The Case Against Fluoride

by Paul Connett

  Whoever was right about the effect of fluoridation on the cancer rate, the furor generated by Yiamouyiannis and Burk’s work led a congressional committee to call for animal studies to determine whether fluoride caused cancer under laboratory conditions. The U. S. National Toxicology Program (NTP), a division of the U. S. Public Health Service (PHS), commissioned Battelle Memorial Institute’s labs to do these studies. Oral, liver, and bone cancer received special attention. The results, which should have been completed and released in 1980, were not released until 1990. When they were released, they led to much press attention, including a cover story and commentary in Newsweek. 19

  It was clear that the NTP results threatened the fluoridation program. We will first examine the NTP study and then see how the PHS, now the DHHS, attempted to contain the issue, with the help of the National Cancer Institute (NCI).

  NTP Animal Study

  The Battelle lab researchers, contracted by the NTP to do the animal study, found a small but statistically significant dose-related increase in osteosarcoma in male (but not female) rats exposed to fluoride. 20 After a government peer review panel downgraded one of the osteosarcomas, this finding was classified as “equivocal evidence of cancer” by Bucher et al. , whose study was finally published in 1991. 21, 22

  In the spring of 1990, with the NTP’s findings attracting media attention, Procter & Gamble, a manufacturer of fluoridated toothpaste, released the findings of its own rat study of fluoride and cancer, which the company had conducted in 1981–1983 and which claimed not to have found this relationship in mice or rats. 23 While Procter & Gamble’s study found several bone tumors in the fluoride-treated animals (versus none in the controls), the results do not achieve statistical significance and were thereby dismissed as random. According to Procter & Gamble’s published report, “All bone neoplasms were considered to be incidental and spontaneous and not related to fluoride treatment, because of their low incidence and random distribution. ”24

  An FDA review of this study appeared in appendix D of the Department of Health and Human Services (DHHS) report of 1991. In its analysis, the FDA identified two additional osteosarcomas in the fluoride-treated rats that were not identified in Procter & Gamble’s published report. According to the FDA, “The adequacy of the gross examination at necropsy was questioned based upon the rat tumors that were not identified by the contract [Procter &Gamble] laboratory. ”25

  However, even with the newly identified osteosarcomas, the FDA noted that the incidence of bone tumors in the Procter & Gamble study still did not achieve statistical significance. The FDA thereby concurred with Procter &Gamble that the bone tumors are incidental. 26

  Other Cancers in the NTP Study

  As well as osteosarcoma, the original Battelle study reported an increase in liver and oral cancers and thyroid follicular cell tumors. However, the same government review panel mentioned above downgraded all the non-bone cancers to a less severe classification via a questionable rationale. 27 When, in May 1990, Dr. William Marcus, chief toxicologist of the Office of Water at the EPA, pointed this out, he was fired. (He was later reinstated, with full pay and compensation, after Labor Secretary Robert Reich ratified a December 3, 1992, ruling by administrative law judge David Clarke that the EPA “retaliated” against Marcus by firing him for scientific reports that recommended removing fluoride from drinking water. )28

  Dr. Marcus pointed out that one of the cancers downgraded was a rare form of liver cancer called hepatocholangiocarcinoma. The peer reviewers examining the slides prepared for microscopic examination claimed that it was not a case of hepatocholangiocarcinoma. However, Dr. Melvin Reuber, an independent pathologist formerly with the National Cancer Institute and the first to describe this rare form of liver cancer, concurred with the pathologist at Battelle Memorial that the downgraded case was indeed a case of hepatocholangiocarcinoma. 29 In light of the importance of this study, in 2000 the union representing professionals working at the EPA headquarters in Washington, D. C. , requested that Congress establish an independent review panel to reexamine those cancer slides. 30 That has so far not been done. The union had previously objected to the way the EPA determined the 1986 safe drinking water standard for fluoride31, 32 (see chapter 20) and also objected to the dismissal of Dr. Marcus for raising red flags on the peer review panel’s downgrading of cancer in the NTP study. 33

  The NTP findings prompted the NCI to review osteosarcoma rates in fluoridated communities in the United States (which Taves had recommended thirteen years earlier). 34 To do that, Hoover et al. used the Surveillance, Epidemiology and End Results (SEER) registries. Both the NTP results and the NCI survey were hastily published in a review by the Department of Health and Human Services titled Review of Fluoride: Benefits and Risks. 35 Thus, Hoover et al. ’s studies were presented to the public before they had been published in a peer-reviewed journal. 36, 37 Publication in a peer-reviewed journal is widely considered by scientists to be essential to acceptance of research findings. Peer review would have been especially important in this case, as the agency publishing this report (DHHS) has a policy of promoting fluoridation.

