Risk Score for Subclinical COPD: The NHLBI Pooled Cohorts Study

Background: Early identification of subclinical COPD is important for risk factor modification. Aim: To describe subclinical disease burden and develop a risk score to inform potential screening programs. Methods: Participants without self-reported respiratory symptoms or prior diagnosis of COPD/asthma were selected from 9 US general population cohorts. A risk score for prevalent Subclinical Airflow Obstruction (SAO; FEV1/FVC Results: Of 33,546 participants (mean age 52 years, 54% female, 44% non-Hispanic White), 4,424 (13.2%) had SAO. The incidence of respiratory events (N at risk 14,024) increased across SAO risk score deciles, and was 3 times higher in subjects with SAO than in those without SAO (179 vs. 54 events/10,000 person-years). The SAO risk score was well calibrated in discovery and validation samples (validation c-statistic 0.81, 95%CI 0.80-0.82). A screening program aimed at asymptomatic adults with SAO risk score >10% would need to perform 4 spirometry exams to identify 1 case of SAO. Conclusions: Adults with SAO demonstrate excess respiratory hospitalization/death. A risk score using easily available clinical risk factors can identify individuals to target in potential screening interventions.