Public perceptions and scientific evidence for perceived harms/risks of community water fluoridation: an examination of online comments pertaining to fluoridation cessation in Calgary in 2011.

Podgorny, Paulina C. ; McLaren, Lindsay

Community water fluoridation refers to the controlled addition of industrial-grade fluoride compounds to public water supplies with the goal of preventing dental decay. (1) In Canada, the optimal fluoride concentration to promote dental health is 0.7 mg/L, with a maximum acceptable concentration (MAC) of 1.5 mg/L. (2) The evidence base for a beneficial effect of fluoride on tooth decay is substantial. In addition to its antibacterial activity against cariogenic bacteria, fluoride exerts a positive effect on tooth remineralization through its absorption into the surface of enamel crystals as it flows over the teeth, protecting against dissolution by bacterial acids. (3) Research also supports the benefits of fluoridated water for preventing tooth decay, in Canada (4,5) and and elsewhere. However, a systematic review of research on water fluoridation and health concluded that overall the evidence base is of low to moderate quality. (6)

Opposition to fluoridation has existed as long as the intervention itself, and the main reasons have not changed: skepticism about its effectiveness, concern about its potential harms and resistance to its intrusive nature. Recently, opposition has materialized in decisions by several communities in Canada to discontinue the practice. (7,8) While the proportion of Canadians exposed to community water fluoridation increased from 6% (1960) to 45% (2007), there has been a decline since 2007 because several large municipalities have discontinued the practice, including Calgary (2011), Waterloo (2010), Windsor (2013), Quebec City (2008) and Moncton (2012).

In terms of the risks/harms associated with fluoridation, research has consistently shown an association between fluoride exposure and risk of dental fluorosis (staining of the tooth enamel). (9) There is less support for other harms. The MAC of 1.5 mg/L set out by Health Canada was identified on the basis that it is "unlikely to cause adverse health effects", including immunotoxicity, developmental toxicity and/or neurotoxicity. (2) A recent report by Public Health England (PHE) confirmed the absence of association between fluoridation status and a range of adverse health outcomes. (10)

Despite these (and other) reports, there remains concern about various harms/risks of fluoridation among some members of the public. (11) Understanding these concerns is important from the point of view of public health communication. For example, if communication is premised on disseminating information from scientific studies, it may be less effective for those who are more influenced by other factors. Sandman describes this phenomenon in his discussion of hazard and outrage in the public perception of risk. Whereas experts may understand risk in terms of magnitude and the probability of an unfavourable event ("hazard"), the public may see risk as a combination of hazard and outrage factors (e.g., fear, dread, misery). (12)

In Calgary, fluoridation was discontinued in May 2011, following a City Council vote. Comments on online forums (i.e., online discussion sites where individuals can hold conversations in the form of posted messages) provide an opportunity to investigate what the public perceives to be the harms/risks of fluoridation. Decisions about fluoridation are made at the municipal level in Canada, within provincial guidelines (if any). Although a systematic study of fluoridation decision-making in Calgary in 2011 has not been undertaken, anecdotal reports suggest that important factors included the need for an infrastructure upgrade (and associated cost), efforts by a veteran councillor to revisit fluoridation, and a City Council characterized by several new councillors and a new mayor. (13-15) Unlike previous instances of fluoridation decision-making in Calgary (i.e., in 1989 and 1998), when a public vote (plebiscite) was undertaken, the 2011 decision was made by City Council. The decision-making process happened quickly, with limited public engagement and essentially none prior to January 2011. (13,14)

The study objective was to identify the health risks, perceived by some members of the public, associated with water fluoridation as expressed in online forums relating to its cessation and aftermath in Calgary (January 2011-) and to systematically examine their scientific basis. Research questions were as follows: 1) which harms/risks are mentioned, 2) for those harms/risks, what kinds of evidence are cited, 3) to what extent is scientific literature cited, and what is its quality, and 4) for a subset of risks/harms, what is known from the broader scientific literature?

METHODS

Identification of online comments about perceived harms/risks of fluoridation

A professional health sciences librarian was consulted to establish an optimal (sensitive and specific) search strategy. Three known links relating to fluoridation in Calgary were used to validate the search strategy; that is, we ensured that the search captured links known to be relevant. (7) The search was conducted in May 2014 and considered materials from January 2011 (as noted above, little if any discussion occurred before then) to May 2014.

First, a series of free-text searches, using Google[TM], was conducted using 1) "fluorid Calgary" (~373,000 results), 2) "fluoride Calgary" (~182,000 results), 3) "fluoridation Calgary" (~24,600 results) and 4) "water fluoridation Calgary" (~12,300 results). Results were listed in order of relevance (i.e., default setting), with no additional filters. Beginning with the top result of each search, each webpage was opened to assess its content. Those webpages that allowed for public comments were included as a potential source. Webpages that did not allow for comments, or had zero comments, were excluded. In viewing multiple pages of results for each search, we observed that the relevance of comments to our study (fluoridation in Calgary) dropped after the 8th or 9th page. We therefore decided to examine all results within the first 10 pages for each search. Many items from the first search (i.e., using the search terms "fluorid Calgary") appeared again in subsequent searches. Duplicate items were omitted.

Next, for each webpage from the original searches, we pursued "related articles", "related links" and any additional webpages cited in the text. Related links that allowed for comments and had at least one comment were added as sources. Most "related links" had either already been identified in the original search, did not have the capacity for comments, had zero comments or were unrelated to water fluoridation. Finally, search engines available on the webpages of the Calgary Herald, the Calgary Sun, The Globe and Mail, the National Post, Canadian Broadcasting Corporation, Metro News, and Maclean's Magazine were used with the search terms "fluoride Calgary" and "fluoridation Calgary". Again, articles that had at least one comment and had not been previously identified were added to the list of sources.

Once an exhaustive list of sources had been assembled, a qualitative judgement was made regarding each source's specific relevance to fluoridation in Calgary. This was done independently by the two authors, with disagreements resolved by discussion. Only sources that pertained to Calgary were used in the next phase of the study.

User comments were extracted from each source. The date of entry of the comment was recorded (if available), along with the full comment text. Individual comments were rated on whether they mentioned the harms/risks of fluoride (yes/no) and on what specific harms/risks were mentioned. Comments that were in favour of water fluoridation or negative comments that did not mention any harms/risks were not considered further. At this stage, it was necessary to develop a working definition of "harms/risks". We operationalized the term to include any possible adverse health effects in humans, including harm to sensitive populations (e.g., children, individuals with existing health conditions); effects on animals; and effects on the environment (e.g., pollution and bio-accumulation). A subset of comments was examined independently by the two authors to gauge agreement and to refine the criteria for judgement in order to permit consistency. The two raters agreed on 94% of the comments (n = 86) from five randomly selected sources. Harms/ risks were recorded verbatim, sorted into categories of similar terms, then further combined into groupings of thematically similar content.

Types of evidence included in online comments about perceived harms/risks of fluoridation

Next, for those user comments that mentioned the harms/risks of fluoridation (identified above), we extracted cited sources of evidence. In other words, we re-examined the comments to determine whether the user provided any evidence for his/her stated harm/risk. We deliberately adopted a broad conceptualization of evidence to include anything that the user identified as support for his/her position (i.e., not just scientific evidence). We grouped similar types of evidence together and computed the frequency of occurrence. For scientific papers (one type of evidence) cited, we retrieved and reviewed the original papers.

