Air pollution-associated changes in biomarkers of diabetes risk.

Background: Ambient particulate matter (PM) and nitrogen oxide (NO) air pollution may be diabetogenic.

Objective: To examine longitudinal associations of short- and longer-term mean PM ≤10 μm (PM), PM ≤2.5 μm (PM), and NO concentrations with five biomarkers of diabetes risk.

Methods: We studied a stratified, random minority oversample of nondiabetic Women's Health Initiative clinical trials participants with biomarkers and geocoded participant address-specific mean air pollution concentrations available at repeated visits (years = 1993-2004; n = 3,915; mean age = 62.7 years; 84% white). We log-transformed the biomarkers, then used multi-level, mixed-effects, longitudinal models weighted for sampling design/attrition and adjusted for sociodemographic, clinical, and meteorological covariates to estimate their associations with air pollutants.

Results: Biomarkers exhibited null to suggestively negative associations with short- and longer-term PM and NO concentrations, e.g., -3.1% (-6.1%, 0.1%), lower homeostatic model assessment of insulin resistance per 10 μg/m increase in 12-month PM. A statistically significant interaction by impaired fasting glucose (IFG) at baseline in this analysis indicated potentially adverse effects only among women with versus without IFG, i.e., 1.4% (-3.5%, 6.5%) versus -4.6% (-7.9%, -1.1%), < 0.05. In contrast, longer-term PM concentrations were largely but not statistically significantly associated with higher biomarkers.

Conclusions: Low-level short-term PM and NO concentrations may have negligible adverse effects on biomarkers of diabetes risk. Although longer-term mean PM concentrations showed primarily null associations with these biomarkers, results suggestively indicated that PM exposure over the range of concentrations experienced in the United States may adversely affect biomarkers of diabetes risk at the population level, as may longer-term mean PM concentrations among women with IFG.