Sunday, April 28, 2013

Food Variety, Calorie Intake, and Weight Gain

Let's kick off this post with a quote from a 2001 review paper (1):
Increased variety in the food supply may contribute to the development and maintenance of obesity.  Thirty-nine studies examining dietary variety, energy intake, and body composition are reviewed. Animal and human studies show that food consumption increases when there is more variety in a meal or diet and that greater dietary variety is associated with increased body weight and fat.
This may seem counterintuitive, since variety in the diet is generally seen as a good thing.  In some ways, it is a good thing, however in this post we'll see that it can have a downside.

When doing my usual rounds of the literature last week, I came across a fascinating paper I hadn't seen before titled "Variety in the Diet Enhances Intake in a Meal and Contributes to the Development of Obesity in the Rat", by Dr. Barbara Rolls and colleagues (2).  As the title suggests, they examined the effect of food palatability and variety on food intake and fat gain in rats.  They tested six different diets:
  1. Regular rat chow
  2. Rat chow plus crackers
  3. Rat chow plus cookies
  4. Rat chow plus chocolate
  5. Rat chow plus crackers, cookies, or chocolate, with each of the palatable foods given in succession (i.e. chow plus crackers, then chow plus cookies, etc.).
  6. Rat chow plus crackers, cookies, and chocolate all at the same time.  
Adding a palatable food to the cage (crackers, cookies, or chocolate) increased food intake, as expected.  Adding multiple palatable foods at once increased food intake further, suggesting that variety in addition to palatability contributed to food intake.

When rats were fed these diets for 7 weeks, they produced major differences in weight gain:

Consistent with the food intake data, each of the palatable foods increased weight and fat gain, however giving them all at once caused the rats to gain weight and fat at an even higher rate.  We already knew from the "cafeteria diet" studies that easy access to a variety of palatable foods can cause rapid fat gain in rats and humans (3, 4), but this study shows that food variety itself is an important part of the equation.
The results of this experiment show that the simultaneous presentation of a variety of palatable foods can lead to hyperphagia [excessive eating] and the development of obesity.
Food variety is a determinant of the palatability and reward value of the meal as a whole.  However, to be more precise, variety-induced overeating is usually explained by a related phenomenon called "sensory-specific satiety".  The idea is this: as a meal progresses, we get tired of eating specific foods, but being' full' on one food doesn't necessarily mean we want to stop eating another food.  If a variety of foods are present on the table, we tend to eat more total food before feeling satisfied.  There is a large amount of evidence supporting the concept of sensory-specific satiety and its relevance to food intake.

What about the effect of food variety on humans?  A number of trials have shown that increasing food variety increases total calorie intake at a single meal, even if the foods all have nearly identical composition (5).  A study published in 2001 showed that this effect is sustained for at least 7 days in humans (6).  Food variety is probably an important part of why human volunteers overeat and gain weight rapidly on a human "cafeteria diet" (7), and why we tend to overeat and gain fat in the modern food environment, which offers easy access to a huge variety of different foods.

I'll close with a quote from the paper:
In the affluent Western Society, food is presented in a succession of simultaneous variety meals, and in this case, the challenge presented by variety to body weight regulation is much greater and could lead to even greater increases in body weight [than was observed in this study].
Update: "DagherStrength" pointed out on twitter that this helps explain "unbuckle your pants syndrome" at all-you-can-eat buffets.  Good point!


Charlie Currie said...

I wonder if vanilla ice cream would become unpalatable if that was all you ate, and would it then lead to weight loss. There has been discussion on other blogs about weight loss on a milk and kefir only diet, so why not ice cream.

Might have to give it a try.


Jim said...

If I were a researcher and I wanted to make a meaningful finding about the relation of food variety to obesity, I would try to get real food variety rather than just using different flavors of starch and sugar. Why not pick from as many disparate edible clades as possible? What would be wrong with dumping the rat chow and trying to use real food? Then I would know if it was really variety driving the obesity.

Gretchen said...

These results don't seem counterintuitive to me. They make sense.