Review of scientific literature for a subset of harms/risks identified in online comments

To consider the broader evidence base, we selected a subset of two (for feasibility reasons) harms/risks groupings for a structured literature review: animals, environment and aquatic life (we focused specifically on aquatic life); and the endocrine system (we focused specifically on the thyroid gland). We selected these two topics because we wished to consider both human and nonhuman organisms, and to avoid duplication of recent published reviews (e.g., cognitive outcomes (16)). The breadth of the aquatic life search necessitated further focus: we decided to focus on phytoplankton. Other foci within aquatic life (e.g., amphibians) would have been equally good choices. Although scientific evidence does not resonate with everyone in terms of their views on fluoridation (i.e., it is not sufficient), it is nonetheless necessary to maintain an up-to-date knowledge base by reviewing and synthesizing published literature on specific topics identified by members of the public as areas of concern. A professional health sciences librarian was consulted in operationalizing and executing searches. Search details are provided in Appendix A. For both topics, we first reviewed titles and abstracts using the inclusion/exclusion criteria described in Appendix B. All titles and abstracts were reviewed by the first author (PP), and a subset was independently reviewed by the second author (LM) to assess and establish consistency.

Papers extracted from both searches were reviewed and summarized. The following information was extracted into a table: study objective, rationale, type of study design, study population (e.g., plant species or human sample), source of fluoride, concentration(s) and exposure time(s), methods, results and implications for community water fluoridation. Methodological limitations, as stated by the author, were also recorded. For each review, a synthesis focused on overall findings, key methodological limitations and relevance to fluoridation at recommended levels.

RESULTS

Of the 55 sources identified from the initial search as being potentially relevant and having comment capacity, 48 (87%) were related specifically to fluoridation in Calgary. Most sources were published or posted during 2013 (n = 24;50%), 35% in 2011 (n = 17), 8% in 2012 (n = 4) and 6% in 2014 (n = 3). The degree of relevance was substantive in 46% of cases (n = 22) and constituted only a mention in the remaining 54% (n = 26). Sources were classified as 1) news/newspaper and magazine articles (n = 24;50%), 2) blogs (n = 9; 19%), 3) public discussion forums (n = 7; 15%), 4) opinion articles or letters to the editor (n = 4; 8%) and 5) others, including petition, video, website and interview (n = 4; 8%). Of the news/newspaper and magazine articles, most were obtained from the National Post (n = 5; 21%) and MacLean's Magazine (13%). The number of user comments per source ranged from 1 to 824, with a mean of 69.

Identification of perceived harms/risks of fluoridation mentioned in online comments

In total, the 48 sources corresponded to 3,330 user comments. Of these, 356 (10.7%) mentioned the harms/risks of fluoride (based on our operational definition). Harms/risks were grouped into 56 categories, which in turn were further combined into 17 thematic groupings. Table 1 shows the 17 thematic groupings, the 56 original categories from which the groupings were formed, examples and the frequency of occurrence.

Types of evidence included in online comments about perceived harms/risks of fluoridation

Table 2 presents the types of evidence cited in user comments about the harms/risks of fluoride, along with examples and frequency of occurrence. Forty-two percent of comments (n = 176) did not cite any evidence. Of those comments that did, the most frequent evidence types were a person viewed as an expert or authority, a generic reference to research, and a website, including YouTube. Less frequently cited types included personal experiences, product labels and non-government or non-profit organizations.

Published literature represented 5.5% of all evidence cited, corresponding to nine papers. (16-24) These papers are summarized in Table 3. Overall, the papers have very significant methodological problems (e.g., no or limited details on methods, limited or unknown measurement of potentially important confounding variables), and their relevance to community water fluoridation at recommended concentrations is limited or unknown.

Review of scientific literature for a subset of harms/risks identified in online comments

Detailed information about all studies retrieved is available as supplemental online material. Attributes of studies are summarized in Table 4. Table 5 shows the correspondence among different units of fluoride, to permit comparison across studies.

Fluoride and Aquatic Plant Life

The search yielded 2,594 unique citations. After applying inclusion and exclusion criteria, we arrived at 15 papers for in-depth review. Four of these were either unavailable in English or could not be retrieved (via Interlibrary Loan), reducing the final set to 11 papers (see Table 4 and Supplementary Table 1).

Collectively, the papers may be summed up as follows: at elevated concentrations (i.e., >2 mg/L) and in some species (e.g., Chlorella pyrenoidosa), the fluoride ion produces visible toxic effects in algae and phytoplankton. These effects include inhibition of growth, photosynthesis, respiration, cell division and protein synthesis, and reduced cellular ATP (adenosine triphosphate) levels, enzyme activity (e.g., enolase) and metabolism. However, the toxic effects are variable, exhibiting fluctuations with fluoride concentration, exposure time, temperature, water pH, water composition (e.g., the presence of other ions), water hardness and season. Many of the studies suggest that fluoride could combine with other constituents present in water (e.g., calcium, magnesium), mediating or enhancing toxicity.

In some cases, positive effects of fluoride were observed. There is minimal growth enhancement observed in some species, suggesting that fluoride may be a nutritional requirement for these plant forms (e.g., Anabaena fertilissima). Further, some species demonstrate resistance to extreme fluoride levels, hence their use in de-fluoridation experiments (e.g., Amphidinium carteri).

Overall, levels of fluoride pollution assessed in the studies reviewed do not seem to pose an immediate threat to the viability and growth of algae and phytoplankton provided sufficient nutrients are available. However, an important limitation is that all studies were carried out in conditions of sufficient nutrients, thus it is important to consider the implications of nutrient insufficiency on toxicity.

Overall, study quality was poor. Most reviews did not include a description of the search strategy, inclusion/exclusion criteria or method of analysis. Frequently, tested fluoride concentrations and exposure times were not justified, and the rationale for the experiments was limited or absent. Most studies did not make explicit references to community water fluoridation.

Fluoride and Thyroid

The initial search of seven databases yielded 955 unique abstracts. After applying inclusion and exclusion criteria, we arrived at 27 papers for in-depth review (see Table 4 and Supplementary Table 2).

Collectively, the papers may be summed up as follows:

Human studies Even at more elevated concentrations (e.g., 4.0 ppm), the fluoride ion did not show toxic effects on the thyroid gland nor did it clearly affect levels of thyroid hormones. Specifically, few studies report serum levels of thyroid hormones (T3, T4, thyroid stimulating hormone) outside the normal range; few studies report any clinical manifestations of thyroid enlargement (e.g., total thyroid volume is not different in children exposed to high [e.g., 4.6 mg/L] and normal [e.g., 0.19 mg/L] fluoride levels);and goitre prevalence does not seem to correlate with fluoride levels in water.

There were numerous methodological limitations identified in human studies, relating especially to a lack of control for other variables (e.g., fluoride consumption in food, presence of other ions and contaminants in water).

Animal studies Animal models testing extremely high fluoride concentrations (e.g., 40-500 ppm) report lower thyroid hormone levels when compared with controls, although these findings are not consistent across studies or species (i.e., rats, mice and rabbits). Most animal studies do not translate their findings to humans. The range of fluoride that showed adverse effects on the thyroid (30-500 ppm) among animals were in all cases substantially higher than recommended concentrations for controlled water fluoridation. There were numerous methodological limitations identified in animal studies, relating especially to a lack of rationale for fluoride concentrations and mode of administration.

DISCUSSION

Overall, the main conclusions are threefold. First, according to online comments surrounding fluoridation cessation in Calgary, Alberta, in 2011, concerns about a wide range of harms/risks of fluoridation were expressed. These concerns relate to the health of humans (diverse body systems), the environment, and nonhuman organisms.

