COPD Assessment Test

Abstract

The Chronic Obstructive Pulmonary Disease (COPD) Assessment Test (CAT) is a validated, patient-completed psychometric instrument engineered to quantify the global impact of COPD on health-related quality of life (HRQoL) and daily functioning. Developed through an international collaboration led by Paul W. Jones and colleagues in 2009, the instrument emerged in response to the clinical need for a short, psychometrically robust alternative to exhaustive research questionnaires such as the St. George’s Respiratory Questionnaire (SGRQ). The CAT consists of 8 items spanning core respiratory symptoms (cough, sputum expectoration, chest tightness, and exertional breathlessness) and functional/psychological impairments (domestic activity limitations, confidence leaving the home, sleep disturbance, and vitality/fatigue).

Each item is evaluated via a 6-point semantic differential scale anchored from 0 to 5, generating a cumulative continuous score ranging from 0 to 40, where higher scores signify greater disease-related health status impairment. Methodologically refined using Item Response Theory (IRT) and Rasch analysis, the CAT demonstrates high unidimensional construct validity, remarkable internal consistency (Cronbach’s α typically ranging between 0.85 and 0.92 across clinical cohorts), and substantial test-retest reliability (intraclass correlation coefficients [ICC] ≥ 0.80). The instrument is widely incorporated into international guidelines, notably the Global Initiative for Chronic Obstructive Lung Disease (GOLD) strategy, for symptomatic classification and longitudinal therapeutic monitoring, featuring an established minimally clinically important difference (MCID) of 2 points.

Keywords

COPD Assessment Test, CAT, Chronic Obstructive Pulmonary Disease, Health-Related Quality of Life, Rasch Analysis, Patient-Reported Outcome Measures, Respiratory Psychometrics, Exertional Dyspnea, GOLD Guidelines, Disease Impact Quantification

Authors

The COPD Assessment Test was conceptualized, developed, and validated by an international, multidisciplinary steering group of academic pulmonologists and patient-reported outcome (PRO) psychometricians:

• Paul W. Jones, PhD, FRCP — Professor of Respiratory Medicine, Division of Clinical Science, St George’s, University of London, United Kingdom.
• Gordon Harding, MSc — Psychometrician and Health Outcomes Researcher, United BioSource Corporation, London, United Kingdom.
• Peter Berry, PhD — Respiratory Clinical Development and Biostatistics, GlaxoSmithKline, Uxbridge, Middlesex, United Kingdom.
• Ingela Wiklund, PhD — Professor and PRO Methodologist, United BioSource Corporation, London, United Kingdom.
• Wan-Cheng Chen, MD — Pulmonologist and Clinical Investigator, Respiratory Therapy Department, Taipei Veterans General Hospital, Taipei, Taiwan.
• Nikolina Kline Leidy, PhD — Senior Vice President and Senior Scientific Director, Center for Health Outcomes Research, United BioSource Corporation, Bethesda, Maryland, USA.

Developmental oversight and funding were facilitated through an unrestricted scientific grant from GlaxoSmithKline, with instrument governance supported by clinical experts to ensure global public accessibility for academic and clinical use.

Purpose

Chronic Obstructive Pulmonary Disease has historically been assessed, graded, and managed primarily using spirometric indices, specifically the forced expiratory volume in one second (FEV₁). However, extensive epidemiological and clinical evidence demonstrates that FEV₁ correlates poorly with the lived experience of patients, capturing only a fraction of the variance in exertional dyspnea, systemic fatigue, daily functional capacity, psychological distress, and overall well-being. Prior to the introduction of the CAT, measuring patient-reported disease impact required extensive, multi-item research batteries, notably the 50-item St. George’s Respiratory Questionnaire (SGRQ) or the 20-item Chronic Respiratory Questionnaire (CRQ). Although psychometrically pristine, these legacy instruments impose prohibitive respondent and administrative burdens, rendering them functionally impractical for routine outpatient consultations, acute inpatient triage, or dynamic longitudinal monitoring.