Robert said...

Dr. Guyenet,

The successive diet (ie. 4 course processed meal) led to lower weight loss than the simultaneous diet despite time being the only difference between them.

Do you think this might be the case of science validating culture? Ie., a French or European style, multi-course meal is superior to an American-style 4-courses on a plate meal?

If so, then educating people to start with veggies and space out portions over time could substantially reduce weight in the human population, even if humans ate a poor quality diet like the successive diet and followed the veggies with fried chicken, french fries, and ice cream?

This method of eating also seems to be supported by the GI-level satiety signal system.

Robert said...

Here is a good link for a recent study showing that as variety increases, pleasurability and portion size goes up. So our preference for food variety may affect meal planning even before we start eating.

Unknown said...

It may be worth distinguishing between variety in a meal and variety in a diet.

bongo said...

This study hits on something (still a waste of time and money) that people who have managed their weight well know, but others may not and that is - drum roll please....

Don't have a bunch of different kinds of junk food available to eat as in don't have it in the house period. Make it a hard and fast rule.

If a "treat" can't be avoided, go to the store, buy the treat, go home and eat as much as necessary and if you haven't finished it, throw the remaining out. Do this every time you need a treat. Don't will power yourself out of it but go to the store every time its treat time. NO treats in the house EVER though!!!!!

Eat good food, may I recommend the Perfect Health Diet. And get enough calories, it is very important to control the cravings. Don't exercise too much but be very active, no extended sitting. Get good sleep.

This snack approach is especially important for kids. Don't deprive them but make a natural barrier to junk food.

One last thing, don't eat processed junk food when you do indulge. No ice cream stabilizers or candy, no trans fat cookies, etc. Buy Hagendaz for ice cream, bake your own cookies with a reasonable recipe, etc.

chris said...

I get what you are saying about variety though most people eat a very monotonous diet on some level. It's all just corn, wheat, & soy. :)

Stephan Guyenet said...

Hi Chuck,

It wouldn't surprise me. However, ice cream is not a nutritionally complete food.

Hi Jim,

It would be interesting to see the experiment replicated with healthier foods.

Rat chow isn't as far off from real food as some people like to claim. It's basically ground whole corn and soybeans, with a bit of meat or fish, a bit of fat and added micronutrients. Rodents do quite well on regular chow, particularly if they're raised in an enriched environment with toys, peers, and physical activity opportunities.

Hi Robert,

I don't know the answer to your question. Thanks for the reference.

Hi Common,

Good point.

Hi Chris,

I hear what you're saying, but the variety that matters here is variety in sensory characteristics, not basic ingredients. You can make a variety of distinct sensory experiences from a palette of corn, wheat, and soy. The brain recognizes them as different foods even though nutritionally they may not be.

Kindke said...

How would we use this evidence to explain why only a subset of the population are carrying excess weight given that all the population has essentially the same access to the same large variety of foods.

Reijo said...

Perhaps this helps in explaining why low carb diets work so well in short term trials?

Carb restriction is an effective way of reducing variety of diet in oppose to low fat diets. In the long run desire for variety (carb dishes) wins.

One more evolutionary strategy to ensure survival.

Stephan Guyenet said...

Hi Kindke,

Differential susceptibility. One could make the same argument about any factor that promotes obesity. Obesity susceptibility is more than half inherited-- put 10 people in the same fattening environment, and some will become obese while others will be lean. However, in a non-fattening environment, these heritable susceptibility factors are less relevant.

Hi Reijo,

I think it plays a role in the effectiveness of most diets, perhaps especially LC diets, which take most common food items off the table.

Gretchen said...

@Reijo: I've been on a LC diet for 17 years. How long does it take for that desire for carby dishes to kick in?

They no longer look tasty to me. My only problem with a LC diet is that it makes it difficult to enjoy potluck dinners where most of the foods are starchy.

Unknown said...

Thanksgiving has been proving this theory for many years!

As have Fairs! However the difference now seems that it's Thanksgiving and Fair food daily for many!

