Screening for lung disease in workers with a low prevalence of chronic obstructive pulmonary disease (COPD) is subject to debate. Examining all workers would lead to unacceptable costs. Surveillance of high-risk groups may be a useful alternative. Data from 314 workers exposed to dust were used to develop a prediction model. Data from a study comprising 96 rubber workers were used to validate the model's performance. COPD was defined as a ratio of forced expiratory volume in 1 second to forced vital capacity outside the 5th percentile. The accuracy of the model was evaluated by receiver operating characteristic curve analysis and by comparison of the observed versus predicted probabilities. Shortness of breath, wheeze, heavy smoking, and work-related lower respiratory symptoms were identified as independent determinants of having COPD. Workers with COPD can be distinguished from those without COPD on the basis of a medical/work-related questionnaire. Occupational physicians can develop a low-cost strategy for detecting workers at risk for COPD.