The COPD Assessment Test was deliberately designed to resolve this clinical-translational gap. Its explicit primary purpose is to provide an administratively brief, easily interpretable, patient-completed measurement tool that yields a reliable, standardized metric of overall COPD-specific health status impairment. Clinically, the CAT fulfills several critical roles:

• Symptom Burden Stratification: In the GOLD management framework, the CAT serves as the primary recommended multidimensional symptom evaluation tool, determining whether a patient exhibits low symptom burden (CAT < 10) or high symptom burden (CAT ≥ 10), which directly drives pharmacological escalation and de-escalation algorithms.
• Facilitating Patient-Clinician Communication: By providing structured anchor points encompassing systemic and psychological dimensions (such as homebound insecurity and fatigue), the scale brings non-spirometric, patient-centered consequences of the disease directly into the clinical dialogue.
• Monitoring Exacerbation Recovery and Trajectory: The CAT tracks acute flare-ups and assesses recovery trajectories following interventions such as systemic corticosteroids, pulmonary rehabilitation, or bronchodilator optimization.
• Clinical Trial Endpoints: It serves as a validated secondary or primary PRO endpoint in interventional clinical trials assessing novel pharmacotherapies, digital health interventions, and non-pharmacological therapies.

Psychological Construct

The primary psychological and health construct measured by the CAT is COPD-specific health status impairment, conceptualized as a unidimensional latent trait reflective of the holistic burden imposed by the disease on an individual’s physical, functional, and psychological everyday life. Although the instrument produces a single composite score, its 8 constituent items purposefully sample four interrelated operational domains of the chronic illness experience:

1. Pulmonary-Specific Somatic Symptoms

The first three items assess classic respiratory symptoms: chronic cough, mucus hypersecretion (sputum/phlegm), and chest tightness. Psychometrically, these items capture the immediate somatic distress caused by airway inflammation, hypersecretion, and bronchoconstriction. For instance, Item 1 evaluates the continuous cognitive and physical irritation of persistent coughing, while Item 3 evaluates visceral sensory discomfort (tightness), which often induces acute illness-related anxiety and panic responses in vulnerable pulmonary patients.

2. Exertional Dyspnea and Physical Limitation

Item 4 examines dyspnea elicited during everyday physical exertion (walking up a hill or ascending a single flight of stairs). Breathlessness is the hallmark symptomatic driver of functional disability in COPD. This item evaluates not merely the mechanical sensation of respiratory insufficiency, but the functional ceiling it imposes upon standard mobility tasks, capturing the initial boundary where physical pathology transforms into behavioral restriction.

3. Functional Autonomy and Psychosocial Agency

Items 5 and 6 expand beyond direct physiological symptoms into behavioral adaptation, social role functioning, and psychological security. Item 5 assesses domestic limitation, indexing how domestic activities of daily living (cleaning, self-care, cooking) are curtailed. Item 6 addresses the psychological construct of task-specific self-efficacy and perceived vulnerability: confidence leaving the home environment. COPD frequently engenders agoraphobia-like avoidance behaviors, where fear of acute breathlessness in public spaces precipitates severe social isolation and dependency.

4. Systemic Physiological Well-Being and Vitality

The terminal items, Items 7 and 8, evaluate systemic manifestations that bridge pulmonary pathology with neuropsychological well-being. Item 7 quantifies sleep disturbance directly attributed to pulmonary symptoms, recognizing nocturnal hypoxemia, orthopnea, and paroxysmal nocturnal cough as major disruptors of restorative architecture. Item 8 measures global energy versus pervasive fatigue/vitality loss. Pervasive fatigue in COPD represents a multi-factorial biological-psychological construct driven by elevated work of breathing, systemic cytokine elevation, deconditioning, and chronic micro-arousals during sleep.

Theoretical Framework

The architecture of the COPD Assessment Test is fundamentally rooted in modern psychometric theory—specifically Rasch Measurement Theory (a unidimensional formulation of Item Response Theory)—combined with the Biopsychosocial Model of chronic illness originally articulated by George Engel (1977) and operationalized for quality of life by Wilson and Cleary (1995).

