Government Guidance on Consultations for Fluoridation Schemes
Readers of the Chief Dental Officer's 'Dear Colleague' letter of guidance on Fluoridation of Drinking Water in February 2008 (Gateway 9361) should be aware of existing scientific challenges to the Government's interpretation of the evidence on fluoridation. The letter as it stands gives a one-sided account of the position.
The principal challenges are in the following places:
- A. NHS Centre for Reviews & Dissemination website What the 'York Review' on the fluoridation of drinking water really found. www.york.ac.uk/inst/crd/fluoridnew (October 2003).
- B. Article by Wilson PM, Sheldon TA. Muddy waters: evidence-based policy making, uncertainty and the "York review" on water fluoridation. Evidence & Policy 2006;2:321-31.
- C. British Medical Journal article by Cheng KK, Chalmers I, Sheldon TA. Adding fluoride to water supplies. BMJ 2007;335:699-702. (6 October 2007)
- D. Letters in response to C. by Professors Stephen T. Holgate and Trevor A. Sheldon ('Interpreting the Newcastle fluoride bioavailability study') and Earl Baldwin of Bewdley ('Addressing the arguments'), responding to the Chief Dental and Medical Officers ('The Department of Health's view'), BMJ 2007;335:840-1. (27 October 2007)
- E. Nuffield Council on Bioethics report Public health: ethical issues, sections 7.43 - 47. www.nuffieldbioethics.org/go/ourwork/publichealth/ introduction (November 2007).
A. was published because the authors of the NHS CRD's 'York review' to which CDO refers in his paragraph 2 were "concerned about the continuing misinterpretations of the evidence" for fluoridation, and wished to republicise their findings. Similar points had been made by Professor Sheldon, the founding director of the CRD who chaired the York review's advisory board, in an open letter of 3 January 2001, and by the four independent scientists on the advisory board of the York review in a letter to the responsible Health Minister of 11 December 2002. (These letters can be seen at www.appgaf.org.uk.)
B. is a peer-reviewed article by a member of the York review team and the chair of the advisory board which oversaw the review. In the section The public response (325 - 7) the authors state that "The 'neutral' researchers on the York review's advisory group ended up thinking that there was insufficient scientific evidence to make a definitive judgement on the benefits and harms of water fluoridation."
They draw attention to the inferior evidential status of the Medical Research Council working group (2002) referred to by CDO in Appendix 1. "Because it gave the impression that the evidence on the impact of fluoride on health inequalities was 'fuller and firmer' that it actually was, the MRC report has often been referred to by politicians and policy makers as an evidence review." And they criticise CDO's claim in the then guidelines for local consultations (2005) that (in their words) the report produced by a pro-fluoridation lobby group [viz. the British Fluoridation Society's One in a Million] was "a very useful source of information".
C. merits reading in full, as it deals with a number of problems surrounding a fluoridation policy. K.K. Cheng is professor of epidemiology at the University of Birmingham; Sir Iain Chalmers was founding director of the UK Cochrane Centre; Trevor Sheldon is professor of Health Services Research at the University of York. The last two were on the advisory board of the York review.
They say "We are concerned that the polarised debates and the way that evidence is harnessed and uncertainties glossed over make it hard for the public and professionals to participate in consultations on an informed basis." They include the Department of Health among bodies whose "objectivity is questionable" over fluoridation.
D. contains strong criticism of the small Newcastle study, relied on by CDO in Appendix 1, 4, by Professor Sheldon and Professor Holgate who is MRC clinical professor of immunopharmacology at the University of Southampton, and expresses surprise at CDO's support for it. Lord Baldwin is an opponent of fluoridation who served on the York advisory board.
E. is a report by a working group chaired by Lord Krebs FRS which examined questions of ethics, science and policy over five public health areas. In its fluoridation section it voices concern over the "potentially misleading account" of the state of the evidence on fluoridation in the publication of One in a Million (2004) by the British Fluoridation Society, the UK Public Health Association, the British Dental Association and the Faculty of Public Health. It also criticises opponents of fluoridation.
While not all these sources criticise the Department explicitly, its line has consistently been to overstate the evidence for fluoridation's effectiveness and safety. Comparison between, for example, the NHS CRD's October 2003 website (A. above) and CDO's view of the evidence contained in his February 2008 letter will show the latter's shortcomings.
Some further points arising from CDO's letter should be made.
1. It is misleading to claim that York "concluded" anything about the effects of water fluoridation (paragraph 1). The reviewers were careful with words, and stressed that nothing could be stated confidently since the "best available evidence" was of moderate to poor quality: "To have clear confidence in the ability to answer the question [on effectiveness] the quality of the evidence would have to be higher" (York 4.9). "The degree to which caries is reduced . . is not clear" (York, Executive Summary, Objective 1; 15% (actually 14.6%) is only a median figure within a possible range of (dis)benefit of 64% to -5%). "The research evidence is of insufficient quality to allow confident statements about other potential harms [than dental fluorosis] or whether there is an impact on social inequalities" (York, Executive Summary, Conclusions). Such confident statements would obviously include assurances of safety.
2. CDO does not mention dental fluorosis, the one reasonably clear harm from fluoridation. Mottling of teeth can be distressing, and York assessed its prevalence in fluoridated areas at nearly 1 child in 2, and 1 in 8 for fluorosis of aesthetic concern.
3. CDO's claim of "accumulating evidence" of benefit to adults should be treated with caution. The reference (2) is to one study, and its quality can only be guessed at. It should be remembered that York in 2000 could not find one reliable study in 50 years of the world's literature on water fluoridation, and these included recent studies. Evidence for adults was negligible: "Only one study addressed the positive effect of fluoridation in the adult population" (York 12.9.1).
