The occupational contribution to chronic obstructive pulmonary disease (COPD) has yet to be put in a global perspective. In the present study, an ecological approach to this question was used, analysing group-level data from 90 sex-specific strata from 45 sites of the Burden of Obstructive Lung Disease study, the Latin American Project for the Investigation of Obstructive Lung Disease and the European Community Respiratory Health Survey follow-up. These data were used to study the association between occupational exposures and COPD Global Initiative for Chronic Obstructive Lung Disease (GOLD) stage II or above. Regression analysis of the sex-specific group-level prevalence rates of COPD at each site against the prevalence of occupational exposure and ever-smoking was performed, taking into account mean smoking pack-yrs and mean age by site, sex, study cohort and sample size. For the entire data set, the prevalence of exposures predicted COPD prevalence (0.8% increase in COPD prevalence per 10% increase in exposure prevalence). By comparison, for every 10% increase in the proportion of the ever-smoking population, the prevalence of COPD GOLD stage II or above increased by 1.3%. Given the observed median population COPD prevalence of 3.4%, the model predicted that a 20% relative reduction in the disease burden (i.e. to a COPD prevalence of 2.7%) could be achieved by a 5.4% reduction in overall smoking rates or an 8.8% reduction in the prevalence of occupational exposures. When the data set was analysed by sex-specific site data, among males, the occupational effect was a 0.8% COPD prevalence increase per 10% change in exposure prevalence; among females, a 1.0% increase in COPD per 10% change in exposure prevalence was observed. Within the limitations of an ecological analysis, these findings support a worldwide association between dusty trades and chronic obstructive pulmonary disease for both females and males, placing this within the context of the dominant role of cigarette smoking in chronic obstructive pulmonary disease causation.