By Ken Perrott 07/02/2017

It seems every time anti-fluoride propagandists present data it is either cherry-picked, distorted or misleading. Often all three. So it is hardly a surprise to find local anti-fluoride propagandists are telling porkies again.

fluoridated vs non-fluoridatedThey have been promoting this graphic claiming it shows people in “non-fluoridated Christchurch have “better teeth.” But the graphic is based on naive cherry-picking of the data, it ignores the effect different ethnic groups have on the data and it uses a single cherry-picked year which fits their bias.

On top of that, axis values have been chosen to exaggerate differences and the labels are incorrect. The “non-fluoridated Christchurch” category uses data for Canterbury and the “fluoridated Hamilton” category uses data of the Waikato.

It seems that several of the commenters on the Fluoride Free NZ Facebook page where this graphic was first used saw the problems and raised them. All they got is insults for their time. These organisations do not seem capable of a rational discussion.

The Ministry of Health data they use is freely available on the MoH website. It provides oral health data for 5-year-old children and year 8 children. The data is presented annually and for different regions.

So let’s have a look at what the data really says – using more normal axis ranges and separating out ethnic groups.


The top graph here is still misleading because it does not take into account the effect of different ethnic groups. However, the correct categories are used and the more rational axis really cuts the exaggerated difference down to size.

In the second graphic the data for Māori and Pacifica have been removed – the MoH describes this group as “Other” – it is mainly Pakeha. We can see that the caries-free % is actually greater for fluoridated Auckland than it is for non-fluoridated Canterbury – exactly the opposite of what the anti-fluoride propagandists were claiming.

It is the same story for Māori – the caries-free % is actually higher in fluoridated Auckland than in non-fluoridated Canterbury.

The problem with the “Total” data is that Pacifica have a large effect – particularly in Auckland where Pacifica are concentrated. Pacifica generally have poorer oral health but are concentrated in fluoridated regions. This drives down the caries-free % figures for the fluoridated areas if the differences are not accommodated.

I referred to this effect of Pacifica on the data in my article A challenge to anti-fluoridationers to justify their misrepresentation of New Zealand research. There I was referring to a similar way anti-fluoride campaigners were misrepresenting data from recent New Zealand research. In this case, they were using data from a paper (Schluter & Lee 2016) and completely ignoring the distortions introduced by inclusion of Pacific – even though the authors had warned against the anomaly introduced by this.

There are other effects which should also be considered in a proper understanding of these data. It is easy to cherry-pick the data for a single year when differences are small – the anti-fluoride people do that a lot. OK if you want to confirm your biases but consideration of the data over multiple years helps indicate trends, identify anomalies and provide an idea of variations in the data. It is also important to consider the numbers in each region. For example, I have not included Pacific in the graphs above because they are concentrated in Auckland and the numbers in Canterbury and Waikato are very low (eg., 45 in Waikato in 2014).


Site Meter