Cross Lydia
Public Dental Service, NHS Lothian, Edinburgh, UK.
Evid Based Dent. 2025 Jun 19. doi: 10.1038/s41432-025-01176-x.
Taylor KW, Eftim SE, Sibrizzi CA, Blain RB, Magnuson K, Hartman PA, Rooney AA, Bucher JR. Fluoride Exposure and Children's IQ Scores: A Systematic Review and Meta-Analysis. JAMA Pediatr 2025; 179: 282-292.
A literature search without language restrictions was undertaken in October 2023 using the following databases: BIOSIS, Embase, PsycINFO, PubMed, Scopus, Web of Science, CNKI, and Wanfang.
Two independent reviewers screened studies based on the title and abstract. Studies were included if the exposure was fluoride and the outcome was a quantitative measure of children's intelligence. In studies with multiple fluoride exposure levels, the highest was considered the exposure and the lowest the reference. Exclusion criteria included case reports, articles without original data, and conference abstracts. Full text articles were then screened. Translation assistance was obtained for non-English studies.
Data was extracted by one extractor and verified by a second. Risk-of-bias was evaluated by two independent trained assessors using the National Toxicology Programme's OHAT approach. Three meta-analyses were conducted: mean-effects, dose-response mean-effects, and regression slopes. Subgroup and sensitivity analyses were undertaken, including subgroup analyses without high risk-of-bias studies.
74 studies were included, of which 59 reported mean IQ scores for group-level exposures and 19 reported regression slopes for individual-level exposures. Children exposed to higher levels of fluoride had statistically significantly lower IQs than those exposed to lower levels (standardised mean difference (SMD), -0.45; 95%CI, -0.57 to -0.33; P < 0.001). A dose-response association was also reported, with lower IQ scores reported with increasing fluoride exposure with a SMD of -0.15 (95%CI, -0.20 to -0.11; P < 0.001) for water fluoride levels and -0.15 (95%CI, -0.23 to -0.07; P < 0.001) for urinary fluoride levels. Fluoride exposure concentrations of <4 mg/L, <2 mg/L, and <1.5 mg/L were analysed. For water fluoride <1.5 mg/L, the association was not statistically significant. In the regression slopes meta-analysis, a 1 mg/L increase in urinary fluoride was associated with a decrease in IQ score of 1.63 points (95%CI, -2.33 to -0.93; P < 0.001); the effect size was smaller when analysis was limited to low risk-of-bias studies (decrease of 1.14 points (95%CI, -1.68 to -0.61; P < 0.001)).
An inverse association was found between fluoride exposure and children's IQ scores, including an inverse dose-response association. This association was less certain at water fluoride concentrations <1.5 mg/L.
泰勒·KW、埃夫蒂姆·SE、西布里齐·CA、布莱恩·RB、马格努森·K、哈特曼·PA、鲁尼·AA、布彻·JR。氟暴露与儿童智商分数：一项系统评价和荟萃分析。《美国医学会儿科学杂志》2025年；179:282 - 292。
2023年10月进行了无语言限制的文献检索，使用了以下数据库：生物学文摘数据库、Embase、心理学文摘数据库、医学期刊数据库、Scopus、科学网、中国知网和万方数据。
两名独立评审员根据标题和摘要筛选研究。如果暴露因素为氟，且结果是对儿童智力的定量测量，则纳入该研究。在有多个氟暴露水平的研究中，最高水平被视为暴露因素，最低水平被视为对照。排除标准包括病例报告、无原始数据的文章和会议摘要。然后对全文进行筛选。对于非英文研究，获取了翻译协助。
由一名提取人员提取数据，并由另一名人员进行核实。两名经过培训的独立评估人员使用美国国家毒理学计划的OHAT方法评估偏倚风险。进行了三项荟萃分析：平均效应、剂量反应平均效应和回归斜率。进行了亚组分析和敏感性分析，包括不纳入高偏倚风险研究的亚组分析。
纳入了74项研究，其中59项报告了群体水平暴露的平均智商分数，19项报告了个体水平暴露的回归斜率。暴露于较高水平氟的儿童的智商在统计学上显著低于暴露于较低水平氟的儿童（标准化平均差（SMD），-0.45；95%置信区间，-0.57至-0.33；P < 0.001）。还报告了剂量反应关联，随着氟暴露增加，智商分数降低，水氟水平的标准化平均差为-0.15（95%置信区间，-0.20至-0.11；P < 0.001），尿氟水平的标准化平均差为-0.15（95%置信区间，-0.23至-0.07；P < 0.001）。分析了氟暴露浓度<4mg/L、<2mg/L和<1.5mg/L的情况。对于水氟<1.5mg/L，该关联无统计学意义。在回归斜率荟萃分析中，尿氟每增加1mg/L，智商分数降低1.63分（95%置信区间，-2.33至-0.93；P < 0.001）；当分析仅限于低偏倚风险研究时，效应量较小（降低1.14分（95%置信区间，-1.68至-0.61；P < 0.001））。
发现氟暴露与儿童智商分数之间存在负相关，包括负剂量反应关联。在水氟浓度<1.5mg/L时，这种关联不太确定。
