Six commonly used COPD screening questionnaires demonstrated significantly worse diagnostic accuracy in never-smokers (AUROCs 0.530-0.574) compared to the general population (AUROCs 0.579-0.653; P<0.05).
Cross-Sectional (n=38,122)
Yes
Do existing COPD screening questionnaires accurately identify COPD in never-smokers compared to the general population?
Commonly used COPD screening questionnaires have suboptimal diagnostic performance in never-smokers, highlighting the need for targeted screening tools for this population.
Effect estimate: AUROC 0.530-0.574 vs 0.579-0.653
p-value: p=<0.05
Abstract Rationale Chronic obstructive pulmonary disease (COPD) in never-smokers represents a significant yet under-recognized burden, particularly in China. Existing screening questionnaires, predominantly developed in populations including smokers, may perform poorly in this subgroup. We evaluated the diagnostic accuracy of six COPD screening questionnaires in a general Chinese population and among never-smokers. Methods This analysis included 38,122 adults aged ≥40 years from the nationwide China Pulmonary Health (CPH) study, comprising 27,130 (71.2%) never-smokers. COPD was defined by post-bronchodilator FEV1/FVC 0.70. The performance of six questionnaires (PUMA, CODE, COPD-SQ, COPD-Q, SCSQ, CAPTURE) was assessed using the area under the receiver operating characteristic curve (AUROC), sensitivity, specificity, and decision curve analysis (DCA) in all participants and never-smokers separately. Results The prevalence of spirometry-defined COPD was 12.3% overall and 8.7% among never-smokers. Non-smoking COPD patients were predominantly female (80.5%), had higher BMI, and reported fewer respiratory symptoms (chronic cough: 12.5% vs. 16.8%; chronic phlegm: 14.9% vs. 22.2%) compared to the overall COPD population. In the general population, the questionnaires showed fair accuracy (AUROCs: 0.579-0.653). However, their performance significantly declined in never-smokers (AUROCs: 0.530-0.574; all P 0.05 by DeLong test). At recommended cut-offs, sensitivity was notably low in never-smokers. DCA indicated limited clinical net benefit for all questionnaires in the never-smoker subgroup. Adjusting cut-off values improved sensitivity but compromised specificity and overall accuracy. Conclusions Commonly used COPD screening questionnaires demonstrate suboptimal and significantly worse diagnostic performance in never-smokers compared to the general population. The distinct demographic and clinical characteristics of non-smoking COPD highlight the urgent need for the development and validation of targeted screening tools for this large and underserved population. This abstract is funded by: Noncommunicable Chronic Diseases-National Science and Technology Major Project (2023ZD0506003)
Yan et al. (Fri,) conducted a cross-sectional in Chronic obstructive pulmonary disease (COPD) (n=38,122). Six COPD screening questionnaires (PUMA, CODE, COPD-SQ, COPD-Q, SCSQ, CAPTURE) vs. General population was evaluated on Diagnostic accuracy (AUROC) of six COPD screening questionnaires (AUROC 0.530-0.574 vs 0.579-0.653, p=<0.05). Six commonly used COPD screening questionnaires demonstrated significantly worse diagnostic accuracy in never-smokers (AUROCs 0.530-0.574) compared to the general population (AUROCs 0.579-0.653; P<0.05).