The Diogenes Study

The best way to lose weight and maintain the loss is a diet moderate in protein and incorporating low glycemic index foods, reports a large study funded in part by the European Union. But is it? A closer look at the results reveals inconsistencies and problems.

Glycemic Index

The Glycemic Index (GI) is a numerical scale that classifies how fast and by how much a  food raises the blood sugar level. Foods low on this index only lead to a moderate rise, while a food with a high GI number has a stronger effect. And the higher the blood sugar level, the more insulin the body produces, with insulin being a hormone that plays a role in the regulation of hunger.

The hypothesis behind the GI is that raising the blood sugar level only moderately will lead to less insulin and in turn to less hunger and more satiety, equaling less consumption of calories.

The Diogenes Study

The above study now, named Diogenes, standing for “diet, obesity and genes“, did the following: four groups of people were randomized and either received a high protein / high GI, low protein / high GI, high protein / low GI or low protein / low GI diet. For two months they underwent  severe and controlled caloric restriction and were then left to their own devices for another four. It was then examined how much weight the groups had gained back and the results showed that the groups with a low GI diet only experienced an average increase of 0.67 kg, while the high GI groups underwent a regain of 1.67 kg.

Apple - Glycemic Index: 38

Now we sail into the first problem: Diets that mainly consist of low GI foods lead to less glycogen in the muscles, glycogen being a form of sugar used as an energy reserve for times when the free sugar (glucose) in the blood has been depleted. One gram of glycogen also stores 3-4 g of water and the average human adult has 600 g of glycogen in his body.

A study from 1996 looked at glycogen concentration in the muscles after 30 days of either a high or low GI diet and found that in those on a low GI diet, the concentration had been lowered by 14%.

If we now hypothesize that the overweight participants of the Diogenes study had the average pool of 600 g before the study was started and each gram of glycogen stores those 3-4 g of water, then more than 400 g  of the low GI groups’ weight loss can be accounted for by the lowered glycogen reserves (14% of 600 g = 84 g and 84 x 4 water g + 84 g glycogen = 420 g).

Differences in Fat-Free Mass

A further look at the Diogenes study’s results reveals that the group on a high protein / high GI diet had an increase of fat-free mass of 1 kg on average, while the high protein / low GI group only experienced an increase of ~0.54 kg. The study doesn’t mention if or how this difference in fat-free mass was accounted for. All that we get is:

A reduction in the glycemic index of 4.7 units resulted in a 0.95-kg difference in body weight between the high-glycemic-index groups and the low-glycemic-index groups. Since there were no differences in fiber intake, the difference in body weight, though small, can be ascribed to a true effect of the glycemic index.

This doesn’t take body composition into consideration, only total weight. Yet there is a huge difference between a 180 lbs man who has a body fat percentage of 10%, and a man with the same weight, but 30% body fat.

If we factor this in and add it to what I outlined above, the small difference becomes even smaller, as another large portion of the weight the high GI groups gained back would have been fat-free mass: 334 g of the regained weight could be due to the difference in glycogen retention and 460 g due to more fat-free mass, equaling a total of 794 g  – almost 2 lbs (1 lb = 454 g) or 3/4th of a  kg.

Calculation Problems

The problem low GI diets have in general is that it is next to impossible to judge where a food truly ranks on the scale. There are lists out there that give numbers, but those are highly unreliable. A white potato may range from a low to a high GI and as foods usually are eaten in combinations, we are further left in the unclear, as a steak (GI 0), potatoes (GI 56-111) and lettuce (GI 15) all have different numbers – which one are we supposed to go by?

And if that wasn’t enough, all these foods also have a “glycemic load”, which means in addition we also are supposed to calculate the amount of a food we eat, to find out by how much it will really raise our insulin. Therefore you need to take a food’s GI number – you’ll just have to decide on one – and multiply it with the amount of carbohydrate in your portion of the food and divide that by 100.

Even the Diogenes authors acknowledge that all this might be a little difficult. One of them told the New York Times:

“Glycemic index is a difficult concept to follow,” said the lead author of the study, Thomas Meinert Larsen, of the University of Copenhagen. “If you look at tables listing glycemic index foods, the data are not very reliable.”

Conflict Of Interest?

Last but not least what remains to mention is the support that the study received and the possible interests involved in it. Mostly paid for by the European Union, Diogenes was also sponsored by various companies, which by itself is not questionable, as conducting large studies requires considerable funds.

But it does raise an eyebrow when one of the study’s authors declares that he receives lecture fees from Nutrition et Santé, a Belgian company, that also sponsored the study through providing their “Modifast” weight loss shakes.

Media Attention

Yet all the above points, that surely merit some closer inspection, haven’t stopped various media sources from proclaiming that a “high protein, low glycemic index diet helps keep pounds off” (Boston Globe), that the study “identifies foods that promote weight maintenance”  (Los Angeles Times) or that we have to “think more protein, fewer carbs to maintain weight loss” (Businessweek). The Times of India goes even further: “Diet study finally solves obesity riddle”.

The full New England Journal of Medicine article of the study is available from the Diogenes website (PDF).

Picture courtesy of Graur Codrin.