Analysis of Paraben and Bisphenol A Exposure in Relation to Food Intake Levels and Health Risk Assessments in the Taiwanese Population

Abstract

Parabens (PBs) and bisphenol A (BPA) are common endocrine-disrupting chemicals commonly encountered in daily life, yet research on their associations with food groups remains limited. To address this, a nationwide study in Taiwan evaluated the exposure risks of PBs and BPA. This study measured urinary concentrations of three PBs—methyl-paraben (MP), ethyl-paraben (EP), and propyl-paraben (PP)—together with BPA using ultra-performance liquid chromatography coupled with quadrupole time-of-flight mass spectrometry (UPLC-QToF-MS), using data from 706 participants in the Nutrition and Health Survey in Taiwan (NAHSIT). The median creatinine-adjusted concentrations for MP, EP, PP, and BPA were 17.54, 0.97, 1.60, and 1.19 µg/g cre., respectively. Multiple linear regression identified a positive linear association was identified between EP exposure and oil group consumption in the ≥ 65 years group (β = 0.194, p = 0.04). G-computation analysis was applied to estimate the population-average causal effects of food group consumption on PBs and BPA levels, revealing distinct dietary associations with both compounds. For PBs, significant positive correlations were observed with fruits (+9.1% for EP in the 19–64 years group, p = 0.036), snacks (+1.7% for MP [p = 0.023] to +3.8% for EP [p = 0.006] in the 12–18 years group), seasoning (+19.6% for MP in the 12–18 years group, p = 0.009), and oils (+55.6% for EP in the ≥ 65 years group, p = 0.025), while negative correlations were noted with livestock (−12.0% for MP in the 12–18 years group, p = 0.030) and other foods (−5.1% for PP in the 19–64 years group, p = 0.034). For BPA, positive correlations were identified with protein (+8.7% in the ≥ 65 years group, p = 0.026) and fish and seafood (+24.4% in the ≥ 65 years group, p = 0.049). A probabilistic health risk assessment was conducted using Monte Carlo simulation to account for variability and uncertainty in exposure estimates, providing a distribution of potential risks and identifying the likelihood of exceeding acceptable daily intakes. Following the probabilistic simulation, the results showed that all hazard indices for PB exposure were below one, indicating no significant risk. However, the hazard quotient of the estimated daily BPA intake exceeded the tolerable level established by EFSA (0.2 ng/kg bw/day) across all age groups, suggesting potential health risks. To our knowledge, this is the first nationwide study to investigate the associations of urinary PBs and BPA with food group consumption in the general population of Taiwan, highlighting its novelty and public health relevance.