IFIN BULLETIN #152: The York Review: The Science and the Spin.
Oct 6, 2000.
By now most of you will have received the news that the York Review is out - probably in the form of sidewinder missiles in your local and national papers. The timing couldn't be better from the promoters point of view...
...a month to go before the Nov 7 elections which have fluoridation on the battot in many cities -San Antonio, Texas, Spokane ,Washington, Ithaca, NY, Salt Lake City, Ut, and many more. We have already been told that the story has made major headlines in many local papers.
...just in time for the "Fluoridation Forum" in Ireland to announce that there is no problem with fluoridation etc etc
Not content to deliver the York Review, the editors of the British Medical Jounral chose to release a study on hip fracture ( Phipps et al) which claims that fluoridation actually REDUCES hip fracture. I will be covering this article in another IFIN bulletin. It is only one of 18 studies done on hip fracture since 1990, and 10 show an increasae in hip fracture and 8 do not. Phipps study is not the final word on this very serious subject and to intimate that it is would be reckless in my view.. She did not estimate total fluoride exposure or measure fluoride levels in the bones. Also in her study she found an increase in wrist fracture (comparable to the decrease in hip fracture) and this increase is hardly commented upon at all simply because it misses significance by the smallest of whiskers, after controlling for 13 variables!
Both articles can easily accessed on the internet. The address of the British Medical Journal is http://www.bmj.com and the York Review is entitled "Systematic Review of Water Fluoridation" by Marian McDonagh et al. The full York report is out on the York University site and can be accessed at http://www.york.ac.uk/inst/crd/fluores.htm.
As expected certain sections of the media are trumpeting the York Review as showing that fluoridation is effective and safe. Nothing could be further from the truth. If one goes beyond the spin in the accompanying press release "The Final Word on Fluoride" from the NHS Centre for Reviews & Dissemination, University of York (CRD) and actually visits the scientific detail of the report, there is - even with all its many limitations -a clear evidence that water fluoridation has been a massive, and potentially dangerous, failure.
You can find my full critique of the final draft of the York Review (and also those by Albert Burgstahler, Bruce Spittle and Dr. Z) on our webapge at http://www.fluoridealert.org and sadly they ignored a great deal of my scientific input. However, the time and effort I put into this analysis iwill not be wasted. That is assuming we can find intelligent scientists and lay people with an open mind and willing to spend more time on this issue than they would put inot a newspaper article.
The report covers four areas: tooth decay; dental fluorosis; bone damage and other health effects. In short, their analysis of tooth decay is very, very weak. Their analysis of dental fluorosis is strong and underlines the massive failure of the fluoridation program to meet one of its key goals. Their analysis of bone damage is seriously flawed and failed to address key points raised by critics. Their analysis of other health effects is far too limited and superficial.
As far as bone damage and other health effects are concerned the report suffered (indeed their whole "systematic" methodology) from an obsession with statistical significance and human epidemiological studies as opposed to a weight of evidence approach which takes into account ALL the available data and evidence which can be gleaned from biochemical, animal and human studies. Their approach seemed more legalistic than scientific: fine perhaps for preparation for a lawsuit to defend against claims of health damage but totally inadequate for the purpose of helping to shape wise public policy in a matter in which it is contemplated to expose a whole population to a known enzyme poison which accumulates in the bone and possibly other calcifying tissues like the pineal gland. With such self-imposed limitations it is preposterous to bill this report as "The Final Word on Fluoride". It is only one small slice of the scientific pie, and a flawed slice at that.
In other words, their "systematic approach" and beautifully presented Forest plots are elegant summaries of data but they are no substitute for a careful scientific analysis of the contents of the different epidemiological studies, coupled with evidence available from other studies. My hunch is that many of the York Team have only a poor understanding of biochemistry and toxicology and very naieve when it comes to the political forces that have manipulated them.
