By Ken Perrott 28/03/2018 26


Slide number 30 from Paul Connett presentation prepared for a talk at NZ Parliament buildings in February 2018.

Continuing my critique of the presentation prepared by Paul Connett for his much-publicised meeting at Parliament Building in February. The meeting attracted only three MPs but his presentation is useful as it presents all the arguments anti-fluoride campaigners rely on at the moment.

My previous articles on this presentation are Anti-fluoride activist commits “Death by PowerPoint” and Paul Connett’s misrepresentation of maternal F exposure study debunked.

In this article, I deal with the argument presented in the slide above. it is an argument repeated again and again by activists. Connett has posted a more detailed list of these studies and his description of them in Fluoride & IQ: The 52 Studiesat the Fluoride Action Network website.

Studies in areas of endemic fluorosis

All the 52 studies comment refers to are from regions of endemic fluorosis in countries like India, China, Mexico and Iran where dietary fluoride intake is above the recommended maximum level. People in these areas suffer a range of health problems and studies show cognitive deficits as one of them. However, a quick survey of Google Scholar shows this concern is well down the list (See Endemic fluorosis and its health effects). Only 5% of the Google Scholar hits related to health effects of endemic fluorosis considered IQ effects.

People in high fluoride areas where fluorosis is endemic suffer a range of health problems. Credit: Xiang (2014)

In, most, but not all, cases the major source of fluoride in the diet is drinking water with high fluoride levels (above the WHO recommended 1.5 mg/L). Paul Connett’s logic is simply to extrapolate to low drinking water fluoride concentrations typical of community water fluoridation (CWF). However, we do not see the other health effects like severe dental fluorosis, skeletal fluorosis, etc., where CWF is used.

His logic also ignores the possibility that cognitive deficits may result from other health problems common in areas of endemic fluorosis. Problems such as premature births and low birth weight, skeletal fluorosis or even the psychological effect of unsightly teeth due to severe dental fluorosis.

Comparing “high” fluoride villages with “low” fluoride villages

This approach is simplistic as it simply compares a population suffering fluorosis with another population not. Yes, the underlying problem is the high dietary intake (mainly from drinking water) in the high fluoride villages – but that does not prove fluoride in drinking water is the direct cause of a problem. The examples discussed above, eg., low birth weights or premature births, could be the direct cause.

It is easy to show statistically significant differences of drinking water fluoride and a whole host of fluorosis related diseases between two villages but that, in itself, does not prove that drinking water fluoride is the direct cause. Nor does it justify extrapolating such results to other low concentrations situations typical of CWF.

Paul Connett’s logic ignores the fact that in most of these studies the “low” fluoride villages (which the studies were treating as the control or normal situations where IQ deficits did not occur) had drinking water fluoride concentrations like that used in CWF. It also ignores, or unjustly attempts to dismiss) studies which show no cognitive deficits related to CWF.

A low fluoride concentration study showing an IQ effect

After making a big thing about the large numbers of studies and being challenged by the high fluoride concentrations involved Connett normally goes into a “yes, but” mode and attempts to transfer that credibility of “large numbers” to the very few studies which report effects at low fluoride concentrations.

He usually makes a big thing of the study by Lin et al (1991):

Lin Fa-Fu, Aihaiti, Zhao Hong-Xin, Lin Jin, Jiang Ji-Yong, M. (1991). THE RELATIONSHIP OF A LOW-IODINE AND HIGH- FLUORIDE .ENVIRONMENT TO SUBCLINICAL CRETINISM lN XINJIANG. Iodine Deficiency Disorder Newsletter, 24–25.

Connett claims this study shows a lower IQ when the drinking water F concentration was 0.88 ppm, but the areas suffered from iodine deficiency which is related to cognitive deficits.

The study I reviewed recent by Bashash et al (2017) (see Paul Connett’s misrepresentation of maternal F exposure study debunked) is also on Connett’s list. He doesn’t mention, however, that while an association of child IQ with prenatal maternal urinary fluoride was reported the paper also reported there was no observed association of child IQ with child urinary fluoride concentrations.

