NHMRC Public Consultations

Skip Navigation and go to Content
Visit NHMRC website

Draft NHMRC Information Paper: Effects of water fluoridation on dental and other health outcomes submission

Personal Details
First Name: 
Last Name: 
Specific Questions
Q1. A) Is the draft Information Paper presented in a manner that is easy to understand? : 
Q1. B) Please provide details regarding your response to Question 1A: 
Plain Language Summary

I do not believe that you have provided a rational explanation for forbidding any critique of the technical reports. This is most unusual. Are you saying that you have put together a team that is infallible on this subject? I believe it is demeaning of the public (both professional and ordinary citizens alike) that you consider they have nothing to offer in correcting or improving on the technical analysis of this team.

I do not believe that a report purportedly designed to protect the health of all of the Australian people should have imposed such restrictive criteria on what evidence would be allowed. No stone should have been left unturned. All available evidence should have been used – on animals, humans, clinical trials, biochemical studies – and then a weight of evidence approached used to assess potential dangers (seeking for red flags); studies should have been included in other languages (e.g. Chines, Russian) not just English, and also bearing in mind the weaknesses of the NHRC 2007 review, studies should not have been limited to post Oct 2006. There are studies published prior to 2006, which take on a greater significance in the light of studies published post Oct 2006, e.g. Xiang et al. 2003a,b) see discussion below.

A risk assessment should have been attempted based on a LOAEL and NOAEL for the end point of lowered IQ (and/or other neurological effects, e.g. ADHD) to determine a safe daily dose that would protect all Australian children from harm from a combination of their estimated exposure to fluoride from all sources including fluoridated water, dental products, natural levels in some foods, air pollution, and pesticide and fertilizer residues on food. A short analysis along these lines is provided below. Even in the simplest of terms such an analysis indicates that Australian children are already exceeding doses of fluoride that would exceed a dose that could lower IQ, even without exposure from fluoridated water. This makes water fluoridation unthinkable.

Any independent scientist - especially a toxicologist or risk assessment specialist - would have to conclude that the way the NHMRC has gone about constructing this review is highly unsatisfactory.

This is particularly worrying when over the last few years (since the 1990s) it has become clear that there is a growing and very extensive body of scientific literature that fluoride is a neurotoxic substance.

As of today there are over 300 animal and human studies that indicate that fluoride can damage the brain. The NHMRC team managed to avoid most of these studies and failed to do a “weight of evidence” analysis to estimate  the potential risk to Australia’s children of being exposed to fluoride from water and other sources.

 Considering that 50 out of 56 studies have indicated that fluoride is associated with a drop of IQ when comparing children living in villages with a high fluoride exposure and those living in a low fluoride village, what was needed was to find a well conducted study which allowed one to determine the lowest observable adverse effect level (LOAEL) and using standard risk assessment methods to determine a safe daily dose that would protect all Australian children (especially the most vulnerable) from this harm (i.e. lowered IQ). The NHRMC failed to do this, even though such studies existed. Even with its arbitrary cut-off date of Oct 2006, the NHMRC had the Harvard meta-analysis in hand (Choi et al, 2012). In this 27 IQ studies were reviewed and out of these 26 showed a loss of IQ, with an average loss of 7 IQ points. As most (all but two) of these studies had fluoride concentrations in the high fluoride villages less than one order of magnitude above the levels used in fluoridated communities in Australia, any reasonable toxicologist or risk assessment specialist would have to conclude that there was sufficient concern to necessitate calculating the lowest level from among all these studies where IQ was lowered.

Had they done so they would have found a well-conducted study which would have enabled them to do this. This was the study by Xiang et al, 2003 a,b. (arbitrarily  excluded because it was published before Oct 2006) and an updated analysis of the same data, Wang et al, 2012).

There were several important features of this Xiang study,

 a) Xiang et al controlled for lead and iodine exposure and subsequently for arsenic exposure);

 b) Xiang et al found an apparent dose response relationship: i.e. as fluoride increased among five sub-groups (A-E), IQ decreased (see their Table 8); and

 c) this dose response relationship held within the high fluoride village alone, thus discounting any confounding factors between the two villages.