4. CDO's paragraph 4 has poor evidential value. No reliable study has shown fluoridation to be the cause of better teeth in one UK area than another: see York for the absence of acceptable evidence on Sandwell and Bolton. This would need high-quality studies that controlled for bias and confounding factors. Claims not based on good science are out of place in public health measures such as this.
5. Claims for "no evidence" also need scrutiny. York found evidence for a number of possible harms including bone problems, cancer and Down's syndrome (see York 8, 9 & 10). But it was poor evidence, matched by equally poor evidence in the other direction. For example, "Whilst there were 11 analyses that found the direction of association of water fluoridation and cancer to be positive (fewer cancers), a further nine analyses found a negative direction of association (more cancers), and two studies found no effect" (York 9.6).
CDO's claim in paragraph 5 "that was [sic] no evidence of any risk to health" therefore stands in contrast to what York reported, which essentially was "no clear association". Other statements by the reviewers included: "Some possible adverse effects . . may take many years to develop and so . . the relationship may go undetected" (York, 1); "High quality research [into adverse effects] . . is needed" (York 12.9.2); "The outcomes related to infant mortality, congenital defects and IQ indicate a need for further high quality research" (York 12.4). The senior York reviewer wrote in a letter to the British Medical Journal on 16th June 2001: "We have been assiduous in our paper, our full report, and our contacts with the media, not to convey a message of no evidence of harm" (BMJ 2000;322:1487-88).
The Department of Health has subsequently conceded this point, at least as far as cancer is concerned, in a written parliamentary answer (Lords Hansard 20 May 2009, WA 323-4).
6. Appendix 1 of CDO's letter addresses the MRC working group report of 2002. MRC conducted a 'narrative' review of the evidence on the way to making recommendations, and in some areas attempted to second-guess the systematic review by York which in its membership, transparency, thoroughness, and the rigour of its systematic approach and assessment criteria was of an altogether higher calibre. This is a point of the greatest importance whenever the two reports differ on fact or emphasis, which they frequently do. Although Government has acknowledged in principle the primacy of York in parliamentary answers (House of Lords, Hansard 30 January 2001, WA 55), in practice Government and other bodies that favour fluoridation prefer the MRC report. Based on what York found, statements on safety in CDO's Appendix 1 paragraphs 1 and 6 cannot be supported.
7. Paragraph 4 of Appendix 1 seriously misrepresents the Newcastle evidence. Reference to the conclusions in section 6 of the published report, Bioavailability of fluoride in drinking water - a human experimental study (June 2004), will show that all caveats and cautions in that section and section 7 have been omitted from CDO's letter, which makes categorical claims for "no evidence for any differences". This letter needs to be read in conjunction with the first BMJ letter referred to in D. above, in which Professors Holgate and Sheldon comment inter alia that "Discarding an outlier (removing 5% of the data) to eliminate an "inconvenient" significant result is not best practice". Although greatly underpowered, the Newcastle study found interesting suggestions of a possible significantly greater effect from water which had been artificially fluoridated. Even the Department of Health in a written parliamentary answer (House of Lords, Hansard, 7 September 2004, WA 137-8) conceded that differences had been shown, reflecting a later version of the study's conclusions (J Dent Res 84(11): 989-993, 2005) based on a different analysis but curiously omitting any reference to the earlier report.
8. It is surprising that CDO in Appendix 2. 21 still relies on the document One in a Million in the light of B. and E. above, and on the British Fluoridation Society's website. The partiality of the BFS can be further gauged from written parliamentary answers confirming its "promotional role" in fluoridation, and the need for the Department to correct the BFS's emphasis in briefings (House of Lords, Hansard, 31 January 2001, WA 66; 12 February 2001, WA 16). The BFS website has a revealing link to the National Alliance for Equity in Dental Health which lists the dozens of PCTs and other public health and dental bodies, many of them likely to play leading roles in consultations on fluoridation schemes, which signed up some time ago as campaigning organisations for fluoridation. Cheng at al's concern in C. above is well founded. Like many writings which claim to tell "the facts", One in a Million is a promotional document and should be read as such.
9. The National Fluoride Information Centre is also commended by CDO in the same paragraph. Although the Department seemed minded to agree to the request from senior scientists involved with the York review to entrust assessments of fluoridation evidence to the Cochrane Collaboration, which has international standing and lies outside the dominant fluoridation culture exemplified on the BFS website, in the end the task went to a body in Manchester with dental involvement. When the NFIC published its website version of York's findings the NHS CRD wrote to suggest changes to reflect more accurately what their reviewers found. These changes were not made, and the website still contains errors to the benefit of fluoridation, and not just of emphasis, which are unfortunate in view of its claim to be academically independent and provide objective information.
10. The core problem is that the York review changed the fluoridation landscape, yet old beliefs remain in place. Fluoridation, as the Australian Dental Association pointed out, is in a sense dentistry's flagship, and has meant too much for too long for health professions to give it up lightly. In consequence their partiality is as marked as that of many of their opponents. Brian Martin's book Scientific Knowledge in Controversy: The Social Dynamics of the Fluoridation Debate (State University of New York Press, 1991) is informative, setting out a social scientist's analysis of the cultural and political forces that have produced passionately held but conflicting views on this topic, little affected by scientific evidence.
The Department, though not always the worst offender, has been consistent in playing up the unproven benefits and playing down the uncertain harms of fluoridation. None of the important critical references in the present letter, all drawn from mainstream publications, was cited in CDO's February letter. What is crucial for decision-makers is that they should address the totality of the evidence, drawing on the best sources, and not accept the Government or anyone else's view without careful scrutiny.
11. Two independent medical researchers, Sir Iain Chalmers and Professor Trevor Sheldon, who were on the advisory board of the York review, have read and endorse the content of this critique.
All-Party Parliamentary Group Against Fluoridation