by Marion Nestle
Apr 28 2016

The Guardian’s article on the “sugar conspiracy”

I mentioned yesterday that whenever something comes out saying that “everything you thought you knew about nutrition is wrong,” it’s a sign that some skepticism may be in order.

Here is another example: The article on the “sugar conspiracy“ by Ian Leslie published in The Guardian.  This strongly criticizes the work of Ancel Keys, whose work was largely but by no means exclusively responsible for the diet-heart hypothesis linking excessive intake of animal fats to heart disease risk.

I love conspiracy theories as much as anyone else and appreciated how the author made his case for this one.  My sense of his article was that it had grains of truth (Gary Taubes and Cristin Kearns, for example, report that Keys had funding from the sugar industry).  But the overall thrust of the article seemed excessively hyperbolic and based on selective picking of the data (cherry-picking).

Going through the piece line by line to identify errors and misinterpretations was not something I thought worth the trouble.

Fortunately, someone else did.

Katherine Docimo Pett, a master’s degree candidate in biochemical and molecular nutrition at Tufts University, who blogs as Nutrition Wonk, sent me her detailed critique of the paper.  She explains:

So I decided to look into the Seven Countries Study and I found a number of occasions where “The Sugar Conspiracy” misinterprets the evidence.  So buckle yourselves up, conspiracy theorists, because in this post, I’m going to cover the history of the diet-heart hypothesis – namely The Seven Countries Study and the subsequent research mentioned in “The Sugar Conspiracy.”

If you can wade through her lengthy analysis, you will be hard pressed to disagree with her conclusion:

In “The Sugar Conspiracy,” the author makes a lot of assumptions about intent, the usefulness of epidemiology, and even the conclusions of papers.  However, upon closer inspection, a lot of his evidence doesn’t hold up under scrutiny.

The idea that Keys claimed in his Seven Countries Study that correlation proved causation is false.  Keys just said that cholesterol is a mediator for heart disease and that saturated fat raises cholesterol, both of which later turned out to be true.  The Menotti “reanalysis” did not find that sugar is more closely correlated with heart disease than fat, and even if it did, it is a simple regression – it controlled for zero confounders, way fewer than were controlled for in the original Seven Countries Study.  Finally, there is plenty of evidence to suggest that substituting saturated fatty foods in favor of unsaturated fats is a good idea [Clarification: she must mean substituting unsaturated for saturated].

It is absolutely worthwhile to debate the merits of all scientific findings or even the merits of an entire field, like epidemiology.  Scientists, even nutrition scientists, do this all the time.  The problem, though, is that if basic facts are actually based on misinformation, you can’t build a real case for or against anything.


Footnote: Sarah Tracy, an historian at the University of Oklahoma, has been working on a biography of Ancel Keys for quite some time.  I can’t wait for her to get it done (she probably can’t either), as it is likely to give us a thoughtful, balanced account of the significance of his work.

  • Do you think this sugar conspiracy is from the vending machines at school? Is the sugar conspiracy derived from sugary soda pops? Or is just derived from a combination of things?

  • George

    I agree that, if Keys is to be posthumously pilloried for the low fat fiasco, then we should be more respectful of his intellectual processes. These also include the signs that he may have modified his ideas later in life – he didn’t necessarily support a lowering of cholesterol by diet in the general population by the use of megadose polyunsaturated fat.

    There are many facts that contradict the cholesterol hypothesis, and books have been filled with the more technical ones, but here are some broad observational “paradoxes”.

    – The prediction mentioned by NutritionWonk, that France would experience a rise in heart disease, because more saturated fats were being eaten didn’t come true.

    – Saturated fat intake rose in Japan, and cholesterol rose with it, but heart disease declined after 1970 just as it did in the US.

    – When Keys first formulated his hypothesis he measured butter consumption in the USA as accounting for 4.8% of fat intake. Plainly butter was not the cause of the epidemic of CHD in the USA at that time.

    – Recent epidemiological studies such as MESA and EPIC-Netherlands show that a higher consumption of saturated fat from dairy is associated with less heart disease, not more. The Malmo Diet and Cancer Study and recent US biomarker studies also show that dairy fat is associated with a reduced incidence of type 2 diabetes. Dairy fat is the most saturated fat in traditional use – more saturated than any meat fat. Low-fat dairy is not so clearly beneficial.

    – Higher vs lower cholesterol has been associated with heart disease in people who have been eating exactly the same diet for some time. Yet in other populations people who have heart attacks have lower cholesterol than people who do not.

    Is there an explanation for these facts?

    Maybe there are things that tend to, on average, raise cholesterol that do cause heart disease. Smoking, genetic conditions like familial hypercholsterolaemia, lead poisoning, vehicle exhausts, persistant organic pollutants, industrial trans fats, obesity, sugar, and so on. Rates of heart disease will rise and fall with exposure to these factors (whether they raise cholesterol in an individual case or not).

    And then there are things that tend to, on average, raise cholesterol but do not cause heart disease. This would include foods like butter, dark chocolate, and salmon.

    To confuse things further, an increase in saturated fat to 20% of energy does not cause a rise in the average cholesterol or LDL level in a population eating a low carb diet (i.e. where fat including saturated fat has been substituted for carbohydrate), so the claim that saturated fat elevates cholesterol does not represent a universal physiological truth. (Tay et al 2008).

  • It was disturbing that many reported the “Sugar Conspiracy” as excellent journalism!

    Thanks for highlighting Katherine’s piece. You’ll also find an excellent (and complimentary) look at the Yudkin’s story by Kevin Klatt here:

    and I took on the sugar conspiracy’s take on the dietary guidelines and reliance on information by Nina Teicholz/Gary Taubes here:
    Sheila Kealey

  • My mother always told me that stop the consuming sugar. but I can’t

  • Always go with mom she is right, but is consuming sugar an addiction for you?

  • normally I used sugar with tea, but tea is an addiction to me

  • Tea is not bad at all. For example, I love Arizona Green Tea, because it’s barely sweetened. But however, the monster is the sugar in it, do you know about sugar alternatives?

  • KentComments

    Arizona Green Tea contains 18 g sugar per 8 oz, and the bottle is 16 oz, so 36 g of sugar — about 9 teaspoons’ full. I’d hardly call that “barely sweetened.”

  • But true, you can’t judge a book by it’s cover. It’s a healthy and safe alternative to a sugary coca cola. I didn’t say you have to consume one Arizona tea everyday, but also, Arizona tea is a great antioxidant.

  • KentComments

    Unfortunately, studies consistently find that bottled green teas’ antioxidants are rapidly degraded while sitting in solution on the shelf:

    In particular, “Honest Green Tea had 27.2 mg of EGCG, and AriZona Green Tea—Ginseng & Honey had only 5.4 mg. Snapple Diet Green Tea had even less—just 3.5 mg.”
    Sweetened, bottled tea is really not meaningfully healthier or safer than soda: it’s all really just sugar and water. The smidgen of residual polyphenols in green tea to give it a health halo, but no real health benefits.

  • Now, EGCG according to research done by me is a good precursor for preventing inflammatory attack on the cells in the immune system. Now polyphenols at some concentration can be unhealthy if denatured and mixed with the wrong foods have some biomechanism effects on the immune system.

  • no, can you plz tell me about them ?

  • Let’s try one. Have you ever heard about Stevia?

  • Pingback: The Guardian's article on the "sugar conspiracy" - The VenusFactor.Reviews Weight Loss Blog()

  • Inc.

    REALLY? Barely sweetened? How old are you 10?

  • Inc.

    Yo really have no clue. The EGCG must be in certain dosages to do any good and must be absorbed by the body. A new born baby is not getting enough if they were to drink Arizona bottled or canned green tea.

  • Why would you come on this site then and provide a non academic response, actually no, I am adjunct from UIC! Keep it academic on my threads!