Wednesday, February 13, 2013

Why the Portland water fluoridation initiative keeps me awake at night

Since August 2012, I have been trying to get answers to my questions regarding community water fluoridation.  That's when I read in the Oregonian that there was a coalition who was pushing for adding fluoride to Portland's drinking water in order to reduce tooth decay.

I had questions related to safety (especially dental fluorosis and the potential link between fluoride and decreased IQ) and tried to engage people involved in the initiative in a dialogue to get some answers, without much luck.

So I was delighted when I finally found a forum where devoted dental health professionals like Kurt L Ferré would listen to my concerns.  In a thread on the coalition Healthy Kids Healthy Portland's Facebook site, I was able to explain in detail why it is relevant to establish a safe level of fluoride in drinking water with respect to children's IQ, and why there are important lessons to be learned from our history with lead poisoning in the U.S.

Unfortunately, the dialogue came to an abrupt ending when all my comments in the thread suddenly were removed.  I was tremendously disappointed.

Fortunately, I had saved the thread on my computer, and I am sharing the part from where I joined the thread below.  If you are interested in stuff like the Null Hypothesis, the importance of population-based studies, what impact a small shift of the Bell curve can have, or just want to see how emotional the discussion about water fluoridation can get, read on!

The Facebook thread

Healthy Kids Healthy Portland
No. Fluoride occurs naturally because of mineral deposits. The fluoridation process is not shown to have negative environmental consequences, as evidenced by the 65 years of its use around the US.
Feb 7 at 9:55pm · Like

Alex Garcia
Uranium naturally occurs and it has polluted our country in the evidence of the last 65 years.
Feb 7 at 10:28pm · Like

Kåre Hultén
Nice non sequitur, Alex.
Feb 7 at 11:08pm · Like

Nate Osborne
It takes two to Tango
Feb 8 at 8:47am · Like

Magnus Carlsson
Kåre Hultén, I think Alex has a point though. The fact that a mineral is naturally occurring and has been used for 65 years does not imply that its use is a safe practise. A relevant parallel is the use of lead in gasoline and paint, a practice that it took decades after evidence showed up before U.S. policy finally caught up. Many European countries had banned lead paint 50 years earlier. We still have not established a safe limit for fluoride in water with respect to children's IQ, have we?
Feb 8 at 11:31am · Like · Edit

Kåre Hultén
Nice attempt at character assassination of fluoride, Magnus. Especially since lead is a harmful toxin, with NO health benefits WHATSOEVER.
Furthermore, lead has never been a nutrient, nor has it ever been used as a public health measure to treat disease, and, therefore, associating lead with fluoride is just a thinly veiled attempt at poisoning the well and strawmanning.

 By the same line of reasoning, fluoride would have to be removed from toothpaste, mouth rinse, bottled mineral water, tea (I could go on). Last I checked, fluoride was a selling point in dental care products.

And like so many times before, you’re JAQ-ing off with your last question: “We still have not established a safe limit for fluoride in water with respect to children's IQ, have we?”.
As far as I know, Magnus, you haven’t demanded the removal of all naturally occurring fluoride from the Bull Run water supply.

Clearly, you have spent a great deal of time studying water fluoridation. But, for someone who claims to be undecided, you seem remarkably inept at finding pro fluoride articles (not that there’s any shortage of them!). Your ineptitude is only balanced by your keen aptitude in finding, and clinging on to, anti fluoride material.
And you cling to the Harvard IQ-study like shipwrecked cling to the floating remains of a sunken ship. A study where the authors themselves, specifically and unequivocally, state that their Chinese meta analysis cannot reasonably be used against the public health fluoridation programs in the US.

I don’t pretend to know everything about fluoride. That’s why I ask healthcare professionals, with relevant fields of expertise, when forming an opinion.

Maybe you should do the same?

ETA: I’m sorry if his comes off as hard and uncompromising, but Magnus and I go way back. I just find his debating style disingenuous and unpalatable, and, for the sake of our friendship, I’ve asked him not to discuss this directly with me, but he keeps calling me out. I guess I now know what that friendship was worth…
Feb 8 at 7:45pm · Like · 1

Healthy Kids Healthy Portland
We agree that lead and uranium are non-sequitors from the actual issue at hand. We also want to foster respectful dialogue here. Surely we can all agree on that?
Feb 8 at 7:52pm · Like · 1

Magnus Carlsson
Kåre Hultén - I am so sorry that I haven't clearly expressed to you last week that I indeed find the idea of water fluoridation questionable, after I started studying the research last summer. I am also truly sad that I have to let you down - I am not asking you any more questions privately, but I cannot stay away from public discussions in the matter. I am concerned about the safety of the practice, and I get more and more frustrated over how difficult it is to openly discuss fluoride safety research with people who propose water fluoridation.

This is why I may sound like a broken record when I keep referring to the Harvard IQ study. My conclusion after reading the study is that neurotoxicity of fluoride should be better researched, and I was intrigued when Healthy Kids Healthy Portland mentioned much higher quality studies that show no relationship between optimal fluoridation and any neurological changes. I keep asking for the references to these studies which could help put the matter to rest, but I get no response.

One thing that I wanted to point out with my reference to the lead article is that we *might* have a situation where water fluoridation is beneficial for dental health, but also results in a some non-obvious negative neurological impact. For lead, it was very tricky to establish its negative impact on children's IQ and turn the evidence into policy changes. The responsible thing to do in the case of water fluoridation is to find out the answer now, either by digging up the missing piece of research, or by conducting it.