Second, a large proportion of comments about harms/risks did not provide any supporting evidence, and of those that did, the sources of evidence were diverse, with scientific research infrequently cited. The nine scientific papers cited were found to have very significant methodological limitations and at best only very limited relevance to community water fluoridation at recommended concentrations. Perhaps most troubling is that, in many cases, information that would permit some readers to gauge the validity of the study was missing entirely, for example, studies having no description of how the participants were sampled, or the absence of a methods section altogether. These are not minor concerns; they are egregious methodological flaws that make it very difficult (in some cases impossible) to have confidence in the reported results.

Third, for the two examples for which we systematically examined the broader research literature, the evidence likewise did not support the perceived harms/risks. On the contrary, the scientific evidence points towards the safety of fluoride at recommended levels (0.7-1.5 mg/L) with respect to these specific harms/risks. Overall, these findings are consistent with Sandman's concept of "outrage", whereby perception of risk, for some members of the public, is influenced by factors other than scientific evidence.

The findings have implications for communication with the public about fluoridation. First, scientific evidence is only one component of why some people support or do not support fluoridation strategies. Communication strategies must accommodate that reality by, for example, incorporating techniques that are not dismissive of expressed concerns. Excellent resources are available for this (e.g., the US-based Campaign for Dental Health, http://ilikemyteeth.org/). Second, to the extent that members of the public consult the scientific literature, it is essential that methodological assessment of new studies, including their relevance to community water fluoridation, is promptly performed and widely disseminated. There are excellent examples of this as well, such as appraisals performed by Peel Public Health in the Peel Region of Ontario (e.g., http://bit.ly/ 1aLhom8).

One limitation of our study is that the comments we examined are limited to a small segment of the population during a particular time frame and do not represent those of the public as a whole. They do, however, resemble the broad cross section of risks/harms identified in other times and places. (11) Although those who are strongly opposed to fluoridation and who may thus contribute to online for a are thought to constitute a small minority of the population, they may have disproportionate impact on plebiscite outcomes, and thus it is important to identify, understand and find ways to address their expressed concerns. Second, because the search was conducted in May 2014, some of the information posted at the time of the 2011 Calgary plebiscite may no longer be available online, and unfortunately we do not know the extent to which this occurred. A third limitation is that for feasibility reasons our broader literature review focused on only two specific harms/ risks among the many identified in this study. Although scientific evidence does not resonate with everyone in terms of influencing support for/opposition to fluoridation (i.e., it is not sufficient), it is nonetheless necessary for informed discussion and decision-making, and thus periodic review and synthesis of existing research on specific harms/risks and fluoride is important; this was the reasoning behind our two reviews (for which recent published reviews do not exist). Fourth, the nature of the evidence base and our review methods are such that potential biases may be present. Most notably, we excluded articles published in non-English language (the proportion of non-English articles on fluoride appears to be non-negligible). Additionally, we did not perform a formal risk of bias assessment, opting instead to focus on major methodological limitations and relevance to community water fluoridation.

Important directions for future research on the subject of public perceptions and fluoridation include analysis of comments in favour of fluoridation, including how disagreements play out in online forums; research into the development and testing of public health communication messages that reflect our findings here; and ongoing systematic reviews of research on other perceived harms/risks that showed up in our sample of comments, such as fluoride's impact on amphibians, issues of the industrial source of fluoride (e.g., sodium fluorosilicate) and concerns about arsenic contamination and lead leaching.

REFERENCES

(1.) Spellman, FR. Mathematics Manual for Water and Wastewater Treatment Plant Operators. Boca Raton, FL: CRC Press, 2004; pp.199-202.

(2.) Health Canada. Summary Report on the Findings of the Oral Health Component of the Canadian Health Measures Survey, 2007-2009. Ottawa, ON: Health Canada, 2010.

(3.) Featherstone JD. Prevention and reversal of dental caries: Role of low level fluoride. Community Dent Oral Epidemiol 1999;27(1):31-40. PMID: 10086924. doi: 10.1111/j.1600-0528.1999.tb01989.x.

(4.) Wolfe J, Ishaque S, Aung YN. The State of Dental Health in Alberta: A Brief Report. Edmonton, AB: University of Alberta, School of Public Health, 2013.

(5.) McLaren L, McIntyre L. Drinking Water Fluoridation in Canada: Review and Synthesis of Published Literature. Prepared for the Public Health Agency of Canada, Ottawa, 2011.

(6.) Iheozor-Ejiofor Z, Worthington HV, Walsh T, O'Malley L, Clarkson JE, Macey R, et al. Water fluoridation for the prevention of dental caries (Review). The Cochrane Library, 2015, Issue 6.

(7.) Understanding public decision-making on community water fluoridation. J Can Dent Assoc 2013;79:d77. PMID: 23763747.

(8.) Rabb-Waytowich D. Water fluoridation in Canada: Past and present. J Can Dent Assoc 2009;75:451-54. PMID: 19627654.

(9.) Levy SM. An update on fluorides and fluorosis. J Can Dent Assoc 2003; 69(5):286-91. PMID: 12734021.

(10.) Public Health England (PHE). Water Fluoridation: Health Monitoring Report for England 2014. London: Public Health England, 2014.

(11.) Carstairs C, Elder R. Expertise, health, and popular opinion: Debating water fluoridation, 1945-80. Can Hist Rev 2008;89(3):345-71. doi: 10.3138/chr.89.3.345.

(12.) Sandman PM. Hazard versus outrage in the public perception of risk. In: Covello VT, McCallum DB, Pavlova MT (Eds), Effective Risk Communication: The Role and Responsibility of Government and Nongovernment Organizations. New York, NY: Plenum, 1989; 45-49.

(13.) "What's Calgary drinking?" The Globe & Mail [Toronto, Canada] 17 Feb. 2011: A25. Canadian Periodicals Index Quarterly. Web Available at: http://www. theglobeandmail.com/globe-debate/whats-calgary-drinking/article566673/ (Accessed October 29, 2015).

(14.) "Calgary's fluoride debate goes public. City committee will decide next move after late-January meeting". CBC News, Posted: 10 Jan. 2011. Available at: http://www.cbc.ca/news/canada/calgary/calgary-s-fluoride-debate-goes-public-1.1015138 (Accessed October 29, 2015).

(15.) McLaren L, Emery JCH, McIntyre L. "Debunking falsehoods about fluoride". Calgary Sun, 6 Feb. 2011.

(16.) Choi AL, Sun G, Zhang Y, Grandjean P. Developmental fluoride neurotoxicity: A systematic review and meta-analysis. Environ Health Perspect 2012;120(10):1362-68. PMID: 22820538. doi: 10.1289/ehp.1104912.

(17.) Duan J, Zhao M, Wang L, Fang D, Wang Y, Wang W. A comparative analysis of the results of multiple tests in patients with chronic industrial fluorosis. Guizhou Med J 1995;18(3):179-80.

(18.) Gazzano E, Bergandi L, Riganti C, Aldieri E, Doublier S, Costamagna C, et al. Fluoride effects: The two faces of Janus. Curr Med Chem 2010;17(22):2431-41. PMID: 20491635. doi: 10.2174/092986710791698503.

(19.) Grandjean P, Landrigan PJ. Neurobehavioural effects of developmental toxicity. Lancet Neurol 2014;13(3):330-38. PMID: 24556010. doi: 10.1016/ S1474-4422(13)70278-3.

(20.) Justus C, Krook LP. Allergy in horses from artificially fluoridated water. Fluoride 2006;39(2):89-94.

(21.) Sandhu R, Lal H, Kundu ZS, Kharb S. Serum fluoride and sialic acid levels in osteosarcoma. Biol Trace Elem Res 2011;144(1-3):1-5. PMID: 19390788. doi: 10. 1007/s12011-009-8382-1.

