Mediating effect of diabetes in the association between long-term PM_2.5 exposure and cancer risk in CHARLS

Long-term exposure to air pollutants and diabetes are both linked to cancer development. However, their combined effect remains unclear. This study examined the relationship between air pollutants and cancer incidence, with diabetes as a potential mediator. Data from 10,590 participants in the 2015 China Health and Retirement Longitudinal Study (CHARLS) were analyzed. Participants were grouped based on cancer diagnosis, and air pollutant exposure levels were estimated using satellite-based spatiotemporal models. Generalized linear regression and restricted cubic spline (RCS) analysis were used to assess the impact of air pollutants and diabetes in covariates-adjusted models. Further analyses, including conditional independence test, mediation effect and sensitivity analysis based on Bayesian networks, were performed to further analyze specific air pollutants. After adjusting for covariates, particulate matter (PM) (PM ≤ 1 μm in aerodynamic diameter [PM₁], PM_2.5, ammonium (NH₄), nitrate (NO₃) and diabetes showed significant associations with cancer incidence. RCS analysis confirmed significant direct effects of PM_2.5 and PM₁₀ on cancer and the mediated effects of diabetes. The interaction between diabetes and both PM_2.5 and PM₁₀ was further supported by conditional independence tests, highlighting diabetes as a significant mediator in the PM_2.5-cancer relationship. This study offers a novel perspective by identifying diabetes as a key intermediary in the association between PM_2.5 exposure and cancer risk, providing evidence that diabetes plays a significant mediating role in air pollutant-related cancer development.