Studies not showing an effect

Connett lists 7 studies which showed no effect on IQ. One of these was the well-known Broadbent et al., (2014) study from New Zealand, which he, of course, proceeds to debunk in an irrational and not very truthful manner.

He does not mention the studies from Canada (Barberio et al. 2017 ) and Sweden (Aggeborn & Öhman 2016) which also show no effect of CWF on IQ.

The 6 other studies listed are all Chinese, and not translated. Interesting because Connett’s Fluoride Action Network invested money and time into translating obscure Chinese papers that could support their argument of harm. They obviously did not bother translating those papers which did not confirm their bias.

Conclusion

So, Connett’s 52 studies are rather a waste of time. Based in areas of endemic fluorosis their findings are not transferable to areas where CWF is used. The quality of most papers is low and, usually, the studies are simply a comparison of two villages, one where fluorosis is endemic and the “control” village where it isn’t but drinking water concentrations are like that used in CWF.

Connett simply is not able to properly evaluate, or in some cases even consider, studies which show no effect of fluoride on IQ or were made in areas where CWF exists and no effects are shown.


26 Responses to “The 52 IQ studies used by anti-fluoride campaigners”

  • Ken, it is unfortunate that you continue to consider that the Bashash et al. study is not of concern. You have no idea of the statistics behind IQ research. Bashash et al. was a very rigorous study funded by the US and was for 12 years.

    The IQ studies that you mention are extremely relevant because they were considered relevant when the US HHS reduced the level of fluoride to 0.7ppm down from the target range of 0.7 – 1.2ppm.

    See this quote from the December 5, National Toxicology Program Committee Meeting

    1hr 07min https://www.youtube.com/watch?v=lytzqSyGV2E

    “I just want to make the comment that both John and I served on the HHS effort that revised what the recommendation was and brought it down. From a high as possibly 1.2 down to .7. And part of that had to do with the fact that when you looked at all of the(IQ) literature there was evidence for effects occurring certainly as low as about 2.5, maybe lower than that and going from 1.2 to 2.5 is only a margin of exposure of about 2 fold. And we know nothing, as I said before about differential susceptibility and vulnerability that occurs within the population. And that was part of the justification for taking it down to .7 which actually was kind of the low end of what had been recommended in 1962 as the low end of the range for public health protection.” Linda S. Birnbaum, Ph.D. Director, NIEHS & NTP

    • Kane – my only concern directly relating to the Bashish study is that it did not include the statistical measurements describing the degree of variance explained by their model which is important for its evaluation and use in predictions. However, they did include the data which enabled an intelligent person to see that very little variance was explained. Data is always useful and I commend them for that. I do not have any problem with the model itself as I found, using a less sophisticated linear regression, essentially the same coefficient for the slope of the best-fit line and its confidence interval that Bashish reported.

      You claim a better understanding of the statistics than I have – so please tell me what your statistical analysis tells you. What degree of confidence and 95% confidence intervals do you place on Paul Connett’s claim that a 1 mg/L increase in maternal-neonatal urinary F causes a drop of 5 – 6 IQ points? Let’s have the benefit of your superior statistical skills.

      As for the factors influencing the US Public Health Service revision of recommendations that can be easily checked. Read:

      Gooche, B. F. (2015). U.S. Public Health Service Recommendation for Fluoride Concentration in Drinking Water for the Prevention of Dental Caries. Public Health Reports, 130(1), 1–14. https://doi.org/10.1177/003335491513000408.

      This lists as the rationale for the new recommendations:

      Importance of community water fluoridation.
      Trends in availability of fluoride sources.
      Relationship between dental caries and fluorosis at varying water fluoridation concentrations, and
      Relationship of water intake and outdoor temperature among children and adolescents in the United States.

      IQ is not mentioned except in relation to submissions and public comments:
      “IQ and other neurological effects. The standard letters and approximately 100 unique responses expressed concern about fluoride’s impact on the brain, specifically citing lower IQ in children.”

