Logo image
Back
Prevalence of abnormal spirometry in individuals with a smoking history and no known obstructive lung disease
Journal article   Peer reviewed

Prevalence of abnormal spirometry in individuals with a smoking history and no known obstructive lung disease

Thuonghien V. Tran, Gregory L. Kinney, Alejandro Comellas, Karin F. Hoth, Arianne K. Baldomero, A. James Mamary, Jeffrey L. Curtis, Nicola Hanania, Richard Casaburi, Kendra A. Young, …
Respiratory medicine, Vol.208, 107126
03/2023
DOI: 10.1016/j.rmed.2023.107126
PMCID: PMC9990311
PMID: 36717002

View Online

Abstract

Introduction Recent evidence suggests a high prevalence of undiagnosed chronic obstructive pulmonary disease (COPD). These individuals are at risk of exacerbations and delayed treatment. We analyzed an at-risk population for the prevalence of abnormal spirometry to provide clarity into who should undergo early spirometry. Methods We analyzed data from the COPDGene study. Participants with ≥10 pack-years of smoking were included. Individuals with self-reported or physician-diagnosed COPD, asthma, chronic bronchitis, emphysema and/or were on inhalers were excluded. Parsimonious multivariable logistic regression models identified factors associated with abnormal spirometry, defined as either airflow obstruction (AFO) or preserved ratio impaired spirometry. Variables were selected for the final model using a stepwise backward variable elimination process which minimized Akaike information criterion (AIC). Similarly, during the 5-year follow-up period, we assessed factors associated with incident diagnosis of COPD. Results Of 5055 individuals, 1064 (21%) had undiagnosed AFO. Age, pack-years, current smoking and a history of acute bronchitis were associated with AFO while body mass index, female sex, and Black race were inversely associated. Among 2800 participants with 5-year follow-up, 532 (19%) had an incident diagnosis of COPD. Associated risk factors included mMRC ≥2, chronic productive cough, respiratory exacerbations during the follow-up period, and abnormal spirometry. Age was inversely associated. Conclusions The prevalence of undiagnosed COPD is high in at-risk populations. We found multiple factors associated with undiagnosed COPD and incident diagnosis of COPD at follow up. These results can be used to identify those at risk for undiagnosed COPD to facilitate earlier diagnosis and treatment.

Details

Metrics

22 Record Views
12 Times Cited - Web of Science
Logo image