镉是动植物生长发育非必需的毒性元素，环境镉暴露可对人体肾脏、骨骼系统等产生多脏器、多系统的器质性损伤，先后被WHO和ATSDR确定为优先研究食品污染物和第七位危害人体健康毒性物质。长期以来，工业生产和农业开发活动导致重金属排放，通过大气沉降、施肥、污水灌溉等方式进入耕地土壤环境中，造成了不同程度的重金属污染。相比其他重金属，土壤中镉迁移性强、毒性强，对区域人群的健康风险更高。因此，为了了解土壤镉污染及其健康风险，并对土壤镉污染健康风险进行预防、控制和管理，国内外学者围绕土壤镉污染健康风险评估开展了大量的研究。然而，我国稻田土壤镉污染健康风险评估总体存在以下四方面的不足：其一，稻田土壤镉污染评价全国性调查数据缺失，系统综述研究未考虑公开发表文献的发表偏倚对综合评价的影响；其二，传统的健康风险评估方法大多未考虑体内组织对不同暴露途径镉的生物利用率差异，且多根据人群集中趋势统计量进行点估计；其三，尿镉和血镉等内暴露指标的生物监测数据获取日趋困难。近年来，考虑体内毒代动力学过程的PBTK模型（physiologically based toxicokinetic model），被应用于关联多介质镉摄入和体内镉负荷，但这一模型对我国高暴露情景人群的适用性，缺少本土化评估数据。因此，基于PBTK模型，探索区域人群土壤镉暴露健康风险评估方法，对开展土壤镉污染健康风险评估和管理具有重要的理论意义和实用价值。因此，针对上述问题，本文首先考虑不同途径镉暴露生物利用率差异，提出了基于PBTK模型的区域人群土壤镉暴露健康风险评估全概率模型体系；其次，为解决PBTK模型适用性评估数据缺失问题，在赣南案例区开展了镉人群外暴露和内暴露水平研究，评估了PBTK模型的适用性；最后，采用Meta分析手段，研究了全国区域稻田土壤镉污染状况，并通过模型预测法，评估了稻田土壤镉经“土-米-人”途径暴露于人体的健康风险。主要内容和研究结果如下：

（1）构建了以PBTK模型与基于过程的稻田土壤镉暴露健康风险评估方法耦合的区域人群土壤镉暴露健康风险评估全概率模型体系，包括区域稻田土壤镉污染评价模型、稻米镉生物富集模型、人群镉暴露评估模型、暴露-反应评估模型和健康风险评估模型。

（2）通过赣南土壤镉暴露案例区开展了镉的人群外暴露和内暴露水平研究，并以概率分布描述人群生理模式和外镉暴露的差异和不确定性，开展人群尿镉预测，评估了概率改进后PBTK模型的适用性。与实测尿镉比较，PBTK模型对人群尿镉的预测效果总体较好，但对异常高值的预测效果较差。模型关键敏感参数包括尿镉排出率，消化道镉吸收系数、日均外暴露水平和血浆-肝脏镉转运系数。

（3）通过赣南土壤镉暴露案例区开展了尿镉和早期肾功能障碍效应的研究，构建了不同年龄段男性和女性尿中镉浓度与β₂微球蛋白（BMGU）异常（≥ 1000 μg/g）发生率的暴露-反应关系，估算了暴露区不同年龄段男性和女性BMGU异常的发生率中位值介于14.2% - 43.5% 和7.7% - 26.4%之间，超额风险水平分别介于1.3% - 7.9% 和1.2% - 8.1% 之间。

（4）采用Meta分析手段分析全国区域稻田土壤镉污染状况发现，稻田土壤镉年际加权平均浓度呈现先增加，后减少，最后基本平稳的变化趋势，但整体变化趋势无显著性差异；稻田土壤镉加权平均浓度、单项污染指数以及地累积指数，均表现出明显的空间变异。整体而言，北方稻区土壤质量整体处于清洁水平，但南方稻区土壤呈现出于不同程度的镉污染，其中矿产资源丰富、矿冶活动强度大的广东、湖南和广西，以及受污水灌溉活动影响的天津和辽宁，稻田土壤镉污染较为严重。

（5）通过模型预测法预测发现，西北地区40 - 49岁、50 - 59岁和60 - 69岁男性和女性稻田土壤镉暴露相关BMGU异常发生率为10.81%（4.70%）、21.27%（9.87%）和29.17%（14.38%）。相比西北地区，不同地区年龄别男性和女性超额风险变化规律均表现为：华南地区 > 西南地区 > 华东地区 > 东北地区 > 华北地区，区域人群稻田土壤镉暴露造成早期肾小管重吸收功能障碍的超额风险介于0.49‰到3.38%之间。

