Tag Archives: O’Brien Institute

Scientific integrity requires critical investigation – not blind acceptance


Some people seem to want to close down any critical discussion of the current research into the relationship between water fluoride and child IQ. They appear to argue that claims made by researchers should not be open to critical review and that the claims be accepted without proper consideration of the data and evidence.

Anti-fluoride campaigners, of course, argue this way any time the research they promote is questioned. After all, they have a bias to confirm and an ideology to support and rely on claims that often don’t stand up to proper consideration. I expect that, but I am concerned to hear these arguments from scientific reviewers of this research.

In the video above Dr William Ghali of the Canadian O’Brien Institute for Public Health counters critiques of some research with the comment: “the studies can’t be undone and they can’t be unpublished.”

Of course they can’t – but they can, and should, be critically considered – not blindly promoted as the best things since sliced bread. Critical consideration is, or should be, the normal scientific reaction to newly published studies.

Dr Ghali is one of the authors of a recent review of the science around community water fluoridation, COMMUNITY WATER FLUORIDATION: A REPORT FOR CALGARY CITY COUNCIL. He made the above comment during his presentation to a recent meeting of the Calgary City Council – the video above contains a section of his presentation (selected and promoted by the Fluoride Action Network (FAN) an anti-fluoride activist organisation).

I am amazed at that comment – and other comments of his. I could understand if he was responding to the research critiques by explaining where they were mistaken or misinterpreted the evidence – we should always consider the factual evidence in our scientific discussions. But he seems personally upset that anyone should pursue a normal scientific critical discussion. He admits to getting angry at:

“The notion that you can just talk away 10 years of research.”

And

“I respect the doers of the research and the deliverers of the evidence and don’t think they should be shot for tough messages.”

Yet he himself is denying respect to scientists who critically discuss research (by authors that he appears to be protective of) and is attempting to “shoot” down researchers who discuss the problems in that research. He accuses scientific critics of attempting to  “simply sweep aside scientific findings because one disagrees with the results.” Yet he attempts to “sweep aside” the normal scientific critique of research – rather than deal specifically and factually with the criticisms themselves.

Sometimes it is necessary to “talk away 10 years of research” if the critical scientific consideration of the research findings show them to be faulty. We are talking about science – not religion.

Nothing sacred about scientific findings

There is nothing sacred about scientific findings – they are and always must be open for critical consideration and critique. Publication in a reputable journal and inclusion of big names in the author list is no guarantee of good science. And all scientific findings must be considered as provisional – most of what is considered factual in science often turn out to be wrong, at least in part. This is how science progresses and critical analysis and scientific critique of published work is key to that development.

Critique of published research is vital and it should never be ignored, “swept aside” or discredited by saying things like “once published it can’t be unpublished” of referring to critiques as “sweeping aside because one disagrees.” Good scientific critique is not swayed by authority or author’s claims but looks at the data, findings and interpretations – critically. It is not an evidence-free “sweeping aside.” In a good-faith open scientific exchange, the response to criticism should be the same.

Having said this I can understand a little of what is driving the two people in their comments in the above video – comments that are critical of scientific commenters but ignore the way the anti-fluoride movement has misused and misrepresented this research. The O’Brien review they were authors of was roundly criticised for its weaknesses when it was made public. On the other hand, the anti-fluoride advocates lavished it with praise – for these very same weaknesses.

Scientists are human (actually very human) and, of course, sensitive to criticism. Even the best scientist will often react defensively and attempt to discredit critics rather than deal with the contents of the criticisms.

Misrepresentations

The only time Dr Ghali gets at all specific in this video section is in his criticisms of the letter sent by 30 academic and health experts to the US National Insitute of Environmental Health Science (NEHS) about the study (see Experts complain to funding body about quality of fluoride-IQ research). This letter expressed concern about the study recently published by Green et al (2019) listing a number of specific scientific limitation of the study (see If at first you don’t succeed . . . statistical manipulation might help and Experts complain to funding body about quality of fluoride-IQ research)  The letter also expressed concern about the poor statistical reporting of the data and lack of transparency regarding methodology.

After listing ten scientific concerns the experts made a specific request:

“We urge NIEHS to ask the Green authors to release their RIF data set and provide a thorough explanation of their analytical methods. Doing so could enable an independent review that would bring clarity and ensure the scientific record is accurate.

Should the Green researchers not voluntarily release their data, please advise us on what the process would be to have the data set released so an independent analysis of the Green data can be conducted.”

But this is how Dr Ghali specifically commented on this important expert’s letter:

“Twenty or so North American academics [actually 30 North Americans and experts from the UK and Australia] wrote to the NIEHS denouncing the recent Canadian study critiquing it on many levels  Making assertions the team at York University refused any access to their data and their refusal to permit reanalysis and they are not being transparent. The allegation is false. The authors are in fact  in an active process of discussing with health Canada a Teflon bias-free process of making the data available for a secondary analysis. And again, there is one thing that gets under my skin are assertions, attacks on messengers.” [My emphasis]

Come off it. On the refusal to make data available the expert’s letter mentions only:

“In recent weeks, at least two of the Green authors have declined to respond affirmatively to requests from other researchers for access to the data and analytical methods they used.”