The Wilson and Cleary Model of Patient Outcomes

Wilson and Cleary conceptualized health outcomes along a progressive causal continuum: biological and physiological variables → symptom status → functional status → general health perceptions → overall quality of life. Traditional respiratory medicine operated almost exclusively at the biological tier (FEV₁, arterial blood gases). The CAT bridges the levels of symptom status (cough, phlegm, tightness, dyspnea), functional status (domestic activities, stair climbing), and general health perceptions (confidence, energy, systemic impact). By anchoring physiological sensations directly to their psychological and functional sequelae, the instrument captures the non-linear cascade through which structural lung damage influences subjective illness perception.

Rasch Measurement Paradigm

From an epistemological and psychometric perspective, the CAT was developed using Rasch analysis rather than Classical Test Theory (CTT). The underlying Rasch paradigm dictates that the likelihood of an individual endorsing a particular severity category on an item is a logistic mathematical function of two independent parameters: the person’s latent level of health impairment (θ) and the specific item’s intrinsic calibration difficulty or severity (β):

P(X_nij = 1 | θ_n, β_i, τ_j) = f(θ_n – β_i – τ_j)

During the candidate item selection phase, an initial pool of 21 candidate items was systematically evaluated in diverse international cohorts. Rasch analysis was applied to test for unidimensionality, item fit (infit and outfit statistics), item difficulty hierarchy, and invariance across language, culture, gender, and age (Differential Item Functioning, DIF). The final 8 items were retained because they demonstrated ordered response thresholds, conformed strictly to equal-interval scale properties along the latent continuum, and exhibited structural invariance across diverse geographic regions.

Validity

The psychometric validity of the COPD Assessment Test has been established across hundreds of independent cross-sectional, longitudinal, and interventional clinical studies.

Construct and Convergent Validity

Convergent validity was initially demonstrated in the seminal validation studies conducted by Jones et al. (2009). The CAT total score demonstrated a strong, statistically significant correlation with the legacy gold standard, the St. George’s Respiratory Questionnaire (SGRQ), with Pearson correlation coefficients consistently falling between r = 0.80 and r = 0.84 (p < 0.0001). Correlations with the SGRQ Symptom, Activity, and Impact subscales were similarly robust (typically r = 0.65 to 0.78). Furthermore, the CAT correlates moderately to strongly with the Modified Medical Research Council (mMRC) dyspnea scale (Spearman’s ρ = 0.61 to 0.67), the Chronic Respiratory Questionnaire (CRQ, r = -0.69 to -0.74), and generic health status measures such as the EuroQol 5-Dimension (EQ-5D, r = -0.58 to -0.68).

Discriminant and Known-Groups Validity

The CAT reliably distinguishes between patient groups differentiated by clinical status. When stratified according to airflow limitation severity under the GOLD classification (Stage I through IV), CAT scores progressively rise, reflecting step-wise deterioration in health status, although the moderate correlation with FEV₁ percent predicted (typically r = -0.24 to -0.35) confirms that CAT measures an independent clinical domain not captured by spirometry alone. More critically, the CAT exhibits exceptional discriminant capacity between stable outpatients and patients experiencing an acute exacerbation of COPD (AECOPD). During acute exacerbation presentations, mean CAT scores routinely rise by 4.5 to 8.0 units above baseline, returning toward baseline levels following therapeutic recovery.

Predictive Validity

Longitudinal cohorts have corroborated the prognostic power of the CAT score. Elevated baseline CAT scores (≥ 21) independently predict an increased risk of future moderate-to-severe exacerbations, higher rates of hospital readmission within 30 and 90 days following discharge, worse clinical trajectory in pulmonary rehabilitation, and elevated all-cause and respiratory-specific mortality, even after multivariable adjustment for age, comorbidities, smoking status, and baseline FEV₁.

Reliability

The CAT demonstrates strong reliability parameters across diverse demographic groups, cultural settings, and disease severities.