Aaron Blaisdell said...

I had the same reaction as some of the other commentors. Why not use real food rather than junk "food" designed to be supernormal appetitive stimuli? A couple of times a month, I eat lunch at a Brazilian Churrascuria, basically an all you can eat meat and salad bar fest. There are usually about 7 or 8 types of meats circulating throughout the lunch, and the salad bar runs the entire length of one long wall of the restaurant, and is stocked full of a variety of salads, vegetables, nuts, fruits, and meat stews. I eat to satiety, which is quite a bit (and more than I typically consume for lunch), but never feel overly stuffed nor do I need to loosen the belt buckle. The entire meal is a great experience (ask Keith and Michelle Norris or Skyler Tanner who joined me for dinner there last year) with a wide variety of real foods, but does not lead to overconsumption. In fact, it leads to underconsumption for the rest of the day, with super later that day usually consisting of something small and light. I know from personal experience that binging on junk food leads to greater hunger and consumption later, not less.

JWO said...

Tyler Cowen has said that any diet restriction will lead to weight loss. He says religious restrictions, low fat, low carb all lead to weight loss.

Never the less variety is that spice of life.

Stephan Guyenet said...

Hi Aaron,

A couple of points. The first is that many people do overeat spectacularly at churrascurias, including myself. I wonder, if you actually measured your total daily kcal intake on a day where you ate at a churrascuria, how it would stack up to a typical day? Sometimes it's hard to keep track of how many kcals you're eating in a situation like that.

That being said, meat specifically is a satiating food. So eating a lot of meat is going to keep you more full than eating a lot of most other types of food. But if it's fatty meat like it usually is at a churrascuria, you can still often take in a lot of kcals before that satiety kicks in.

Hi Mark,

I agree-- the variety factor puts in the last piece of the puzzle for me re: Thanksgiving overeating.

Hi Jim,


Aaron Blaisdell said...

Hi Stephen,

I agree that I eat at a churrascuria as if it were a feast (it is!). I don't doubt I consume many more kcal in that one meal than I do in a typical meal. Nevertheless, what I find interesting is the personal observation that despite the large feast, a) I rarely feel bloated or lethargic after the feast, and b) I feel much less hungry the rest of the day and even into the following day. This is quite a contrast with the way I would feel after I would eat at a buffet of SAD foods when I was following the SAD diet. I would feel bloated and lethargic afterward, and feel hungry or get "the munchies" such as cravings for snack-type foods (chips, cookies, etc.) only a few hours after the feast. It seems very apparent that the two types of meals affect me in quite different ways, both in the short and long term. I have not done a proper experiment of what would be the effect of chronic eating at churrascurias, but after completing one meal at a churrascuria, it is difficult to bring myself to want to eat at one for a while. This is all personal anecdotal observation, but seems to jive with what I've been gleaning from the scientific literature about the metabolic and appetite effects of "junk" foods (highly processed, carb-heavy) versus whole foods (low processed, more balanced in macronutrient ratios). I also don't mean to imply any generalities with regards to macronutrient ratios. There will likely be a large degree of individuality in responses to macronutrient ratios that come from whole foods.

Reijo said...

Gretchen, I'm happy that you do well. I do know many else like you, but even more those who do not cope with low carb diets for longer than half a year or so. That's my experience in clinical context.

I think this phenomenon is also clearly visible in clinical LC studies lasting longer than a year. Same applies to low fat diets though.

Robert said...

This new study is off-topic but related to leptin signalling in the hypothalmus.

Gretchen said...

@Reijo. Most people don't follow *any* diet for more than 6 months. That's why I don't think it's fair to blame LC diets for noncompliance.

Jane said...

Gretchen, the problem with low carb diets as I see it is that they often do not seem to correct the disorder they were adopted to correct: glucose intolerance. It looks like beta cell malfunction is difficult to cure on low carb diets. Could I ask you please, is the reason you are eating low carb after 17 years because you still can't tolerate starchy foods? Do you get symptoms of reactive hypoglycemia from them?