(22.) Sawan RM, Leite GA, Saraiva MC, Barbosa F, Tanus-Santos JE, Gerlach RF. Fluoride increases lead concentrations in whole blood and in calcified tissues from lead-exposed rats. Toxicology 2010;271(1):21-26. PMID: 20188782. doi: 10.1016/j.tox.2010.02.002.

(23.) Susheela AK, Gupta R, Ganesh K, Brahmankar S, Bhasin S, Gupta G. Effective interventional approach to control anaemia in pregnant women. Curr Sci 2010;98(10):1320-30.

(24.) Tang QQ, Du J, Ma HH, Jiang SJ, Zhou XJ. Fluoride and children's intelligence: A meta-analysis. Biol Trace Elem Res 2008;126(1-3):115-20. PMID: 18695947. doi: 10.1007/s12011-008-8204-x.

Received: February 26, 2015

Accepted: June 5, 2015

Appendix A

Search Strategy: Fluoride and Pollution/Aquatic Life (July 2014)

Databases

Aqualine Biological Abstracts (OVID to 2005) CAB Abstracts Environment Abstracts Environment Complete Environmental Sciences and Pollution Management MEDLINE Pollution Abstracts ToxLine Web of Science Zoological Record

Total abstracts (before de-duplication): 4155 Total abstracts (after de-duplication): 2594

Search strings and total number of results by database: Aqualine

1. TI (algae or aquatic life* or aquatic mammal* or aquatic organism* or aquatic plants or fish or salmon or water plants or water pollut* or dolphin* or sharks or tuna or whales) OR AB (algae or aquatic life* or aquatic mammal* or aquatic organism* or aquatic plants or fish or salmon or water plants or water pollut* or dolphin* or sharks or tuna or whales)

2. TI (fluoride* or fluoridation) OR AB (fluoride* or fluoridation)

3. 1 and 2

4. Limit 3 to scholarly peer review journals

Total: 293 results

Biological Abstracts (OVID to 2005)

1. (algae or Aquatic life* or Aquatic mammal* or aquatic organism* or aquatic plants or Fish or Salmon or Water plants or water pollut*).tw.

2. (dolphin* or sharks or tuna or whales).tw.

3. 1 or 2

4. fluorid*.tw.

5. 3 and 4

Total: 437 results

CAB Abstracts

1. exp Fishes/

2. exp Aquatic organisms/

3. exp Cetacea/

4. (algae or Aquatic life* or Aquatic mammal* or aquatic organism* or aquatic plants or Fish or Salmon or Water plants or water pollut*).tw.

5. (dolphin* or sharks or tuna or whales).tw.

6. water pollution/ or exp water pollution, chemical/

7. exp algae/

8. exp aquatic plants/

9. 1 or 2 or 3 or 4 or 5 or 6 or 7 or 8

10. fluorid*.tw.

11. 9 and 10

Total: 878 results

Environment Abstracts

1. TI (algae or aquatic life* or aquatic mammal* or aquatic organism* or aquatic plants or fish or salmon or water plants or water pollut* or dolphin* or sharks or tuna or whales)

2. TI (fluoride* or fluoridation)

3. 1 and 2

4. Limit 3 to scholarly peer review journals

Total: 11 results

Environment Complete

1. TI (algae or aquatic life* or aquatic mammal* or aquatic organism* or aquatic plants or fish or salmon or water plants or water pollut* or dolphin* or sharks or tuna or whales) OR AB (algae or aquatic life* or aquatic mammal* or aquatic organism* or aquatic plants or fish or salmon or water plants or water pollut* or dolphin* or sharks or tuna or whales) OR SU (algae or aquatic life* or aquatic mammal* or aquatic organism* or aquatic plants or fish or salmon or water plants or water pollut* or dolphin* or sharks or tuna or whales)

2. TI (fluoride* or fluoridation) OR AB (fluoride* or fluoridation) OR SU (fluoride* or fluoridation)

3. 1 and 2

4. Limit 3 to Scholarly Peer Reviewed Journals

Total: 444 results

Environmental Science and Pollution Management

5. TI (algae or aquatic life* or aquatic mammal* or aquatic organism* or aquatic plants or fish or salmon or water plants or water pollut* or dolphin* or sharks or tuna or whales) OR AB (algae or aquatic life* or aquatic mammal* or aquatic organism* or aquatic plants or fish or salmon or water plants or water pollut* or dolphin* or sharks or tuna or whales)

6. TI (fluoride* or fluoridation) OR AB (fluoride* or fluoridation)

7. 1 and 2

8. Limit 3 to scholarly peer review journals

Total: 514 results

MEDLINE

1. exp Fishes/

2. exp Aquatic organisms/

3. exp Cetacea/

4. (algae or Aquatic life* or Aquatic mammal* or aquatic organism* or aquatic plants or Fish or Salmon or Water plants or water pollut*).tw.

5. (dolphin* or sharks or tuna or whales).tw.

6. water pollution/ or exp water pollution, chemical/

7. 1 or 2 or 3 or 4 or 5 or 6

8. exp Fluorides/ae, to [Adverse Effects, Toxicity]

9. Fluoridation/ae [Adverse Effects]

10. fluorid*.tw.

11. 8 or 9 or 10

12. 7 and 11

Total: 358 results

Pollution Abstracts

1. TI (algae or aquatic life* or aquatic mammal* or aquatic organism* or aquatic plants or fish or salmon or water plants or water pollut* or dolphin* or sharks or tuna or whales) OR AB (algae or aquatic life* or aquatic mammal* or aquatic organism* or aquatic plants or fish or salmon or water plants or water pollut* or dolphin* or sharks or tuna or whales)

2. TI (fluoride* or fluoridation) OR AB (fluoride* or fluoridation)

3. 1 and 2

4. Limit 3 to scholarly peer review journals

Total: 21 results

ToxLine

1. (algae or aquatic life* or aquatic mammal* or aquatic organism* or aquatic plants or fish or salmon or water plants or water pollut* or dolphin* or sharks or tuna or whales)

2. (fluoride* or fluoridation)

3. 1 and 2

4. Limit 3 to scholarly peer review journals

Total: 335 results

Zoological Record

1. (algae or Aquatic life* or Aquatic mammal* or aquatic organism* or aquatic plants or Fish or Salmon or Water plants or water pollut*).tw.

2. (dolphin* or sharks or tuna or whales).tw.

3. 1 or 2

4. fluorid*.tw.

5. 3 and 4

Total: 26 results

Web of Science (Science Citation Index and Social Science Citation Index)

1. SU (algae or aquatic life* or aquatic mammal* or aquatic organism* or aquatic plants or fish or salmon or water plants or water pollut* or dolphin* or sharks or tuna or whales)

2. SU (fluoride* or fluoridation)

3. 1 and 2

Total: 838 results

Search Strategy: Fluoride and Thyroid (July 2014)

Databases

MEDLINE PubMED EMBASE Global Health CINAHL ToxLine Web of Science

Total abstracts (before de-duplication): 1616 Total abstracts (after de-duplication): 955

Search strings and number of results by database: MEDLINE

1. exp Fluorides/ae, to [Adverse Effects, Toxicity]

2. Fluoridation/ae [Adverse Effects]

3. fluorid*.tw.

4. 1 or 2 or 3

5. Thyroid Gland/

6. exp Thyroid Diseases/

7. (euthyroid* or goiter* or goitre* or hyperthyroid* or hypothyroid* or thyroid* or thyrotoxicos*).tw.