      Please don’t disingenuously attempt to conflate the mass commenting and submitting typical of organisations like yours with actual science.

      The report explains why those submissions were irrelevant. They certainly did not provide a reason for the recommended changes.

  • Ken, did you not see the quote from Dr Linda Birnbaum? The IQ studies were a part of the reason to reduce the level of fluoride concentration to 0.7ppm. Incidentally, Auckland Council followed Fluoride Free NZ’s advice to lower the fluoride concentration level and ignore the NZ MOH.

    See this quote from the December 5, National Toxicology Program Committee Meeting

    1hr 07min https://www.youtube.com/watch?v=lytzqSyGV2E

    “I just want to make the comment that both John and I served on the HHS effort that revised what the recommendation was and brought it down. From a high as possibly 1.2 down to .7. And part of that had to do with the fact that when you looked at all of the(IQ) literature there was evidence for effects occurring certainly as low as about 2.5, maybe lower than that and going from 1.2 to 2.5 is only a margin of exposure of about 2 fold. And we know nothing, as I said before about differential susceptibility and vulnerability that occurs within the population. And that was part of the justification for taking it down to .7 which actually was kind of the low end of what had been recommended in 1962 as the low end of the range for public health protection.” Linda S. Birnbaum, Ph.D. Director, NIEHS & NTP

    • Kane – I am aware that anti-fluoride propagandists continually misrepresent the discussions and plans for research of the NTP as established findings (which, of course, they aren’t) – but did you not read the relevant report which describes in 16 pages the rationale for the change in recommendations?:

      Gooche, B. F. (2015). U.S. Public Health Service Recommendation for Fluoride Concentration in Drinking Water for the Prevention of Dental Caries. Public Health Reports, 130(1), 1–14. https://doi.org/10.1177/003335491513000408

  • Ken, do you understand the concept of ‘Margin of Safety’?

    Birnbaum was quite clear that IQ was a part of the reason to reduce the level to 0.7ppm: “And that was part of the justification for taking it down to .7”

    • Kane, the problem for you is that whatever has been said vaguely in a discussion or exchange nothing of this is reported in the official documents. Therefore your claim is simply wrong.

      By the way, I am still waiting for your description of the significance and CI for Connett’s claim based on you implied high degree of statistical skills.

      It is telling that you ignore my request.

  • Ken, Auckland Council has had to purchase 25% less fluoride this year compared with previous years. Watercare began to reduce the fluoride concentration in January 2016. A win for common sense and a good start at ceasing fluoridation all together.

    • Yes, I know Kane, you appeared before the Auckland City Council – does that appeal to your Walter Mitty vanity.

      The fact remains the Auckland City Councils levels for fluoride in drinking water remain within the guidelines recommended by the MoH – as do the levels for all other water treatment plants where CWF exists in NZ.

      Meanwhile – can you answer my question about the accuracy and statistical significance of the Paul Connett’s widely publicised claims about the effects of CWF on child IQ?

  • Ken, it is best to go to the experts on this issue. You and I are not qualified to comment on the statistical analysis.

    Dr. Howard Hu, lead author said, “This is a very rigorous epidemiology study. You just can’t deny it. It’s directly related to whether fluoride is a risk for the neurodevelopment of children. So, to say it has no relevance to the folks in the U.S. seems disingenuous.” –National Post 9/20/17 . . . “We tested for all the things we could think of that could act on neurodevelopment. But we haven’t found anything else that was a potential confounder.” – CTV News 9/19/17

    Dr. Phillipe Grandjean, world-renowned scientist/author on neurotoxicity, unaffiliated with this study said: “I think this study is a red flag. And when you take it into consideration with the Chinese studies, I think the time is way overdue for a broad-scale evaluation of fluoride exposure.” – Medscape 10/2/17

    • Kane, I take your response to mean you feel you are not skilled enough to comment on the statistics involved.

