Objectives

We aimed to assess the association between long-term exposure to ambient PM₁₀ and risk of diabetes incidence, based on the “Jinchang Cohort” platform in the Northwest of China.

Methods

We selected 19884 subjects who had not yet developed diabetes in the baseline and had completed survey information from “Jinchang Cohort”. The residential address was used to match the nearest pollution monitoring station for each subject, and the average concentration of PM₁₀ from baseline to follow-up were used as an estimate of individual exposure level. Cox regression model and restricted cubic splines functions were used to evaluate the effects of PM₁₀ on the incidence of diabetes and the dose-response relationship after adjusting for confounding covariates.

Results

We observed 791 new-onset diabetics with a total follow-up of 45254.16 person-years (incidence rate of 17.48 per 1000 person-years). The risk of diabetes incidence increased by 17% (HR = 1.17, 95%CI: 1.08–1.26) per 10μg/m³ increase in environmental PM₁₀, and the risk rises gradually with the rise of PM₁₀ concentration. Comparing with the first quartile of PM₁₀, the fully adjusted HRs (95%CI) for incident diabetes from the second to the fourth quartile of PM₁₀ were 1.15 (95%CI: 0.93–1.43), 1.50 (95%CI: 1.22–1.84) and 1.44 (95%CI: 1.15–1.79), respectively (P for trend<0.001). Stratified analyses suggested that the risk of diabetes incidence associated with ambient PM₁₀ was higher in female, young to middle-aged people, overweight and obese subjects, and subjects with FPG level at baseline lower than 5.6 mmol/L.

Conclusions

Long-term exposure to ambient PM₁₀ significantly associated with a higher risk of diabetes development. Some urgent strategies may be advocated to reduce air pollution that can aid in preventing the prevalence of diabetes in the population.