8. 5 or 6 or 7

9. 4 and 8

10. limit 9 to animals

11. limit 9 to (animals and humans)

12. 10 not 11

13. 9 not 12

Total: 144 results

PubMED

1. Fluorides[MeSH] OR Fluoridation[MeSH]

2. fluorid*[tiab]

3. 1 or 2 or 3

4. Thyroid Gland[MeSH]

5. Thyroid Diseases[MeSH]

6. (euthyroid* or goiter* or goitre* or hyperthyroid* or hypothyroid* or thyroid* or thyrotoxicos*)[tiab]

7. 5 or 6 or 71

8. 4 and 8

Total: 400 results

EMBASE

1. fluoridation/ae [Adverse Drug Reaction]

2. fluoride/ae, to [Adverse Drug Reaction, Drug Toxicity]

3. fluorid*.tw.

4. 1 or 2 or 3

5. exp thyroid gland/

6. exp thyroid disease/

7. (euthyroid* or goiter* or goitre* or hyperthyroid* or hypothyroid* or thyroid* or thyrotoxicos*).tw.

8. 5 or 6 or 7

9. 4 and 8

Total: 326 results

Global Health

1. exp thyroid gland/

2. exp thyroid disease/

3. (euthyroid* or goiter* or goitre* or hyperthyroid* or hypothyroid* or thyroid* or thyrotoxicos*).tw.

4. 1 or 2 or 3

5. fluoride/ or fluorides/

6. fluoridation/

7. fluorid*.tw.

8. 5 or 6 or 7

9. 4 and 8

Total: 138 results

CINAHL

1. (MH "Fluoridation") OR (MH "Fluorides+")

2. TI fluorid* OR AB fluorid*

3. 1 or 2

4. (MH "Thyroid Neoplasms") OR (MH "Thyroid Diseases+") OR (MH "Thyroid Hormones+") OR (MH "Thyroid Gland") OR (MH "Hypothyroidism+") OR (MH "Hyperthyroidism +") OR (MH "Graves' Disease+") OR (MH "Goiter+") OR (MH "Thyrotoxicosis+")

5. TI (euthyroid* or goiter* or goitre* or hyperthyroid* or hypothyroid* or thyroid* or thyrotoxicos*) OR AB (euthyroid* or goiter* or goitre* or hyperthyroid* or hypothyroid* or thyroid* or thyrotoxicos*)

6. 4 or 5

7. 3 and 6

Total: 17 results

Toxline

1. SUfluorid*

2. SU (euthyroid* or goiter* or goitre* or hyperthyroid* or hypothyroid* or thyroid* or thyrotoxicos*)

3. 1 and 2

Total: 411 results

Web of Science (Science Citation Index)

1. TOPIC fluorid*

2. TOPIC (euthyroid* or goiter* or goitre* or hyperthyroid* or hypothyroid* or thyroid* or thyrotoxicos*)

3. 1 and 2

Total: 180 results

Appendix B

Inclusion/Exclusion Criteria

A. Inclusion/Exclusion Criteria: Fluoride and Aquatic Life

1. Must include a specific, named harm vs. generic "pollution" or "contamination" or "accumulation".

2. Harmful outcome must pertain to aquatic plants and/or phytoplankton.

3. If study pertains to aquatic animals (e.g., salmon) or non-aquatic life, such as humans or sheep, exclude.

4. Fluoride vis-a-vis specific, named harm must be primary (vs. peripheral) focus. For example, if fluoride is one of a number of compounds being studied, exclude.

5. Focus must be on the direct effect of fluoride on the specific harm, versus interactive effect. For example, if the focus is how fluoride interacts with aluminum, exclude.

6. All forms (i.e., ions, compounds) and sources (i.e., "natural" and artifical, e.g., industrial waste, controlled addition at water systems) of fluoride should be considered for inclusion.

7. If primary focus is processes of defluoridation, exclude.

8. Must be original, empirical research (letters to the editor, editorials and commentaries are excluded).

B. Inclusion/Exclusion Criteria: Fluoride and Thyroid

1. Fluoride must be primary focus (vs. peripheral). For example, if fluoride is one of a number of factors being studied, exclude, unless it is a study of a range of outcomes related to either naturally high fluoride levels in a region's drinking water or artificially fluoridated water. Include if fluoride is one of only TWO compounds under investigation.

2. Focus must be on the direct effect of fluoride on the specific harm, versus interactive effect.

3. The specific harm must be explicitly related to the thyroid. If the paper considers the effects of fluoride on multiple organs/systems, exclude, unless the thyroid is one of only TWO systems being investigated (e.g., skeleton and thyroid).

4. Must be human science focused (have human application) versus veterinary sciences or related (e.g., tadpoles).

5. Exclude papers that are mechanism-focused.

6. If the study considers the effects on thyroid hormones, include if focus is on production; exclude if focus is on circulating hormones without direct reference to the thyroid or production of the hormones.

7. Must be original, empirical research (letters to the editor, editorials, and commentaries are excluded).

Paulina C. Podgorny, BHSc, Lindsay McLaren, PhD

Department of Community Health Sciences, University of Calgary, Calgary, AB

Correspondence: Lindsay McLaren, PhD, Department of Community Health Sciences, University of Calgary, TRW3, 3280 Hospital Dr. NW, Calgary, AB T2N 4Z6, Tel: 403-210-9424, E-mail: lmclaren@ucalgary.ca

Acknowledgements: P. Podgorny acknowledges her support through an O'Brien Centre Summer Studentship (OCSS) from the University of Calgary. L. McLaren holds an Applied Public Health Chair funded by the Canadian Institutes of Health Research (Institute of Population and Public Health; Institute of Musculoskeletal Health and Arthritis), the Public Health Agency of Canada, and Alberta Innovates--Health Solutions. We are grateful to Diane Lorenzetti, Research Librarian, Department of Community Health Sciences, University of Calgary; and to Lorraine Toews, Librarian, Veterinary Medicine and Bachelor of Health Sciences, University of Calgary, for their expertise and assistance with the literature searches.

Conflict of Interest: None to declare.

Table 1. Harms-risks mentioned in online comments pertaining
to community water fluoridation cessation in Calgary,
January 2011-May 2014, ordered by frequency

Thematic          Original          Example terms       Frequency
grouping          harm/risk                             (%) of
                  category                              mention
                                                        (total
                                                        = 639)

Generic: toxic/   Chemicals         "Contaminant"       187 (29.3%)
poisonous/        including
chemicals/        build-up in       "Deadly poison"
hazardous/        body
contaminant                         "Toxic"
(including        Contaminant
deadly)                             "May be fatal
                  Death             if swallowed or
                                    inhaled"
                  Hazardous
                                    "Noxious waste"
                  Noxious

                  Poison

                  Toxic

Generic:          Dangerous,        "Dangerous          83 (13.0%)
unhealthy/        detrimental       substance"
damaging to
health            Epidemic          "Detrimental
                                    substance"
                  Harmful
                                    "Harmful side
                  Multiple          effects"
                  body systems
                                    "Serious
                  Overexposure      abnormalities
                                    of several body
                  Sick people       systems"

                  Side effects      "Health
                                    problems"
                  Unhealthy
                                    "Linked to
                                    several serious
                                    health
                                    conditions"

Dental,           Dental            "Staining of        66 (10.3%)
including but     fluorosis         teeth due to
not limited to                      mottling of
fluorosis         Dental, not       tooth enamel"
                  fluorosis
                                    "Discoloration"

                                    "Rotting"

                                    "Not good for
                                    porcelain caps
                                    as it makes
                                    them crack over
                                    time"

                                    "Weakens teeth"