      This is underlined by your inability to understand that neither of the quotes you give relate to the specific question I put to you about the accuracy and statistical significance of the Paul Connett’s widely publicised claims about the effects of CWF on child IQ?

      Incidentally, this sort of appeal to authority rather than facts and science is wearing rather thin. For example, I do not consider Granjean a reputable authority on this matter for a number of reasons. The main one being his refusal as chief editor to even allow consideration of my paper – Perrott, K. W. (2017). Fluoridation and attention deficit hyperactivity disorder – a critique of Malin and Till ( 2015 ). Br Dent J. https://doi.org/10.1038/sj.bdj.2017.988.

      That decision was unethical because he did not allow a critique in the journal where Malin & Till published. This thereby denied them the normal right of reply. And his reason for not allowing publication relate to his own bias and political activity on the question.

      I complained to the company publishing the journal and they dealt with it. Maybe a slap on the wrist with a wet bus ticket – but clearly a reprimand for unethical behavior.

      Scientists are used to dealing with facts and reason – appeal to authority in this manner does not impress them. Espoecially as they are likely to know a bit more about the “authorities” than you do.

  • Ken, I said I wasn’t qualified and neither are you. Why don’t you respond to the Bashash paper directly and advise the authors that they are wrong? Do you disagree with Howard Hu’s comment?

    “This is a very rigorous epidemiology study. You just can’t deny it. It’s directly related to whether fluoride is a risk for the neurodevelopment of children. So, to say it has no relevance to the folks in the U.S. seems disingenuous.” –National Post 9/20/17 . . . “We tested for all the things we could think of that could act on neurodevelopment. But we haven’t found anything else that was a potential confounder.” – CTV News 9/19/17

    • Hu’s comments do not relate at all to the issue we are discussing – they are simply a political promotion of his work which, unfortunately, many scientists must resort to these days.

      I do not think the authors are wrong at all. They are not the ones making the claim about the effects of fluoridation in New Zeland – that was Paul Connett. Paul is wrong.

      Of course, as a scientist, there will be minor details of any paper I review that I can and often do critique. In the Bashash case, there are niggles (eg the poor technique for determining urinary F, the fact that fluorosis presence was not included in the analysis, etc.) But these are inevitable with what one is given in life.

      I think Bashash could have provided more of their statistical analytical results to show that the sort of prediction made by Connett is flawed. And yes, this is the subject of a response to the Bashash paper that I have already written. It is currently being reviewed by peers locally and will be submitted when that, and resulting revisions, is complete.

      Hopefully, this will enable Bashash et al to clarify the limits on the precision of predictions made using their model.

      Incidentally, incomplete reporting of statistical analysis data and incorrect reliance on p-values are common problems with published papers. I guess many authors, and many reviewers, do not have experienced statisticians to advise them. One advantage I had in my career is that we worked closely with statisticians and they always served on review committees. That experience was invaluable.

  • Interesting that you have rubbished both Grandjean and Hu today in your ramblings.

    It must be difficult to be the ‘only scientist’ that can understand this issue.

    • I have not “rubbished” Hu – but yes I have criticised Grandjean for his unethical behavior as the chief editor of the journal involved. I also formally complained about that behaviour and I have been told that he has been dealt with – whatever that means.

      But this experience reinforces for me how unreliable he is as an “authority” – although I am not so constrained that I pin my hat on “authorities” – as you well know.

  • Ken, you seem to be ideologically driven. Have you ever been a member of the NZ Communist Party (or similary named political organisation)?

  • So were you a member Ken? I think it is relevant from a ideological perspective to understand motives.

  • Thank you for your continued work to out the misrepresentation of science from this cult of anti-fluoridation.

  • Ken and Den, another fluoride study published. It just keeps flooding out.

    http://oem.bmj.com/content/75/Suppl_1/A10.1

    Abstract
    Background/aim Recent studies report an inverse association between fluoride (F) exposure and IQ in children, but few included individual measures of exposure or assessed associations with prenatal exposure using a prospective study design.