Cadmium is a non-essential toxic element for the growth and development of plants and animals. Exposure to environmental cadmium can cause damage to multiple organs and systems like kidney and skeletal system. It has been determined to be the prior researched food contaminants and the seventh toxic substances that endanger human health respectively by WHO and ATSDR. For a long time, heavy metal emissions have been caused by industrial production and agricultural development activities, and they have entered the soil environment through atmospheric sedimentation, fertilization, sewage irrigation, etc., causing heavy metal pollution to varying degrees. Compared with other heavy metals, the cadmium in the soil is highly mobile and toxic, which poses a higher health risk to regional populations. Therefore, in order to understand soil cadmium pollution and its health risks, and to prevent, control and manage the health risks of soil cadmium pollution, national and international scholars have carried out a lot of studies around the health risk evaluation of soil cadmium pollution. However, the health risk evaluation of cadmium pollution in paddy soils in China generally has the following four shortcomings: First, the national survey data for cadmium pollution evaluation in paddy soils is lacking, and the systematic review does not consider publishing the impact of bias on the comprehensive evaluation of published literature; Second, most traditional health risk evaluation methods do not take into account how body tissue have the differences in the bio-availability of cadmium via different exposure ways, and point estimation is based on the central trend statistics of the population. Third, obtaining the biological monitoring data is becoming increasingly difficult such as urine cadmium and blood cadmium. In recent years, a PBTK model considering in body toxicokinetic processes has been applied to correlate multimedia cadmium intake and cadmium load in the body, but it is lack of localization data for China's high-exposure scenario population. Therefore, it is of great theoretical and practical value to explore the health risk evaluating methods for soil cadmium exposure in the regional population based on the PBTK (physiologically based toxicokinetic) model. Therefore, in order to address the above issues, first I considers the differences in the bioavailability of cadmium exposure from different pathways, and proposes a PBTK model based on a probabilistic health risk evaluation system for soil cadmium exposure in the regional population. Secondly, in order to solve the lack of applicability evaluation data for the PBTK model, a study on the external and internal exposure levels of the cadmium population was conducted in the case area in southern Jiangxi Province in China, and the applicability of the PBTK model was evaluated. Finally, the meta-analysis method was used to study the cadmium pollution conditions in the paddy soils across the country, and through the model prediction method I evaluate health risks when paddy soil cadmium is exposed to human through the "soil-rice-human" pathway. The main contents and results of this study are as follows:

(1) A model system has been build up based on full-probability model for regional population soil cadmium exposure health risk evaluation coupled with process-based paddy soil cadmium exposure health risk evaluation methods as well as a PBTK model, including regional paddy soil cadmium pollution evaluation model, rice cadmium bio-accumulation model, population cadmium exposure evaluation model, exposure-response evaluation model and health risk evaluation model.

(2) The study on the external and internal exposure levels of cadmium has been conducted through the case study of soil cadmium exposure in southern Jiangxi, and the differences and uncertainties of the population's physical model and external cadmium exposure were described with probability distribution. The applicability of the improved PBTK model is shown. Compared with the measured urinary cadmium, the PBTK model has a good overall predictive effect on urinary cadmium, but it has an ineffective prediction effect on abnormally high values. The model's key sensitive parameters include urinary cadmium excretion rate, digestive tract cadmium absorption coefficient, daily average external exposure level, and plasma-liver cadmium transport coefficient.

(3) The study of the effects of cadmium in urine and early renal dysfunction was carried out through the case study area. It builds an exposure-response relationship between urinary cadmium concentration and the incidence of β2-microglobulin (BMGU) abnormalities ( 1000 μg/g Cre) in men and women at different ages. It estimates that the he median incidence of BMGU abnormalities in men and women at different ages in the exposed area is estimated to be between 14.2% - 43.5% and 7.7% - 26.4%, and the excess risk levels are between 1.3% - 7.9% and 1.2% - 8.1%.

(4) Analysis of cadmium pollution in paddy soils across the country has adopted meta-analysis. It has been found that the inter-annual weighted average concentration of cadmium in the paddy soil increased first, then decreased, and finally stayed at a stable trend, but there was no significant difference in the overall change trend; Paddy soil weighted average concentration of cadmium, the individual pollution index, and the geoaccumulation index all showed significant spatial variation. On the whole, the soil quality in the northern paddy area is at a clean level, but the soil in the southern rice area is exposed to different levels of cadmium pollution. The cadmium pollution in the paddy soil was relatively serious, especially in Guangdong, Hunan, and Guangxi, which are rich in mineral resources and intensive mining activities, as well as in Tianjin and Liaoning where sewage irrigation activities have had the influence.

(5) Model prediction methods have been used to assess the health risks of rice field soil cadmium exposure to the human body through the "soil-rice-human" way. It was found that the incidence of BMGU anomalies due to soil cadmium exposure in the paddy fields of men and women aged 40-49, 50-59, and 60-69 years in Northwest China was 10.81% (4.70%), 21.27% (9.87%) and 29.17% (14.38%). Compared with the Northwest region, the changes in the excess risk of men and women in different age groups in different regions are as follows: South China > Southwest > East China > Northeast > North China. This indicates that the excess risk of early renal tubular reabsorption dysfunction caused by cadmium exposure in the paddy soil of the regional population is between 0.49 ‰ and 3.38%.