Tuesday, October 01, 2013

Study: People Who Eat More Sugar Weigh Less

These are my marinated strawberries to which I added a few teaspoons of
sugar, some cinnamon, and a splash of red wine.
Something doesn't make sense about the recent advice to avoid sugar because it causes obesity. Judge Judy says, "If it doesn't make sense, it's not true."

If sugar makes people fat, why did this study...

Dietary Composition And Fat To Sugar Ratios In Relation To Obesity, International Journal of Obesity and Related Metabolic Disorders, December 1994

... find that the more sugar people ate, the less they weighed. This was a large sampling and it broke intake into 5 categories, not just "high" and "low." It found a dose-response relationship among these categories. As sugar intake increased, obesity decreased. Such a trend adds validity to the finding.
"The aim of this work was to investigate the relationship between dietary composition and prevalent overweight and obesity in a middle-aged Scottish population. ... The subjects were 11,626 men and women aged 25-64 who participated in the baseline Scottish Heart Health and MONICA studies.

The following were measured: (1) the prevalence of overweight (BMI 25-28.6 for women and 25-30 for men) and obesity (BMI > 28.6 for women, and > 30 for men) according to intake fifths of carbohydrates (starch, total, extrinsic, intrinsic and milk sugars) and fat to carbohydrate ratios; (2) the percentage of the variance in BMI explained by multivariate analysis models which included each of the sugar variables and total energy intake.

The overall prevalence of overweight and obesity in the Scottish population were 43 and 11% for men and 38 and 14% for women respectively. Their prevalence increased from the lowest to the highest fifth of Fat:ES intake, respectively for men and women, from 5 to 18.5% and from 13 to 26%. The prevalence of overweight and obesity declined from the lowest to the highest fifth of total carbohydrate."
My opinion - you have to look at the whole diet. Dietary fat is an integral player. In this study, as participants' sugar intake went up, their relative fat intake went down, as did their weight. As their fat intake went up, their relative sugar intake went down ... and their weight went up.

This is exactly what we saw in the Cuban study. When Cubans entered their Special Period in the early 1990s, they lost weight. Their diet had become very low in fat (13%) and high in carbohydrate (77%); focusing heavily on sugar and rice.

I think the reason recent studies are finding sugar problematic is because they are studing it against a backdrop of high fat intake (20% and greater).

I also think it's unwise to assume sugar is benign and consume it ad libitum if fat intake remains high. The Cuban fat intake of 13% was low; it was Pritikin/Ornish/Esselstyn territory. You can't get there by eating much if any animal food, or much if any added oil. The macronutrient ratios of the Cubans resembled that which the long-lived Okinawans ate ... high in carbohydrate, low in fat.

Related:
Sugar And Diabetes, Part II
Sugar And Diabetes
Is Sugar All That Bad?
________

13 comments:

Healthy Longevity said...

Typically people who criticize the observational studies that have found an inverse association with sugar intake and obesity point out the possibility that obese people were underreporting sugar intake. However, if this was the case it would be expected that obese people would have also underreported fat intake resulting in an inverse association with fat intake and obesity which was not found in this study. As shown in randomized controlled trials, sugar typically only increases body weight when consumed in a hypercaloric diet. This suggests that the findings from this study can be in part explained by a lower caloric intake as the result of a lower fat diet.
http://www.ncbi.nlm.nih.gov/pubmed/22351714

There are a number of other studies that cast doubt on the hypothesis that sugar intake has significantly adverse effects on body fat composition independent of caloric intake.

In the Women’s Health Initiative the control group who were advised to lower fat intake significantly increased the intake of sugar and showed trends towards lower body weight and waist circumference.
http://ajcn.nutrition.org/content/91/4/860.full

Kempner’s rice-fruit diet which was based on rice, fruit, fruit juice and sugar resulted in significant weight loss in most patients.
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC436162/

The Kuna Indians from Panama who also consume a starchy diet that is relatively rich in fruit and added sugar have been observed to be slim.
http://www.ncbi.nlm.nih.gov/pubmed/16794446

Healthy Longevity said...

Typically people who criticize the observational studies that have found an inverse association with sugar intake and obesity point out the possibility that obese people were underreporting sugar intake. However, if this was the case it would be expected that obese people would have also underreported fat intake resulting in an inverse association with fat intake and obesity which was not found in this study. As shown in randomized controlled trials, sugar typically only increases body weight when consumed in a hypercaloric diet. This suggests that the findings from this study can be in part explained by a lower caloric intake as the result of a lower fat diet.
http://www.ncbi.nlm.nih.gov/pubmed/22351714

There are a number of other studies that cast doubt on the hypothesis that sugar intake has significantly adverse effects on body fat composition independent of caloric intake.