    Methods This study utilised the Early Life Exposures in Mexico to Environmental Toxicants (ELEMENT) birth cohort and archived pregnancy samples to study prenatal F exposure and its association with subsequent child neurobehavioral outcomes at ages 1, 2 and 3 years. A Generalised Mixed Model (GMM) was used to model the association between mean creatinine-adjusted urinary F (MUFcr), averaged over three trimesters, and Mental Development Index (MDI), a subscale of the Bayley Scales of Infant Development-II (BSID-II) test, among 401 mother-infant pairs. The analysis controlled for maternal age, education, marital status, ELEMENT cohort, child’s sex, and child’s age.

    Results The median MUFcr was 0.835 mg/L (minimum: 0.195, maximum: 3.673). MUFcr was significantly inversely associated with offspring MDI scores, with an increase in MUFcr of 0.5 mg/L (roughly the interquartile range value) corresponding to a decrease in MDI of −1.20 points (95% CI: −2.19,–0.20).

    Conclusion Our findings add to our team’s recently published report on prenatal fluoride and cognition at ages 4 and 6–12 years by suggesting that higher in utero exposure to F has an adverse impact on offspring cognitive development that can be detected earlier, in the first three years of life.

    • Yes, Kane, I have had the abstract for about a week now and was wondering when the anti-fluoride campaigners would latch on to it.

      However, the word went our from FAN headquarters this morning with a newsletter from Paul Connett alerting the troops. So I fully expect to see plenty of social media traffic on this in the coming days. Talk about trolls and “Fake News.”

      I, of course, do not like to comment on abstracts or draw conclusions from them. Unfortunately, this is the abstract of what was probably a poster paper presented at the 3rd Early Career Researchers Conference on Environmental Epidemiology. The meeting was in Freising, Germany, on 19-20 March 2018. https://www.isee-young.eu/

      So it might be some time before a proper evaluation is possible. However, I have been meaning to correspond with Thomas about her data so I may see if I can get the data she used for this poster.

      So, Kane, there is a bit of a mystery and I wonder if you, with your contacts, can solve the mystery.

      I have had Deena Thomas’s PhD thesis for some time and have actually read it. She, unusually, does not include data (hence my interest in corresponding with her) but these are the conclusions she made in her thesis about the early childhood (years 1 – 3) data.

      “Neither maternal urinary or plasma fluoride was associated with offspring MDI scores”

      And

      “This analysis suggests that maternal intake of fluoride during pregnancy does not have a strong impact on offspring cognitive development in the first three years of life.”

      And

      “Maternal intake of fluoride during pregnancy does not have any measurable effects on cognition in early life.”

      Now that is exactly the opposite of the conclusions she presented in the poster:

      “Our findings add to our team’s recently published report on prenatal fluoride and cognition at ages 4 and 6–12 years by suggesting that higher in utero exposure to F has an adverse impact on offspring cognitive development that can be detected earlier, in the first three years of life.”

      So you can understand why I wonder what is going on.

      One clue is that she appears to have removed 30 mother/child data points for the poster. This could explain the different conclusions.

      Hopefully, I will get some answers from her about why she removed data and changed her conclusions.

      However, I strongly suspect with the CIs and best-fit coefficient she quotes that we will be looking at a similar scatter of data to that presented by Bashash et al, and another relatively large standard error which will make accurate predictions impossible.

      Anyway, if you can find out why the conclusions were changed and the 30 data points removed, please let me know.

  • http://oem.bmj.com/content/75/Suppl_1/A10.1

    Thomas et al., concluded “Our findings add to our team’s recently published report on prenatal fluoride and cognition at ages 4 and 6–12 years [the Bashash et al 2017 paper] by suggesting that higher in utero exposure to F has an adverse impact on offspring cognitive development that can be detected earlier, in the first three years of life.”

  • This might help explain Ken.

    http://fluoridefree.org.nz/fluoridation-defender-ken-perrott-is-wrong-again/

    Fluoridation defender Ken Perrott is wrong again.
    He does not understand epidemiological studies of neurotoxins.

