Unveiling causal connections: Long-term particulate matter exposure and type 2 diabetes mellitus mortality in Southern China

Abstract

Evidence of the potential causal links between long-term exposure to particulate matters (PM, i.e., PM₁, PM_2.5, and PM_1-2.5) and T2DM mortality based on large cohorts is limited. In contrast, the existing evidence usually suffers from inherent bias with the traditional association assessment. A prospective cohort of 580,757 participants in the southern region of China were recruited during 2009 and 2015 and followed up through December 2020. PM exposure at each residential address was estimated by linking to the well-established high-resolution simulation dataset. Hazard ratios (HRs) were calculated using time-varying marginal structural Cox models, an established causal inference approach, after adjusting for potential confounders. During follow-up, a total of 717 subjects died from T2DM. For every 1 μg/m³ increase in PM_2.5, the adjusted HRs and 95% confidence interval (CI) for T2DM mortality was 1.036 (1.019-1.053). Similarly, for every 1 μg/m³ increase in PM₁ and PM_1-2.5, the adjusted HRs and 95% CIs were 1.032 (1.003-1.062) and 1.085 (1.054-1.116), respectively. Additionally, we observed a generally more pronounced impact among individuals with lower levels of education or lower residential greenness which as measured by the Normalized Difference Vegetation Index (NDVI). We identified substantial interactions between NDVI and PM₁ (P-interaction = 0.003), NDVI and PM_2.5 (P-interaction = 0.019), as well as education levels and PM₁ (P-interaction = 0.049). The study emphasizes the need to consider environmental and socio-economic factors in strategies to reduce T2DM mortality. We found that PM₁, PM_2.5, and PM_1-2.5 heighten the peril of T2DM mortality, with education and green space exposure roles in modifying it.

Keywords: T2DM; causal modeling; effect modification; large cohort; long-term PM exposure.

Figure 1.

Directed acyclic graph (DAG) for the association between air pollution and T2DM, created with the help of dagitty.net The DAG is a graphic representation of the assumption of how variables are causally related. Combined with the backdoor criterion, DAG can be used to determine the minimally sufficient confounders set that can eliminate all potential confounding. The green node at the bottom is the exposure variable, and the blue node with I in the center is the outcome variable. The pink nodes represent observed confounders, and the other blue nodes are ancestors of the outcome. The directed edges represent the causal relationship between pairs of variables, which start from the cause and end at the outcome. The green and pink lines represent the causal and biasing paths, respectively. More details on DAG can be found elsewhere. According to the final DAG and back door criteria, the minimal sufficient set of confounders includes sex (men or women), age (years), ethnicity (Han or Minority), education level (illiterate or semiliterate, elementary school, middle school, high school or college or above), marital status (never married, married, widowed or divorce), medical insurance (for urban workers, for urban residents, the new rural cooperative medical insurance, others), exercise frequency (very low, low, moderate, high), smoking status (never, ever or current), alcohol assumption (never, once a week, occasionally, daily).

Figure 2.

HRs (95% CIs) of the mortality risk due to T2DM with each 1 μg/m³ increase in exposure to PMs. Abbreviations: Model 0: A model with only air pollution exposure. Model 1: A model with air pollution exposure, age, and gender; Model 2: A model with air pollution exposure and all potential confounders. Model 3: A model with air pollution exposure and potential confounders using IPWs with a generalized estimating equation. HR, hazard ratio; PM₁, particulate matter with an aerodynamic diameter ≤ 1 μm; PM_2.5, particulate matter with an aerodynamic diameter ≤ 2.5 μm; PM_1–2.5, particulate matter with an aerodynamic diameter greater than 1μm and less than 2.5 μm.

Figure 3.

Hazard ratios and 95% confidence interval of T2DM mortality risk associated with each 1μg/m³ in long-term exposure to ambient particulate matter, stratified by demographic and lifestyle factors. Note: The effects were estimated under the causal inference model with adjustment for age, sex, education, marital status, physical activity, smoke, and NDVI (500m). All stratified estimates were adjusted for the remaining covariates. The interquartile range of NDVI is 0.0995–0.5328. The NDVI value is 0.0995–0.1875 for the first quartile, 0.0995–0.2112 for the second quartile, 0.0995–0.2449 for the third quartile, and 0.0995–0.5328 for the fourth quartile. Abbreviations: PM₁, particulate matter with an aerodynamic diameter ≤ 1 μm; PM_2.5, particulate matter with an aerodynamic diameter ≤ 2.5 μm; PM_1–2.5, particulate matter with an aerodynamic diameter greater than 1 μm and less than 2.5 μm; HR, hazard ratio; CI, confidence interval; NDVI, normalized difference vegetation index.