Validation of symptom-based COPD questionnaires in Japanese subjects

Background and objective:  Symptom-based questionnaires may be helpful in diagnosing patients with COPD. The aim of this study was to determine whether two COPD questionnaires designed in Western countries were applicable to Japanese and other Asian patients. Methods:  The participants were Japanese people aged 40 years and over. Each subject answered questions on demographics and symptoms and underwent spirometry before and after administration of a bronchodilator. Questionnaire A was designed to identify previously undiagnosed COPD and questionnaire B was designed to differentiate between COPD and asthma. Results:  The numbers of COPD patients who answered questionnaires A and B were 33 of 169 (19.5%) and 112 of 168 (66.7%), respectively. Comparison of the COPD group with the non-COPD group revealed a significant difference in total score in both questionnaire A and questionnaire B (both P < 0.001). The area under the receiver operating characteristic curve (AUC-ROC) for questionnaire A was 0.791. With a cut-off value of 16.5 points, the sensitivity and specificity were 0.939 and 0.404, and with a 19.5-point cut-off, sensitivity and specificity were 0.848 and 0.647, respectively. The AUC-ROC for questionnaire B was 0.765. With cut-off values of 18.5 and 24.5 points, the respective sensitivities and specificities were 0.946 and 0.393, and 0.741 and 0.607. Conclusions:  A simple self-administered questionnaire can help to diagnose COPD in Japanese subjects. When these questionnaires are used in Japan, cut-off values should be set somewhat higher than in Western countries.