The point is, recent research suggests beta cell malfunction is caused primarily by iron overload and manganese deficiency, and low carb diets may not be the best way to correct this.

'Manganese supplementation protects against diet-induced diabetes in wild type mice by enhancing insulin secretion'

'Iron and diabetes risk' (thanks Stephan I got this one from you)

Low carb diets often have a lot of meat and saturated fat. Meat, especially beef, is very high in iron and very low in manganese. And saturated fat has been found to promote iron absorption and inhibit manganese absorption.

Gretchen said...

@Jane I have diabetes, and LC diets are the best way for most people to control BG levels. I eat less meat on a LC diet than I ate before, because I'm simply eating less food.

Mice have been cured of diabetes a zillion times.

Kirk said...

The 'simultaneous' presentation of high-reward foods shows a remarkable increase. As I consider how to apply this concept, it implies that I would gain more weight by combining foods, and yet would gain less weight by eating the same foods separated by a time interval. For example, instead of eating a piece of buttered gluten-free toast, eat an un-doctored piece of toast, and then, after a period of time, eat the butter. Do you have a gut-feel for what would be an effective time interval?

Medjoub said...


I'm curious how you manage to get high levels of manganese in your diet, as so many of the consistently high sources also contain anti-nutrients? Pineapple? Spinach?

Not being combative, just curious.

Jane said...

Hi Medjoub
I get most of it from grains and nuts. I eat wholemeal bread instead of meat, and the antinutrients don't bother me because I have no micronutrient deficiencies. Actually they aren't really antinutrients at all, in the sense of something the plant puts there because it doesn't want you to eat it. All of them have sensible functions, and we have sensible systems for dealing with them, which require micronutrients.

This question of getting enough manganese is a very interesting one, because the authors of the paper I linked about diabetic mice found that the Mn had to be injected because putting it in the drinking water didn't work. They think it was excreted. We know Mn undergoes enterohepatic circulation, meaning Mn that's actually needed gets excreted in the bile and re-absorbed. It may be that the saturated fat in the high fat diet prevented re-absorption.

Why on earth would we want to excrete all that precious Mn? Well it's possible that the purpose is to encourage friendly gut bacteria. Some kinds of Lactobacilli like Mn a lot. Gut bacteria are known to help mineral absorption, so it looks like they're supposed to give it back.

That paper about diabetic mice is interesting in another respect. The authors had found earlier in a mouse model of hemochromatosis, that the excess iron prevented entry of Mn into mitochondria, so MnSOD had no Mn and didn't work. This could explain why hemochromatosis patients get diabetes. In the more recent paper (and this has only just been published, so it hasn't had time to create a sensation) they found that wild type mice on a high fat diet also had MnSOD with no Mn. Now MnSOD isn't just important in pancreatic beta cells, it's important everywhere. And iron overload has been found in many common diseases. So prevention of Mn entry into mitochondria by iron could be a fundamental cause of disease, ANY disease.

It's kind of interesting that white flour has had most of its Mn removed ... and replaced with iron.

Jane said...

Gretchen, humans have also been cured of diabetes. Insulin treatment will do it, for instance. It may be that insulin makes beta cells work better by providing them with the nutrients they need for autophagy, which we know is what keeps beta cells healthy.

These nutrients include magnesium, known to be regulated by insulin, and manganese, which can enter cells in exchange for magnesium.

Jane said...

'Short-term intensive insulin therapy in type 2 diabetes mellitus: a systematic review and meta-analysis'

'In summary, our findings suggest that short-term intensive insulin therapy improves β-cell function and insulin resistance in patients with early type 2 diabetes mellitus. With this treatment, a high proportion of patients achieve long-term drug-free glycaemic remission...'

Gretchen said...

@Jane. This is not a diabetes board, so I don't want to debate you here. I don't agree with some of your comments. And beware of the word "cure."

If you want to discuss this privately, I'd be happy to do so. My e-mail is on my profile.

Jane said...

