Aims To investigate the association between airflow limitation and occupational exposure and to compare a fixed limit with an age adjusted limit for airflow limitation.

Methods 3924 employees in 24 Norwegian smelters and related workplaces were investigated annually over 5 years (n=16 570) using spirometry and a respiratory questionnaire on smoking habits and job category. Employees working full time on the production line were classified as line operators; subjects who never worked on the production line were regarded as non-exposed. A job exposure matrix (JEM) was available in most smelters. Airflow limitation was expressed as (i) forced expiratory volume in 1 s/forced vital capacity (FEV₁/FVC ratio) <0.7 and (ii) FEV₁/FVC ratio less than the lower limit of normal (LLN). Longitudinal analyses on the prevalence of airflow limitation were performed using a generalised linear mixed model.

Results The prevalence of airflow limitation was stable during follow-up. The OR of airflow limitation during follow-up was 1.24 (95% CI 1.03 to 1.50) in line operators compared with unexposed subjects for FEV₁/FVC ratio <0.7 and 1.44 (1.14 to 1.81) for FEV₁/FVC ratio <LLN. A dose–response relationship was also found between prevalence of airflow limitation and dust exposure. The annual decline in FEV₁ was −77.5 (SD 99.9) ml/year in subjects with FEV₁/FVC ratio <0.7 and −83.8 (114.2) ml/year in subjects with FEV₁/FVC ratio <LLN.

Conclusion Prevalence of airflow limitation was higher in exposed than non-exposed employees. Both the FEV₁/FVC ratio <0.7 and FEV₁/FVC ratio <LLN identify subjects with accelerated annual decline in FEV₁.