Methylene diphenyl diisocyanate (MDI), widely used to produce polyurethane foams and coatings, is a common cause of occupational asthma. However, few studies have assessed the association of chronic MDI exposure with longitudinal change in FEV1 , and results have been inconsistent. This study evaluated longitudinal change in lung function in a cohort of MDI-exposed workers at a polyurethane factory, and assessed the impact of host and exposure factors. Study subjects included employees who had worked at the factory for one or more years over a 17-year period (n=108). FEV1 values, height, weight, job category, smoking information, and history of asthma symptoms were extracted from medical records and questionnaires. Annual FEV1 change was computed for each worker by fitting a least squares slope through periodic spirometry measurements. Multivariate linear regression with backwards elimination was used to assess predictors of FEV1 change and adjust for confounding variables. The mean age of the 108 workers at baseline was 36 yrs, with mean follow-up of 7 yrs. The mean FEV1 change was a decline of 29.2 ml/yr. In the final model, FEV1 change was significantly associated with smoking (p = 0.004), weight gain (p=0.002), and height squared (p=0.040). Mean adjusted FEV1 changes attributable to smoking (loss of 56.3 ml/yr) and weight gain (loss of 8.1 ml/yr) were consistent with values from previous studies. FEV1 change was not associated with job exposure category, gender, age at baseline, ethnicity, or self-reported history of asthma symptoms. Only nine workers reported a history of asthma symptoms. Twenty-four workers left the factory over the study period. Failure to detect an exposure-related decline in FEV1 may be due to a healthy worker effect.