People are exposed to various chemicals contained in consumer products for which the risks are poorly characterized. There is growing evidence that exposure to endocrine disrupting chemicals (EDCs) through product use potentially affects development, behavior, and reproduction. However, limited information is available about common combinations of chemicals based on their appearance and potential health effects. The present study listed the ingredients contained in 11064 household chemical products from a publicly available database, and identified EDCs related to estrogenicity, androgenicity, thyroid hormone disruption, and changes in steroidogenesis. Association rule mining was applied to the dataset to identify frequent combinations of chemicals or commonly occurring EDCs contained in a single product. Among the target products, ingredient names were matched with 1241 chemical identifiers. A total of 293 chemicals were related to endocrine disruption, and nearly two-thirds of the products contained more than one of these chemicals. Cleaning products, synthetic detergents, fabric softeners, air fresheners, and deodorants have several hotspots for fragrances, isothiazolinones, glycol ethers, and parabens. The three most prevalent EDCs in household chemical products were added to act as fragrances and preservatives. The present study demonstrated that commonly occurring chemical combinations can be derived using an association rule mining algorithm. The results of this study will be useful in prioritizing chemical combinations and developing management plans for EDC mixture in consumer products.
Keywords: Association rule mining; Endocrine disrupting chemical; Fragrance; Ingredient; Products.
Copyright © 2022 The Authors. Published by Elsevier Inc. All rights reserved.