Content of review 1, reviewed on October 30, 2020

Overall this study was well written, with some strengths and weaknesses. There were some minor revisions suggested below. There were also some major flaws in the method, results and discussion and conclusion which, if are not revised deem the conclusions of the study invalid. The following suggestions should be considered by the authors:

Title and abstract: The title was interesting, informative and was an accurate summary of the paper. The paper's aim addressed a clearly focussed issue and the methods were adequately outlined in the abstract. The results of the study aligned with the aim.

References: The references used in this study were relevant with no key studies missing. The majority of references were current with approximately 50% of the cited references from the last 5 years (2015-2020). The references were cited correctly.

Introduction: The introduction provided a brief summary on the existing literature on the topic, however it assumed the reader understood the indications for community water fluoridation. Providing background information on the benefits of water fluoridation in caries prevention would have provided more context to the study. Additionally, more detail on how maternal urinary fluoride (MUF) was a reflection of fluoride exposure from drinking water would have assisted the reader in understanding the research. The introduction clearly outlined the research question and justified the need for the study based on the gap in the current literature on the topic.

Methods: Overall the method described a complex cohort study and was easy to follow. The method clearly outlined the subject selection which appeared appropriate. The study used a flowchart to present the inclusion criteria of the sample population. The flowchart was confusing as some participants were included in more than one section which made it difficult for the reader to follow. Explaining the number of participants that fall into two or more categories would make the flow chart easier to understand. As there was not sufficient detail on the demographic used in the study, the results were not generalisable. The exposure of daily fluoride intake in mothers was measured through matching the participant’s postal codes to estimate the water fluoride concentration for each woman. This had low reliability as it relied on information from the water treatment plants to determine the local water fluoride concentration which was highly variable. The study also used urinary fluoride measurements in an attempt to increase reliability, however these were also variable. The outcomes were measured appropriately. The method provides sufficient detail however, to improve the reproducibility of the study, the questionnaire used to determine the mother’s daily fluoride intake should have been included in the appendix.

Results: The data was presented in an appropriate and logical way. The tables and figures were clearly presented, although the data presented figure 3 was not discussed in the results section. This data revealed no statistically significant IQ differences between children from nonfluoridated and fluoridated communities and no overall relationship between MUF and IQ. Therefore, the results are biased as results that contradicted the conclusions of the paper were ignored. The results also indicated that boys are more susceptible to having a lower IQ due to fluoride exposure than girls. This is not a practically meaningful result as the authors failed to explore the biological and developmental differences at this age between boys and girls, which is likely the cause of this outcome.

Discussion and conclusion: The discussion clearly stated the major conclusion from the study which was ‘that estimated maternal exposure to higher fluoride levels during pregnancy was associated with lower IQ scores in children’. However, this conclusion was invalid due to the biased results previously discussed in the results. The discussion considered some limitations of the study, but failed to address the limitations in the analysis of the study. The conclusion stated that the study indicates the need to reduce fluoride intake during pregnancy. This implied a causation relationship between the variables, however, the study violated at least three of the Hill criteria of causality, so this conclusion is invalid.


    © 2020 the Reviewer.