So I would like to respectfully disagree that lead is a non sequitur in this discussion. Since we are already quoting the authors of the Harvard IQ study from sources outside the actual paper, let me finish with another one: “Fluoride seems to fit in with lead, mercury, and other poisons that cause chemical brain drain,” Grandjean says. “The effect of each toxicant may seem small, but the combined damage on a population scale can be serious, especially because the brain power of the next generation is crucial to all of us.”
Feb 8 at 11:10pm · Like · Edit

Healthy Kids Healthy Portland
First off: That study was not funded by Harvard. It was a project of several doctors who attended Harvard. The methodology they used is questionable, and they have publicly denounced the use of that study in discussion of fluoridation:
Harvard scientists: Data on fluoride, IQ not applicable in U.S. | Wichita Eagle
Feb 8 at 11:13pm · Edited · Like

Magnus Carlsson
Healthy Kids Healthy Portland - thank you for providing the link to the article. I don't understand why the study cannot be used in a discussion about fluoridation and its underlying research or lack thereof. To quote the article:

'Two of the scientists who compiled the Harvard study on fluoride said it really doesn’t address the safety of fluoridation levels typical of American drinking water.

“These results do not allow us to make any judgment regarding possible levels of risk at levels of exposure typical for water fluoridation in the U.S.,” the researchers said in an e-mail response to questions from The Eagle. “On the other hand, neither can it be concluded that no risk is present.”

The researchers noted that the fluoride levels they studied were much higher than what is found in fluoridated water in the United States and recommended “further research to clarify what role fluoride exposure levels may play in possible adverse effects on brain development, so that future risk assessments can properly take into regard this possible hazard.”'
Feb 8 at 11:26pm · Like · Edit

Healthy Kids Healthy Portland
The study they conducted was on children being exposed to much-higher-than-optimal fluoride content in water. We support the optimal fluoridation of Portland's water to 0.7ppm.
Feb 8 at 11:30pm · Like

Magnus Carlsson
Thank you for your response. Can you also provide the reference to the higher quality studies that show no relationship between optimal fluoridation and any neurological changes, that you referred to earlier?
Feb 9 at 12:06am · Like · Edit

Healthy Kids Healthy Portland
The 2006 NRE study on the EPA's drinking water standards for fluoride is where the recommendation of 0.7 - 1.2ppm comes from .
Fluoride in Drinking Water: A Scientific Review of EPA's Standards
Feb 9 at 12:11am · Like

Healthy Kids Healthy Portland
Here's the CDC statement on that report:
Feb 9 at 12:11am · Like

Magnus Carlsson
Thanks again, Healthy Kids Healthy Portland, I really appreciate that you take the time!

I must have have misunderstood your earlier comments when you talked about the mixed quality of the Chinese IQ studies that the Harvard meta-analysis covered. You were referring to a much higher quality study of optimal fluoridation and neurological changes. However, the 2006 review of EPA' standards does not provide any such study in itself, it is a review of the scientific basis for EPA's maximum-contaminant-level goal (MCLG) of 4 mg/L. (The review's main conclusion was that the MCLG should be lowered, since its current level does not protect children against severe dental fluorosis, and is not likely to protect against bone fractures, page 3.)

The review's findings regarding fluoride's neurotoxicity is in alignment with the Harvard IQ study: there is a number of Chinese studies of quality that it found difficult to assess, but they are consistent in that they correlate fluoride exposure with IQ deficits (page 8 ). Just like the Harvard study, the NRC review finds that there is ground for additional research on the effects of fluoride on intelligence.

This brings me back to what I was asking for before: I think the responsible action now is to demand high quality studies that establish a safe limit for fluoride in drinking water, so that we know it doesn't impact children's IQ.
Feb 9 at 11:31am · Edited · Like · Edit

Magnus Carlsson
Kåre Hultén brought up some more interesting questions when I made my controversial parallel to the history of lead. Let me try to rephrase them, and please correct me if I am wrong.

As Kåre mentioned, we are already exposed to fluoride from the Bull Run, mineral water, tea, dental care products, and so on. The implicit question is: if we now find out (Heaven forbid) that fluoride has neurotoxic effects, shall we just stop drinking and eating? How could it be possible in the first place, that we have a toxin everywhere?

It might not be that bad for grown-ups. But the brain is extra sensitive to neurotoxins during its initial critical development. When it comes to fluoride in the drinking water, there is an important subgroup: infants who are given formula instead of breast milk. By adjusting the drinking water to optimal fluoride levels, we have a 600% increase in the amount of fluoride these infants get exposed to through the water (I believe Bull Run today has 0.1 mg/L, perhaps someone can confirm, and we are targeting 0.7 mg/L). We don't know whether or not that increase has any effect on children's IQ.

The 0.7 mg/L exposure that formula-fed infants receive should also be compared with the 2-8 microgram/L that breast-fed infants receive, a difference in orders of magnitude. There is an active plasma-milk barrier that keeps fluoride several times lower in the mother's milk compared to the mother's blood. In the paper below, it was suggested that the newborn is actively protected against fluoride exposure from the breast milk.
No evidence of transfer of fluoride from plasma to breast milk.
Feb 9 at 4:49pm · Like · Edit

Healthy Kids Healthy Portland
Infants growing up currently in Vancouver and Beaverton, where the water is fluoridated, do not show any evidence of problems with fluoride intake. According to this study, fluoride is known to cross the placenta because it is a vital mineral nutrient in the development of milk teeth.
[The passage of fluorid... [J Gynecol Obstet Biol Reprod (Paris). 1990] - PubMed - NCBI
Feb 9 at 7:06pm · Like

