonary rehabilitation programs. The CAT-total score had a very high accuracy and could be a tool considered for use in pulmonary rehabilitation programs to screen for fatigue in people with COPD. This may potentially promote further assessment and management of fatigue in people with COPD.

Author Contributions

Concept/idea/research design: Z. Reizes, R.J. McNamara, Z. McKeough

Writing: Z. Reizes, R.J. McNamara, M. Dale, Z. McKeough

Data collection: Z. Reizes, R.J. McNamara

Data analysis: Z. Reizes, R.J. McNamara, M. Dale, Z. McKeough

Project management: Z. Reizes

Providing participants: R.J. McNamara

Providing facilities / equipment: R.J. McNamara, Z. McKeough

Providing institutional liaisons: Z. McKeough

Consultation (including review of manuscript before submitting): R.J. McNamara, M. Dale, Z. McKeough

Funding

There are no funders to report for this study.

Ethics Approval

This study was reviewed by the South Eastern Sydney Local District Human Research Ethics Committee and deemed to be a quality improvement study not requiring independent ethics review in accordance with the NSW Health Guideline GL2007_020 Human Research Ethics Committees—Quality Improvement and Ethics Review: A Practice Guide for NSW.

Disclosures

The authors completed the ICMJE Form for Disclosure of Potential Conflicts of Interest and reported no conflicts of interest.

Data Availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.