Flawed comparison of cancer mortality between fluoridated and non-fluoridated U.S. cities doesn’t provide sufficient evidence that fluoride causes cancer

Scientific evidence has demonstrated that community water fluoridation is an effective means of preventing tooth decay, with the U.S. Centers for Disease Control and Prevention hailing it as one of the greatest public health achievements. However, water fluoridation is opposed by some who believe that it causes many negative health effects. Common claims revolve around the effect of water fluoridation on child development and IQ, which Health Feedback covered in previous reviews and found such claims to be unsubstantiated by reliable evidence.

Another popular claim among opponents of water fluoridation is that fluoride causes cancer. Research by biochemists Dean Burk and John Yiamouyiannis, published in papers during the 1970s, is commonly cited as evidence for this claim. This Facebook video is one example in which Burk can be found talking about this research. Copies of this video, which shows part of an interview in the Dutch television program “Hier en Nu” dating to the seventies, can also be found on YouTube.

In brief, Burk asserted that cancer deaths spiked in the ten largest fluoridated cities in the U.S. compared to the ten largest non-fluoridated cities, and that before water fluoridation was introduced, cancer deaths were similar between the two groups. He called this “one of the most conclusive bits of scientific and biological research” that fluoride killed people through cancer.

However, the evidence provided by Burk is highly flawed, and his claim that fluoride causes cancer is unsupported by reliable evidence, as we will explain below.

Burk and Yiamouyiannis’ comparison of U.S. cities doesn’t compare like with like, fails to account for confounding factors

This study on fluoridated and non-fluoridated cities was among the research examined by the Working Party on Fluoridation of Water and Cancer, a group convened by the U.K. Department of Health and Social Security and led by professor of social medicine George Knox. Among this group were epidemiology experts from the Institute of Cancer Research and the Medical Research Council.

The Knox Report, which was published in 1985 and looked at more than a hundred published papers on the subject, pointed out significant problems with Burk and Yiamouyiannis’ conclusions. A detailed discussion of the pair’s research can be found in Chapter 3 of the report. Notably, the pair failed to ensure that the populations they were comparing were similar to each other and whether factors apart from water fluoridation could account for the higher cancer mortality rates in the fluoridated cities.

The group commented that:

“The main problem in interpreting the results from simple comparisons of cancer mortality in different areas is whether the groups being considered are similar enough to bear comparison one with another. If the groups differ in features other than fluoridation status, then it would be necessary to ask whether any of those features provide a more likely explanation of the observed differences in cancer mortality.”

The report found that the fluoridated cities included tended to be situated to the east of the country and to have been established longer than non-fluoridated cities, and that this is likely to contribute to socioeconomic differences that may influence cancer risk, independent of fluoridation.

It delved into more detail on this in Chapter 9, explaining that fluoridation in the U.S. began earliest in areas that were relatively disadvantaged in socioeconomic terms and where the elderly population was growing particularly fast. These areas therefore had crude cancer mortality rates that rose faster than other areas that didn’t have the same features.

Moreover, in one comparison, Yiamouyiannis excluded three non-fluoridated cities with higher cancer rates as he deemed them “aberrant” owing to waterborne contaminants that caused the higher cancer rate. However, the report stated that Yiamouyiannis didn’t provide evidence to support this assumption nor did he appear to have made any effort to determine whether this was true. Simply put, this appeared to be an instance of cherry-picking. From the report:

“It is clear that the exclusion of three ‘aberrant’ cities has no adequate basis; it is an example of a well-recognised error of method which is certain to alter the results in the direction of the hypothesis supported by Yiamouyiannis.

The Working Party concluded that:

“there is no substantiated evidence from studies of human populations that fluoride or fluoridation causes cancer, or increases mortality from cancer, whether for cancer as a whole or for cancer at individual sites.”

It also concluded that studies which showed the opposite contained “elementary errors”, particularly in failing to use standard, tried-and-true approaches that would allow the study to properly account for “important demographic and socioeconomic differences between populations”.

More recent evaluations haven’t established an association between water fluoridation and cancer risk, but the quality of the evidence so far is variable

In the U.K., the National Health Service Centre for Reviews and Dissemination published a systematic review of public water fluoridation in 2000, which included an evaluation of more than 20 studies examining a potential association between cancer incidence and water fluoridation.