Magnus Carlsson
Thank you for the link, it sounds interesting! Regarding infants in our fluoridated areas: without proper case studies we cannot know if IQ is affected.
Feb 9 at 7:20pm · Like · Edit

Healthy Kids Healthy Portland
This is a misleading claim, Magnus. The 2006 NRC report exhaustively looked into the total effects of fluoride on human health and found no reason to be concerned for IQ.
Feb 9 at 7:33pm · Like

Magnus Carlsson
I am sorry if I come across as misleading! Let me quote from Page 8 of the 2006 NRC report I was referring to before:

"A few epidemiologic studies of Chinese populations have reported IQ deficits in children exposed to fluoride at 2.5 to 4 mg/L in drinking water. Although the studies lacked sufficient detail for the committee to fully assess their quality and relevance to U.S. populations, the consistency of the results appears significant enough to warrant additional research on the effects of fluoride on intelligence."

Can you point me to specific pages in the report that can help me understand that it found no reason to be concerned for IQ?

I am sorry for bringing up the 2012 Harvard IQ meta-analysis once again, but it reports on Page 16:

"Although the studies were generally of insufficient quality, the consistency of their findings adds support to existing evidence of fluoride-associated cognitive deficits, and suggests that potential developmental neurotoxicity of fluoride should be a high research priority."

It is the collection of evidence together with the lack of high-quality IQ studies that I am concerned about. Please help me understand what you find misleading in my line of reasoning, because I find this important. Thank you!
Feb 9 at 8:09pm · Like · Edit

Kurt L Ferré
Magnus, you are using the typical ploy of the anti-fluoride activists: that being the use of the Null Hypothesis.

Null hypothesis - Wikipedia, the free encyclopedia "The null hypothesis can never be proven. Data, such as the results of an observation or experiment, can only reject or fail to reject a null hypothesis. For example, if comparison of two groups (for example, comparing subjects treated with a medication with untreated subjects) reveals no statistically significant difference between the two, it does not prove that there really is no difference; it only shows that the results were not sufficient to reject the null hypothesis."
Feb 10 at 2:00am · Like

Kurt L Ferré
Magnus, you seem to want 100% certainty of fluoridation. What I can tell you is that since the beginning of fluoridation over 68 years ago, the average IQ in the United States has risen 15 points. Let's look at some of the fine educational institutions in the United States: Harvard, Columbia, Georgetown, Duke, U. of Michigan, U. of Chicago, Stanford, UC-Berkeley, U. of Washington, and closer to home, Oregon State, Willamette University, and Pacific University (to name just a few) are all located in a long-standing fluoridated community. Prove to me Magnus, that any of the professors, their families, the citizens in those communities, have any IQ deficiencies. I can tell you that so many of my friends who were born and raised with fluoridation just shake their heads in disbelief that reduced IQ has become the "allegation du jour" of the anti-fluoride activists.
Feb 10 at 2:08am · Like · 1

Magnus Carlsson
Kurt, I truly appreciate that you take your time to listen to my concerns, and I am delighted that you also brought this discussion straight to the core of the issue!

Many people and organizations who support water fluoridation start off by presenting it as being safe, and it is not uncommon for us to see the wording that water fluoridation has been "proven safe". Such wordings push the buttons of all anti-fluoride activists of course, and are easily understood as "water fluoridation is 100% safe", and prompts questions like "show us the proof then that it is 100% safe!" That's how we get to the dead end of the null hypothesis.

I am not asking for 100% safety, and there are no studies that talk about 0% or 100%, but instead we talk about confidence intervals and statistical significance. Why am I so focused on safety anyways? Since I grew up in Sweden where water fluoridation is illegal, I found the practice foreign when I heard about the Portland plans in August last year. I have attempted to review the current research about fluoride safety, and by the way, Kurt, you have been helpful in increasing my understanding about fluorosilicates, for example. I still have some remaining questions that keep me awake at night though.

The rise of IQ test scores over the years brings up a number of interesting questions. Out of many influential factors, the one particularly relevant here is children's exposure to toxic chemicals. Again, I must recommend Bellinger & Bellinger's paper about childhood lead poisoning, because it explains so well what I am trying to get across. And since I broke the rules and mentioned the L-word, I will try to explain carefully why the paper is relevant, so that it doesn't come across as another strawman argument.

According to the paper, one of the factors that slowed down the efforts to eliminate lead poisoning was a prevailing focus on patients instead of populations. If an individual showed no symptoms, it was assumed that he or she was not lead poisoned. Then, in the early 1970s, population-based studies showed that there was something called "subclinical lead poisoning". The lesson is that it is impossible to examine a person (professor or not) and learn if this person's mental function has been affected by a small amount of lead poisoning. What if fluoride in the water would happen to have a similar effect on mental function? We would not be able to tell by looking at individuals.

What population-based studies showed was that if a large number of children were exposed to lead, the IQ distribution of the population was shifted down. When you shift a Bell curve slightly like this (I am going with the same example of 5 IQ points as the article), nothing dramatically happens with the population around the average, and the shift could easily be dismissed as noise. But look what happens closer to the tails: we get double the number of children with IQ under 70, and we cut the number of kids above 130 in half. The best that could be said about this situation is that it guarantees jobs for teachers in special education.

But as I said, Bellinger & Bellinger really explains this much better, so if you have not read their paper yet, please do!

What is then the connection with fluoride? I would hate to learn down the road that fluoride also has a negative impact on IQ, but I cannot rule it out because I am missing a crucial piece of research.

