Associations between smoking, anthropometric factors, dust exposure, and longitudinal changes in lung function in steel workers were examined. The cohort consisted of 541 male steel workers who had had at least two valid spirometric evaluations separated by a period of at least 4 years (yr). Associations between longitudinal changes in 1 second forced expiratory volume (FEV1), forced vital capacity (FVC), and the FEV/FVC ratio and body weight, smoking status, and dust exposure, as estimated from the subjects' job histories, were examined by logistic regression techniques. A total of 203 subjects were current smokers, 175 were former smokers, and 163 were never smokers. The mean followup period was 6.1yr and each subject had an average of 4.5 spirometric tests. The mean age of the subjects at the midpoint of followup was 39.5yr. Across all subjects, FEV1 and FVC values decreased by an average of 44 and 50 milliliters per yr (ml/yr), respectively. When the data were stratified by smoking status, FEV1 and FVC decreased by 52 and 54ml/yr for current smokers, 43 and 53ml/yr for former smokers, and by 36 and 43ml/yr for never smokers, respectively. The decreases in lung function appeared to accelerate with increasing age. Increasing weight was strongly correlated with accelerated decreases in the spirometric variables. A matched pair comparison analysis was performed using workers whose FEV1s decreased by more than 60ml/yr (rapid decliners) versus those whose FEV1s decreased by lesser amounts (slow decliners) with respect to smoking and body weight gain during the study period. The mean slopes for FVC, FEV1, and the FEV1/FVC ratio were -96ml/yr, -95ml/yr, and -0.40%/yr for the rapid decliners and 5ml/yr, 10ml/yr, and 0.10%/yr for the slow decliners. The rapid decliners gained an average of 4.313 kilograms (kg) in weight over the followup period whereas the slow decliners showed only a 1.044kg weight gain. These produced a slope in weight gain of 0.708kg per yr (kg/yr) for the rapid decliners and 0.191kg/yr for the slow decliners. The authors conclude that in this population of steel workers, weight gain, aging, and cigarette smoking are associated with longitudinal decreases in pulmonary function.