RESUMO
OBJECTIVE: To assess the measurement properties of the COPD assessment test (CAT) in a randomized trial comparing a face-to-face interview (FFI) with a telephone interview (TI). METHODS: A randomized study was conducted at two teaching hospitals in Fortaleza, Brazil. Patients were randomly assigned to answer the CAT questionnaire either in a FFI or by TI. The two groups were assessed for internal consistency reliability, cross-sectional validity and test-retest reliability. All patients performed spirometry and answered the modified medical research council dyspnea scale and the St. George's respiratory questionnaire (SGRQ). RESULTS: The total scores of the CAT questionnaire were similar for face-to-face and TI groups, 20.71 (95 % CI 18-23.4) versus 20.81 (95 % CI 19.31-21.7), respectively. For both mode of administration, we found good internal consistency reliability, the Cronbach's alpha ranged from 0.74 (95 % CI 0.61-0.84) to 0.89 (95 % CI 0.84-0.93) for the TI and FFI, respectively. In general, moderate-to-high correlations of CAT with SGRQ were observed, independent of the administration format. For the test-retest reliability, the intraclass correlation coefficients were very similar for both FFI and TI group 0.96 (95 % CI 0.93-0.97) versus 0.98 (95 % CI 0.96-0.98), respectively. CONCLUSION: This study demonstrated that the CAT questionnaire administration either in a FFI or by TI presents moderate-to-high measurement properties. This provides support for the use of both modes of questionnaire administration.