The Association of Air Pollution and Blood Glucose Levels

Abstract

Background: The number of Diabetes Mellitus (DM) patients worldwide is increasing rapidly. Whether air pollution induces DM or whether diabetic patient are more susceptible, is unclear. The evidence on association of air pollution with blood glucose (BG) levels is limited. We investigated this association among DM patients. Methods: We included all BG tests of DM patients (>18 years) residing within 20 km from an urban monitoring station in southern Israel during 2003-2012. DM diagnosis was assigned in the presence of one of the following: at least one BG>200 mg/dL; hospitalization diagnosis; Hemoglobin A1C>6.5%; dispensing of anti-diabetic medications. Log-transformed BG levels were modeled by Generalized Estimation Equation model with adjustment to age, gender, hospitalizations, anti-diabetic medications and ambient temperature. Environmental exposures data were retrieved from monitoring stations on dates of a relevant blood test, and included 3-days moving average pollutants and pollutants concentration beyond the 90% percentile. Results: We identified 28,186 DM patients (4% of the adult population), with a total of 622,752 tests. The Inter Quartile Range (IQR) Sulfur Dioxide (SO2) concentration ranged between 1.23-2.24 ppb. The IQR Nitrogen Dioxide (NO2) concentration ranged between 6.45-12.52 ppb. An IQR increase in 3 day averaged SO2 and 3 day averaged NO2 were significantly associated with the increase of blood glucose levels [%exp B (95% Confidence Interval, p value)]: 0.44% (0.26%-0.59%, p<0.001) and 0.36% (0.24%-0.55%, p<0.001) increase, respectively. SO2 concentration beyond 90% percentile was associated with 0.9% (0.6%-1.1%, p<0.001) increase in BG levels. Conclusions: Our preliminary results showed positive associations between NO2 and SO2 and BG levels among diabetic patients. In attempt to reduce a possible exposure measurement error, we are currently working on refining exposure estimates by using a novel satellite-based technique.