Behavioural and   Attention         "Inability to       50 (7.8%)
cognitive         deficit           concentrate"
                  hyperactivity
                  disorder          "Lowered IQ" or
                                    "IQ reduction"
                  Alzheimer's
                                    "Anti-
                  Anti-depressant   depressant
                                    effect on
                  Autism            people"

                  Depression        "Makes you a
                                    slug"
                  Dyslexia
                                    "Effects on
                  Energy            mental
                                    performance"
                  IQ

                  Irritability

                  Concentration

Generic:          Children--        "Harm to young      38 (5.9%)
children,         generic           children"
elderly
                  Elderly people    "Overexposure
                  --generic         in formula-fed
                                    infants"

                                    "The elderly"

                                    "Affects
                                    seniors"

                                    "Babies"

Animals,          Animals           "Spinal             37 (5.8%)
environment and   Environment       collapse in
aquatic life                        guppies"

                                    "Cattle dropped
                                    dead"

                                    "Harms salmon
                                    and other
                                    aquatic life"

                                    "Harm to the
                                    downstream
                                    environment"

                                    "Damaging
                                    environmental
                                    pollutant"

Bones and         Skeletal          "Degenerative       36 (5.6%)
skeleton          fluorosis         problems in
                                    bone, bone
                  Bone problems     tissue"
                  (break,
                  fracture)         "Broken hips
                                    from brittle
                  Arthritis         bones"

                                    "Painful and
                                    debilitating
                                    fluorosis in
                                    the joints"

                                    Arthritis as
                                    "subclinical
                                    skeletal
                                    fluorosis"

                                    "Affects
                                    skeleton"

Endocrine         Pineal gland      "Thyroid            36 (5.6%)
system                              suppressant"
                  Thyroid
                                    "Goitre
                                    development"

                                    "Pineal gland
                                    damage"

                                    "Suppresses
                                    thyroid by
                                    competing with
                                    iodine for
                                    absorption in
                                    the thyroid"

                                    "Overactive
                                    thyroid"

Cancer            Cancer            "Rats started       34 (5.3%)
                                    to develop
                                    small tumours
                                    in their
                                    brains ...
                                    tumours
                                    developing in
                                    their stomach
                                    lining"

                                    "Bladder
                                    cancer"

                                    "Bone cancer"

                                    "Carcinogenicity"

                                    "Osteosarcoma"

Brain and         Brain and CNS     "Destroys the       24 (3.8%)
central nervous                     brain"
system (CNS)      Neurotoxin
                                    "Calcinates
                                    parts of the
                                    brain"

                                    "Abnormal
                                    development of
                                    the central
                                    nervous system
                                    in fetuses and
                                    young children"

                                    "Neurotoxin
                                    which impairs
                                    brain function"

                                    "Central
                                    nervous system
                                    effects"

Urinary system    Bladder           "Those with         23 (3.6%)
                                    kidney
                  Dehydration       impairment"

                  Kidney disease    "Anyone with
                                    kidney stones"

                                    "Stress on the
                                    kidneys"

                                    "Serious
                                    bladder
                                    problems and
                                    surgeries"

                                    "People on
                                    kidney
                                    dialysis"

Immune system     Allergy           "The immune         8 (1.3%)
                                    compromised"
                  Immune system
                                    "Allergic
                  Irritation        reactions"

                  Oxidative         "Causes
                  stress            irritation to
                                    skin, eyes"
                  Sensitivity
                                    "Causes
                  Ulcers            oxidative
                                    stress by
                                    interfering
                                    with the body's
                                    defence
                                    mechanisms
                                    against
                                    reactive oxygen
                                    species"

                                    "Mouth ulcers"

Digestive         Digestive         "Upset stomach"     6 (0.9%)
system            system
                                    "Digestive
                  Liver             systems of
                                    fluoride
                                    poisoning"

                                    "Stomach
                                    problems"

                                    "Not good for
                                    your liver"

                                    "Nausea"

Others            Diabetes          "People with        4 (0.6%)
                                    diabetes" or
                  Genetic           "Diabetics"

                  Hair              "Genetic
                                    damage"

                                    "Hair loss"

Circulatory       Cardiovascular    "Coronary           3 (0.5%)
system            system            artery disease"

                                    "Scarring
                                    arterial walls"

                                    "Affects
                                    circulatory
                                    system"

                                    "Harmful to
                                    many arteries"

                                    "Affects
                                    heart"

Respiratory       Respiratory       "Chronic            2 (0.3%)
system            and               cough"
                  related
                                    "Affects
                                    respiratory
                                    system"

Pregnancy and     Pregnancy and     "Sudden infant      2 (0.3%)
related           related           deaths"

                                    "Anemia in
                                    pregnant women"

                                    "Premature
                                    births"

                                    "Low baby birth
                                    weight"

Table 2. Types of evidence cited in online comments, ordered
by frequency

Type of evidence       Examples and/or        Frequency (%) of
                       sample quotes          mention (total =
                                                    416)

None (no evidence)     N/A                      176 (42.3%)

Person viewed as an    "Dr. James Beck"
expert or authority,
including personal     "Christopher Bryson
expertise and          (journalist)"
credentials
                       "14 Nobel prize
                       winners in either         53 (12.7%)
                       medicine or
                       chemistry"

                       "I am a practicing
                       dental hygienist"

                       "I'm a nutritionist
                       that practices
                       alternative
                       medicine"

Generic reference to   "Feel free to ask         30 (7.2%)
research               for sources"

                       "Read the facts
                       about what happens
                       to kids who get too
                       much fluoride before
                       the age of 8"

                       "Numerous scientific
                       studies"

                       "According to new
                       research"

                       "More and more
                       science is showing"

Website, including     http://slw eb.org/        30 (7.2%)
YouTube                bibliography.html
                       ("A Bibliography
                       of Scientific
                       Literature on
                       Fluoride")

                       http://
                       www.nofluoride.com
                       /presentations/
                       Nobel%20Prize%20
                       Winners.pdf ("14
                       Nobel Prize
                       Winners who
                       object to
                       fluoridation")

                       www.fluoridealert
                       .org (Website of
                       The Fluoride
                       Action Network)

                       http://en.wikipedia
                       .org/wiki/Dental_
                       fluorosis
                       (Wikipedia page
                       for dental
                       fluorosis)

                       http://cof-cof.ca/
                       convincing-canadian
                       -studies-
                       demonstrating-water
                       -fluoridations-
                       questionable-
                       merit/ (A list
                       of "Convincing
                       Canadian studies
                       demonstrating water
                       fluoridation's
                       questionable
                       merit")

Personal experience    "We have lost 8           25 (6.0%)
                       horses and 4 dogs
                       from the
                       consumption,
                       accumulation and
                       systemic effects of
                       this product"

                       "It destroyed my
                       thyroid"

                       "I've started to
                       lose a lot of hair
                       ... my thyroid
                       started acting up
                       and I developed a
                       goitre"

                       "I've been fighting
                       health problems"

                       "I don't use tap
                       water for the
                       guppies either, it
                       was causing their
                       spines to collapse"

Government report or   "Food and Drug Act"       23 (5.5%)
organization,
including government   "The Safe Drinking
acts and regulations   Water Act (2002)"

                       "The Ontario Safe
                       Drinking Water Act,
                       Section 19, in
                       effect January 2013"
                       National Research
                       Council

                       "The Canadian Health
                       Measures Survey
                       (CHMS) released data
                       ..."

