A Probabilistic Framework for Multimedia Dioxin Risk Assessment in Susceptible Populations

In this study, a probabilistic framework was developed to address uncertainty in multimedia risk assessment for susceptible populations while capturing population variability, illustrated by dioxin emissions from municipal solid waste incinerators. The framework integrates emission estimation, AERMOD dispersion modeling, and MEPAS multimedia modeling. Cancer risks were evaluated with dose-based slope factors and age-dependent adjustment factors (ADAFs) for early-life susceptibility. Three assessment approaches were compared: deterministic without ADAF adjustment, deterministic with ADAF adjustment, and probabilistic with ADAF adjustment. For the general population (n = 7058), stratified into adults and two susceptible subgroups, the P95 risks across all simulated grids (n = 139,287) increased from 5.5 × 10−9 to 1.4 × 10−8 (2.54-fold) after ADAF adjustment and further to 6.6 × 10−8 (an additional 4.71-fold) when inter-individual variability in IR/BW and CR/BW was incorporated through bivariate Monte Carlo simulation. For school children (n = 1879), the corresponding P95 risks increased from 6.9 × 10−10 to 3.0 × 10−9 (4.35-fold) and then to 6.3 × 10−9 (an additional 2.1-fold). The stepwise increase in risk across these approaches illustrates how the proposed framework reduces uncertainty and quantifies variability. Additional analysis examined uncertainty associated with gas–particle partitioning, ADAF adjustment, and inhalation slope factor extrapolation.