Polychlorinated dibenzo-p-dioxins and dibenzofurans (PCDD/Fs) are highly toxic, persistent organic pollutants that bioaccumulate through food webs and constitute a major human exposure pathway. National-scale dietary exposure assessments in China remain uncertain due to limited monitoring data, assumptions of complete bioaccessibility, and heterogeneity in regional emissions and dietary patterns. In this study, we developed a China-specific multimedia food-chain model to quantify environmental transfer, food-chain accumulation, and human exposure of PCDD/Fs across six major urban agglomerations, including the Beijing-Tianjin-Hebei (BTH), Yangtze River Delta (YRD), and Pearl River Delta (PRD). The model integrates regional environmental concentrations, population dietary patterns, and cooking-related bioaccessibility for more accurate exposure estimation. Model predictions agree well with observations (n = 224), with 87.9% of mass concentrations and 91.5% of toxic equivalents (TEQs) within one order of magnitude, and significant correlations (r = 0.60 for mass and r = 0.84 for TEQ). BTH, YRD, and PRD regions exhibited the highest contamination and exposure levels, reflecting intensive industrial activities. Animal-derived foods dominated dietary exposure (∼92.7%), mainly eggs, milk, fish, and pork. Urban populations experienced 1.92-6.51 times higher exposure than rural populations, with larger disparities in industrialized regions. Children aged 2-5 years had the highest exposure. Incorporating cooking-related bioaccessibility reduced the daily dose by 86.6 ± 1.7%, 75.7 ± 0.62%, and 80.6 ± 0.95% under boiling, frying, and Chinese-style cooking, respectively. This study highlights spatial disparities in dietary exposure and the importance of considering cooking processes in exposure assessments, providing insights for region-specific food safety management and pollution control.
Keywords: Cooking bioaccessibility; Dietary exposure; Health risk; Multimedia food-chain model; PCDD/Fs; Urban–rural disparity.
Copyright © 2026 Elsevier Ltd. All rights reserved.