Childhood diseases correspond to major public health issues. A large number of studies using different approaches provide evidence regarding effects of environmental exposures, encompassed in the exposome, on children's health. We aimed to summarize the overall level of evidence (LoE) from all streams of evidence regarding exposome effects on child health.

For 88 selected chemical and urban factors, we retrieved the conclusions of agency reports or literature reviews published between 2015 and 2021 regarding effects on child health, including cardiovascular, metabolic, neurodevelopmental, respiratory and other health outcomes. Adapted versions of PRISMA flowchart and AMSTAR-2 tool were used to select and assess the quality of the systematic reviews retrieved from PubMed and SCOPUS databases.

For each factor-outcome pair, conclusions in three streams of evidence (epidemiological, toxicological and mechanistic, the latter corresponding to in vitro and in silico approaches) were translated into stream-specific LoEs and then combined into an overall LoE ranging from “very unlikely” to “very likely”.

The 88 environmental factors were implied in 611 factor-outcome pairs. Forty-four pairs (7%), corresponding to 16 factors, had a very likely overall LoE (≥80%); 127 pairs (21%), corresponding to 49 factors, had a likely or more overall LoE (≥60%). For 81 pairs (13%), no evidence was available in agency reports or published reviews, while for 275 pairs (45%), corresponding to 68 factors, the overall LoE was very unlikely (<20%). Exposure factors with the greatest number of associated health outcomes with a high overall LoE were HCB, PCBs, temperature (8 outcomes), PFOA (7 outcomes), PFOS, cotinine (6 outcomes), arsenic, lead (5 outcomes), bisphenols A and S, PFNA and PM_2.5 (4 outcomes), DDT, DDE and DDD, PFHxA, PFDA, green space, UV radiation (3 outcomes).

We developed an approach to extract and summarize the existing evidence about effects of environmental factors on health. The plausibility database built for children's health can be used to identify research gaps, conduct quantitative risk assessment studies. It could be expanded to consider a larger fraction of the exposome and other age groups and should be updated on a regular basis.

儿童疾病对应重大公共卫生问题。使用不同方法的大量研究提供了有关暴露组中包含的环境暴露对儿童健康影响的证据。我们的目的是总结有关暴露组对儿童健康影响的所有证据流的总体证据水平 (LoE)。

对于 88 个选定的化学和城市因素，我们检索了 2015 年至 2021 年间发表的有关对儿童健康影响的机构报告或文献综述的结论，包括心血管、代谢、神经发育、呼吸和其他健康结果。使用 PRISMA 流程图和 AMSTAR-2 工具的改编版本来选择和评估从 PubMed 和 SCOPUS 数据库检索的系统评价的质量。

对于每个因素-结果对，三个证据流（流行病学、毒理学和机制，后者对应于体外和计算机方法）的结论被转化为特定流的 LoE，然后组合成总体 LoE，范围从“非常不可能” ”改为“很有可能”。

611 个因素-结果对中隐含了 88 个环境因素。对应 16 个因素的 44 对 (7%) 具有很可能的总体 LoE (≥80%)；127 对 (21%)，对应于 49 个因素，具有可能或更高的总体 LoE (≥60%)。对于 81 对 (13%)，机构报告或发表的评论中没有提供证据，而对于 275 对 (45%)，对应 68 个因素，总体 LoE 的可能性很小 (<20%)。与总体 LoE 较高的相关健康结果数量最多的暴露因素包括 HCB、PCB、温度（8 个结果）、PFOA（7 个结果）、PFOS、可替宁（6 个结果）、砷、铅（5 个结果）、双酚 A S、PFNA 和 PM _2.5（4 个结果）、DDT、DDE 和 DDD、PFHxA、PFDA、绿地、紫外线辐射（3 个结果）。

我们开发了一种方法来提取和总结有关环境因素对健康影响的现有证据。为儿童健康建立的合理性数据库可用于识别研究差距，进行定量风险评估研究。它可以扩展到考虑更大比例的暴露组和其他年龄组，并且应该定期更新。