In the Women’s Health Initiative the control group who were advised to lower fat intake significantly increased the intake of sugar and showed trends towards lower body weight and waist circumference.
http://ajcn.nutrition.org/content/91/4/860.full

Kempner’s rice-fruit diet which was based on rice, fruit, fruit juice and sugar resulted in significant weight loss in most patients.
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC436162/

The Kuna Indians from Panama who also consume a starchy diet that is relatively rich in fruit and added sugar have been observed to be slim.
http://www.ncbi.nlm.nih.gov/pubmed/16794446

RB said...

Bix, I like your strawberry picture. I think adding cinnamon is fine but for my taste I think strawberries are sweet enough without adding extra sugar.

Seems these studies contradict the Gary Taubes' carbohydrates make us fat theory.

Just saying sugar is too general. It would be nice to know if the sugar consumption is from whole fruits and other whole plants, fruit juices, milk/dairy, refined sugar, HFCS, sugar added to processed foods, etc.

Bix said...

I always appreciate your input, Healthy Longevity. There isn't enough time in my days to read up on all this!

Regarding: Underreporting...

As I've seen, using quartiles or quintiles attenuates confounding by underreporting. The stepped feature allows you to look for a trend, a dose-response, which is relative.

If I may be blunt, the word "underreporting" is sometimes invoked without consideration. It's a word that's heard and repeated. It sometimes works against the invoker! (I'm not referring to you HL or to any people who visit this blog. Not yet anyway.)

Let me look at this for a second. Say, the groups are honest, and...

Group 1 eats 100 grams carb/day
Group 2 eats 110 grams carb/day
Group 3 eats 120 grams carb/day
Group 4 eats 130 grams carb/day
Group 5 eats 140 grams carb/day

Say they are underreporting...

Group 1 actually eats 120
Group 2 actually eats 130
Group 3 actually eats 140
Group 4 actually eats 150
Group 5 actually eats 160

You would still find the same trend, if everyone underreported. And that trend was relative... as you rose in quintiles, you weighed less. What mattered was the relativity, not the absolute number. However, note that even if people were underreporting, they were in fact eating more sugar and losing weight!

Or, let's say the opposite is true. That sugar really does cause weight gain. This study found that weight went down as you rose in quintiles, so sugar consumption must have gone down as you rose. But to be assigned to a quintile, you had to say you were eating more. I'm not sure how that would work.

Group 1 actually ate 150, but said they ate 110 (underreporting)
Group 2 actually ate 140, but said they ate 120 (underreporting, but not to the same degree)
Group 3 actually ate 130, but said they ate 130 (not underreporting)
Group 4 actually ate 120, but said they ate 140 (overreporting)
Group 5 actually ate 110, but said they ate 150 (overreporting, but not to the same degree)

So, if sugar really does cause weight gain, in this study, there would have to have been a mix of underreporting and overreporting, and the amount of that -/+ would not have been consistent, perhaps relative to the amount actually eaten? Those constraints are very specific, and less likely to hold as the number of participants grows.

Bix said...

Thanks RB. I like sugar on a few fruits, sometimes strawberries, always cranberries. I make an apricot compote with both honey and sugar. Most other fruits I just eat as is.

Travis Jensen said...

Or it could be that you cannot successfully control for enough factors in observational studies. Perhaps as sugar consumption went down, wheat consumption went up faster, which will cause much the same effect on the body. Maybe these low-sugar people were chowing on low-fat yogurts, which would be unlikely to be added to the "sugar" list, yet are as full of sugar as soda. There are just too many variables for it to say anything.

Observational studies like this are bunk at proving ANYTHING. Even if it had said that there was a direct relationship between sugar and weight, it still wouldn't tell us that sugar causes weight gain. Dietary fat as the primary source of weight gain has only ever been implicated through crappy studies like this that cannot control for variables (Have toast with your eggs? How do we know it was the eggs and not the toast, then?). But if you have a pet theory (dietary fat causes heart disease (no, wait, it's just saturated fat), animal protein will kill you, fructose will kill you, people are only meant to be veg*ns, etc), observational studies are great for allowing you to have a media splash by getting your statistics to say anything you want.

You are right, though, that it is the entire diet that matters, and, more problematic, it varies by individual. I cannot eat carbohydrates without gaining weight, PERIOD. When I cut them out completely, I have absolutely no problem maintaining a healthy weight. Sugar and other processed carbs are bad for us in too high of a quantity, where "too high" is an individual number that changes with age. My "too high" is around 40g/day; somebody else's might be 150g. A few people might be OK with the 350g recommended by the USDA, but given even thin people can become diabetic, my guess is that there is a limit to how long that can occur before that dose becomes toxic.