The review concluded that “Overall, from the research evidence presented no association was detected between water fluoridation and mortality from any cancer, or from bone or thyroid cancers specifically.” However, the review found that the majority of the studies were from the “lowest level of evidence […] with the highest risk of bias”.

The relatively low quality of many studies on this subject also affected the conclusions of the National Research Council, which is part of the U.S. National Academies. In a report published in 2006, the committee that evaluated the evidence stated that “the evidence on the potential of fluoride to initiate or promote cancers, particularly of the bone, is tentative and mixed.”

Therefore, further research with better-designed studies is needed in order to better understand the relationship between fluoride levels and cancer risk.

Because fluoride tends to accumulate in tissues that are high in calcium, special attention has been paid to studying a potential relationship between water fluoridation and bone cancer.

Concerns over bone cancer risk grew after a 1990 study by the National Toxicology Program (NTP), run by the U.S. Department of Health and Human Services, which reported higher rates of osteosarcoma, a relatively rare type of cancer affecting the bones, in male rats fed with high doses of fluoride through the drinking water. However, the NTP failed to replicate the results in a two-year follow-up study, which found no association between fluoride and osteosarcoma in either male or female rats. This raises questions about the reliability of the first result.

Partial results from a study spanning more than ten years were published by researchers at Harvard, which reported an association between fluoride in drinking water and osteosarcoma in boys, but not in girls. The study revived concerns over the safety of water fluoridation programs. However, other researchers cautioned that the study was still ongoing, and that these results were based on cases between 1989 and 1992, whereas cases between 1993 and 2000 weren’t included.

A study that included the later cases was published in 2011. It reported no association between fluoride and osteosarcoma risk. This study was done by comparing fluoride levels in bone samples taken close to the tumor site with that of samples taken from a distant site in the same patient, or with bone samples taken from patients with a different type of bone tumor^[1].

This approach of measuring fluoride intake is advantageous, since it more accurately reflects an individual’s fluoride exposure than measuring fluoride exposure based on estimated drinking water intake.

However, the drawback here is that since the control samples were taken from patients who either had a different type of bone cancer or had benign (non-cancerous) tumors, the study would be unable to detect an association between fluoride and bone cancers in general, although the authors didn’t consider this to be of concern, stating that “There is no published evidence of such an association”.

A 2014 study in the U.K. that examined more than 3,000 osteosarcoma and Ewing sarcoma cases found no association between drinking water fluoride levels and either cancer^[2].

In a 2019 publication, the World Health Organization stated that “there is no evidence in recent and peer-reviewed publications that fluoride levels in drinking water aimed at controlling dental caries is associated with increased risk of bone cancers in humans”.

Conclusion

In summary, there is no reliable evidence showing that water fluoridation increases cancer risk. Burk’s claim is based on research that didn’t account for confounding factors, and therefore doesn’t provide a sound basis for his claim. Studies that do account for confounding factors didn’t detect an association between fluoridation and cancer. Better-designed studies on how fluoride levels might influence cancer risk is needed in order to improve our understanding of fluoride’s effects on health, but at the moment, there’s no evidence indicating that the amount of fluoride present in drinking water is harmful or linked to cancer.

READ MORE

Burk also promoted the use of laetrile, a synthetic form of amygdalin also incorrectly called “vitamin B17”, for treating cancer. However, there’s no evidence that laetrile is effective against cancer and it can even be harmful by inducing cyanide poisoning. His colleague Yiamouyiannis, who was diagnosed with colorectal cancer in 1999, sought out laetrile and other alternative cancer treatments in lieu of conventional treatments like surgery and chemotherapy. He died the following year from cancer.

REFERENCES

• 1 – Kim et al. (2011) An Assessment of Bone Fluoride and Osteosarcoma. Journal of Dental Research.
• 2 – Blakey et al. (2014) Is fluoride a risk factor for bone cancer? Small area analysis of osteosarcoma and Ewing sarcoma diagnosed among 0–49-year-olds in Great Britain, 1980–2005. International Journal of Epidemiology.