Does the Vitamin and Mineral Content of Food Influence Our Food Intake and Body Fatness?

The Claim: We Overeat Because Our Diet is Low in Vitamins and Minerals

We know that animals, including humans, seek certain properties of food.  Humans are naturally attracted to food that's high in fat, sugar, starch, and protein, and tend to be less enthusiastic about low-calorie foods that don't have these properties, like vegetables (1).  Think cookies vs. plain carrots.

In certain cases, the human body is able to detect a nutritional need and take steps to correct it.  For example, people who are placed on a calorie-restricted diet become hungry and are motivated to make up for the calorie shortfall (2, 3).  People who are placed on a low-protein diet crave protein and eat more of it after the restriction is lifted (4).  Humans and many other animals also crave and seek salt, which supplies the essential minerals sodium and chlorine, although today most of us eat much more of it than we need to.  At certain times, we may crave something sweet or acidic, and pregnant women are well known to have specific food cravings and aversions, although explanations for this remain speculative.  Research suggests that certain animals have the ability to correct mineral deficiencies by selecting foods rich in the missing mineral (5).

These observations have led to a long-standing idea that the human body is able to detect vitamin and mineral (micronutrient) status and take steps to correct a deficit.  This has led to the secondary idea that nutrient-poor food leads to overeating, as the body attempts to make up for low nutrient density by eating more food.  In other words, we overeat because our food doesn't supply the micronutrients our bodies need, and eating a micronutrient-rich diet corrects this and allows us to eat less and lose body fat.  These ideas are very intuitive, but intuition doesn't always get you very far in biology.  Let's see how they hold up to scrutiny.

The Evidence

For this hypothesis to be correct, the human body would have to:

1. Have the ability to evaluate its own micronutrient status for a variety of micronutrients.
2. Have the ability to detect the micronutrient content of food, whether instinctively or by learning.  
3. Have the ability to promote the consumption of foods that contain the needed micronutrients, or eat more food in general.
4. Prioritize micronutrient status over calorie balance.

In short, I'm not aware of much scientific evidence that supports any of this in humans.  Micronutrient deficiency does have physiological effects on the human body of course, but for the most part I'm not aware of clear evidence that those effects are able to influence behavior in a way that would aim to correct the specific deficiency.

Micronutrient deficiencies are often associated with a reduction of appetite, not usually an increase, suggesting that the body does not generally attempt to increase overall food intake to correct a deficiency.  This is also true of animals, which tend to decrease overall food intake and lose weight on nutrient-deficient diets (6).  Correcting the deficiency restores appetite and body weight.

We know the human body can detect deficiencies of calories and protein, and respond accordingly, and at least for calories (and protein to a lesser extent) the systems that regulate this are fairly well understood (7).  For micronutrients, there is little evidence that such a detection and regulatory system exists.  We know that the body can detect sodium chloride (salt), because we can taste it.  Beyond that, I'm not aware of any clear evidence that the human body can detect the micronutrient content of foods or respond to deficiencies by altering behavior.  The fact that some animals can do it does leave open the possibility that we simply haven't uncovered this effect in humans yet.

We can detect many food properties that are indirectly associated with micronutrients.  For example, sweetness and tartness would have been associated with potassium and vitamin C in the wild, because fruit is rich in both.  The flavor and aroma of cooked meat (glutamate and volatiles) would have been associated with B vitamins, iron, and zinc.  Some animals can learn which foods are poor and rich sources of minerals (8).  Yet in the modern world, these potential nutritional guidelines are easily fooled by added sugars, glutamate, and flavorings, uncoupling the natural association between flavor/aroma and nutrition.  Humans do seek variety in the diet*, perhaps for nutritional reasons, but I'm not aware of evidence that this behavior is modified by the body's nutritional status or the actual micronutrient content of foods.

Another problem with this hypothesis as an explanation for obesity is that modern Americans have relatively good micronutrient status by global and recent historical standards.  Although we may not eat an optimal amount of all micronutrients, frank deficiency is uncommon, in large part because much of our food is fortified.  The picture was quite different a century ago.  Before food fortification, deficiency of iodine (goiter), vitamin D (rickets), niacin (pellagra), and vitamin C (scurvy) were common in specific regions of the United States.  Yet this was true at a time when the prevalence of obesity was much lower than it is today-- these people were evidently not eating more food to make up for deficient micronutrient status.

