This study investigated the occurrence, sources, and health risks of 21 per- and polyfluoroalkyl substances (PFAS) in commercially available shellfish and crustaceans from Zhejiang Province, China. Among the 306 samples analyzed, 87.9% contained at least one detectable PFAS. Perfluorooctanoic acid (PFOA) was the most frequently detected PFAS (64.7%), followed by perfluorooctanesulfonic acid (PFOS) (53.8%), perfluorononanoic acid (PFNA) (52.9%), and perfluorodecanoic acid (PFDA) (50.0%). The total PFAS in shellfish and crustaceans ranged from ND to 0.86 to 173 ng/g wet weight, with a median of 4.11 ng/g ww; the median concentration of total PFAS followed this order: marine crustaceans > fresh-water crustaceans > bivalves. Estimation of the human intake of adult consumers, the estimated daily intake (EDI) of Σ₂₁ PFAS ranged from 0.01 to 15.7 ng/kg bw/day; 0.31% of the adult study population had Σ₄PFAS exposure levels resulting in Hazard Quotient (HQ) values > 1, which may represent a potential public health concern for these individuals. Long-term exposure risks for specific PFCAs such as perfluoroundecanoic acid (PFUdA) and perfluorotridecanoic acid (PFTrDA) merit concern.

To understand the epidemiological characteristics and patterns of mushroom poisoning in Zhejiang Province from 2012 to 2023, and to overcome the limitations of previous descriptive studies in precise early warning and spatial identification, this study explored the feasibility of identifying spatial distribution characteristics and high-risk spatiotemporal clusters. First, descriptive epidemiological analysis was conducted on 2276 cases from the Foodborne Disease Case Surveillance System and 408 outbreaks from the Foodborne Disease Outbreak Surveillance System reported over the 12-year period to clarify the basic characteristics and trends of poisoning. Subsequently, spatial autocorrelation analysis (Moran’s I) was employed to reveal spatial dependence and clustering patterns. Finally, spatiotemporal scan statistics (SatScan) were used to precisely identify high-risk spatiotemporal clusters, systematically analyzing the spatiotemporal distribution and clustering patterns of mushroom poisoning cases. The results showed a distinct summer–autumn seasonal peak (June–October), attributed to the subtropical monsoon climate with high temperatures and abundant rainfall, which is conducive to mushroom growth. Farmers were the most affected population (47.93%), and homes were the primary poisoning locations (71.7%), reflecting widespread foraging habits and insufficient risk awareness in rural areas. Chlorophyllum molybdites (36.27%) and Russula japonica (10.05%) were the dominant poisoning mushroom species, with gastrointestinal symptoms being the predominant clinical manifestation (84.07%). Spatial analysis revealed significant spatiotemporal clustering of mushroom poisoning in Zhejiang Province. The global Moran’s I index showed significant positive autocorrelation in some years (p < 0.05), with local hotspots mainly distributed in western Zhejiang counties. This pattern is driven by a dual model of environmental suitability and behavioral risk, resulting from the high forest coverage and humid climate of the western Zhejiang mountainous areas providing suitable habitats, combined with long-standing foraging habits among local residents. Retrospective spatiotemporal scanning identified high-risk clusters for each year from 2018 to 2023, with the Lishui area in 2023 being the most significant cluster (Relative Risk (RR) = 15.44, Log-Likelihood Ratio (LLR) = 114.49). The results confirm that mushroom poisoning in Zhejiang Province exhibits a stable and identifiable spatiotemporal clustering pattern, providing a quantitative basis for precise health education and targeted prevention and control in high-risk counties of western Zhejiang during June–October, thereby shifting the approach from passive reporting to targeted intervention.