Predicting Lung Function Trajectory and COPD Risk

Forecasting future lung function will significantly benefit patients at risk of chronic obstructive pulmonary disease (COPD), but lung function trajectory decline largely differs from patient to patient. In this study, researchers described the creation of an individualized prediction model of lung function trajectory and airflow limitation risk in the general population. The Framingham Offspring Cohort provided data on 4,167 patients aged ≥ 20 years with at least two valid spirometry assessments. The primary and secondary outcomes were pre-bronchodilator forced expiratory volume at 1 second (FEV1) and the risk of airflow limitation (defined as FEV1/forced vital capacity < Lower Limit of Normal), respectively. Individualized predictions were made using mixed-effects regression models, and a machine-learning algorithm aided in establishing essential predictors. Two independent multi-center cohorts were employed to validate the model. Using 20 common predictors, the model successfully explained 79% of variation in FEV1 decline in the derivation cohort. The model exhibited a low error rate in FEV1 decline prediction and effectively predicted airflow limitation risk. A web-based application for accessing the model is available and allows prediction based on changeable predictor sets. The authors concluded that their model may be useful in patients at high risk of COPD who could benefit from interventional therapies. Read more