This is why I found it relevant to quote one of the authors of the Harvard study earlier in this thread. Grandjean really captures the essence of the connection that I now have tried to make with Bellinger & Bellinger's paper, so please allow me to repeat the quote: “Fluoride seems to fit in with lead, mercury, and other poisons that cause chemical brain drain,” Grandjean says. “The effect of each toxicant may seem small, but the combined damage on a population scale can be serious, especially because the brain power of the next generation is crucial to all of us.”

Kurt, I see an underlying question in your comment: how could water fluoridation possibly be bringing IQ down when we have rolled out fluoridation in the U.S. over the last 68 years, yet at the same time we have seen IQ test scores increase? We simply don't know today what the IQ would have been without fluoridation. Even higher, perhaps? The same? Or lower? There are so many factors involved, and fluoride may or may not be one of them.

Now I hope that I have explained a bit clearer why I find the IQ studies so disturbing.

To summarize:

1. We are seeing an increasing number of foreign studies of uncertain quality that show a correlation between fluoride in drinking water and decreased IQ; the 2006 NRC review suggested that this connection should be further researched, and the Harvard IQ meta-analysis showed that the correlation was statistically significant, and the authors suggest that further studies should be a research priority.

2. I am asking for a high-quality case study that properly handles confounding factors and that can answer: is there a statistical significance between fluoride in drinking water and IQ test score? If there is - can we establish a safe limit?

Please point out if I am missing something in my line of reasoning, because that might help me sleep at night. Thank you!
JCI - Childhood lead poisoning: the torturous path from science to policy
Yesterday at 9:05am · Like · Edit

Magnus Carlsson
Healthy Kids Healthy Portland, you write that "According to this study, fluoride is known to cross the placenta because it is a vital mineral nutrient in the development of milk teeth." That is a remarkable conclusion. I was not able to find this conclusion in the abstract of the study, and I currently have no access to the full paper. Can you confirm that the full paper contains the claim that fluoride is a vital mineral nutrient in the development of milk teeth? Thank you!
2 minutes ago · Like · Edit 

Final words

Within ten minutes or so after my last post, all my comments were removed from the thread.

If you can point out what I am missing in my reasoning, please let me know!

If you can help me get in touch with professionals in public health in other ways, that would be fantastic.

Please consider sharing this message, because I think the open questions around water fluoridation safety are hugely important.


I just came across a fresh post called Fluoridated Water and brains.  Read it, it is by Philippe Grandjean, and it appears that the Wichita Eagle did not check the facts with the authors.


  1. I have read the discussion and must say I am dismayed at the attempts to dismiss your concerns. The big problem seems to me that there is not much quality information about the subject out there - either pro or con.

    On tablet now so will comment on some things later. Here is one interesting piece I came across as I was tunneling into this particular rabbit hole: I was surprised to learn that in Europe there is improvement in battling caries over the last few decades but this is true whether you flour I date the water or not! Fluoride has to do with it but more likely in the form of toothpaste.

  2. Stefan, it is indeed so much noise out there on both sides. I found the 2006 NRC review to be a good starting point.

    Here is a WHO world map giving a picture of the dental health situation in 2003. Europe, where I believe only 3% of the population has water fluoridation, is comparable with the U.S., I think. So if it turns out that water fluoridation really is a brain drain, the U.S. can get inspiration from Europe.

  3. While I don't give much credence for IQ as it is used for determining who is worthy of a good school I do think it is possible to measure cognitive ability with tools like it.

    Moving a large group seven point would indeed be a huge effect if it could be attributed to a single cause. Most of the weight to the shift would come from the mean (just the nature of Bell curves) but I think it is reasonable not to think of it as a uniform 7 point shift - to me it would make more sense if you see a consistent shift vertically in a cumulative distribution view.

  4. Magnus- Outstanding job finding your way through the minefield of fluoridation, and asking the right questions. Three points to add to your concerns:

    1.) Grandjean's systematic review says: "From the geographical distribution of the studies, it seems unlikely that fluoride-attributed neurotoxicity could be due to other water contaminants [such as lead and arsenic]."

    2.) Among the 27 studies, all but one study showed ... estimates that indicated an inverse association ...." Inverse meaning the higher fluoride exposure group had lower IQ. So, 26 of 27 studies all found a lowering of IQ in the higher fluoride group. The average IQ drop was 7 points. The remarkable consistency and the size of the effect are indeed worrying.

    3.) Amongst the 27 studies, the water concentration of the high fluoride groups ranged from 0.88 to 11.5 mg/L. Three studies had concentrations no higher than 2.0 mg/L, and eleven studies had concentrations below 3.0 mg/L.

    So the frequently heard argument by defenders of fluoridation that the exposure levels are irrelevant to artificial water fluoridation (0.7-1.2 mg/L) is unfounded. A factor of 4 is an inadequate margin of safety, especially since no threshold has been established by any high quality studies, below which no lowering of IQ occurs.

    The US EPA routinely requires a 10-fold safety factor, and that is when a well-established no-effect threshold is demonstrated. From the existing body of evidence on fluoride and IQ, after a 10X safety-factor is applied, the safe concentration would be less than 0.3 mg/L, or well below the level that fluoridation proponents want to put into drinking water.

    The reason for a safety factor is because there will always be some people who are much more sensitive or who are exposed to much higher levels of any toxicant, than the average level. When the scientific evidence is weak for a no-effect threshold, as with fluoride and IQ, the 10X safety factor is especially important.

  5. Anonymous - thank you for your feedback - good points, and good to have them recorded here!

    I would have looked for opportunities to bring them up in my exchange, had I not been banned by Healthy Kids Healthy Portland.