It did not “denounce” the study (scientific critique is not “denouncing”), and it definitely did not assert the whole team was refusing any access. It simply pointed out that no one at that stage had reacted positively to the request for access to the data. That is not, as Dr Ghali claims an “attack on messengers.” Nor is it, as he claims, a “false allegation.”

A respectful and scientifically ethical response to the expert’s letter would be for  Dr Ghali to consider and respond to the list of ten limitations of the study described in the letter. But instead, he has misrepresented the letter and made a false allegation himself regarding the request for access to data.

Where is the scientific integrity in that?

As an aside, I am a bit cynical about the authors’ claim that they are “discussing with health Canada a Teflon bias-free process of making the data available for a secondary analysis.” Dr Ghali appears to be in more intimate contact with the authors than the rest of the scientific community because this is the first I have heard of the authors’ response. But I fear the “Teflon bias-free process” referred to may, in the end, be a bureaucratic solution which makes the data available to only a select “trusted” few for their presumed approval.

The problem of transparency

Dr Ghali also misrepresents the letter by claiming it accuses the authors of lack of transparency. Yes, it expresses concern about the lack of statistical and methodological information but refers to this as a general problem in scientific publications, particularly where statistical analyses are involved. It even cites a published paper on this (Prager et al. 2019: Improving transparency and scientific rigor in academic publishing. Brain Behav. 9(1): e01141).

Another example relates to reliance on p-values:

“The American Statistical Association has established six principles on the use and analysis of p-values, one of which states: “Proper inference requires full reporting and transparency.” By releasing the data and a detailed explanation of their analytical methods, the Green authors would enable the scientific community to better assess whether their choice of p-value was appropriate.”

All this is simply part of a good-faith scientific critique which should be normal in science and should never be squashed or prevented. Remember Ghali himself said messengers should not be shot for delivering a message.

But if we are to discuss the problem of transparency I am really concerned at the unwillingness of the authors, and their scientific defenders, to participate in a free good-faith scientific exchange on their findings.

I guess they can not be blamed for promoting their own research while being silent about its limitations, or for the fact that the journal which published their paper has a policy of not publishing any critiques of published paper after 4 weeks. But why should they promote their findings on social media but refuse to enter into any discussion on it?

For example, Rivka Green, the first author of the paper, opened a Twitter account where she promoted the paper. But when some discussion of the limitations started she withdrew and closed the account down.

In another example, a biostats PhD student at Pittsburgh university was making some general comments about the data in the Green paper on Twitter soon after its publication. But two of the authors approached his university department and supervisors and he was forced to delete his tweets. (This is information from the student  himself who is wary about going public because of this unpleasant exposure to academic politics and he is unsure of the consequences of making further comments).

I have had personal experience of the lack of transparency by Dr Chrsitine Till’s group (involved in the study reported by Green et al. 2019) and its supporters. My own critique of one of the early papers from the group (Malin & Till 2015) was denied consideration for publication in the publishing journal by the Chief Editor, Prof Grandjean, who publicly identifies with the group and the anti-fluoride movement (see Fluoridation not associated with ADHD – a myth put to rest). My critique was eventually published in another journal: (see Perrott 2018: Fluoridation and attention deficit hyperactivity disorder a critique of Malin and Till (2015). British Dental Journal, 223(11), 819–822). Christine Till is aware of this critique but purposely ignores it whenever she or her coauthors cite Malin & Till (2015) in their publications (see, for example, ADHD and fluoride – wishful thinking supported by statistical manipulation?).

And what about the lack of transparency displayed by Dr Ghali himself. He misrepresented the expert’s letter – but was also very selective in referring to other reviews of this study. For example, in the video above, he mentioned the CADTH (Canadian Agency for Drugs and Technologies in Health) review  on possible neurological effects of fluoride which was very critical of the Green et al. (2019) paper and quotes from two sections of the review which said:

“The evidence is weak due to multiple limitations  . . ” (p 5)  and “further well conducted research is needed to reduce uncertainty. ” (p 14)

But he ignores completely a more damning statement in the CADTH review which says:

“The study by Green et al., 2019 concluded that “maternal exposure to higher levels of fluoride during pregnancy was associated with lower IQ scores in children aged 3 to 4 years.” (p. E1) This conclusion was not supported by the data” (page 12)

A disclaimer

I am very conscious that I have relied on only one section of Dr Ghali’s presentation to the Calgary City Council. And that this section was cherry-picked by FAN to present him as an ally in their anti-fluoride campaign. I have not had the time to look at the full video of his presentation yet – it is available on YouTube: Dr. Ghali (O’Brien Institute) – Full Calgary Presentation on Fluoride. However, I think the comments made on this specific section of his presentation stand by themselves and needed a response.

Dr Ghali may well have made criticisms of the misrepresentation of this research by FAN, by the anti-fluoride campaigners also presenting to the Calgary City Council and by anti-fluoride campaigners in general. After all, FAN, which made the selection for this video and is promoting it is hardly likely to include such criticism.

So to be fair to Dr Ghali and to support the proper good-faith scientific exchange I am talking about I will email him and offer him the right of reply to this article.

Similar articles