Gretchen, I tried to email you and delivery failed. I need to know which of my comments you don't agree with, and why.

Gretchen said...

@Jane. Oops. I forgot to change my profile when I changed ISPs. It's OK now.

I'm about to leave on vacation and won't be back for several weeks. If you're still interested then, remind me.

Unknown said...

If calories are really king in chronic disease risk than how do you explain the results of The Look AHEAD: Action for Health in Diabetes trial?
As qouted on

The Look AHEAD: Action for Health in Diabetes trial has been halted two years early. Here’s Gina Kolata in The New York Times:

The study randomly assigned 5,145 overweight or obese people with Type 2 diabetes to either a rigorous diet and exercise regimen or to sessions in which they got general health information. The diet involved 1,200 to 1,500 calories a day for those weighing less than 250 pounds and 1,500 to 1,800 calories a day for those weighing more. The exercise program was at least 175 minutes a week of moderate exercise.

But 11 years after the study began, researchers concluded it was futile to continue — the two groups had nearly identical rates of heart attacks, strokes and cardiovascular deaths.

It’s clearly a negative result for “eat less, move more” as a health strategy for obese diabetics.

Was “Eat Less Move More” Harmful?

A few Paleo bloggers are not surprised; indeed, Peter Dobromylskyj speculates that all-cause mortality – which Ms. Kolata and the NIH press release do not report – may have been higher in the “eat less, move more” intervention group:

It seems very likely to me that more people died in the intervention group than in the usual care group, but p was > 0.05.

Call me a cynic, but I think they stopped the trial because they could see where that p number was heading".


Stephan Guyenet said...

Hi Marcus,

I will refer you to the large DPP study, in which a combination of weight loss (low-fat calorie-restricted diet) and exercise caused a 58% reduction in diabetes incidence in pre-diabetic people. Multiple studies in several countries/racial groups have confirmed this result.

Peter's speculations about Look AHEAD are baseless and self-serving. There is no evidence that event rate was higher in the intervention group, in fact I'm 100% certain that there was no statistically significant effect, as I will explain. The problem with the Look AHEAD study is that it was statistically underpowered-- the event rate was lower than expected so they didn't have the numbers to make meaningful statistical comparisons. They terminated the study because there was no hope of obtaining a statistically significant result, given the small number of events. This was stated in the press releases for those who were willing to make the effort to look into it.

That being said, it's possible that once you develop diabetes, the approach of weight loss and exercise will be less helpful than prior to developing diabetes. Still, we know that the approach of weight loss and exercise is extremely effective at preventing diabetes. This has not been demonstrated for any other intervention, dietary or otherwise.

Gretchen said...

As far as I know, there's no evidence that the DPP prevented diabetes, meaning that the people who lost weight were diabetes-free forever. That's what "prevent" means to me.

It may simply have delayed the clinical onset in those with a strong genetic predisposition.

Of course delay is good. But I think many people misunderstand the meaning of prevention.

The real question is whether those who lost the weight lived any longer, and I don't think there's evidence about that yet. If there is, I'd love to have a reference.

Stephan Guyenet said...

The reduction in diabetes prevalence was significant at 10 years follow-up, which is remarkable considering adherence to the intervention was poor at that point. "Forever" is a strong word (particularly since it's impractical to design an intervention study that expects volunteers to adhere to an intervention forever), but the DPP intervention prevented diabetes about as well as could have been hoped for a controlled trial. When compared to the outcomes of other disease prevention trials, the DPP study stands out as one that clearly and strongly supported the hypothesis.

The DPP study was not designed with mortality as an endpoint. Total mortality of course is an important metric, but it's not the only way to measure health. People with diabetes live a long time now with proper care, but it's still a condition most people would rather not have.

Sanjeev said...

> A few Paleo bloggers are not surprised; indeed, Peter Dobromylskyj speculates ... Ms. Kolata and the NIH press release do not report

Gee ... I wonder why PETRO, of all people, thinks of UNreported / deleted data.

why, why, WHY? most curious.

yeah, NO