Potential causal links of long-term PM_2.5 components exposure with diabetes incidence and mortality in the United States

There is limited evidence on the relationship of diabetes burden with fine particulate matter (PM_2.5) and its components, which is not conducive to sustainable development in the context of rapid urbanization. To obtain relevant clues in the United States (US), we collected annual county-level diabetes incidence and mortality, concentrations of PM_2.5 and five major components (including elemental carbon, organic carbon, sulfate, nitrate, and ammonium), temperature, and socioeconomic factors during 2008-2017. Through an integrating method of difference-in-differences approach and quantile G-computation, we observed that (i) long-term PM_2.5 components mixture exposure was associated with diabetes mortality, but not incidence, with percent risk increase (IR%) of 3.58% (95%CI: 1.84%, 5.36%); (ii) among the five components of PM_2.5, sulfate was estimated to have the largest weight (0.519); (iii) the effect of PM_2.5 and its components mixture was higher when the summer mean temperature was 2 or 3 degrees below the 10-year average temperature; (iv) in counties with higher health insurance coverage, nitrate was the most important component (with the greatest weight of 0.829). Our findings suggest that long-term PM_2.5 exposure is associated with increased diabetes mortality, and reducing sulfate and nitrate emission could effectively alleviate the burden of PM_2.5-related diabetes mortality in the US.