近年来随着经济社会的发展，空气污染已经成为了一个全球性的重要健康问题。中国是世界上最大的发展中国家，污染水平高于欧美等发达国家，现有研究已证实了空气污染与多种呼吸系统疾病的相关性，加剧发作频率，并且对胎儿和新生儿的健康有重要影响。中国二孩政策全面开放以来，新生儿数量得到极大提升，保障新生儿的健康发育与成长就显得尤为重要。通过GIS时空分析的方法，将空气污染、温度时空数据与GIS结合起来，更好地分析得到影响新生儿健康的决定因素，进而可以为污染、温度暴露下新生儿疾病的监测防护提供科学依据。本文主要针对影响新生儿健康的若干问题，研究了空气污染与新生儿黄疸，空气污染与怀孕初期稽留流产和温度变化与稽留流产的关系与影响。主要取得以下成果：

采用广义相加模型首次系统量化分析了空气污染与黄疸胆红素峰值关系，并结合地理探测器对于空气污染暴露时间和胆红素峰值的关系做了进一步研究，结果表明高浓度的空气污染及长时间的污染暴露会增加新生儿黄疸风险。该研究结果是对空气污染与新生儿黄疸相关研究的重要补充。

采用Logistic回归结合限制性立方样条研究分析了空气污染与稽留流产关系，得到了二者之间的非线性剂量反应关系及定量结果，证实了较高浓度空气污染暴露会增加孕妇怀孕初期稽留流产风险，该研究结果对于空气污染环境政策的制定具有借鉴意义。

研究了距离主干道路远近对稽留流产风险的影响。结果表明距离主干道路越近，发生稽留流产的风险较大，该研究结论对于主干道路附近居民空气污染防护意识的提高具有促进作用。

利用分布滞后非线性模型对于稽留流产风险与温度变化关系进行了研究。使用双阈值分布滞后模型，采用多次循环迭代的方法得到冷阈值和热阈值，估计了温度低于冷阈值每降低1°C的冷效应和温度高于热阈值每升高1°C的热效应影响，结果表明高温或低温对稽留流产风险增加具有促进作用。该研究首次系统量化分析了温度变化与稽留流产关系，对于温度与其它健康结局的研究具有借鉴意义。

    在分布滞后非线性模型的基础上，对不同温度范围可归因风险进行了估计；定义了热浪和寒潮指示变量，对于热浪、寒潮对稽留流产的附加影响进行了研究。结果表明低温天气所导致的稽留流产人数最多，热浪、寒潮对于稽留流产发生均为促进作用，并且寒潮影响要更持久。与其它针对热浪、寒潮研究方法相比，本文方法考虑更为全面，结果较准确。

With the development of economy and society in recent years, air pollution has become an important global health problem. China is the largest developing country in the world, and its pollution level is higher than that of developed countries such as Europe and the United States. Existing research has confirmed the correlation between air pollution and a variety of respiratory diseases, exacerbating the frequency of attacks, and has an important impact on the health of the fetus and newborn. Since the two-child policy in China has been fully liberalized, the number of newborns has greatly increased, and it is especially important to ensure the healthy development and growth of newborns. Through the method of GIS spatio-temporal analysis, air pollution, temperature, spatio-temporal data are combined with GIS to better analyze and determine the determinants affecting the health of newborns. Furthermore, it can provide scientific basis for the monitoring and protection of neonatal diseases under pollution and temperature exposure. This thesis focuses on several issues affecting the health of newborns, and studies the relationship and impact of air pollution and neonatal jaundice, air pollution and missed abortion in early pregnancy and temperature changes and missed abortion. The main results are as follows:

Generalized Additive Model (GAM) was used for the first time to quantitatively analyze the relationship between air pollution and the peak value of bilirubin in jaundice. Furthermore, the relationship between the exposure time of air pollution and the peak value of bilirubin was further studied in combination with geographic detectors. The results indicate that high levels of air pollution and prolonged exposure to pollution can increase the risk of neonatal jaundice. The results of this study are an important complement to the research related to air pollution and neonatal jaundice.

Logistic regression and Restricted Cubic Spline (RCS) were used to analyze the relationship between air pollution and missed abortion, and the nonlinear dose-response relationship and quantitative results were obtained. It is confirmed that higher concentrations of air pollution exposure will increase the risk of miscarriage of pregnant women in the early stages of pregnancy. The results of this study have reference significance for the formulation of air pollution environmental policies.

The effect of the distance from the main road on the risk of miscarriage was investigated. The results show that the closer to the main road, the greater the risk of miscarriage. The conclusion of the study has a promoting effect on the awareness of air pollution protection of residents living near the main roads.

The relationship between the risk of missed miscarriage and temperature changes was investigated using a Distributed Lag Nonlinear Model (DLNM). The double threshold DLNM model was used to obtain the cold and hot thresholds using multiple iterations. The cold effect for each 1°C decrease in temperature below the cold threshold and the hot effect for each 1°C increase in temperature above the high threshold were estimated. The results show that high or low temperature can promote the increased risk of miscarriage. This study systematically and quantitatively analyzes the relationship between temperature changes and missed abortion for the first time, which is of reference significance for the study of temperature and other health outcomes.

    Based on the DLNM, the attributable risks in different temperature ranges were estimated. The indicator variables of heat wave and cold spell were defined, and the additional effects of heat waves and cold waves on missed abortion were investigated. The results show that the number of miscarriages due to moderate cold was the largest, and heat waves and cold spells all promoted the occurrence of miscarriages. The effects of cold spells are more permanent relative to heat waves. Compared with other research methods for heat waves and cold spells, the method in this thesis is more comprehensive and the results are more accurate.