Anxiety is a frequent comorbidity in patients with Chronic Obstructive Pulmonary Disease (COPD) and with important repercussions, such as functional impairment, decreased quality of life and adherence to treatment, and increased risk of hospitalization. In this context, the Anxiety Inventory for Respiratory Diseases (AIR), the only instrument specifically developed to investigate anxiety symptoms in this population, was cross-culturally adapted for use in Brazil. Still, the measurement properties of the face-to-face version have not yet been investigated.

To investigate the internal consistency, test-retest and inter-rater reliability, convergent and divergent validity of the face-to-face AIR in patients with COPD.

This is a cross-sectional methodological study. On the first day, after signing the Free and Informed Consent Form, the Mini-Mental State Examination, the sociodemographic and clinical questionnaire were applied, and anthropometric data were collected, followed by spirometry. Subsequently, patients answered the Modified Medical Research Council (mMRC), the COPD Assessment Test (CAT), the AIR, the Hospital Anxiety and Depression Scale (HADS), the Beck Anxiety Inventory (BAI), and the London Chest Activity of Daily Living (LCADL). To determine the test-retest and inter-rater reliability of the IAR, seven days after the first application of the questionnaire, the leading researcher reapplied the scale with a control form via telephone contact, and a second rater contacted patients 48 hours later. The internal consistency of the AIR was measured using Cronbach's alpha and reliability using the Intraclass Correlation Coefficient (ICC). Spearman's correlation test (rho) was used to determine validity (p<0.05).

Twelve individuals [7 female, age 60.5 (min. 53.7-max.70.8) years] participated in the study; nine were included in the test-retest reliability analyses and eight in the inter-rater reliability. For internal consistency, Cronbach's α was 0.94. For test-retest reliability, the ICC was 0.73; for inter-rater reliability, it was 0.88. Regarding convergent validity, significant correlations of high magnitude were observed between the AIR and the HADS anxiety domain (rho = 0.82), depression domain (rho = 0.87), and the total score (rho = 0.79), and of moderate magnitude between the AIR and the BAI (rho = 0.70). For divergent validity, non-significant correlations of weak magnitude were observed between AIR and LCADL (rho = 0.39), AIR and CAT (rho = 0.10), and AIR and MRC (rho = 0. 17).

The study's preliminary results indicate that the AIR demonstrates adequate measurement properties for assessing anxiety symptoms in patients with COPD. The study will be continued to expand the sample number.

This study will provide a measurement instrument for investigating anxiety symptoms in patients with COPD with appropriate measurement properties, which may facilitate early identification and proper treatment.