The report, published in the October issue of Diabetes Care, is among the first large-scale population-based studies to link diabetes prevalence with air pollution. It is consistent with prior laboratory studies finding an increase in insulin resistance, a precursor to diabetes, in obese mice exposed to particulates, and an increase in markers of inflammation (which may contribute to insulin resistance) in both the mice and obese diabetic patients after particulate exposure.

Like the laboratory studies, the current study focused on fine particulates of 0.1-2.5 nanometers in size (known as PM2.5), a main component of haze, smoke and motor vehicle exhaust. The investigators obtained county-by-county data on PM2.5 pollution from the EPA, covering every county in the contiguous United States for 2004 and 2005.

They then combined the EPA data with data from the Centers for Disease Control (CDC) and the U.S. Census to ascertain the prevalence of adult diabetes and to adjust for known diabetes risk factors, including obesity, exercise, geographic latitude, ethnicity and population density (a measure of urbanization).

In all analyses, there was a strong and consistent association between diabetes prevalence and PM2.5 concentrations. For every 10 microgram per cubic meter increase in PM2.5 exposure, there was a 1 percent increase in diabetes prevalence. This finding was seen in both 2004 and 2005, and remained consistent and significant when differing estimates of PM2.5 exposure were used.

Even among counties falling within EPA limits for exposure, those with highest versus the lowest levels of PM2.5 pollution had a more than 20 percent increase in diabetes prevalence, which remained after controlling for diabetes risk factors.