  6. It has been established since before 1936 that Fluoride in an aqueous solution is a protoplasmic poison, therefore, depending on any number of affecting compounds, molecules and inorganic substances present, it is a cell destroyer par excellence.

    The Journal of the American Medical Association, Sept 18, 1943, editorial, said:
    “Fluorides are general protoplasmic poisons.”

    This concurs with an earlier declaration contained in another professional Journal.

    The Journal of the American Dental Association, Volume 23, page 568, April, 1936, titled "Fluorine in relation to bone and tooth development" by Floyd DeEds, Phd,
    where that previous statement corroborated this research.

  7. Here is an example of a good quality study showing no relationship between fluoridation and IQ:

    The National Research Council reviewed the same studies as Choi et. al., and determined that they were all too poor quality to conclude anything. Bazian Ltd., a British research institute found the same thing:

    1. Bazian Ltd. was hired by a British public health authority (Southampton) who were aggressively promoting fluoridation. Bazian was hired to try to counteract the scientific evidence the opponents of fluoridation had presented. Hired consultants such as Bazian will never bite the hand that feeds them. Bazian presented some valid criticisms of the Chinese IQ studies, but also made some glaring errors in their critique.

      The NRC 2006 report did not conclude the IQ studies were too poor to conclude anything. You either haven't read the NRC report or you are just making things up as you go. Haven't you figured out Magnus is incredibly detail oriented, rational, and cautious in his approach to these questions. Making things up won't get you far here, and will kill your credibility.

  8. Mel, I highly appreciate your attention to this matter, and thank you for the pointers!

    Here it the problem as I see it, taking your references into account.

    We are accumulating indications that fluoride in the drinking water has developmental neurotoxic effects. As Grandjean and Landrigan have explained in their 2006 Lancet review, we know that brain development is extremely sensitive in the fetus and early childhood, and chemicals are able to permanently disturb one or many steps of the development, where each step is critical and cannot be redone later if it is disturbed. Grandjean and Landrigan point out that fluoride is one of the obvious candidates for being included in the list of neurotoxic substances, and later research only strengthens the indications.

    So how do we know what is a safe limit of fluoride exposure for pregnant mothers and for formula-fed infants? We are lacking high-quality IQ studies to establish a margin of safety.

    The Ferguson study from 1986 measured the following variables. Number of years the child has resided in a fluoridated area, morbidity, conduct disorder, and a number of control variables: maternal education, maternal age, child's ethnicity, family type and socioeconomic status.

    The problem is that the study did not control for the mother's exposure to fluoride during pregnancy, and we know nothing about the children's exposure to fluoride during the first six months before which the blood-brain barrier that provides protection against toxic chemicals is incomplete. Were the children breast fed or formula fed?

    Moreover, it was not IQ that was measured by directly testing the children. It was conduct disorder, which was assessed by maternal and teacher ratings based on the Rutter and Conners behavior scales. I have seen no research saying that a subtle decrease in IQ will be picked up by such indirect ratings.

    Finally, since the study was on water fluoridation levels and says nothing about the actual exposure to fluoride, it fails to establish any margin of safety.

    So even though this study is high quality in the sense that it controls for a number of relevant variables, it cannot be used to dismiss the indications of potential early developmental neurotoxic effects. In short: we need to stop looking under the lamppost.

    Now, what about the NRC review? Its assessments are actually consistent with the Harvard review. Both reviews point out that although there are many uncertainties with the studies, the consistency of the results warrants additional research (page 8 in the NRC review, page 16 in the Harvard review).

    The Harvard review substantially extends the NRC review by doing a meta-analysis. The result of the analysis strengthens the indications by establishing a statistical significant decrease of IQ test scores, and the analysis was based on over 27 studies that were considered to be independent of each other.

    Finally, we have the review conducted by Bazian. It points out what was already documented in the NRC and the Harvard reviews: the Chinese studies lack control for confounding variables, and Bazian points out that the impairment in IQ can be explained by other factors. This is obvious, and only the missing high-quality studies can increase our understanding.

    The Bazian study also pointed out weaknesses in an earlier meta-analysis by Tang et. al., but as far as I can see, these weaknesses were addressed by the Harvard meta-analysis (including doing proper pooling and addressing heterogenity).

    [My reply is continued in a later comment.]

  9. [continued]

    Mel, suppose for a second that fluoride in drinking water really is impairing brain development. How much stronger do the indications need to be before we take the precautionary approach given in the conclusion of the Grandjean and Landrigan review? Here it is:

    "Prevention of neurodevelopmental disorders of chemical origin will need new approaches to control chemical exposures. The vulnerability of the human nervous system and its special susceptibility during early development suggest that protection of the developing brain should be a paramount goal of public health protection. The high level of proof needed for chemical control legislation has resulted in a slow pace of interventions to prevent exposures to lead and other recognised hazards. Instead, exposure limits for chemicals should be set at values that recognise the unique sensitivity of pregnant women and young children, and they should aim at protecting brain development. This precautionary approach, which is now beginning to be used in the EU, would mean that early indications of a potential for a serious toxic effect, such as developmental neurotoxicity, should lead to strict regulation, which could later be relaxed, should subsequent documentation show less harm than anticipated. As physicians, we should use prudence when counselling our patients, especially pregnant mothers, about avoidance of exposures to chemicals of unknown and untested neurotoxic potential."

  10. Hi Magnus,

    I think we would all agree that additional research is good. But, in public policy there will always be some uncertainty in results, and we need to make policy decisions to support the public good even when uncertainty exists.

