Famous Low-GI scientists skewered on shonky sugar study

[rory robertson]

Good morning. I am arguing near and far for the retraction of a deeply flawed paper - The Australian Paradox - by Dr Alan Barclay and Professor Jennie Brand-Miller.

Background: The University of Sydney scientists last year claimed a profound scientific observation: "an inverse relationship" between sugar consumption and obesity, the Australian Paradox! Yet their four sets of valid sugar indicators trend up not down in their own charts (see Figures 1-4 in http://www.australianparadox.com ). Huh?

Anyway, the authors were skewered this week by a widely respected journalist who documented less-than-ideal scholarly conduct (shall we say) on the way to their published rebuttal of my critique: http://www.smh.com.au/business/pesky-ec ... z21hP5IgYR

Did you notice the slippery disappearing made-up false claim of up to 14kg per person per annum worth of raw sugar being consumed by ethanol production. The significance of that claim is that it tended to shift the trend for (human) consumption in Figure 1 from up to down.

But the made-up excuse "crashed and burned" because the real answer is zero. Ouch! (The key "Ethanol/Cars consumed a big chunk of the available sugar" excuse is found at the bottom of page 2 at http://www.australianparadox.com/pdf/RE ... ERTSON.pdf while the authors' published rebuttal - sans key argument - is at http://www.australianparadox.com/pdf/nu ... 1-s003.pdf )

So, it turns out that the so-called Australian Paradox is not even a puzzle - it's just plain wrong, because sugar consumption and obesity have tended to move in the same direction.

In my opinion, THE key question is WHY the exact moment that Dr Alan Barclay and Professor Jennie Brand-Miller agreed with each other that it was time to quietly RETRACT without acknowledgement their made-up false claim about ethanol production/cars consuming big chunk of the available sugar, somehow was NOT also the right moment to agree to formally correct or retract their entire error-ridden paper?

The unreasonable delay in removing the clearly false claim of "The Australian Paradox!" - an extraordinary "inverse relationship" between sugar consumption and obesity - from the scientific record increasingly has reasonable people wondering when a series of inadvertent errors deliberately left uncorrected becomes scientific misconduct? (http://en.wikipedia.org/wiki/Scientific_misconduct ) Any thoughts, anyone?

Best wishes,
Rory Robertson (former fattie)

 

[borofergie]

More great work Rory, it looks like you really have them on the ropes...

Here is what happens if you plot sugar consumption (FAO data) vs Obesity (BMI>25, WHO data) for every country in the world:

Now I know from your work that the FAO data is kinda dodgy, but the fact that you can get a strong correlation from a dirty dataset probably suggests that the correlation between real sugar consumption and obesity would be even stronger.

What this says is that 46% of the varition in obesity can be predicted* by a nation's sugar consumption alone (Pearson Correlation Coeff 0.67). If I add wheat consumption into this, to make a "refined carbohydrate index" the r2 value goes above 50%.


*correlation does not necessarily imply causation, yada, yada, yada.

 

[dawnmc]

The graph looks rather like the original Ancel Keys study on fat consumption before he doctored it to look more like his hypothesis.

 

[rory robertson]

borofergie...I hope you are well. It's surprising we still keep being told in Australia that sugar is not a particular problem. Blind Freddie and all that.

So I've read Gary Taubes's Good Calories, Bad Calories since last we spoke, so I now "get" better where you are coming from (and agree).

Anyway, I can tell you that securing a retraction is a bit like watching the grass grow: you assume it eventually will happen but progress seems glacial. So I'll just keep chippin' away.

I was thinking on the topic of formal "Retraction" the other day and came across a piece in the prestigious science journal Nature. It noted that “It is reassuring that retractions are so rare, for behind at least half of them lies some shocking tale of scientific misconduct — plagiarism, altered images or faked data — and the other half [sic] are admissions of embarrassing mistakes. But retraction notices are increasing rapidly. In the early 2000s, only about 30 retraction notices appeared annually. This year, the Web of Science is on track to index more than 400...” (http://www.nature.com/news/2011/111005/ ... 8026a.html ) .

So formal retractions are rising rapidly, and there is plenty of room for one more! The article observes that “Perhaps surprisingly, scientists and editors broadly welcome the trend.” In any case, what was fascinating is that in the “Comments” section below that online story, I stumbled across a distinguished scientist at the University of Sydney who'd had a terrible time trying to maintain the integrity of the scientific record in his field, as the ignored referee of what he viewed as a deeply flawed paper:

“The paper in question was scientifically flawed, grossly distorted the literature in the field clearly to progress the authors' stated personal agenda [relating to male circumcism]. The statistical analyses were erroneous. I recommended rejection, but after being sent a revised version in which virtually none of my extensive list of initial criticisms was addressed I stated that extensive corrections were required. When it then appeared on-line in advance... The paper in question should be retracted. I can understand that the Editors might be embarrassed that this paper slipped through in to print. But their failure to take any steps to date to address the situation that has arisen now leads to an impression of complicity. Is there a higher authority individuals such as me can turn to for advice?” (http://www.nature.com/news/2011/111005/ ... 8026a.html ).

That sounds familiar! No, Professor Brian J. Morris wasn't one of the referees of Australian Paradox before publication; they were either non-existent, incompetent or - as in Professor Morris’s case - ignored. It's good to know, however, that at least one distinguished scientist at the University of Sydney is still fighting the good fight (http://www.physiol.usyd.edu.au/~brianm/ ). Where are the rest?

Does anyone else also find the contrast appalling: one University of Sydney science Professor devoting himself to preserving the integrity of the scientific record, while just across the way at the University of Sydney there's another with a sloppy, factually incorrect paper refusing to correct the scientific record?

How would rubbish papers like Australian Paradox get published in the first place? After all, I have been advised that the fact that one of the authors also was the "Guest Editor" of the relevant "Special Issue" of the journal would not have compromised quality control.