Internal Consistency

In the original validation trial encompassing cohorts in the United Kingdom and the United States, Cronbach’s α coefficient was calculated at 0.88. Subsequent international validation initiatives across Europe, Asia, South America, and Africa have consistently documented Cronbach’s α values ranging from 0.85 to 0.92. This indicates that while the items evaluate distinct facets of impairment (from physical phlegm to psychological confidence), they reliably measure a single underlying construct without excessive item redundancy.

Test-Retest Reliability

The temporal stability of the CAT in stable, non-exacerbating COPD patients has been evaluated across multiple retest intervals ranging from 2 hours to 2 weeks. The intraclass correlation coefficient (ICC) consistently ranges between 0.80 and 0.96. In the primary derivation paper, stable patients demonstrated a test-retest correlation coefficient of r = 0.80, with a mean score difference between testing visits of fewer than 0.5 units, confirming that spontaneous measurement noise is negligible in the absence of clinical change.

Measurement Error and Responsiveness (MCID)

The Standard Error of Measurement (SEM) for the total score is estimated at approximately 2.0 to 2.4 points. Across anchor-based and distribution-based studies, the Minimally Clinically Important Difference (MCID) has been established at 2.0 points. A decline of ≥ 2 points indicates a clinically meaningful improvement in response to bronchodilators, anti-inflammatory therapies, or pulmonary rehabilitation, whereas a sustained rise of ≥ 2 points indicates clinically meaningful clinical deterioration.

Factor Analysis

The latent structural integrity of the CAT has been scrutinized using both classical Exploratory Factor Analysis (EFA), Confirmatory Factor Analysis (CFA), and Item Response Theory Rasch modeling.

Unidimensionality and Rasch Diagnostics

During scale construction, principal component analysis of residuals within the Rasch framework confirmed that the primary latent dimension accounted for over 50% of total variance. The Rasch infit and outfit mean square (MNSQ) statistics for all 8 items fell cleanly within the acceptable psychometric boundary of 0.70 to 1.30 (observed values ranged from 0.77 to 1.25). This confirms that no single item introduces unacceptable distortion or multidimensional contamination. Analysis of residual correlations revealed minimal local dependency between items (all residual correlations < 0.25), proving that each item provides unique informational value along the continuum of latent trait impairment.

Confirmatory Factor Analysis (CFA)

Subsequent independent validation studies applying CFA have uniformly supported a single-factor, unidimensional structural model. Goodness-of-fit parameters across representative large cohorts consistently fulfill rigorous structural equation modeling standards:

• Comparative Fit Index (CFI): ≥ 0.96
• Tucker-Lewis Index (TLI): ≥ 0.95
• Root Mean Square Error of Approximation (RMSEA): 0.045 – 0.062 (90% CI [0.035, 0.071])
• Standardized Root Mean Square Residual (SRMR): ≤ 0.040

Standardized factor loadings across all 8 items are consistently high, ranging from λ = 0.60 to 0.85. Although minor secondary covariance is occasionally observed between somatic items (Item 1 [cough] and Item 2 [phlegm]) and between functional items (Item 5 [domestic activity] and Item 6 [leaving the home]), bifurcated or two-factor models (e.g., partitioning items into ‘Physical Symptoms’ vs. ‘Functional Consequences’) fail to demonstrate statistically superior fit compared to the parsimonious, mathematically robust single-factor model.

