BACKGROUND: Even though the diagnosis of chronic obstructive pulmonary disease (COPD) is easy and based mainly on spirometry and symptoms, the prevalence of underdiagnosis is extremely high. The use of simple screening tools (e.g., questionnaires, hand-held spirometers) has been proved to be a simple method for case finding of COPD. Nevertheless the most appropriate target group of the general population has not been specified yet.OBJECTIVES: The aim of the present study was to compare 3 screening questionnaires among smokers aged >40 years in the primary care setting.METHODS: We excluded all subjects with a previous medical diagnosis of bronchial asthma or chronic pulmonary disease other than COPD. All participants were in a stable clinical condition, filled in the International Primary Care Airways Group (IPAG) questionnaire, the COPD Population Screener (COPD-PS) questionnaire, and the Lung Function Questionnaire (LFQ) and underwent spirometry. Medical diagnosis of COPD was established by an experienced pulmonologist.RESULTS: We studied 3,234 subjects during a 3.5-year period. COPD prevalence was 10.9% (52.1% underdiagnosis). All 3 questionnaires showed extremely high negative predictive values (94-96%), so in this case the diagnosis of COPD could be safely excluded. The area under the curve was similar across the 3 questionnaires (AUCROC: 0.794-0.809). The COPD-PS questionnaire demonstrated the highest positive predictive value (41%) compared to the other 2. On the other hand, the IPAG questionnaire and LFQ demonstrated higher sensitivities than COPD-PS resulting in lower percentages of missed cases.CONCLUSIONS: Three validated screening questionnaires for COPD demonstrated different diagnostic characteristics.