Thursday, July 22, 2010

Low-Fat Vegan Diet Outperforms Conventional ADA Diabetes Diet

Participants, all of whom had type 2 diabetes, were assigned to either a low-fat vegan diet or a conventional diabetes diet (2003 American Diabetes Association (ADA) Guidelines), which they followed for 22 weeks.1

Here were each group's goals:

The vegan diet (~10% of energy from fat, 15% protein, and 75% carbohydrate) consisted of vegetables, fruits, grains, and legumes. Participants were asked to avoid animal products and added fats and to favor low–glycemic index foods, such as beans and green vegetables. Portion sizes, energy intake, and carbohydrate intake were unrestricted.

The conventional diabetes diet (15–20% protein, <7% saturated fat, 60–70% carbohydrate and monounsaturated fats, and cholesterol ≤200 mg/day) was individualized, based on body weight and plasma lipid concentrations.

Here's what participants ate over the course of 22 weeks (which did not quite meet their goals). Click to enlarge:*





*Not shown:
Vegan Diet: Caloric intake 1425 kcal/day, Cholesterol 24 mg/day
Diabetes Diet: Caloric intake 1392 kcal/day, Cholesterol 189 mg/day

________

What happens when people with diabetes eat a high-carb diet ... consuming 70% of their calories as carbohydrates?

Here's how each group fared (all of the following reached levels of significance):

Among participants whose diabetes medications remained unchanged:
A1C fell 1.23 points in the vegan group
A1C fell 0.38 points in the ADA group
(A1C is short for HbA1c, a measure of blood glucose over the last 3 months)

Reduction in diabetes medications:
43% (21 of 49) of the vegan group
26% (13 of 50) of the ADA group

Body weight decreased:
6.5 kg [14.3 lbs] in the vegan group
3.1 kg [6.8 lbs] in the ADA group

Among those who did not change lipid-lowering medications:
LDL cholesterol fell 21.2% in the vegan group
LDL cholesterol fell 10.7% in the ADA group

Reductions in urinary albumin:2
15.9 mg/24h in the vegan group
10.9 mg/24 h in the ADA group

In sum, those on the high-carb vegan diet had lower blood sugars (and so could reduce their meds more), lower LDL cholesterol, improved kidney function and over double the weight loss. There were also significantly greater reductions in BMI, waist circumference, and total cholesterol in the vegan group compared to the ADA group. There was no significant difference in exercise between groups, so these changes weren't because vegans were overtly spending more calories. ("Overtly" is key, since, as previously discussed, some diets lead to greater expenditure of calories from digestion and thermogenesis.)

Gary Taubes in his book, Good Calories, Bad Calories said, "Carbohydrates make us fat and ultimately cause obesity. The fewer carbohydrates we consume, the leaner we will be." He also said, "Refined carbohydrates, starches, and sugars are the dietary cause of coronary heart disease and diabetes." This study did not support his claims.

This is noteworthy ... Participants in the vegan group were allowed unrestricted consumption. They could eat as many calories and as much carbohydrate as they wanted, as long as they didn't eat from certain food groups. Participants in the conventional diet group had to limit their caloric intake, count calories, and control portion sizes. Even with unrestricted food intake and a higher calorie consumption, the vegan group lost more than twice as much weight.
________
1 A Low-Fat Vegan Diet Improves Glycemic Control And Cardiovascular Risk Factors In A Randomized Clinical Trial In Individuals With Type 2 Diabetes, Diabetes Care, August 2006
2 People with diabetes suffer microvascular complications involving the kidneys that allow passage of protein into urine. The lower the amount of the protein albumin that leaks into the urine, the healthier the kidneys.

Charts: Bix. Data from study.

19 comments:

Bix said...

Barnard's work is telling. I don't think you need to eat 100% vegan to achieve these benefits though. Notice his vegan group still averaged 24mg/day cholesterol. Only animals make cholesterol, so his vegan group wasn't 100%. They were eating almost double their fat goal too. And those were amounts they let be known ... they were likely higher. Still had better HbA1c's.

I thought to revisit this study since I saw he extended his analysis of these groups and published a 74-week run-down recently. A year and a half later and the vegan group was still trending better.

bee said...

interesting. however what i find intriguing is that the people on the "diabetic diet" were allowed 60-70% carbs.

i'm a vegan who's trying to increase my protein intake and this comparison is very interesting.

Bix said...

I saw that too, bee, the "60–70% carbohydrate and monounsaturated fats." I suppose if you excluded mono fat, olive oil, you could get to the higher carb range.

Bix said...

A difficulty in reducing carbohydrate in your diet, if that is your goal, is that by default it increases the amounts of fat and protein you eat. For low income people, that invariably means consuming more cheap meat. Many environmental pollutants (pesticides, endocrine disrupters) are soluble in fat, and appear in larger quantities in the fat of animals, especially those animals fed cheap, contaminated fodder, a practice one might see on a confined animal feeding operation, that is, a factory farm.

Steve Parker, M.D. said...

The vegan diet was also easier for participants to follow. I'm guessing the vegans age relatively low-glycemic-index, which would help with glucose control.

I wish the researchers had made it easier to compare baseline hemoglobin A1c levels of the two groups. It appears the vegan group baseline was 8.1% versus 7.9% in the other group.


-Steve

Steve Parker, M.D. said...

My use of "age" above should have been "ate." Could you fix it, Bix? Thanks.

-Steve

Bix said...

Blogger doesn't let me edit comments, Steve. Although you may be able to delete and repost it.

Dr. Mel said...

