This study identified predictors of chronic obstructive pulmonary disease (COPD) in office and school workers, a subset of the population generally assumed to have a low prevalence of COPD and limited exposure to chemicals and dusts at work. Logistic regression was used to develop a multivariate model for COPD using National Health and Nutrition Examination Survey (NHANES III) data on 2486 US office and school workers. COPD was defined as FEV1/FVC<70% and FEV1 <80% predicted. The population prevalence of COPD among US office and school workers was estimated at 4.7% compared with 8.1 % in other US workers. As expected, current smoking was a strong predictor of COPD. After adjusting for age, race, sex, and current smoking, workers with COPD had significantly (p<0.2) greater mean duration of employment, pack years, alcohol intake, body mass index (BMI), coffee consumption, dietary fat and cholesterol. Workers with COPD had lower mean serum selenium and energy use during exercise. In multivariate modeling, employment duration, cigarette smoking, age, race, BMI, alcohol, use of propane fuel at home, and serum antioxidant (selenium) were found to explain COPD. Home heating with propane fuel was associated with a 4.4-fold risk of COPD (95% CI 1.2-15.8) in these study subjects. Duration of employment as an office or school worker for 10 or more years (vs. < 10 years} was associated with a 1.8-fold (95% CI 1.0-3.3) excess risk of COPD. Multivariate modeling revealed several potential risk factors for COPD amenable to modification. The predictive model suggests that, after controlling for confounding factors, working as an office or school worker for 45 years is associated with a 2.2-fold excess risk of COPD.
American Journal of Respiratory and Critical Care Medicine, 2003 International Conference, The American Thoracic Society, Seattle, WA, May 16-21, 2003