TOOTH DECAY. They admit that "water fluoridation does appear to reduce the incidence of caries (tooth decay), but it is smaller reduction than previously reported". Later their press release indicates that they found "an average 15% reduction in tooth decay". 15% is indeed a much lower percentage than that claimed by ardent promoters of fluoridation (40-60%), however hold the applause because they then make the extraordinary claim that this amounts to "an average 2.25 less decayed missing and filled primary/permanent teeth amongst children living in fluoridated areas compared to non-fluoridated areas". This latter figure cannot be correct - the average DMFTs for 12 year olds in many non-fluoridated countries is les than 2 - and on inspection of their study method we find that they have made an extremely silly mistake. They have averaged the DMFT differences from a number of different studies performed at different times over a long period (1961-1997). It has been well established that over this time period tooth decay has been declining in both fluoridated and non-fluoridated communities and countries, for a variety of reasons, and thus to average differences in DMFT determined at different times does not make sense. For reasons, I do not understand they chose to insist on only longitudinal studies (studies on the same communites at two different time points), and not studies on teeth in different communities and countries at the same time in history. By choosing longitudinal over cross-sectional studies they managed to exclude some of the largest and best surveys ever conducted. For example they excluded the NIDR study of 35,000 children in 84 communities in the US (Yiamouyiannis, 1990 and Brunelle and Carlos, 1990) which found practically no difference in DMFTs in fluoridated and non-flfuoridated communities and only an 18% difference in DMFS, which is about one half of a tooth surface; Gray's studies in Canada, which showed less tooth decay in proivinces in Canada with less % fluoridation; Colquhoun's studies in New Zealand of 49,000 children (1987, 1994) which showed little difference in DMFTs for 5 year old and 12 year olds; Teotia and Teotia (1994) who looked at over 400,000 children in India and found that tooth decay went up as fluoride concentration went up and went sddown whne calcium went up. Other key studies by Diesendorf (1986) and de Liefde (1998) were also ignored. These authors showed that tooth decay continues to go down even after the maximum benefits atttributable to fluoride have been achieved.
It will be extremely interesting to hear how they defend the 2.25 difference in DMFT.
DENTAL FLUOROSIS. In the press release they state that the reduction in tooth decay "comes at the expense of an increase in the prevalence of fluorosis (mottled teeth)." Later they give a partial quantification of this when they state that "At a fluoride level of 1 part per million, an estimated 12.5% of exposed people would have fluorosis that they would find aesthetically concerning, although fluoride from other sources may also be playing a part".
It is here that we see the "spin" at work. In the BMJ article the actual figures are stated as follows: "the pooled estimate of the prevalence of fluorosis at a water fluoride concentration of 1.0 ppm was 48% (95% confidence inrterval 40% to 57%) and for fluorosis of aesthetic concern 12.5% (7.0% to 21.5%). There was, however, considerable heterogeniety between results of different studies."
The fluorosis of "aesthetic concern" is a PR term for the moderate and serious classifications of dental fluorosis. The mottling covers over 50% of the tooth surfaces and get colored orange brown and black and leads to pitting. In the US it can cost $1000 to crown a tooth to conceal this damage to the enamel.
To put these numbers of 48% total fluorosis (nearly one in two children) and 12.5% moderate and severe fluorosis (one in eight children) in perspective, the early promoters of fluordation in the US set two goals for their program a) to reduce dental decay and b) hold dental fluorosis to 10% of the children in its MILDEST FORM. In other words, the fluoridation program has MASSIVELY failed on one of its two key goals.
To partly excuse this by saying that "fluoride from other sources may also be playing a part" is unacceptable. Those who advocate putting fluoride into the water are not getting fluoride out of toothpaste, or taking steps to minimize the multiplier effect when fluoridated water is used to prepare processed food and beverages. Those that promote fluoridation know that they are going to INCREASE dental fluorosis to even more unacceptable levels than those causedd by swallowing fluoride from other sources.
