OBJECTIVE

Air pollution is associated with an increased risk for cardiovascular events. Many of the biological pathways involved could also promote diabetes mellitus (DM). We therefore investigated the association between DM prevalence and exposure to traffic-related air pollution (nitrogen dioxide [NO 2]).

METHODS

Study participants were patients who attended two respiratory clinics in Hamilton (n = 5228) and Toronto (n = 2406). The diagnosis of DM was ascertained by linkage to administrative databases of the Ontario universal Health Insurance Plan for patients aged 40 years and above. Geographic Information systems methodology was used to assign individual estimates of NO2 based on a network of samplers in each city. Logistic regression was used to estimate the relations between NO2 exposures and the odds of DM diagnosis.

RESULTS

After adjusting for age, body mass index, and neighborhood income there were positive effects in women on the odds ratio for DM for each 1 ppb NO2 exposure in Toronto (OR 1.055, 95% CI: 0.99 to 1.11) and Hamilton (OR 1.029, 95% CI: 0.98 to 1.08). In a meta-analytic model including both cities, there was a significant effect in women (OR = 1.04; 95% CI: 1.00 to 1.08). Across the inter-quartile range (approximately 4 ppb NO2) there was nearly a 17% increase in the odds of DM for women. There were no positive associations among men.

CONCLUSIONS

Exposure to NO2, a marker of traffic-related air pollutants, was associated with DM prevalence among women. Exposure estimate errors in men may explain the apparent gender difference. These results suggest that common air pollutants are associated with DM and warrant more investigation to determine if this is a cause-and-effect relationship.