Instrument / Measurement Tool

• Test Type: Patient-Reported Outcome Measure (PROM) / Self-administered clinical health status questionnaire.
• Target Population: Adults and elderly individuals diagnosed with Chronic Obstructive Pulmonary Disease (COPD); adaptable to chronic bronchitis and emphysema phenotypes.
• Administration Format: Paper-and-pencil self-report, digital/electronic interface (e-PRO), tablet-based clinic check-in, or clinician-facilitated interview.
• Time Required for Completion: Approximately 2 to 3 minutes.
• Number of Items: 8 items.
• Response Scale: 6-point semantic differential scale (scored integer values 0 to 5) framed between two contrasting semantic statements representing opposing polarities of impairment.
• Scoring Rules: Each item is scored from 0 (lowest impairment/absence of symptom) to 5 (maximal impairment/constant symptom). The total score is computed as the direct algebraic sum of all 8 items: Total Score = ∑(Item 1 to Item 8). No items are reverse-scored.
• Score Range: 0 to 40 points, where 0 indicates zero self-perceived disease impact and 40 represents catastrophic disease impact.
• Severity Categorization and Clinical Stratification:
  • Score < 10 (Low Impact): Most days are good; patient can undertake most desired activities despite mild symptoms. Emphasizes smoking cessation, annual vaccinations, and exercise.
  • Score 10 – 20 (Medium Impact): COPD is one of the main problems in the patient’s daily life. Cough, phlegm, and breathlessness restrict activities; physical exhaustion occurs regularly. Threshold for pharmacological escalation under GOLD guidelines.
  • Score 21 – 30 (High Impact): Severe disease impact. Routine domestic chores are curtailed; walking up stairs or leaving the house is severely restricted; frequent sleep disruption. Indication for specialist referral, combination bronchodilators, and pulmonary rehabilitation.
  • Score > 30 (Very High Impact): Profound, catastrophic impairment. Patient is largely homebound, severely restricted in self-care, and suffers from unremitting exhaustion and respiratory distress. Requires intensive multidisciplinary intervention and advanced therapy evaluation.

Permissions & Fee and Test Year

The COPD Assessment Test was officially published in 2009 by Paul W. Jones and colleagues. The copyright is held by the COPD Assessment Test multi-stakeholder governance structure (originally funded and managed in collaboration with GlaxoSmithKline Services Unlimited).

The questionnaire is freely available for routine clinical practice and non-funded academic research without royalty fees. To maintain psychometric integrity and uniform linguistic validation, users are required to utilize authorized, validated translations and may not modify item phrasing, scoring weights, or formatting. Information regarding authorized electronic implementations, commercial licensing for pharmaceutical clinical trials, and certified language translations is managed through the official CAT website (www.catestonline.org) and Mapi Research Trust / ePROVIDE platform.

References

• Engel, G. L. (1977). The need for a new medical model: A challenge for biomedicine. Science, 196(4286), 129–136. https://doi.org/10.1126/science.847460
• Global Initiative for Chronic Obstructive Lung Disease (GOLD). (2024). Global strategy for the diagnosis, management, and prevention of chronic obstructive pulmonary disease (2024 report). GOLD. https://goldcopd.org/
• Jones, P. W., Brusselle, G., Dal Negro, R. W., Ferrer, M., Hussain, N. Y., Kardos, P., Lusuardi, M., Price, D., Sitter, H., Wencker, M., & Roberts, J. (2011). Properties of the COPD assessment test in a cross-sectional European study. European Respiratory Journal, 38(1), 29–35. https://doi.org/10.1183/09031936.00177210
• Jones, P. W., Harding, G., Berry, P., Wiklund, I., Chen, W. C., & Kline Leidy, N. (2009). Development and first validation of the COPD Assessment Test. European Respiratory Journal, 34(3), 648–654. https://doi.org/10.1183/09031936.00102509
• Jones, P. W., Tabberer, M., & Chen, W. H. (2011). Creating scenarios of the impact of COPD and their relationship to COPD Assessment Test (CAT™) scores. BMC Pulmonary Medicine, 11(1), Article 42. https://doi.org/10.1186/1471-2466-11-42
• Kon, S. S., Canavan, J. L., Jones, S. E., Nolan, C. M., Clark, A. L., Dickson, M. J., Haselden, B. M., Polkey, M. I., & Man, W. D. C. (2014). Minimum clinically important difference for the COPD Assessment Test: A prospective analysis. The Lancet Respiratory Medicine, 2(3), 195–203. https://doi.org/10.1016/S2213-2600(14)70001-3
• Wilson, I. B., & Cleary, P. D. (1995). Linking clinical variables with health-related quality of life: A conceptual model of patient outcomes. JAMA, 273(1), 59–65. https://doi.org/10.1001/jama.1995.03520250075037

Items of the Scale