November 25, 2010
High Protein, Low Glycemic Index To Avoid Weight Gain

More protein and lower glycemic index carbs work best to keep the weight off.

Researchers at the Faculty of Life Sciences (LIFE), University of Copenhagen, can now unveil the results of the world's largest diet study: If you want to lose weight, you should maintain a diet that is high in proteins with more lean meat, low-fat dairy products and beans and fewer finely refined starch calories such as white bread and white rice. With this diet, you can also eat until you are full without counting calories and without gaining weight. Finally, the extensive study concludes that the official dietary recommendations are not sufficient for preventing obesity.

After going on an 800 calories/day diet the participants were put on different maintenance diets to see which diet would keep the weight off.

A total of 772 European families participated, comprising 938 adult family members and 827 children. The overweight adults initially followed an 800 kcal/day diet for eight weeks, losing an average of 11 kg. They were then randomly assigned to one of five different low-fat diet types which they followed for six months in order to test which diet was most effective at preventing weight regain.

The results are plain to see in this chart. The Low Protein-High Glycemic Index (LP-HGI) diet was worst. The High Protein-Low Glycemic Index (HP-LGI) diet was best.


This is right out of the Paleo Diet game book. Stay away from those high glycemic index grains. Eat more meat.

If you want to eat some carbs that are low in glycemic index then check out this searchable glycemic index food database. Check out your foods in that database. Note that rices span a very wide range of glycemic index values. The sticky rice in Chinese restaurants is some of the highest glycemic index food you can eat. But Uncle Ben's Converted Rice is one of the very low glycemic index foods.

Share |      Randall Parker, 2010 November 25 07:07 PM  Aging Diet Weight Studies

Brett Bellmore said at November 26, 2010 6:41 AM:

I'm willing to try it out, losing the weight I gained during chemo is turning out to be extraordinarily difficult. And I do eat an awful lot of white rice, my wife being from the Philippines.

But something about that graph bothers me: I noticed that every last diet represented on that chart showed an initial weight drop, followed by a climb. Including the high protein, low hypoglycemic index diet. The graph just stops when the HP-LGI diet gets back up to the starting weight. If you'd stopped the study at 12 weeks, the LP-HGI diet would have looked like a winner, too. And at 26 weeks your weight is trending up in ALL the diets.

What happened after 26 weeks?

And do you perhaps want to alternate between the LP-HGI and HP-HGI diets, switching back and forth as each becomes ineffective? Maybe it's just the case that NO diet works forever?

Brett Bellmore said at November 26, 2010 6:45 AM:

Or perhaps HP-LGI and HP-HGI? Just eat lots of meat all the time, and alternate between high and low glycemic index foods? I think I might try that, I do like meat.

Mark said at November 26, 2010 7:14 AM:

But are these foods (white flour, white rice) causing additional weight gain because they are high glycemic or because they are generally a poor source of micronutrients? I have found fruit to always cause weight loss, yet it is a fairly highly glycemic food. However, fruit is full of micronutrients. Scientists who structure these studies all too often focus merely on the macronutrient (e.g. grams of fat, etc) structure of the diet with paying any attention to the micronutrient structure of the diet. (Note to those interested in nutrition - research what happens to humans when you get a deficiency of magnesium and/or copper. It ain't pretty.)

Randall Parker said at November 26, 2010 9:23 AM:


My guess is that people kept trying to restrict their calories when they shifted to the second diet. But eventually their focus on keeping the weight off weakened and they started responding to their hunger pangs.

Randall Parker said at November 26, 2010 9:47 AM:


Go to that Glycemic Index database and search on some fruits you eat. Apples have a glycemic index range of the 30s and 40s. That's low.

Lou Pagnucco said at November 26, 2010 10:15 AM:

Also eating carbohydrates with carb blockers (e.g., green tea, white bean extract, apple vinegar, hibiscus tea, ...) and/or viscous fiber may lower GI. Also, food preparation matters - leaving cooked potatoes in the fridge overnight significantly lowers GI.

Bruce said at November 26, 2010 11:10 AM:

"The overweight adults initially followed an 800 kcal/day diet for eight weeks ... "

As Brett says, all people on an 800kcal/day diet ended up on tack to gain it all back. Some diets quicker than others.

I wonder what would happen if they went on 4 x 2 week 800kcal diets and then went on HP-LGI in between?

