Mediating factors explaining the associations between polycyclic aromatic hydrocarbons exposure, low socioeconomic status and diabetes: A structural equation modeling approach.


Department of Occupational and Environmental Health, Huazhong University of Science and Technology, Hangkong Road 13, Wuhan 430030, Hubei, PR China; Key Laboratory of Environment and Health, Ministry of Education & Ministry of Environmental Protection, State Key Laboratory of Environmental Health (Incubating), School of Public Health, Tongji Medical College, Huazhong University of Science and Technology, Hangkong Road 13, Wuhan 430030, Hubei, PR China.. Electronic address: [Email]


Exposure to polycyclic aromatic hydrocarbons (PAHs) is linked with increased risk of diabetes, whereas socioeconomic status (SES) may contribute to the development of diabetes. However, the mechanisms underlying the relationships between them are unclear. We used structural equation modeling (SEM) to identify mediating factors in the associations of PAHs exposure, low SES with diabetes risk. Data were collected from 2751 Wuhan participants at baseline from the Wuhan-Zhuhai Cohort Study (n = 3053). They answered the questionnaires regarding socio-demographic, participated physical examinations and provided urine samples for measurements of urinary monohydroxy-polycyclic aromatic hydrocarbons (OH-PAHs) levels. SEM was used to identify the mediating factors (such as hypertension, body mass index (BMI), triglycerides (TG) and total cholesterol (TCHO)) in the associations of low SES or PAHs exposure with diabetes risk. We observed that partial effect of PAHs exposure (β = 0.281, p = 0.034), BMI (β = 0.182, p = 0.000), TG (β = 0.358, p = 0.000), TCHO (β = 0.203, p = 0.009) or hypertension (β = 0.385, p = 0.000) on diabetes was directive. Moreover, low SES also exhibited a directive effect on PAHs exposure (β = -0.084, p = 0.000), BMI (β = 0.301, p = 0.000), hypertension (β = 0.134, p = 0.003) and TG (β = 0.087, p = 0.001). PAHs exposure directly affected TCHO levels (β = 0.080, p = 0.002) and TG (β = 0.076, p = 0.017). The proportion of the effect of PAHs exposure on diabetes mediated by TG and TCHO was 15.6%. The proportion of the effect of low SES on diabetes mediated by BMI, hypertension and TG was 89.1%. The results suggested that low SES increased diabetes risk, which may be partially explained by BMI, hypertension and triglycerides, and exposure to high levels of PAHs may have indirect contribution to increased risk for diabetes with dyslipidemia.