Furthermore, because of the authors preoccupation with statistical manipulations they fail to acknowledge what dental fluorosis actually is and what it is telling us. Pam DenBesten (1999) has shown that dental fluorosis is caused by fluoride poisoning an enzyme which is involved in laying down the tooth enamel. This enzyme is supposed to remove the last little bit of protein from between the mineral prisms before they fuse to form the smooth enamel. In the presence of fluoride this enzyme does not work properly and little bits of protein are left in place and hence the mottling of the enamel. What dental fluorosis is telling us is that a child has been overdosed on fluoride and what it is warning us is that other even more important enzymes may be being poisoned as well, like the enzymes in the pineal gland (Luke, 1998).
It is very important that the British government and other decision makers elsewhere understand these warning signals. It is very sad that they have been partially concealed by "spin".
BONE DAMAGE. This area is another area where the unwillingness of the York team to take a more holistic approach to scientific evidence becomes manifest. Much of the momentum to look at a possible association between water fluoridation and increased hip fracture in the elderly came from trials with osteoporoiss patients. In the late eighties patients with osteoporosis were given high doses of fluoride to increase the hardness of their bones - their bone mineral density - under the notion that the harder their bones became the stronger they would become. However, even though their bone mineral density was increased, they became more brittle especially when exposed to torsional stress. The treatment may have lowered the number of compression fractures of the vertebrae but it led to an INCREASE not a decrease in hip fracture (Heglund and Gallagher, 1988 and Riggs, 1990).
Since 1990, there have been 18 studies (4 unpublished) on hip fracture that have been reported. 10 show an association between water fluoridation and increased hip fracture in the elderly and 8 do not. Despite the critique I sent ot the York team, they failed to present the postive association identified in two of the these studies ( Kurtio, 1999 and Li, 1999) This is particularly serious in the case of Li et al (1999) for these authors appear to have picked up a dose response relationship between fluoride levels as they rise from 1 ppm to 4 ppm and over. I assume the reason they reported Li as showing no association is because they were only interested in comparing the community at 1 ppm and the community with low fluoride levels. However, this is inconsistent with the use of the Sowers data. Again, their limited use of this important data serves the interests of those who want to believe that increased fluoride exposure does not increase hip fracture, but it does not serve honest science or the public interest very well. This is especially so, since their commentary does not make clear the associations that both Kurtio and Li actually found. Only those who pursue the fine print in the tables or in the original articles find this out.
I was also particularly disappointed that they continued to present all the bone data together, when there are very plausible biological and physical reasons why different bones will respond differently to increasing fluoride levels.
Hip fracture in the elderly is a very serious matter. One in four are dead within one year of the fracture. 50% never regain an independent existence. In the US it costs about $10 billion a year. Thus this issue demands more than the balancing out of positive and negative studies. This is precisely where a weight of evidence approach is needed. We know that at high exposures ( treatment of osteoprposis patients) fluoride makes hip bones more brittle. Secondly, we have what appears to be a dose response relation with hip fracture and levels of fluoride above 1 ppm. Moreover, we know the safety margin for this end point is not very high (less than 4, and probably less than 2) and we also know that it is the dose which is important from all sources, and we know that once fluoride is in the water we cannot control the doses and finally, people will be exposed for a whole lifetime.
So the authors do a serious disservice to the British taxpayers who paid for this study when they cavalierly state that "There was no clear evidence of other potential health adverse effects". In my view, it wasn't clear because they didn't look in the right places and they didn't look in the appropriate way.
While the York Team makes it clear (although the press has not) that the vast number of studies on fluoridation are poorly designed (especially the ones that launched fluoridation in the 40's - PC) they should have been much clearer about the limitations of their own analysis. They appear to be allowing their report to be portrayed as the "scientific analysis" as opposed to the "ethical and environmental analysis", when in actual fact it is only a partial scientific analysis and a fairly weak one at that.