Mark said at November 26, 2010 12:48 PM:


Thanks. Yes, apples are fairly low on the GI index. But potatoes are high, and they generally cause weight loss. (Eaten in the proper form - not french fries, of course.) But to get back to the study, I think it's important to note what these studies never say in this age of Nutrition CrimeThink: saturated fat and cholesterol serve essential bodily functions: nerve signaling, brain function, sex hormones, etc. The body has little use for the crap that is stuffed into a Lean Cuisine or any other low-fat processed food. This is never discussed in a nutrition study, which usually just repeats calories consumed, fat grams consumed, etc rather than the quality of the food consumed.

Randall, in case you missed it, check out Ray Peat's website. It's pretty interesting. (It's somewhat over my head at times, but perhaps some more scientifically inclined readers can chime in.)

ziel said at November 26, 2010 5:30 PM:

Note the relatively low indexes for durum pasta (which is what almost all commercial pasta is - durum wheat pasta). I have often argued that whole wheat pasta is an unnecessary flavor compromise, since durum is a very hard wheat and not like refined white flour - and these GI values bear that out. Unless for some reason you really like whole wheat pasta, stick with the real stuff - especially if you can find good brands from Italy (~$4.50 per lb, vs ~$1.50 - very much worth it).

Randall Parker said at November 26, 2010 6:50 PM:


Yes, it is not widely known that the wheat for pasta has much lower glycemic index than the wheat for bread. Similarly, the range for rice glycemic index is very broad. But I'm skeptical on eating wheat at all. Aside from using about 2-4 slices of whole wheat bread per week (for hamburgers) I rarely eat wheat.

The durum wheat and lower glycemic index rices are high in amylose. That's key. An Italian group in Bologna is attempting to develop a higher amylose wheat and their commentary about their attempt mentions lower glycemic index as one of the motivations:

This project aims to modify the ratio between the amylose and amylopectin content in starch to increase the amylose content, both in durum wheat and common wheat. Various studies have shown that the assumption of wheat-based foods with these characteristics brings considerable health benefits. In fact, a high percentage of amylose in food leads to the formation of “resistant starch”, stimulating the activity of the intestinal flora and inducing the production of protective substances against colon cancer. Besides, foods with “high amylose” have a low glycemic index, which is beneficial to 25% of the insulin-resistant population and, by stimulating the formation of HDL cholesterol, prevents cardiovascular disease.


If you haven't already read it I recommend Stephan Guyenet's Potatoes and Human Health, Part I, Potatoes and Human Health, Part II, Potatoes and Human Health, Part III, and Sweet Potatoes. Note what Guyenet says at the end of the Sweet Potatoes post. There are some unknowns here about the types of sweet and regular potatoes people eat today versus what more primitive (and free of heart disease) cultures ate.

Randall Parker said at November 26, 2010 7:13 PM:

Here's another Italian research group chasing higher amylose wheat. They used RNA interference to block amylopectin synthesis in order to boost amylose.

It is probably possible to genetically engineer grains to be good for our health. Add in the efforts to boost DHA in grains and boost proteins as well and one could get a pretty good meal from genetically engineered grains. The Euro anti-GMO enviro-weenies would be upset. But heart disease would plummet.

ziel said at November 26, 2010 7:41 PM:

Randall, interesting about the amylose. Over at Razib's, Cochran left a comment that a "population approaching 7 billion makes widespread application of a paleo diet non-viable".

Jerry Martinson said at November 27, 2010 5:16 AM:

I propose a $0.04 tax per Glycemic Load. That way a banana would be taxed $0.40, Strawberries $0.04, but 100g of french bread would be taxed $1.92 and a snickers bar $0.96. A can of coke would be $0.64, rice milk $1.16, cranberry juice $0.96. French fries (without much oil) would be $0.80. Pasta would raise more tax money than Exxon.

You could also tax fat calories (exempting n-3 or monosats) at $0.08 per gram. A big mac would then be 22*$0.04+33*$0.08 = $3.00 tax.

The tax revenues could go to food stamps where allotment is based not only on financial need but also BMI.

bbartlog said at November 27, 2010 5:49 AM:

'They were then randomly assigned to one of five different low-fat diet types'

Only low-fat diets, eh? Good job really considering all the possibilities and evaluating the entire search space, there! Not! Telling me what the best low-fat diet is is akin to advising me which soda will rot my teeth the slowest. This is a useless study.

mike said at November 28, 2010 6:25 PM:

after 26 weeks the spread is only +/- 1 kg. not very much difference after the initial loss of 11 kg in just 8 weeks. can it mean that much then.

Post a comment
Name (not anon or anonymous):
Email Address:
Remember info?

Go Read More Posts On FuturePundit
Site Traffic Info
The contents of this site are copyright ©