Beverage consumption modifies the risk of type 2 diabetes associated with ambient air pollution exposure

Background Evidence on how beverage consumption modifies associations between air pollution (AP) exposure with the type 2 diabetes (T2D) risk remains scarce, which we aimed to investigate in this study. Methods A total of 77,278 adults from the UK Biobank cohort were enrolled. Annual average concentrations of fine particulate matter (PM2.5 and PM10) and nitrogen oxides (NO2 and NOX) were estimated to represent the long-term AP exposure using the land use regression model. The consumption of beverages (alcoholic beverages, juice, sugar-sweetened beverages [SSB], coffee, and tea) was estimated with the 24-hour dietary assessment. The AP-T2D and beverage-T2D risks were assessed using Cox regression models. Modifying effects of beverage consumption on AP-T2D associations were evaluated through stratified analysis and heterogeneity test. Results During a median follow-up of 12.19 years, 1486 T2D events were recorded. One interquartile range increase of PM2.5, NO2, and NOX raised the T2D risk with the hazard ratios (HR) and 95 % confidence intervals (95 % CI) being 1.09 (1.03, 1.16), 1.14 (1.06, 1.21), and 1.09 (1.04, 1.15), respectively. For beverages, compared with non-consumption, daily consumption (>0 cup) of red wine, > 0-3 cups of white wine, ground coffee, and herbal tea, and > 0-1 cup of spirits were associated with a 13 %-37 % reduced T2D risk, while > 0 cup of SSB were associated with a 21 %-122 % elevated T2D risk. Beverage consumption modified AP-T2D associations, as compared with non-consumption, > 0-3 cups of red wine, white wine, ground coffee, and herbal tea had a lower attenuated T2D risk associated with NO2 and/or NOX. Conversely, those with > 1 cup of SSB had a higher T2D risk associated with both NO2 and NOX (P-heter <0.05). Conclusions This study highlights the significant role of beverage consumption in mitigating or exacerbating the T2D risk associated with long-term NO2 and NOX exposure.