If our bodies can detect micronutrients in food and act to correct deficiencies, and this is a major driver of behavior, why don't people crave vegetables and liver rather than chips and soda?  Why eat a larger quantity of micronutrient-poor food to try to make up the nutrient shortfall when it would be much more effective to just swap out the junk for nutrient-dense foods that are readily available?  Yet many of these nutrient-dense foods remain unappealing to most people of all weights, while nutrient-poor foods remain appealing, suggesting that what humans seek from food is primarily something other than micronutrients.

Randomized Controlled Trials

The best test of this hypothesis would be to determine if 1) reducing the micronutrient content of the diet independently of other factors increases food intake in lean people, or if 2) increasing the micronutrient content of the diet independently of other factors decreases food intake in overweight people.  The second prediction has been tested by using multivitamin/mineral supplementation in randomized controlled trials.

The first trial I came across was published in 2010 (9).  In a group of obese Chinese women, a multivitamin/mineral supplement caused no significant change in the food-derived intake of calories, protein, fat, carbohydrate, or any of the 15 measured micronutrients.  However, over 26 weeks, the supplement did increase calorie expenditure, modestly reduce fat mass, and improve markers of metabolic and cardiovascular health.  This study does not support the idea that we eat to obtain micronutrients, but it does suggest a possible intriguing effect of micronutrient status on calorie expenditure, metabolic health, and body weight, at least in obese Chinese women.

I found this result difficult to accept uncritically, because it's hard for me to believe that an effect this large on markers of metabolic health wouldn't have been noticed in all the other studies involving multivitamin/mineral supplementation.  It turns out, there is other evidence that relates to this question.  My favorite study was published in 2008 and combined an observational study with a randomized controlled trial (10).  The observational study, like many others, showed that people who habitually take multivitamin/mineral supplements tend to be leaner than people who don't.  However, as with many diet/lifestyle variables, these people had an overall healthier lifestyle pattern that suggested they were more health-conscious, casting doubt on how much of their leanness could be attributed directly to the supplements as opposed to some other variable associated with it.

To test this hypothesis more directly, the investigators administered a multivitamin or placebo to obese men and women during a 12-week weight loss intervention.  Changes in fat mass and energy expenditure didn't differ between groups.  Women, but not men, had a reduced "fasting desire to eat" in the supplement group, which was one of four measures they used to measure food-related motivation.  The other three measures, "hunger", "fullness", and "prospective food consumption", didn't differ between supplement and placebo groups in either gender.  In my view, this study does not support the idea that micronutrient status plays a significant role in appetite and body weight, or increases the effectiveness of fat loss efforts.

The Bottom Line

There's little evidence that the human body is able to detect its own micronutrient status, detect the micronutrient content of foods, or respond behaviorally to manage micronutrient status, aside from sodium chloride**.  Neither humans nor other animals appear to eat more calories to make up for a micronutrient shortfall, or eat less when the micronutrient density of the diet is increased, and therefore I doubt that the diet's overall micronutrient density plays a direct role in overeating, although specific micronutrients could still have biological effects that tip the scales in one direction or the other.

There is an intriguing suggestion that micronutrient supplementation can influence calorie expenditure and body fatness, but this will require independent confirmation before I can accept it.  Also, we may have more to learn about human food selection in response to nutritional requirements, since we know that some animals are capable of it.

Regardless of this, eating nutrient-dense whole foods remains a good idea, and it does help with fat loss, even if that has nothing to do with vitamins and minerals.  Similarly, eating nutrient-poor highly processed foods can lead to overeating, but we don't have to invoke micronutrients to explain that.  It's probably more related to the fact that this food is seductive, palatable, calorie-dense, and provides little satiety per calorie, making it easy to overeat and continue eating independently of hunger.  There may be other reasons as well, such as effects on the gut microbiota.


* Clear evidence of this comes from research on sensory-specific satiety (11).

** This implies that NaCl was an important and scarce nutrient in our evolutionary past.  This remains true of many wild animals and a few human groups with no access to concentrated NaCl.  Humans may have an elevated need for NaCl because we sweat more than other animals, but our current intake in affluent nations far exceeds this requirement.