A risk-prediction tool for chronic obstructive pulmonary disease (COPD) has been developed and validated for use in the general population, according to study results published in CHEST.

Using the Transparent Reporting of a multivariate prediction model for Individual Prognosis or Diagnosis (or, TRIPOD), researchers sought to develop and validate a personalized tool for predicting changes in 20-year prebronchodilator forced expiratory volume in 1 second (FEV₁) and determine the risk for airflow limitation in the general population without COPD.

The derivation cohort comprised participants in the Framingham Offspring cohort (n=5124) who were aged ≥20 years at baseline and had valid spirometry measurements at ≥2 different time points. There were 2 validation cohorts. The first included participants in the Coronary Artery Risk Development in Young Adults study who were aged 18 to 30 years with spirometry measurements at baseline and at 3 subsequent time points thereafter over 20 years (n=5115). The second cohort included participants in the Atherosclerosis Risk in Communities study who were aged ≥45 years and had spirometry data recorded at baseline and 3 years later (n=15,792). Only data from participants aged 20 to 65 years were used to enable direct comparisons with the derivation cohort.

In the derivation cohort, 78.9% of the variance in FEV₁ was explained by the model using 20 machine-selected predictors, including age, sex, height, presence of respiratory symptoms, and certain laboratory tests, all measured at baseline. A slight acceleration of declining FEV₁ over time was indicated by the square term of the follow-up year entering the model with a small negative coefficient. There was low error in predicting FEV₁ decline and high-discriminative power in predicting the risk for airflow limitation in the 2 validation datasets.

Related Articles

Study limitations included the Framingham Offspring cohort that was predominantly white and recent changes in the distribution of some risk factors. Also, the derivation cohort lacked some potentially important variables such as genetic deficiencies, air pollution, physical exercise, and family history of COPD.

“This easy-to-implement tool can be harnessed to identify individuals at a high risk of developing COPD, and target them for closer monitoring and application of preventive and therapeutic interventions,” the researchers concluded. “Another important utility of the model is in clinical trial design. Identification and recruitment of individuals with a rapid lung function decline can efficiently reduce the sample size required for clinical trials that aim at evaluation interventions that potentially impact lung function decline.”

Reference

Chen W, Sin DD, FitzGerald JM, Safari A, Adibi A, Sadatsafavi M. An individualized prediction model for long-term lung function trajectory and risk of COPD in the general population [published online September 19, 2019]. CHEST. doi:10.1016/j.chest.2019.09.003