    The question becomes what step is in the interest of the public good? And, the benefits of fluoridation in reduced tooth decay are about as strong evidence as you get in public policy. There have been more than 100 studies showing that fluoridation reduces tooth decay by at least 25%, and three studies have shown that Medicaid dental costs are cut in half from fluoridation. So, we could literally cover twice as many people with dental coverage with fluoridated water. Fluoridation is simply a game-changer in dental health for kids and adults alike.

    But, you are asking specific questions about fluoridation and cognitive development. And, I would say clearly that no researcher is actually saying there is credible evidence that fluoride has any effect on the brain, especially not at the recommended level of 0.7ppm. All the researchers that have reviewed the studies in China and elsewhere have said they are poor quality studies. It is very logical, and in fact, is very likely that confounding factors are distorting all these results. The studies look at naturally-occurring sources of fluoride. If there is high levels of natural fluoride in the water that comes from soil, coal burning, etc., then this will be correlated with lots of other things in the water including arsenic and lead, which are known to effect cognitive development. If these things are not controlled for, then you can not conclude that there is credible evidence for an impact on cognition. And, that is exactly how the researchers phrased the results, that the meta-analysis could not eliminate the possibility that fluoride affected intelligence, but in no way said that there was credible evidence that fluoride actually did affect cognition. They suggested more research, and that is certainly appropriate.

    But, as I pointed out there is a good quality study looking at fluoridation at the recommended level on cognition that showed no impact. I disagree with your point about it not addressing IQ. IQ is a poor measure of cognitive development. It varies by race and class, and has been shown to increase consistently over time, which is known as the Flynn effect. That is why studies in industrialized countries don't use IQ to assess cognitive development. It is the wrong measure to use. Also, it doesn't make sense to control for variables like maternal water use, unless you are suggesting that drinking lower levels of fluoride would increase kids intelligence. Confounding factors are factors that distort the results.

    The best evidence supports a conclusion that fluoridation at the recommended level has no impact on cognition. Now, I agree that there is always some uncertainty in science about impacts. But, the precautionary principle was never meant to stop a policy that has clear proven health benefits because of some unproven health concern. In fact, the precautionary principle has always been phrased in a way that says there is an obligation to act when there is a risk to health because of inaction. Clearly, if we do not fluoridate the water, there will be real pain and suffering from dental decay among children, and therefore, the course of action that best protects the public's health is fluoridation.

    If we used the argument that there is an unproven risk of harm to apply to all policies, we simply would not have any public policy whatsoever. We would not have public schools because there is a potential for kids to get the flu, or even worse get shot. And, certainly, we could not offer medical care because there is a potential for something to go wrong.

    I am a big fan of the precautionary principle myself, but I believe invoking the precautionary principle is inappropriate to stall a policy that is supported by every major health organization that has taken position.

  11. I would also point you specifically to the discussion in the paper you linked to above by Grandjean and Landrigan where they say specifically that fluoride has not been shown yet to be toxic to people. They mention three epidemiological studies. Two of them show lowered IQ at high levels, and one study actually shows increased IQ at levels consistent with water fluoridation in the U.S. But, they say all of these studies did not control for confounding factors. This evidence is inconsistent and flawed. More research is good, but there still is no credible evidence of harm due to water fluoridated at the recommended level.

  12. Mel, I am struggling with following your reasoning---could I ask you to ground your assessments in more detail?

    You say that it is very likely that confounding factors are distorting the results, and that high fluoride levels are correlated with lots of other things in the water which are known to effect cognitive development. I am unable to find any research supporting these claims. Could you help me finding this research?

    You say that the researchers phrased the results like "the meta-analysis could not eliminate the possibility that fluoride affected intelligence". Please show me where you find such a phrasing. I cannot find this phrasing consistent with the wordings in the Harvard review: "The results suggest that fluoride may be a developmental neurotoxicant that affects brain development at exposures much below those that can cause toxicity in adults (Grandjean 1982)", and "In conclusion, our results support the possibility of adverse effects of fluoride exposures on children’s neurodevelopment."

    You say that IQ is a poor measure of cognitive development, and that studies in industrial countries don't use IQ to assess cognitive development. Can you tell me how you came to this conclusion? IQ studies have been instrumental in realizing that certain chemicals are developmental neurotoxines at low concentrations. One example that deals with low-level exposure to lead is reference 22 in Grandjean's and Landrigan's Lancet review.

    I realize that I wasn't clear in my comments about the Ferguson study regarding the child's exposure to fluoride. The important point is that the brain development is particularly sensitive and easily disturbed before the child is born and up to six month's age, and if we are concerned about effects during this critical time period, studies that don't say anything about exposure during pregnancy and up to six month's age have limited value for proving safety. So yes, the pregnant mother's exposure to fluoride is indeed important and may influence the child's brain development. I will explain this in more detail below.

    You say that one study mentioned in Grandjean's and Landrigan's review shows increased IQ at levels consistent with U.S. water fluoridation. Could you tell me which study this is, and explain your conclusion in detail?

  13. Let me clarify what Grandjean's and Landrigan's precautionary approach is about and what it isn't about, and why the approach is important.

    Their precautionary approach is specifically about preventing disruptions to brain developent in children caused by neurotoxicants in the environment. As Grandjean and Landrigan have explained, the history has been that we expose the developing brain to chemicals until we find out any side effects. The problem with this wait-and-see approach is that proving that something is causing brain development disruptions is extremely difficult and takes a long time. If it turns out that we had a problem, we have already caused serious and permanent damage to a whole generation of children. What makes the problem even bigger is that children are exposed to a number of potential neurotoxins, and their combined interaction is unknown. That is one reason why we recommend pregnant mothers to minimize their overall exposure to the neurotoxic substances we are aware of.