  Hoover et al. Survey of the SEER Registries

  In 1991 Hoover et al. examined the nine SEER cancer registries (registries that cover about 10 percent of the U. S. population) for bone cancer and found a greater incidence of osteosarcoma in young males (not young females) in fluoridated versus non-fluoridated counties. 38 However, the same authors, using a subset of the data, claimed that the greater incidence was unrelated to duration of exposure and discounted the original finding. 39

  Today, more credence is given to Hoover et al. ’s first finding than their second. This is largely because, by the time the NCI authors had used a subset of the data and divided it among four different durations of exposure, there were so few cases left in each grouping that the study lacked any statistical power. 40–43

  However, Hoover’s second finding certainly helped fluoridation promoters to allay concern about this issue. Promoters included the authors of the DHHS report, who continued to argue that fluoridation was safe and effective. Here is what the DHHS authors had to say in their abstract about the animal studies (including the NTP study) as well as the human surveys on the question of a possible connection between fluoridation and cancer:

  Taken together, the only two methodologically acceptable animal studies [the NTP study and the earlier study by Procter & Gamble] available at this time fail to establish an association between fluoride and cancer. In humans, optimal fluoridation of drinking water does not pose a detectable cancer risk as evidenced by extensive human epidemiological data reported to date, including new epidemiological studies [the reviews of the SEER data by Hoover et al. ] prepared for this report. No trends in cancer risk, including the risk of osteosarcoma, were attributed to the introduction and duration of water fluoridation. 44

  Somehow the osteosarcoma finding in male rats reported in the NTP study and the designation by peer reviewers of “equivocal evidence” of cancer have been lost from public view. Nor does the DHHS mention Hoover et al. ’s initial finding of a greater incidence in osteosarcoma in young males in fluoridated communities compared to non-fluoridated communities, which they later discounted. Someone reading only the abstract of the DHHS report would get a very different view of the issue than someone who studied the details of both the animal and the human studies.

  So to what extent was this DHHS report rushed through in an effort to protect the fluoridation program from this latest threat? An indication of the mind-set of those who set up this review is given in the last paragraph of the February 16, 1990, letter Dr. James Mason (assistant secretary for health and acting surgeon general) sent to the federal employees who made up the review panel: “Given the tangible public health benefits of fluorides in reducing tooth decay, the rarity of the tumor type in humans that is implicated by the study, and the preliminary nature of the NTP findings, our current policy supporting fluoridation must b
e maintained until your review is finished. ”45

  McGuire et al. 1991

  Two months after the DHHS review was published, a small study by S. M. McGuire et al. came out in the April 1991 issue of the Journal of the American Dental Association (JADA). 46 One of the coauthors of this study was Professor Chester Douglass, chair of the department of oral health policy and epidemiology at the Harvard School of Dental Medicine (see more below and under “Bassin 2001”). Even though this paper was a very small, preliminary study, it was given the full treatment by JADA. It was made the cover story, complete with a cover picture showing mountains and a lake seen through a huge glass of water.

  It is clear that McGuire, Douglass, and the other authors of this study knew that a finding that fluoride might cause osteosarcoma would threaten the fluoridation program. Nor do they hide their concerns about such an eventuality, as the following quotes make clear:

  An incorrect inference implicating systemic fluoride carcinogenicity and its removal from our water systems would be detrimental to the oral health of most Americans, particularly those who cannot afford to pay for increasingly expensive restorative dental care. . . 47

  Because of its strengthening action, fluoride has been widely accepted as the responsible agent for the dramatic declines in the tooth decay rates of U. S. children and adolescents. . . A disruption in the delivery of fluoride through municipal water systems would increase decay rates over time. . . Linking of fluoride ingestion and cancer initiation could result in a large-scale defluoridation of municipal water systems under the Delaney clause. 48

  Luckily for those who believed in the fluoridation program, the authors did not find that fluoridation was associated with an increase in osteosarcoma. In fact, they found that the very opposite might be the case, stating that “fluoridation at recommended levels may provide a protective effect against the formation of osteosarcoma” [emphasis added]. 49 This speculative finding allowed the authors to reach the conclusion that they (and the JADA editors and the ADA) clearly wanted out of this study: “Given present knowledge, every effort should be made to continue the practice of fluoridating community water supplies. ”50

  Chester Douglass, coauthor of the JADA article, received a sizable grant from the National Institute of Environmental Heath Sciences (NIEHS) to continue researching the fluoride-osteosarcoma issue after the article was published. It is surprising that the NIEHS would choose any dental researcher to oversee research on a life-or-death issue that might be related to the fluoridation program, let alone someone who had already expressed his understanding that a positive finding on the fluoridation-osteosarcoma connection would sabotage the practice that he favored so strongly. Douglass’s further involvement in this issue is discussed under “Bassin 2001” and subsequent sections below.