Oh, and appealing to Judge Judy as an authoritative source? Really? ;)

Bix said...

Hi Travis,

The study assessed various types of carbohydrates (starch, total sugar, extrinsic sugar, intrinsic sugar, milk sugars). So, it addressed your concern about sugar and wheat. It found that as total carbohydrate (also, total sugar, and extrinsic sugar) went up, the prevalence of overweight and obesity went down. Also, as fat intake went up, overweight and obesity went up with it.

Hill and Prentice said virtually the same thing:

"Metabolic studies show that diets high in fat are more likely to result in body fat accumulation than are diets high in carbohydrate. There is no indication that simple sugars differ from complex sugars in this regard.

Further, although high intake of dietary fat is positively associated with indexes of obesity, high intake of sugar is negatively associated with indexes of obesity.

There is ample reason to associate high-fat diets with obesity but, at present, no reason to associate high-sugar diets with obesity."


Regarding your comment about proof...
No study proves. Science doesn't prove, it disproves. You ask a research question, formulate a hypothesis, then design a study to test the hypothesis. You try to disprove the hypothesis. If the findings can be reproduced over and over, the hypothesis stands and becomes a theory. So, even a finding from a clinical trial with a control group does not prove something. It has never been proven that cigarette smoking causes lung cancer, only that smoking increases the risk.

I think that low-carb diets are effective for weight loss. I don't think they are good for long-term health. Since they are meat-heavy, they increase the risk for cancer.

However, I happen to agree with you when you say processed carbs are bad for us. I think the carbs to eat are minimally processed ... squashes, pumpkins, broccoli, carrots, yams, peaches, apples, zucchini, corn, watermelon, potatoes, peas, green beans, oats, lentils, cantaloupe, rice, kale, onions, tomatoes, peppers, spinach, bananas, cucumbers, blueberries, quinoa, cabbage, turnips, oranges, beets, kidney beans, cauliflower, cherries, garbanzos, mangoes, etc. Things like boxed breakfast cereals and all forms are bread aren't a good idea.

Anonymous said...
This comment has been removed by a blog administrator.
Angela and Melinda said...

That is by far one of the rudest comments I've ever seen on a food blog. You should be deeply ashamed, Anonymous. (No wonder you're anonymous.)

Bix said...

Sure was out of the blue, that one. A lot of anger there.

Thomas Anthony said...

Why is that comment even still there?

Thomas Anthony said...

Some very interesting info here. I don't know a lot about diabetes, but started looking more into when my dad told me he his doctor told him not to eat bananas because they were high in sugar, and my dad was apparently having some blood sugar issues (prediabetic). He told me that as he asked for a second helping of bacon to go with his pancakes I had cooked for him.
I've never heard that sugar causes diabetes which is a very vague notion to begin with (of course it doesn't 90-100% of our population would have diabetes then) but taken to excess, high coloric intake, lack of activity, add that to other stresses and even an aging body, and common sense would indicate that a lot of sugar, (or caloric regardless of source) consumption at least could have a significant role to play. As an avoidance behavior, seems to me the best way to navigate the often confused and jargon-filled seas of nutrition science is by following the mantra of moderation. A balanced diet, resist processed "factory" foodstuffs, appropriate caloric intake, and regular excercise. The rest is up to genetics and fate perhaps. I suppose one could go several steps further and only eat raw organic vegetables; this has worked for some.

Bix said...

I agree with a lot of that, Thomas.

Diabetes is a terrible disease. I've worked in it for 17 years. There's no part of the body it leaves unscathed. It slowly robs a patient of their sight, their kidney function, their heart function, their fingers and toes and feet and legs. It increases their risks for cancer and mental illness. They lose the sense of touch, of hearing. They can't digest well. They're tired but don't sleep well. It's a terrible, insidious disease.

As I've seen, 2 things contribute to the development of type 2 diabetes. They are... the environment and lifestyle. Lifestyle includes diet. Diet is key, both in diabetes' development, and in its resolution.

It is unfortunate that some people are saying Tom Hanks' type 2 diabetes was genetic. This is a teachable moment that's being frittered away. Lifestyle is the deciding factor, not genes. The global prevalence of diabetes is rising faster than our genome could have mutated. There are case studies of immigrants from China or Japan, who do not have diabetes, whose risk skyrockets when they come here. They get diabetes at a much higher rate than their peers at home.