Abstract
COPD definition requires post-bronchodilator (BD) spirometry. In epidemiological studies where post-BD spirometry was not performed, COPD is generally defined using pre-BD spirometry. However, the extent of misclassification in this definition is unknown.
Using the data of 3,332 subjects who underwent spirometry both in the European Community Respiratory Health Survey II (pre-BD) (1999/2001) and III (pre + post-BD) (2010/2014), we assessed the validity of a definition of pre-BD “chronic” (present at both ECRHS II and III) airflow obstruction, towards post-BD (“fixed”) obstruction at ECRHS III (the gold standard). Airflow obstruction was defined as FEV1/FVC<lower limit of normal.
Subjects were 54.1±7.1 year-old at baseline, 53% were women. Chronic and fixed obstruction were present in 163 (4.9%) and 188 (5.6%) individuals, respectively. Subjects who had obstruction according to both definitions (n=136) were more likely to be smokers (73 vs 54%) and to have asthma (62 vs 19%) than subjects without chronic/fixed obstruction (n=3,117) (both p<0.002). The definition of chronic obstruction had sensitivity=83.4% [95%CI] (76.8-88.8%), specificity=98.4% (97.9-98.8%), positive predictive value=72.3% (65.4-78.6%), and negative predictive value=99.1% (98.8-99.4%). Sensitivity was higher in subjects with a history of asthma (92.3 vs 72.2%) than in subjects without, and the opposite was true for specificity (94.8 vs. 99.2%, respectively) (both p<0.001). Predictive values were similar between subjects with/without asthma (p>0.05).
In young adults, pre-BD airflow obstruction confirmed at two occasions 9 years apart correctly identified the large majority of subjects with post-BD airflow obstruction.
- Copyright ©ERS 2015