  Cohn 1992

  In 1992, Perry Cohn, working for the New Jersey Health Department, reported a significant increase in osteosarcoma in young males in fluoridated communities in three New Jersey counties—again, not in young females. 51 Most significantly, Cohn suggested that there might be a time frame when young boys are particularly vulnerable to fluoride’s carcinogenic effect:

  If rapidly growing bone in adolescent males is most susceptible to the development of osteosarcomas (Glass and Fraumeni, 1970), it is possible that fluoride acts as a cancer promoter during a narrow window of susceptibility. The interplay of hormonal influences and the intensity of the growth spurts may be potent influences. Since fluoride is toxic to cells and a variety of enzymes at high concentrations (reviewed by Kaminsky et al. , 1990; and Department of Health and Human Services, 1991), it may exert tumor-promoting effects in the osteoblast cell microenvironment during bone deposition. Genetic predisposition may also play a role. 52–55 [emphasis added]

  Other Osteosarcoma Studies

  Other epidemiological studies of various sizes and quality have failed to find a relationship between fluoridation and osteosarcoma. 56–60 A full review of these and other studies on osteosarcoma is included in submissions to the National Research Council by the Fluoride Action Network (FAN). 61, 62 FAN’s submissions were triggered by the discovery of a DMD thesis by Elise Bassin, discussed in the next section.

  Bassin 2001

  Dr. Elise Bassin investigated a possible relationship between osteosarcoma and exposure to fluoride as part of her doctorate in dental medicine thesis at the Harvard Dental School. 63 Suspecting a possible window of vulnerability for this problem, as Cohn had conjectured in 1992, 64 Bassin examined osteosarcoma rates as a function of the age at which boys were exposed to fluoride.

  In a matched case-control study, Bassin reported what she described as a “robust finding” that young boys exposed to fluoride in their sixth to eighth years (which corresponds to the mid-childhood growth spurt) had a fivefold-to sevenfold increased risk of contracting osteosarcoma by the age of twenty. 65

  It is extraordinary that, after Bassin’s thesis was successfully defended in 2001, it was not followed up with a swift publication of her results, or any kind of statement made to warn the scientific community and the public about her findings. After all, if she was correct, a chemical that was given daily to millions of Americans in their drinking water might actually be killing people. For several years Bassin’s thesis disappeared from view.

  Professor Chester Douglass, coauthor of the 1991 McGuire et al. study, was Bassin’s research sponsor and signed off on her thesis. Clearly, he knew the very serious implications of her findings for the future of fluoridation, as indicated in our discussion in “McGuire et al. 1991” above. 66 However, it appears that even though he was given several opportunities to do so over the next three to four years, he failed to warn his colleagues about Bassin’s findings. The first opportunity occurred when he gave a presentation before the British Fluoridation Society (BFS) in 2002. He told the audience that his studies found no relationship between fluoride and osteosarcoma, but somehow he failed to mention his own graduate student’s findings. This is how the BFS described his report in its piece on fluoridation titled “One in a Million, ” which is still posted on its Web site today: “Professor Chester Douglass of Harvard University presented preliminary results, as yet unpublished, from that and a separate National Cancer Institute study by Hoover et al. , at a symposium held at the Royal College of Physicians, London, in November 2002. These two large case-control studies showed no association between fluoride exposure and osteosarcoma. ”67 Eight years later (as of July 2010) these promised studies have still not been published (see the discussion in the next section) and meanwhile, in the period 2006 to 2010, the BFS has not felt the need to mention the existence of Bassin’s study.

  Douglass’s second opportunity to mention Bassin’s findings came when he was invited to send a summary of his work to the NRC panel (which published its report in March 2006). 68 He again asserted that his work showed no significant association between fluoride and osteosarcoma. In his short statement he did include a citation to Bassin’s thesis as a footnote but did not indicate to the NRC that her findings contradicted what he had said. 69 He sent a similar statement to his funders at the NIEHS.

  Bassin’s Thesis Discovered

  Eventually, Bassin’s thesis surfaced in the Harvard Medical School Rare Books Room, where Michael Connett accessed it in January 2005. The resulting public release of this material triggered a demand by the Environmental Working Group (EWG) for an official enquiry into Douglass’s behavior by the NIEHS, the body that had funded this work. This led to a great deal of press attention, including a lengthy article in the Wall Street Journal. 70 The NIEHS passed on the request for an enquiry to Harvard. After a year, Harvard produced a short statement declaring that Douglass did not “intentionally” hide Bassin’s findings. 71 Harvard has refused to provide any arguments or explanations supporting this finding, despite repeated requests from alumni, other U. S. citizens, and even members of Congress.

  Bassin’s findings were published in May 2006 in the journal C
ancer Causes and Control. 72 However, the same issue of the journal published a letter coauthored by Douglass downplaying the significance of her findings. 73 It is interesting to contrast Douglass’s failure to warn the public of Bassin’s findings between 2001 and 2005 with his alacrity in claiming that her findings might be “premature” on the same day that her article appeared in press.

  In their letter, Douglass and Kaumudi Joshipura pointed out that Bassin’s findings were based on a subset of a larger cohort and claimed that the larger cohort did not support her thesis. However, they provided no evidence that her methodology had been applied to the larger cohort, nor is it clear that it has ever been so applied. The letter further claimed that Douglass’s larger study (to be coauthored by Robert Hoover, mentioned previously in connection with his critique of the Yiamouyiannis and Burk study and the NCI review of the SEER cancer registries) was “currently being prepared for publication. ” The NRC report had cited a January 3, 2006, communication from Douglass that their study “was expected to be reported in the Summer 2006. ”74 It is now over four years since Douglass made that promise, and the study has not yet been published. Douglass has since retired from his position at Harvard.