Health risk and temporal trend of dietary potentially toxic elements exposure in the residents of the Shenzhen metropolis, China, between 2005 and 2017: a risk assessment based on probabilistic estimation

Dietary potentially toxic elements (PTEs) exposure in developing countries is of great concern. Probabilistic estimation exhibits great superiority in risk assessment by dealing with the variability and uncertainty of the parameters. Here, a probabilistic estimation based on two dimensions, PTEs in foods and food intake, was conducted. A total of 13 foods were collected from Shenzhen markets during 2005–2017, and the concentrations of Pb, Cd, Hg, and As were detected. A total of 853 residents from 245 households participated in a total diet study. The mean concentrations of Pb, Cd, Hg and As were 0.046, 0.0196, 0.0038, and 0.029 mg kg −1 in cereals, 0.042, 0.0174, 0.0027, and 0.014 mg kg −1 in vegetables, 0.044, 0.0237, 0.0056, and 0.021 mg kg −1 in meat, and 0.081, 0.1035, 0.0257, and 0.680 mg kg −1 in aquatic products, respectively. The probability density function showed that the 95th percentiles of the Pb, Cd, Hg, As hazard quotients (HQ) and the hazard index (HI) were 0.68, 1.57, 0.38, 5.81 and 7.51, respectively. Cumulative probability and sensitivity analysis showed that cereals and vegetables contributed most to Pb and Cd exposure; aquatic products to Hg exposure; and cereals and aquatic products to As exposure. The results showed that Shenzhen residents were at risk of exposure to Cd, As, and four PTEs in combination, although a temporal decreasing trend was observed. The probabilistic estimation used here reveals a complete picture of multiple PTEs exposure risk and identifies major contributing food categories, providing a valuable means for risk assessment.