Just came across this in today's Times: http://www.nytimes.com/roomfordebate/2012/04/17/is-veganism-good-for-everyone/?ref=dining

Bix said...

I like this that Erika Kendall said:

"The more that we research and discuss meat, the easier it becomes for us to be wary of its origins and nutritional value.
...
In a rush to give consumers value, food producers have jeopardized the quality of most meat out there."

Dr. Mel said...

The Brian Patton essay was interesting in that while cats are obligate carnivores, ursines (e.g., grizzly bears) and canines are omnivores, eating quite a bit of vegetative material as well as meat. Dogs can also be healthy as vegetarians. But if we follow Patton's logic, then our closest relatives should also be vegan, but they are not. Jane Goodall was the first person to observe chimpanzee meat-eating, and now there's some evidence provided by Goodall (not proven yet, but suggestive) that gorillas and bonobos also eat vertebrates like monkeys and antelope. So Patton's argument doesn't hold up.

Dr. Mel said...

Here's a Fish & Wildlife Service summary of what grizzlies eat. I can't imagine where Patton got his image of grizzlies. ". Scavengers by nature, grizzlies spend most of their waking hours searching for food. Forbs, roots, tubers,
grasses, berries and other vegetation, and insects comprise most of the bear’s diet. But grizzlies are very adaptable,
finding and subsisting on a variety of foods if necessary. The grizzly diet can include small rodents, fish, carrion, and
even garbage and human food if it is easily available."

Bix said...

That Fish & Wildlife Service diet of the grizzly, before the adaptable part, sounds mighty fine indeed. I eat like a grizzly!

Bix said...

I'll never forget how astonished I was to see a deer eat a baby bird:

http://fanaticcook.blogspot.com/2011/05/deer-eats-bird.html

Bix said...

This is so weird how people, people who are read in the field, think eating eggs and dairy is more ethical than eating meat.

From the post you linked, Mel:

http://www.nytimes.com/roomfordebate/2012/04/17/is-veganism-good-for-everyone/meat-is-brain-food

Drew Ramsey (assistant clinical professor of psychiatry at Columbia University) said:

"Those with ethical objections to killing animals can meet all their animal-nutrient needs with dairy products from grass-fed cows, organic eggs from cage-free chickens and occasional servings of mussels or oysters."

Yet Peter Singer and Jim Mason, in The Way We Eat say that consuming dairy products may be one of the most unethical actions we perform, whether the milk is organic, grass-fed, etc. or not:

The Ethics of Dairy Consumption

Eggs aren't much better, in that we, literally, I'm sorry to say, toss male chicks into a grinder or plastic bag.

Anonymous said...

Not sure what is meant by "fiber" on the graph. Carbs include starches and fiber, so what is included in "fiber"? Or are you intending "carbohydrates" to mean "starchy carbohydrates"?

Bix said...

Fiber here is a subclass of carbohydrate. If it was not part of "carbohydrate" they would have had to note that, which they didn't.

There are many types of fiber - cellulose and lignin, resistant starch and resistant dextrins, inulin, waxes and gums, chitins, pectins, beta-glucans, sugar alcohols, oligosaccharides.

Even what you call "starchy carbohydrates" can contain "fiber," that is, indigestible carbohydrate, e.g. resistant starch. The degree of digestion depends upon the individual, their disease state, other foods eaten, and the chemical make-up of the carbohydrate when eaten, which can, for example, change to become more resistant to digestion when it is heated then cooled. For example, hot pasta right out of the pot has more available carbohydrate than that same pasta which has been cooled. Cooled pasta and potatoes have more resistant starch and less available carb for digestion. I guess the same pasta would then have fewer calories. This is a problem with food tables that list calories, let alone the variability of the food source.

K. said...

Bix, I would have preferred it if the two groups were demographically similar. They are not.

For example, the vegan group was 55% female, the ADA group 66%. One can reasonably ask whether there are gender differences affecting the results. Also, women are on average lighter, so half the weight loss may actually correspond to a less marked difference in percentage weight loss.

31% of the vegan group were retired, but only 8% of the ADA group. Are retirees, with more time and less stress, better able to comply with a diet?

For me, perhaps the biggest difference that makes it hard to compare the groups directly was their BMIs. The vegan group mean was 33.9, while the ADA group mean was 35.9. Furthermore, 30 (61%) of the vegan group was obese, compared with 43 (86%) of the ADA group.
That's almost 1 1/2 times the number of obese people in the ADA group. It may be that any intervention will have less success in obese people.

Of course, the reason for all these differences was that the participants were assigned randomly to the diets. As far as I can see, no analyses of subgroups was carried out.

Actually, something else troubles me much more about this study. I'll leave that for a separate comment.

K. said...

Bix, did my comment on this post about a day ago get through? It hasn't appeared and isn't so easy to reconstruct.

Bix said...

For differences between groups check the P values in Table 1. You will see they are statistically similar.

All studies have demographic differences among groups. Randomization is employed to lessen the effect of differences. Making unscientific assumptions about differences among individuals and groups to support your view is confirmation bias. And you make a lot of assumptions here...

- You say "women are on average lighter." Yet you say the group with more women in it had higher BMIs.

- You say retirees have more time (?) and less stress (?). It could also be said retirees have more physical and mental disabilities, less income, less mobility, less social support. You could say that. But without data you don't know. Nor do you know how any of these assumptions interact with the intervention.

- You suggest that "any intervention will have less success in obese people." Whether this is true or not, there was no statistically significant difference in BMI between groups in this study.

If you have justifiable concerns about the validity of this study, it would be honorable of you to raise them with the authors, editors, and publishers.