Objectives: To determine longitudinal changes in the prevalence of chronic respiratory symptoms among Canadian grain workers.
Design: Data on respiratory symptoms, smoking status, and pulmonary function were obtained approximately every 3 years (termed cycle) over 15 years beginning in 1978 from five regions of Canada.
Participants: The number of grain workers participating in each cycle were as follows: cycle 1 (n = 5,702); cycle 2 (n = 5,491); cycle 3 (n = 3,713); cycle 4 (n = 2,847); and cycle 5 (n = 3,079).
Results: A procedure based on generalized estimating equations (PROC GENMOD; SAS Institute; Cary, NC) was used to fit marginal models to determine risk factors influencing the prevalence of chronic respiratory symptoms (wheeze, dyspnea, sputum, and cough). The prevalence (predicted probability based on the final model) of chronic respiratory symptoms had an increasing trend with increasing number of years in the grain industry from cycle 1 to cycle 3 (before dust control) for all three smoking categories (current smokers, ex-smokers, and nonsmokers). For cycle 4 and cycle 5 (after dust control), there was a reduction in the prevalence of these respiratory symptoms. For example, in cycle 1, the prevalence of chronic wheeze among current smoking grain workers increased from 12% (for those in the industry for < 5 years) to 44% (for those in the industry for > 35 years); in cycle 5, the prevalence of chronic wheeze among current smoking grain workers increased from 9% (for those in the industry for < 5 years) to 28% (for those in the industry for > 35 years). Similar trends were observed for ex-smokers and nonsmokers and for other chronic respiratory symptoms.
Conclusions: Our results indicate that grain dust control was effective in reducing the prevalence of chronic respiratory symptoms among grain workers in all smoking and exposure categories.