Study or article in    "Sawan, et al.           23 (5.5%) *
peer-reviewed          (Toxicology 2/2010):
journal                Water fluoride
                       chemicals boost lead
                       absorption in lab
                       animals' bones,
                       teeth and blood"

                       "Tang el al.,
                       "Fluoride and
                       Children's
                       Intelligence: A
                       Meta-analysis" in
                       Biological Trace
                       Element Research"

My own research/       "I have done              17 (4.1%)
reading                extensive research
                       on fluoride"

                       "I read an article"

                       "I have read
                       reports"

Product label          "If you look at the       13 (3.1%)
                       bag labels Sodium
                       Fluoride you will
                       see the skull and
                       cross bones"

                       "Warning labels on
                       toothpaste"

                       "Material Safety
                       Data Sheet"

Documentary,           "Christopher              13 (3.1%)
magazine, or book      Bryson's 'The
                       Fluoride Deception'"

                       The Case Against
                       Fluoride

                       "Book published in
                       1961, The
                       Fluoridation
                       Experiment, by
                       Exner, Waldbott &
                       Rorty"

                       "Time magazine
                       listed fluoride as
                       one of the 'Top Ten
                       Common Household
                       Toxins' and
                       described fluoride
                       as both 'neurotoxic
                       and potentially
                       tumorigenic if
                       swallowed.'"

NGO/non-profit         The Council of            11 (2.6%)
organization           Canadians (Windsor
                       Chapter)

                       Sierra Club

                       Columbia
                       Riverkeepers

                       The Canadian
                       Association of
                       Physicians for the
                       Environment

                       The National Kidney
                       Foundation

Newspaper              "Editorial that            2 (0.5%)
                       appeared in the
                       Windsor Star"

                       "Howard University's
                       student newspaper,
                       The Hilltop"

* Corresponds to 9 separate published articles, some of
which were mentioned multiple times.

Table 3. Summary of published scientific papers * cited in
online comments

Citation (#)     Synopsis                 Key issues/concerns

Choi et al.      Systematically           Quality of original
(16)             reviewed research on     studies is quite poor
                 fluoride and             (e.g., information on
                 neuro-developmental      child's sex and
                 delays published         parental education was
                 between 1980 and 2011,   not reported in >80%
                 including studies from   of studies, and only
                 rural China that         7% [n = 2] of studies
                 examine naturally        reported household
                 occurring high levels    income; most reports
                 of fluoride. Studies     were fairly brief and
                 contained high and       complete information
                 reference exposure       on covariates was not
                 groups (final n = 27;    available; most
                 2 from Iran and the      studies did not report
                 rest from China).        age adjustment of the
                 Overall results          cognitive test
                 support association      scores). All original
                 between high fluoride    studies appear to be
                 exposure and lower       cross-sectional
                 intelligence (based on   comparisons of
                 standardized mortality   fluoridated and
                 ratio, pooled risk       non-fluoridated
                 ratio). Finding was      groups. Fluoride
                 robust to different      concentrations in the
                 study exclusions.        high exposure group
                                          were in most cases
                                          higher than
                                          recommended (0.7 ppm)
                                          and maximum (1.5 ppm)
                                          levels: range >2 to 11
                                          ppm.

Duan et al.      Various tests (i.e.,     Paper was translated
(17)             cognitive ability,       from Chinese by the
                 electroencephalograph,   Fluoride Action
                 neurological history     Network. No
                 taking and physical      information on how
                 examinations) were       participants were
                 administered among       sampled/selected
                 three groups of          within each group.
                 individuals: 1) n =72    Though authors report
                 men with chronic         that economic status,
                 industrial fluorosis     lifestyle habits and
                 who worked or had        other factors were
                 worked in the            similar across the
                 electrolysis workroom    groups, no data are
                 at an aluminum           presented, and there
                 production facility;     is no information on
                 2) n = 43 men who had    how these were
                 worked in same           measured nor what
                 environment for less     "lifestyle habits" and
                 time and whose           "other factors"
                 condition did not meet   entail. Limited
                 the diagnosis for        information is
                 industrial fluorosis;    provided on the tests
                 and 3) n = 42 healthy    and what the results
                 persons. Economic        mean (e.g., from the
                 status, lifestyle        electroencephalogram,
                 habits and other         the proportion
                 factors were similar     moderately abnormal
                 across groups. Across    and mildly abnormal
                 tests, the worst         are reported for each
                 outcomes were observed   group, but it is not
                 in group A (longest      clear what these
                 exposure) followed by    categories mean or how
                 group B (shorter         they were assigned).
                 exposure), and the       Whether the findings
                 best results were in     have relevance to
                 group C (healthy         fluoridated drinking
                 group). Exposure was     water and to what
                 confirmed by air         extent is unclear.
                 quality tests in the
                 facility and from
                 fingernail samples.

Gazzano et       Review article (134      No methods section; it
al. (18)         references) anchored     is unclear how the
                 in the observation       authors selected the
                 that fluoride's          studies reviewed.
                 behaviour in the human   Article is very
                 organism makes it a      technical, therefore
                 classic example of a     not accessible to a
                 double-edged sword.      non-expert reader.
                 The rationale for the
                 article appears to be
                 the proposed insertion
                 of fluoride in the
                 preparation of
                 biomaterials to
                 improve their
                 integration in the
                 bone, which demands
                 understanding of the
                 safety of fluoride in
                 terms of prolonged
                 exposure to living
                 tissue. The review
                 covers the following
                 areas: 1) fluoride
                 metabolism and types
                 of exposure; 2)
                 mechanisms of fluoride
                 action on dental
                 caries onset; 3)
                 fluoride application
                 in caries prevention;
                 4) effects of fluoride
                 on bone; 5) acute and
                 chronic effects of
                 fluoride; 6)
                 activation of G
                 proteins and kinases
                 and inhibition of
                 phosphatases; 7)
                 fluoride inhibition of
                 many other enzymes;
                 and 8) fluoride and
                 oxidative stress.
                 Concludes by
                 presenting a model for
                 "Is there a unifying
                 hypothesis for
                 fluoride effects?"

Grandjean &      Review (115              Search-selection
Landrigan (19)   references) that is an   methods are only
                 update of a 2006         briefly described.
                 review of the            Because the
                 developmental            identification of
                 neurotoxicity of         fluoride as a newly
                 industrial chemicals.    recognized
                 The 2006 report          developmental
                 identified five          neurotoxicant is based
                 industrial chemicals     entirely on Choi, (16)
                 that could be reliably   this paper suffers
                 classified as            from the same
                 neurodevelopmental       limitations as above
                 toxicants: lead,         in terms of the low-
                 methylmercury,           unknown quality of the
                 arsenic,                 original studies. The
                 polychlorinated          section on "newly
                 biphenyls and toluene.   recognized
                 The authors are          developmental
                 concerned that           neurotoxicants" begins
                 subclinical toxicity     with a paragraph about
                 may be widespread        "powerful
                 before it is realized    epidemiological
                 that concentrations      methods" such as
                 thought to be safe are   prospective birth
                 shown, by                cohort studies, which
                 epidemiological          gives the misleading
                 research, to be too      impression that the
                 high. They also cite     information that
                 examples of early        follows is based on
                 warning signs of         those methods. The
                 subclinical              studies in the Choi
                 neurotoxicity being      paper, (16) as noted
                 ignored or dismissed.    above, are all cross-
                 Fluoride is recognized   sectional, mostly do
                 as one of the "newly     not include any
                 recognized               covariate information
                 developmental            and pertain to higher
                 neurotoxicants". This    than recommended
                 is based entirely on     fluoride levels.
                 the Choi (16) paper
                 above.