From this dose relationship it can be shown that the lowest concentration where IQ was lowered was 1.53 ppm (sub group B). From subsequent calculations by Wang et al this equates to a daily dose of less than 1.4  mg/day (Figure 2, Wang et al., 2012) . Even without going to a further step of estimating a NOAEL (this is usually set at 10 times less than the LOAEL, if no data is available) and without using a safety factor to protect against the range of sensitivity expected in a large population (a default value of 10 is used), it can be shown that some Australian children are ALREADY exceeding this daily dose from other sources (including swallowing fluoridated toothpaste and tea consumption), making the deliberate addition of more fluoride to this daily dose from fluoridated water unthinkable.

For those who would argue that the Xiang et al (2003 a,b) study only represents one study it should be noted that at least 10 other IQ studies of the 27 reviewed by Harvard ( Choi et al, 2012) would offer no adequate margin of safety  to protect all of Australia’s children if similar calculations were performed on their results for the level of fluoride where IQ was lowered.

 The NHMRC’s Ignoring this analysis – or failing to attempt one of its own - is reckless in my view and telling the Australian people that fluoridation poses no risk to the brain (or any other tissues) is scientifically unsupportable and simply reflects the contrived way the analysis was conducted as discussed above.

This recklessness is compounded by that fact that proponents of fluoridation can offer no randomly controlled study (the gold standard for human epidemiological studies) conducted – either at the individual or community level -  that demonstrates that swallowing fluoride actually lowers tooth decay.

My son Michael Connett is submitting a far more extensive analysis of fluoride’s impact on the brain and this should be compared side by side with the NHMRC’s review of the same subject. Should Australian citizens and scientists make this comparison for themselves they will see the NHMRC report for what it is: a public relations exercise constructed to support a practice (i.e. government policy) not a truly independent exercise to examine the risks that water fluoridation policy actually poses.


Choi et al., 2012.

Xiang et al.  (2003,a,b).

Wang et al. (2012).

Q2. A) Is it clear how NHMRC reviewed the health and dental effects of water fluoridation? : 
Q2. B) Please provide details regarding you response to Question 2A: 

It is not clear on how the Cochrane review on effectiveness (2015) was treated. Originally it was meant to stand alone as the critique on which the NHMRC would solely rely – but it appears to have been downgraded to a secondary role and the analysis given over to known pro-fluoridation analysts

Q5. Is there additional evidence on any other health effects of water fluoridation that should be considered in the draft Information Paper?: 

In my previous comments I pointed out that as far as determining the potential health risks of water fluoridation was concerned I did not believe that limiting studies to human epidemiological studies was sufficient. I felt that the NHMRC should have considered ALL the evidence that might indicate fluoride’s potential risks to brain development, including animal studies and biochemical studies. As the NHMRC review panel chose to use only human epidemiological studies, I would like to ask why it is that the NHMRC was seeking advice from the US agencies NIEHS and NTP on how to go about doing systematic reviews of animal studies on fluoride’s neurotoxicology. [NHMRC has removed potentially commercial-in-confidence information] When the NTP performed such a review they reported that the evidence of fluoride’s neurotoxicology in animals ranged from low to moderate? (See Systematic Literature Review on the Effects of Fluoride on Learning and Memory in Animal Studies. NTP Research Report 1. National Toxicology Program. http://fluoridealert.org/wp-content/uploads/ntp-2016.pdf Does that conclusion not provide supporting evidence for the biological plausibility that fluoride is neurotoxic in humans? And does it not indicate that the NHMRC’s decision to exclude animal (and other studies) a poor one and that their final analysis is inadequate for identifying the potential risks that fluoride poses to the brain. Again this commentary goes to the weakness of the review of fluoridation’s health effects and the process by which it was organized and communicated to the public.

Page reviewed: 4 July, 2017