What is the discriminative accuracy of various thresholds for the ratio of the forced expiratory volume in the first second to the forced vital capacity (FEV1:FVC) for predicting chronic obstructive pulmonary disease (COPD)-related hospitalization and mortality?
Among 24 207 participants from 4 US general population–based cohorts, the optimal fixed threshold for discriminating COPD-related events was 0.71 (C statistic for the optimal fixed threshold, 0.696). The discriminative accuracy of the 0.71 threshold was not significantly different than that of the 0.70 threshold (difference, 0.001) but it was more accurate than a lower-limit-of-normal threshold derived from population-based reference equations (difference between the optimal ratio threshold vs the model using the LLN threshold, 0.034). The 0.70 threshold provided optimal discrimination in a subgroup analysis of ever smokers and in adjusted models.
These results support the use of FEV1:FVC less than 0.70 to identify individuals at risk of clinically significant COPD.
According to numerous current guidelines, the diagnosis of chronic obstructive pulmonary disease (COPD) requires a ratio of the forced expiratory volume in the first second to the forced vital capacity (FEV1:FVC) of less than 0.70, yet this fixed threshold is based on expert opinion and remains controversial.
To determine the discriminative accuracy of various FEV1:FVC fixed thresholds for predicting COPD-related hospitalization and mortality.
Design, Setting, and Participants
The National Heart, Lung, and Blood Institute (NHLBI) Pooled Cohorts Study harmonized and pooled data from 4 US general population–based cohorts (Atherosclerosis Risk in Communities Study; Cardiovascular Health Study; Health, Aging, and Body Composition Study; and Multi-Ethnic Study of Atherosclerosis). Participants aged 45 to 102 years were enrolled from 1987 to 2000 and received follow-up longitudinally through 2016.
Presence of airflow obstruction, which was defined by a baseline FEV1:FVC less than a range of fixed thresholds (0.75 to 0.65) or less than the lower limit of normal as defined by Global Lung Initiative reference equations (LLN).
Main Outcomes and Measures
The primary outcome was a composite of COPD hospitalization and COPD-related mortality, defined by adjudication or administrative criteria. The optimal fixed FEV1:FVC threshold was defined by the best discrimination for these COPD-related events as indexed using the Harrell C statistic from unadjusted Cox proportional hazards models. Differences in C statistics were compared with respect to less than 0.70 and less than LLN thresholds using a nonparametric approach.
Among 24 207 adults in the pooled cohort (mean [SD] age at enrollment, 63 [10.5] years; 12 990 [54%] women; 16 794 [69%] non-Hispanic white; 15 181 [63%] ever smokers), complete follow-up was available for 11 077 (77%) at 15 years. During a median follow-up of 15 years, 3925 participants experienced COPD-related events over 340 757 person-years of follow-up (incidence density rate, 11.5 per 1000 person-years), including 3563 COPD-related hospitalizations and 447 COPD-related deaths. With respect to discrimination of COPD-related events, the optimal fixed threshold (0.71; C statistic for optimal fixed threshold, 0.696) was not significantly different from the 0.70 threshold (difference, 0.001 [95% CI, −0.002 to 0.004]) but was more accurate than the LLN threshold (difference, 0.034 [95% CI, 0.028 to 0.041]). The 0.70 threshold provided optimal discrimination in the subgroup analysis of ever smokers and in adjusted models.
Conclusions and Relevance
Defining airflow obstruction as FEV1:FVC less than 0.70 provided discrimination of COPD-related hospitalization and mortality that was not significantly different or was more accurate than other fixed thresholds and the LLN. These results support the use of FEV1:FVC less than 0.70 to identify individuals at risk of clinically significant COPD.
Bhatt SP, Balte PP, Schwartz JE, et al. Discriminative Accuracy of FEV1:FVC Thresholds for COPD-Related Hospitalization and Mortality. JAMA. 2019;321(24):2438–2447. doi:10.1001/jama.2019.7233
Browse and subscribe to JAMA Network podcasts!
Customize your JAMA Network experience by selecting one or more topics from the list below.
Create a personal account or sign in to: