Read Original Story

Is sugar “definitely toxic”?


5 Star


Sugar Is Definitely Toxic, a New Study Says

Our Review Summary

sugarThis news story reports that a diet study, conducted on 43 overweight minority children and teenagers, is believed by the researchers who did it to definitively prove that sugar — in and of itself — is metabolically toxic regardless of how much or little of it is eaten, and irrespective of the weight of the eaters. It does a solid job of describing the design of the study and discussing why the evidence it provides is qualitatively different from previous research. But we thought the framing of the story was off kilter, as it leads with lots of glowing discussion about the strength and importance of the research, while putting important caveats and limitations far down in the piece. We also think the story could have been more thorough in its discussion of those limitations — including the fact that 43 is a small number of participants, the study was very short in duration, and it had no control group. But while there was room for improvement, readers who make it to the end of the piece should come away with a reasonably thorough understanding of the study’s findings and what they might mean.


Why This Matters

It’s hard to imagine that any American family could be completely uninformed about the health-related risks of consuming high fructose corn syrup and table sugar (sucrose), major components of sweet drinks, and a supermarket full of processed foods. But as the article points out, it has been difficult to design studies that show that such “added sugar” is an independent toxic substance even when total calories and weight stay the same. That difficulty, in turn, has made it hard to say with certainty whether people who reduced their sugar consumption and were improving their metabolic health (e.g. blood pressure, liver function, cholesterol levels, fat accumulation in the abdominal region) were getting the benefit from less sugar or from fewer overall calories. Given the growing rates of diabetes, fatty liver disease and heart disease in ever-younger populations, adding to evidence that not all calories are equal in harm or benefit is more than welcome by physicians and parents. And studies like this one to add to a growing body of evidence that restricting sugar may be good public health policy. If additional studies, longer in duration and larger in scope are done, this could result in policy change that may help keep everyone healthier.


Does the story adequately discuss the costs of the intervention?

Not Applicable

Many sugar-rich packaged and prepared foods are more expensive than home-made foods.But the study was not designed to compare the costs of diets, and most people know how much different types of food that contain sugar cost. So we’ll rate this Not Applicable. On the other hand, it would have been useful to cite estimates of the cost of treating the disorders linked to metabolic syndrome components, such as high blood pressure, as well as the costs of treating the chronic diseases that are a consequence of the syndrome, such as diabetes and heart disease.

Does the story adequately quantify the benefits of the treatment/test/product/procedure?

Not Satisfactory

The quantification of the results does not quite clear our bar here. The story says this:

Overall, their fasting blood sugar levels dropped by 53%, along with the amount of insulin their bodies produced since insulin is normally needed to break down carbohydrates and sugars. Their triglyceride and LDL levels also declined and, most importantly, they showed less fat in their liver.

We would have preferred more numbers documenting the changes for each outcome, in absolute terms (the “53%” figure is a relative reduction), along with a comment or some description of how meaningful the changes were clinically.

Does the story adequately explain/quantify the harms of the intervention?

Not Applicable

It’s hard to say what “harms” or risks would result from cutting sugar from childrens’ diets. We’ll rate this Not Applicable.

Does the story seem to grasp the quality of the evidence?


The story does a fine job of explaining how the study was done and why the evidence it provides is meaningful. It explains that compared with previous research, which was observational in nature and couldn’t prove cause and effect, the new study suggests that cutting out sugar may be causing the beneficial metabolic changes observed. We do have a problem with the story’s framing, however, which frontloads discussion about how important the findings are, while burying discussion of limitations far down in the text. Experts point out some problems with the study, but we think the story could have gone further — explaining for example that the study was small and time-limited and that it had no control group. Though the story feels slightly unbalanced, we’ll acknowledge that it did make an attempt to critically evaluate the findings — hence the Satisfactory rating.

Does the story commit disease-mongering?


As noted, increased sugar consumption, especially in children, has long been linked to alarmingly high and growing rates of obesity, diabetes and other chronic diseases in children and adults. There is no disease mongering here, and indeed, some numbers (prevalence) would have helped the story.

Does the story use independent sources and identify conflicts of interest?


The story includes restraining comments from two independent experts, although those comments come far down in the story and should have been placed higher. The story perhaps also could have mentioned that Dr. Lustig is author of a 2012 book called “Fat chance: Beating the Odds Against Sugar, Processed Food, Obesity, and Disease,” and so may have an intellectual and financial conflict of interest when it comes to sugar.  The study was funded by the NIH and Touro University, as well as UCSF — another aspect that could have been mentioned for context.

Does the story compare the new approach with existing alternatives?


The article did a fairly good job of describing the limitations of previous dietary studies, compared to this one, especially as it relates to the effects of cutting sugar vs. losing weight. It would have also been helpful to mention the role of exercise in controlling metabolic disease.

Does the story establish the availability of the treatment/test/product/procedure?


The story explains that the dietary changes they made in the study could be made by anyone. The story says:

“We took chicken teriyaki out, and put turkey hot dogs in. We took sweetened yogurt out, and put baked potato chips in. We took pastries out and put bagels in,” says Lustig. “So there was no change in [the children’s] weight and no change in calories.”

Does the story establish the true novelty of the approach?


It’s always risky at best to call something a first of its kind as this story does, without citing any evidence for that or being more specific about what, exactly, was a “first.” But the story does give the general sense of what’s new and important about the study — the fact that it suggests beneficial changes of reducing sugar independent of weight loss. And it also draws a line between previous correlation studies and this experiment. Close call here but we’ll give the benefit of the doubt.

Does the story appear to rely solely or largely on a news release?


There’s enough original reporting here for us to be sure this story didn’t rely on a news release.

Total Score: 7 of 8 Satisfactory

Comments (5)

Please note, comments are no longer published through this website. All previously made comments are still archived and available for viewing through select posts.


October 30, 2015 at 4:29 pm

I would like to see the same tests but with a group of lowfat vegans studied too. I think it is more complicated than sugar=bad, I think sugar becomes bad when the body is dealing with more fat than it is evolved for. I assert that the fat creates insulin resistance and then that starts a cascade of unhealty events related to stress, such as storage of energy as fat. When it comes to health and diet, one can’t look at one thing, everything needs to be looked at in relation to eachother. **And not funded by industry.



    November 2, 2015 at 6:55 am

    Why lowfat vegan instead of high-fat vegan?
    I realize that you say “…when the body is dealing with more fat than it is evolved for.” but do you mean body fat or dietary fat?
    Thus, if body fat amount is your claim of the variable that affects how sugar affects these biomarkers, then lowfat makes no sense, or, alternatively, you could do an arm of lowfat vegan and an arm of high fat vegan.
    If your emphasis is on dietary fat, then utilizing a wide variety of foods (including animal foods) w/o much fat content would test that.
    After all, some did lose weight and some did not.
    Bagels suck, but I guess that they are offering typical American food with the added sugar as the only altered variable.
    Personally, I think that the data point to a fat soluble nutrient – vitamin K – as being involved in health in ways that are not currently appreciated, including obesity, diabetes, CVD, and more.
    I believe that we have not asked the right questions and that vitamin K repletion – which is almost unknown in our population – can be found with a wide variety of dietary patterns.
    We have defined vitamin K as vitamin K1, and it is found in some wonderful foods, but in leafy greens it is poorly bioavailable. Instead, we need to look at vitamin K2. It is in diet in a wide variety of (lost) foods but we endogenously make MK-4, a process with which we muck with trans fatty acids because these novel fats are hydrogenated high K1 oils (usually) and this alters their vitamin K content to become dihydrophylloquinone, which impairs making MK-4 and which is poor in activating vitamin K-dependent proteins. Some drugs foul up this biosynthesis, too.
    Look into vitamin K2 and such endpoints as diabetes, CVD, cancer and more.
    It is quite remarkable, but generally not pursued. There is no Rx vitamin K2. So no money to fund studies. Also, there are currently two forms of supplemental vitamin K2 (MK-4 and MK-7). I am critical of emphasis on these two forms and wish that more studies would look at ALL long chain menaquinones because the EPIC cohort that found the linear Inverse dose response of K2 intake and cancer risk noted that the most prominent form was MK-9, high in fermented dairy foods. Right now, Japan tends to study MK-4 and use it clinically whereas the Netherlands and some others are emphasizing MK-7. MK-7 is representative of long chain menaquinones, but we need to look at all of them comparatively and where they are found in food.


Trevor Butterworth

November 2, 2015 at 6:51 am

I’m baffled by your score. The study is a statistical train wreck—and Time doesn’t even know how to ask the right questions of the design and data. Compare the Guardian’s coverage of this topic to Time—and the general critical reception to the study by nutrition researchers.


    Kevin Lomangino

    November 2, 2015 at 10:33 am

    Thanks for your note Trevor. The star score for this Time story benefits from several close calls on our review criteria. We emphasize that it’s our reviewer comments — and not the star score itself — that readers should pay most attention to when reading our story reviews. Yes, we wish that skeptical voices had been included higher up in the story as indicated in our review. But both the story and our review point to many of the same caveats and limitations that are highlighted in the report — including the weight loss that occurred in these children and the lack of a control group. While the story certainly could have been framed differently and more skeptically, and those caveats could have been placed higher up, we grade generously with our criteria. And the ratings on those criteria are ultimately what drive the star score. We think that anyone who read the review would find that we had some significant concerns with the coverage as suggested by the following: “But we thought the framing of the story was off kilter, as it leads with lots of glowing discussion about the strength and importance of the research, while putting important caveats and limitations far down in the piece. We also think the story could have been more thorough in its discussion of those limitations — including the fact that 43 is a small number of participants, the study was very short in duration, and it had no control group.”


    Gary Schwitzer

    November 2, 2015 at 1:43 pm


    We often tell audiences that, just as you can’t judge a book by its cover, you can’t judge one of the stories we reviewed simply by its star score. In either case – the book or the reviews – you need to read the fuller content. In our review, you’ll see clear comments of reservation about where/how any reference to limitations appeared.

    We’ve had other 5-star stories – which means only that they were graded satisfactory on most of our criteria – that, nonetheless leave gaping holes. And we’ve had 2-star stories that may seem to warrant more credit than the grades on our criteria dictate.

    As Kevin Lomangino suggested in his response to you, we try not to bludgeon journalists but to reach out with constructive suggestions about how things could have been done better.

    Gary Schwitzer