Harmonization of Respiratory Data From 9 US Population-Based Cohorts

Abstract

Chronic lower respiratory diseases (CLRDs) are the fourth leading cause of death in the United States. To support investigations into CLRD risk determinants and new approaches to primary prevention, we aimed to harmonize and pool respiratory data from US general population-based cohorts. Data were obtained from prospective cohorts that performed prebronchodilator spirometry and were harmonized following 2005 ATS/ERS standards. In cohorts conducting follow-up for noncardiovascular events, CLRD events were defined as hospitalizations/deaths adjudicated as CLRD-related or assigned relevant administrative codes. Coding and variable names were applied uniformly. The pooled sample included 65,251 adults in 9 cohorts followed-up for CLRD-related mortality over 653,380 person-years during 1983-2016. Average baseline age was 52 years; 56% were female; 49% were never-smokers; and racial/ethnic composition was 44% white, 22% black, 28% Hispanic/Latino, and 5% American Indian. Over 96% had complete data on smoking, clinical CLRD diagnoses, and dyspnea. After excluding invalid spirometry examinations (13%), there were 105,696 valid examinations (median, 2 per participant). Of 29,351 participants followed for CLRD hospitalizations, median follow-up was 14 years; only 5% were lost to follow-up at 10 years. The NHLBI Pooled Cohorts Study provides a harmonization standard applied to a large, US population-based sample that may be used to advance epidemiologic research on CLRD.

Integrating Clinical and Epidemiologic Data on Allergic Diseases Across Birth Cohorts: A Harmonization Study in the Mechanisms of the Development of Allergy Project

“…This manuscript supports and extends previous and ongoing initiatives on data harmonization (15-21, [43][44][45][46]. Two novel features of our harmonization process are:…”

Whole genome sequence analysis of pulmonary function and COPD in 19,996 multi-ethnic participants

“…Our study used pre-bronchodilator pulmonary function from a broad range of cohorts. We leveraged extensive quality control and harmonization efforts for pulmonary function using standardized criteria 48 , and previously demonstrated that the effect of using pre-vs post-bronchodilator measures has minimal impact on genetic association 49 . Our analysis of cross-sectional data, adjusting for age and smoking exposure, does not specifically address determinants of change in lung function over time or other phenotypes that may be important in COPD.…”

Discriminative Accuracy of FEV1:FVC Thresholds for COPD-Related Hospitalization and Mortality

“…The National Heart, Lung, and Blood Institute (NHLBI) Pooled Cohorts Study harmonized data from 9 US general populationbased studies that collected spirometry data. 13 The current report is limited to 4 cohorts that completed follow-up for COPDrelated clinical events: the Atherosclerosis Risk in Communities Study (ARIC) 14 ; Cardiovascular Health Study (CHS) 15 ; Health, Aging and Body Composition Study (Health ABC) 16 ; and the Multi-Ethnic Study of Atherosclerosis (MESA). 17 CHS and Health ABC were designed to study older adults, whereas ARIC and MESA included both middle-aged and older adults (eTable 1 in the Supplement).…”

“…21 Lifetime smoking status was classified as never or ever by comparison of self-reported smoking status over all available examinations. 13 Height was measured using standard methods. Due to the extensive quality control and harmonization efforts performed in the NHLBI Pooled Cohorts Study, 13 missing covariate data at enrollment were rare (<1% [eFigure 1 in the Supplement]).…”

“…13 Height was measured using standard methods. Due to the extensive quality control and harmonization efforts performed in the NHLBI Pooled Cohorts Study, 13 missing covariate data at enrollment were rare (<1% [eFigure 1 in the Supplement]).…”