Justus &         Authors build on         The paper is an
Krook (20)       another paper recently   account of a personal
                 published in the same    experience; as such,
                 journal, which           it lacks the
                 demonstrated fluoride    systematic nature of
                 poisoning in horses      rigorous research, and
                 that consumed            thus it is more
                 artificially             difficult to rule out
                 fluoridated water.       alternative
                 This paper focuses on    explanations.
                 allergy as another
                 expression of
                 fluorosis in horses.
                 The horses were not
                 likely exposed to
                 other sources of
                 fluoride. Over the
                 years, 2 of 11 horses
                 exposed to the water
                 developed allergy
                 (skin lesions), and
                 the two cases are
                 presented. In the
                 first horse, for
                 example, the lesions
                 were reduced when the
                 horse consumed snow
                 instead of fluoridated
                 water and disappeared
                 when it consumed water
                 from a different
                 source. Eventually,
                 the horse was taken
                 off fluoridated water
                 altogether and the
                 lesions ceased
                 entirely.

Sandhu et al.    The study aimed to       The age/and sex/
(21)             examine serum            matching is only
                 fluoride, among other    mentioned in the
                 things, in 25            abstract (not in the
                 osteosarcoma patients    methods section). No
                 as well as age- and      information about how
                 sex-matched controls     the individuals were
                 with 1) bone-forming     sampled/selected. No
                 tumours other than       other information
                 osteosarcoma (n = 25)    about the three groups
                 and 2) musculoskeletal   (covariates) to permit
                 pain (n = 25). Found     assessment of how
                 that serum fluoride      similar/different they
                 levels were              were on other
                 significantly higher     variables. Conclusion
                 in the osteosarcoma      is thus overstated:
                 group than in the two    "this report proves a
                 control groups. The      link between raised
                 authors acknowledged     fluoride levels in
                 studies that show a      serum and
                 link between fluoride    osteosarcoma".
                 in drinking water and    Implications for
                 osteosarcoma as well     fluoridated drinking
                 as those that did not    water at recommended
                 show a link.             levels are unclear.

Sawan et al.     Authors aimed to test    The fluoride
(22)             whether administration   concentration far
                 of fluorosilicic acid    exceeds recommended
                 could increase blood     (0.7 mg/L) and maximum
                 lead content and         (1.5 mg/L) levels for
                 mineralized tissue       drinking water.
                 lead concentration in    Relevance of findings
                 rats exposed to low      to community water
                 levels of lead from      fluoridation is
                 the beginning of         minimal, if any.
                 gestation
                 (silicofluoride is the
                 fluoride compound used
                 most commonly for
                 fluoridated water in
                 the US, Canada and
                 other countries). The
                 fluoride concentration
                 for the control and
                 lead-only groups was 1
                 mg/L; for the fluoride
                 and fluoride + lead
                 group it was 100 mg/L.
                 The authors note that
                 this fluoride
                 concentration produces
                 plasma fluoride levels
                 that are comparable to
                 those commonly found
                 in humans chronically
                 exposed to 8 mg/L in
                 drinking water (which
                 far exceeds
                 recommended and
                 maximum levels).
                 Higher blood lead
                 concentrations were
                 reported in the
                 fluoride + lead group
                 compared with the
                 lead-only group, and
                 lead concentrations in
                 calcified tissues were
                 significantly higher
                 in the fluoride + lead
                 group than in the
                 lead-only group. No
                 significant
                 differences in
                 fluoride
                 concentrations in
                 calcified tissues were
                 found between these
                 groups.

Susheela et      Article examined         The paper is poorly
al. (23)         (among other things)     written/organized, so
                 whether, among anemic    is difficult to
                 pregnant women with      follow. Unclear
                 urinary fluoride         whether the effect is
                 beyond 1.0 mg/L, an      attributable to the
                 intervention to reduce   reduction in fluoride,
                 fluoride intake          the improved nutrition
                 reduced pre-term         or a combination of
                 births and low birth     the two. Although
                 weight. The              covariate data were
                 intervention included    collected, it does not
                 counselling on how to    appear that the
                 avoid fluoride in        authors examined
                 water and food.          whether results could
                 Eligible women were      be explained by
                 randomly assigned to     covariates (though
                 intervention vs.         random allocation is a
                 control groups.          strength).
                 Information on
                 confounding factors
                 was gathered: diet,
                 economic status,
                 literacy status,
                 employment status,
                 first pregnancy,
                 miscarriage and other
                 problems, other
                 ailments, and use of
                 folic acid and iron
                 supplements. The
                 intervention group,
                 compared with the
                 control group, had
                 higher (better)
                 hemoglobin and higher
                 birth weight/lower%
                 low birth weight.

Tang et al.      A systematic review of   No information at all
(24)             studies from China,      on covariates/
                 written in English or    confounders in
                 Chinese, on the          original studies or
                 association between      any other
                 fluoride and             methodological detail
                 intelligence/IQ,         of the original
                 published between 1988   studies.
                 and 2008 (the "earlier
                 review" cited by Choi
                 (16). Among the 16
                 included "case
                 control" studies, the
                 authors found that
                 children in
                 fluoridated areas had
                 increased risk of
                 lower IQ
                 (meta-analysis,
                 sensitivity analyses).

* All studies are published in peer-reviewed journals.

Table 4. Attributes of the final set of papers from the
aquatic plants/phytoplankton (n = 11) and thyroid
(n = 27) literature searches (see Supplementary Tables
1 and 2 for details of individual studies)

Topic of review   Publication       Main countries
                  years (range)     of origin

Fluoride and      1962-1999 (55%)   Canada (45%)
aquatic plants/   2000-2011 (45%)   India (18%)
phytoplankton                       Other (36%)

Fluoride and      1960-1999 (56%)   India (30%)
thyroid           2000-2014 (44%)   Tunisia (11%)
                                    China (11%)
                                    United States (7%)
                                    Switzerland (7%)
                                    Turkey (7%)
                                    Other (26%)

Topic of review   Study type            Fluoride forms
                                        administered/
                                        tested

Fluoride and      Experimental (55%)    Sodium fluoride
aquatic plants/   Review (36%)          (NaF), hydrogen
phytoplankton     Observational (9%)    fluoride (HF)
                                        and ammonium
                                        fluoride
                                        (N[H.sub.4]F)
                                        (only NaF is
                                        commonly used
                                        for water
                                        fluoridation).

Fluoride and      Observational (44%)   Sodium fluoride
thyroid           Experimental (48%)    (NaF) Naturally
                  Review (8%)           occurring
                                        fluoride (e.g.,
                                        studies of
                                        children in
                                        areas of high
                                        naturally
                                        occurring
                                        fluoride in the
                                        water).

Topic of review   Levels/             Measured
                  concentration       parameters
                  of fluoride and
                  exposure times

Fluoride and      0 to 1900 mg/L      Growth
aquatic plants/   Exposure times      (measured as
phytoplankton     ranged from hours   cell number
                  (minimum 5) to      and/or
                  days (maximum       absorbance),
                  36).                oxygen
                                      exchange, ATP
                                      levels,
                                      chlorophyll
                                      content, enzyme
                                      activity (i.e.,
                                      carbonic
                                      anhydrase) and
                                      respiratory
                                      activity.

Fluoride and      0 to 500 ppm        Thyroid volume,
thyroid                               thyroid weight,
                                      thyroid hormone
                                      levels in serum
                                      (T3, T4, and
                                      thyroid
                                      stimulating
                                      hormone),
                                      clinical
                                      thyroid
                                      enlargement and
                                      goitre
                                      prevalence.

Table 5. Unit conversion chart (for fluoride concentrations)

Standard   Equivalent    Recommended fluoride concentration
unit          unit          in drinking water (converted)

1 mg/L        1 ppm                  0.7-1.5 ppm
             0.05 mM              0.035 mM-0.075 mM
           50 [micro]M         35 [micro]M-75 [micro]M