    So in our context of health policies, the precautionary approach is not a general one where we demand that policies are proven safe before we implement them. The approach is particularly about protecting brain development, and tells us that we need to act immediately if we see signs of disturbances in brain development. "Early indications of a potential for serious toxic effects" as they write, should lead to strict regulation which could later be relaxed.

    As long as we do not take the precautionary approach seriously, we are dealing with the "silent pandemic", as Grandjean and Landrigan call it, where millions of children have impaired brain development that we could have prevented. The pandemic is silent because it is not apparent by any standard health statistics.

    So when we see indications that fluoride in the drinking water may cause brain development disruption, we should not wait for the proof, according to the approach. And the Harvard review has precisely pointed out that we have indications of a potential for serious toxic effects.

    You suggest that the precautionary approach cannot be used to stop policies with clear proven health benefits, or if there is a risk to health because of inaction. But we are trying to solve a dental health problem for which there are other successful approaches (just look at what Europe is doing). And the tradeoff you are suggesting means that if the evidence is true, children's brain development will be impaired for which there is no cure at all. Can you explain how it makes sense to ignore the precautionary approach here?

    Let me try and make my concern a bit more concrete, so we can discuss the various uncertainties we are facing. I will present two hypotheses, and we can discuss how they will explain the evidence we are seeing, and how likely the explanations are.

    The first hypothesis is one that I have tried to distill from your response. Feel free to adjust it or come up with some other hypothesis that you think is better.

    Hypothesis A:

    "The results of the Harvard review can be explained by confounding factors that were not taken into account, such as other things in the water including arsenic and lead."

    How does Hypothesis A explain the evidence, and how likely is the hypothesis?

    Although Hypothesis A in theory fits with the Harvard review, there is a problem with the consistency of the results of the underlying studies, and the following note from the review:

    "Large tracts of China have superficial fluoride-rich minerals with little, if any, likelihood of contamination by other neurotoxicants that would be associated with fluoride concentrations in drinking water. From the geographical distribution of the studies, it seems unlikely that fluoride-attributed neurotoxicity could be due to other water contaminants."

    This suggests that hypothesis A would have to find some other factors and explain why these are likely to correlate with the high fluoride in the drinking water in a consistent way.

  14. Hypothesis B:

    "Children exposed to fluoride during pregnancy and the first six months will have their brain development disturbed, resulting in lower cognitive function. The dose response is unknown, but it may be that exposure resulting from optimal water fluoridation can result in IQ reduction of about 0.5 to 1 point."

    In this hypothesis, there is a time window during which the brain development can be affected, and the following reasons support such a time window: 1) As the Harvard review points out, the placenta does not prevent fluoride from entering the fetus. 2) As the Lancet review points out, the blood-brain barrier is not fully developed until six months of age, which is important because once this barrier is in place, it reduces fluoride transfer to the brain. 3) In early development, the brain undergoes unique stages (see the Lancet review) and even if fluoride only disturbs one such stage, cognitive function will be irreversibly disturbed.

    The presence of this time window means that fluoride exposure after the first six months may have much lower effect on cognitive function.

    Let's see how hypothesis B holds up against the evidence we have.

    B can explain the result of the Harvard meta-analysis: it is not unreasonable to assume that the children being tested in the underlying studies lived in the same area when their mothers were pregnant.

    Do we have any evidence that can refute Hypothesis B? The Ferguson study is a candidate, since it shows no significant behavior rating between kids in fluoridated areas compared to non-fluoridated ones. However there are at least two reasons why this study could have missed the effect caused by the hypothesis: 1) It does not measure the important variables: exposure during pregnancy and the first six months. We do not know to what extent the children's mothers got exposed to fluoride, and we do not now if the infants were breast fed (which would have prevented almost all fluoride exposure), or if they were formula fed. Since these variables are unknown, the result may have been blurred enough that any significant difference would vanish. 2) The study's measure for cognitive function uses behavior assessments, and I know no research that shows that a lack of behavior disturbance implies a lack of cognitive function disturbance. (I would be interested in learning about such research.)

    I cannot see that Hypothesis A is less likely than Hypothesis B.

    Can you, Mel? If so, please explain in detail, and in particular, I would be interested in knowing how many times more likely you think Hypothesis A is over B. As I mentioned, pick some other variant for Hypothesis A if you prefer.

    Thank you!

  15. Correction: I meant to say that I cannot see that Hypothesis A is more likely than Hypothesis B.

  16. Magnus,

    Chinese water supplies are heavily polluted. You can just read this summary, to see the dramatic problems with water pollution:

    More than two thirds of water supplies are contaminated by human or industrial waste, and the World Bank has said water supply and pollution issues in China could have "catastrophic consequences for future generations."

    There are many potentially confounding factors in water supplies that weren't measured in the studies including contagious diseases, lead, arsenic, and iodine levels. That is one of the reasons why all the Chinese studies are very flawed. You simply can't make any real conclusions from studies that have not controlled for confounding factors.

    And in relation to the Fergusson study, they actually measured water intake among kids at birth, at 4 months, and then in one year increments up until age seven. And, they found that the kids drinking fluoridated water were rated slightly better by their teachers, but there wasn't enough of a difference to be statistically significant. This is a good quality study, looking at water fluoridated at the recommended level and showed a conclusion that contradicts any effect on intelligence.

    One good study in epidemiology is worth more than a thousand bad studies, and this study is way better quality than all the Chinese studies that you are bringing up.

    It is important that as a society, we rely on good quality research not poorly designed studies to make important public policy decisions. That is why all the trusted health authorities that have looked at fluoridation are supporting it.