    Ken Perrott seems to think the results of the Bashash study should look like those in laboratory experiments where everything can be controlled, even to the extent of having genetically identical cloned animals. Humans don’t work that way. You can’t get a population of humans that all have the same IQ score and then vary just their fluoride exposure.

    Ken is digging himself deeper and deeper into a hole by maintaining that the large degree of scatter in graphs of fluoride versus IQ in the Bashash study means its findings are of questionable validity.

    He doesn’t seem to understand that IQ tests are designed to have scores with a relatively wide degree of variance. They are commonly normalized so that the population mean is 100 and the standard deviation is 15 points. The standard deviation (SD) is a measure of the degree of variance. Such IQ tests are also expected to have a roughly normal, or bell-shaped, distribution of scores about the mean. For any group with a mean score of 100 and an SD of 15, there will be a few children on the extreme “tails” of the distribution with scores up around 130 or down around 70, and most of the rest will be somewhere in between. That gives an expected overall range for most samples of children of about 60 IQ points.

    Furthermore, numerous studies have shown that genetic variation explains most of the variation in IQ scores: more than 80% of the variance. Epidemiological studies are not able to control for individual variation in genetics, so studies of neurotoxins and other environmental factors that affect IQ will always have a large degree of scatter.

    To illustrate what is typical in developmental neurotoxin studies, in terms of degree of scatter of data, here are Neurotoxin study scattergrams of IQ versus lead and mercury. These studies all found the neurotoxin to have a large and statistically significant effect. The studies were by respected researchers, published in high quality journals, and all concluded that they found clear evidence that the neurotoxin reduced IQ by the estimated amounts.

    If Ken continues to argue that the wide scatter in the Bashash study invalidates the conclusion that it found clear evidence of an effect, then he’s going to have to challenge all these other studies of lead and mercury effects on IQ too. We look forward to hearing back from Ken about fluoride after he has successfully debunked the studies that show that lead and mercury lower the IQ of children. Then we can all rest easier because we will no longer have to worry about the neurotoxicity of lead and mercury, let alone fluoride.

    Ken also needs to look up the difference between a Prediction Interval and a Confidence Interval. He is incorrectly using Prediction Intervals to assess the confidence in, or the validity of, the findings of the Bashash study. Confidence Intervals are the proper measures, as used by the authors of the paper.

    Here is a concise explanation of the difference:

    A Prediction Interval is used for predicting the chance that a single observation (a single person in the case of the Bashash study) will have an outcome (IQ score) that falls within the range of the Prediction Interval.

    A Confidence Interval is used for assessing the probability that the entire population will have the relationship (the dose-response relationship) that falls within the Confidence Interval.

    A Prediction Interval might be of interest to a clinician who was asked by their patient: “Given my individual urine fluoride level, age, socio-economic status, smoking history, IQ, lead, mercury, and other factors, can you predict what the IQ of my child will be at age 4 years, and what is the “margin of error” for your prediction?” Most clinicians would say this is an impossible question to answer, but if they were to do the calculations based on the Bashash results they would have to use the Prediction Interval to calculate the margin of error, and it would understandably be very wide.

    In contrast, the Confidence Interval is what is used in all epidemiological studies to assess the confidence one has that the results of the study reliably predict what the true relationship is in the entire population. The result of the Bashash study is the predicted dose-response relationship for the entire sample, which is about 6 IQ points lost per 1 mg/L increase in urine F for the Bashash study. Note that this does represent “all the data”, despite Perrott’s repeated erroneous statements that it does not represent “the data as a whole”. The multivariate regression analysis by which the dose-response relationship was calculated used all the data.

    • No, Kane, it doesn’t. It doesn’t even refer to Deena Thomas’s abstract.

      You must be having a senior moment.

    • I what way, Kane?

      by the way, why does Mary not take up my offer of a right of reply?

      If she did this article would be posted on my blog and here and get more coverage than hidden away on your website or in comments.