    1. Mel, I doubt you have read a single one of the 36 or so Chinese and other nation original papers that looked at fluoride and IQ. Many of them were conducted in rural areas with no industry for miles around. Some of them explicitly stated there was industrial activity anywhere nearby, and others explicitly measured potential confounders such as lead, arsenic, and iodine. Choi et al analyzed a subset of the studies that excluded any where where arsenic or iodine was identified as a possible uncontrolled confounder. They found a very similar drop in IQ due to fluoride in this higher quality subset of studies.

      Choi et al explicitly stated that the remarkable CONSISTENCY of the findings, together with the diversity of the studies, is evidence that systematic bias (from confounding or any other bias) was unlikely to be occurring across the 26 of 27 studies which ALL found clear drops in IQ with fluoride exposure.

      The single New Zealand study you like to cite has some serious weaknesses. It is not sufficient evidence to dismiss the need for higher quality research to determine the actual dose-response between fluoride and neurotoxicity.

      Subjecting 200 million Americans to a chemical which the authoritative NRC 2006 report and one of the world's leading experts in developmental neurotoxicity, Dr. Grandjean, have both concluded needs to be tested further to see whether effects are occurring at the levels of exposure existing in the US, goes directly against the precautionary principle.

  17. Hi Mel,

    Yes, relying on good quality research to make important public policy decisions---that is precisely what I am advocating too.

    That is why I am asking you to seriously consider the results of not only the Fergusson study, but also the Harvard review and the Lancet review. Or do you consider only the Fergusson study being of high quality?

    It does not make sense to me to dismiss the Chinese fluoride studies because of water pollution. If we did that, why did the Harvard researchers bother doing a meta-analysis anyways based on these studies, strengthening the indications of fluoride being a neurotoxin? When you dismiss the Chinese studies, you also dismiss the Harvard review and its results.

    I gave you two scenarios that are more or less consistent with the results in the Harvard review and the Fergusson study. I cannot say that one is vastly more likely than the other and I would like to ask you to explain in more detail how the Fergusson study can rule out Hypothesis B. Please recall that I gave two independent reasons, and even if only one of them holds up, that is enough for Hypothesis B to "fly under the radar" of the Fergusson study. Until I understand you better here, I have to disagree that we can conclude from the Fergusson study that fluoride at the recommended level would contradict any effect on intelligence.

    Our main problem is: one of the scenarios I presented is that fluoride in the water disturbs brain development. I don't see that we have anything that can let us easily dismiss the precautionary approach motivated in the Lancet review.

  18. Hello again Mel,

    I just realized that there is a remarkable omission with the Fergusson study.

    We already discussed the variables that were taken into account, and the only one that attempted to estimate fluoride exposure was the number of years the children were living in the area that had water fluoridation.

    The study does not have any estimate of the total fluoride exposure, something which many of the IQ studies used in the Harvard review attempt to estimate by looking at dental fluorosis and urine levels.

    The data in the Fergusson study comes from Reference 6 ("Relationships between exposure to additional fluoride, social background and dental health in 7-year-old children" by Fergusson and Horwood), and that paper shows that total fluoride exposure is a complex variable to measure. Fergusson and Horwood noted that there was a negative correlation between the use of fluoride tablets and exposure to fluoridated water. Presumably, parents in the non-fluoridated area followed recommendations to use fluoride tablets to compensate for the lack of fluoride in the water.

    However, Fergusson's behavior study does not account for the fluoride exposure through tablets. By looking at the numbers in the Fergsson & Horwood study, one can see that there were 70 children that were given fluoride tablets for the full seven years, which means they got them even during their first year when brain development is extra sensitive.

    The omission of tablet exposure means that it is impossible to draw any conclusions from the Fergussson study regarding potential brain development disturbance caused by water fluoridation.

    (On a more general note, the Fergusson study has another limitation in our context in that it does not account for any safety factor---see also Michael Connett's comment where he suggests a 10X factor.)

    Also, Mel, since you suggested a theory around the fact that the Chinese studies did not account for confounding factors such as arsenic, lead and iodine. Look at the study published in 2012 made on school children in India: This study took into account a number of variables and looked at levels of lead, arsenic, iodine and fluoride in urine, in addition to fluoride in the water. Iodine, arsenic and lead differences were not significant, but urine fluoride was. Intelligence differences were significant when measured against fluoride levels in the water, and also when measured against fluoride in the urine.

    It just gets more and more difficult for me to fit your hypothesis with the data.

    I would love to hear your thoughts here, Mel.

  19. Mel,

    I would like to ask you a favor.

    Two weeks ago, I sent an email to Damien Fair and asked about his assessments of the Harvard IQ review. Since the IQ review and the questions we have been discussing is in Damien's area of expertise and given Damien's role at Upstream, I would very much appreciate if he could speak up here.

    However, I haven't heard anything back from Damien. Can you see if he is interested?

    Thank you!

  20. I was heavily involved in the campaign against the pro-fluoride camp here in Wichita, at the time that slanderous article was published in the Wichita Eagle, which claimed the Harvard scientists denounced their own study.

    Since then, the Harvard scientists involved in that study have denounced the Wichita Eagle's article, essentially calling out the articles author, Dion Lefler, as the liar we local Wichitans all knew him to be.

    See the story here

    And the direct comments of Harvard scientist, Philippe Grandjean, MD here:

    1. Thanks for the pointers regarding the Wichita Eagle article, Jonathan. As you can see in my other post, people who should know better continue to refer to this unfortunate misrepresentation.