On Good Fat, Bad Fat, and the Problems With BMI

Today, the terms “obese” and “overweight” have precise medical definitions based on how much someone weighs relative to their height, but this is a modern formulation. Until recently, people knew their height but almost no one could measure their weight, so whether you were thin, plump, fat, stout, or corpulent was a judgment call. That began to change after the invention and spread of the penny scale in the late nineteenth century, so named because you paid a penny to find out your weight. Within a few decades, inexpensive bathroom scales were invented, and suddenly average people could learn their weight. An obvious next question was how much should they weigh?

The first scientist to address this question was a Belgian polymath named Adolphe Quetelet (1796–1874). Quetelet was uninterested in obesity and instead wanted to use recently developed spring-based scales along with new statistical methods to describe a “normal man.” Since tall people generally weigh more than short people, what, he asked, is a typical weight for a given height? After collecting data on a sample of European men, he calculated that body weight (in kilograms) is usually proportional to the square of height (in meters). Thus two people with the same ratio of weight to height squared (kg/m2) have effectively the same weight for a given height. This ratio was initially termed “Quetelet’s index.”

That kind of miscategorization highlights BMI’s most disturbing shortcomings: it is reductive, potentially misleading diagnostically, and stigmatizing.

Quetelet didn’t intend his index to measure obesity, but after the invention and spread of mass-produced scales, insurance companies realized Quetelet’s index could help them boost profits. If you are selling life insurance, you can make more money by charging more or denying coverage for people with a lot of extra body fat who might be more likely to die prematurely. The first standardized weight tables were thus developed by the Connecticut Mutual Life Insurance Company, which initially deemed anyone whose Quetelet’s index was in the top 20 percent of a sample of seven hundred thousand policy holders too risky to insure. For the next fifty years, insurance companies, not doctors, were the chief arbiters of corpulence.

Then in 1972, the influential American physiologist Ancel Keys wondered how well Quetelet’s index predicts a person’s body fat percentage. Using a sample of nearly 7,500 American men (less than 2 percent were Black and 14 percent were Asian), Keys and colleagues showed that Quetelet’s index, which he renamed the “body mass index” (BMI), correlates strongly with levels of body fat.11 Despite Keys’s warning that the “characterization of persons in terms of desirable weight percentage has resulted in attributing to ‘overweight’ some tendencies to ill health and death that are actually only related to age,” BMI became the worldwide standard for categorizing people’s body fat levels. Today most healthcare providers and researchers classify everyone according to six categories set in stone by the World Health Organization (WHO):

Underweight: <18.5 kg/m2
Normal: 18.5–24.9 kg/m2
Overweight: 25.0–29.9 kg/m2
Obese (Class I): 30–34.9 kg/m2
Obese (Class II): 35–39.9 kg/m2
Obese (Class III): >40 kg/m2

It’s often said that an accurate measurement is worth a thousand expert opinions, but I’m not so sure. Just because a measure is accurate (or, for that matter, precise) doesn’t mean you understand its meaning. Despite their widespread use for decades, BMI categories have serious flaws. One glaring drawback is that BMI doesn’t measure what proportion of a person’s weight is fat versus muscle, bone, or other tissues. Very muscular people are sometimes miscategorized as overweight.

In addition, percentages of fat and other tissues differ among individuals depending on age, sex, and race or ethnic background. An analysis of nearly twelve thousand Americans found that for the same level of body fat, age, and sex, the BMIs of Black Americans averaged 1.3 kg/m2 higher compared to Americans of European descent, while the BMIs of Asians was between 1.9 and 3.2 kg/m2 lower. That means BMI typically overestimates the percentage of body fat in Blacks and underestimates it in Asians. Based on percentage of body fat, a Black person with a BMI of 26 is likely to be miscategorized as overweight instead of normal.

If you ask (as I have) hunter-gatherers what they prefer to eat, more often than not the answer is meat, especially fatty meat.

That kind of miscategorization highlights BMI’s most disturbing shortcomings: it is reductive, potentially misleading diagnostically, and stigmatizing. Research shows that labeling individuals as obese or overweight based on this single ratio can make them feel judged, marginalized, and devalued. What’s more, even if someone’s BMI accurately reflects their body fat percentage, BMI is not a complete measure of health.

At the very least, that also requires considering blood pressure, resting heart rate, cholesterol levels, blood sugar level, liver function, inflammation, and fitness. It’s also overly reductionist to consider fat as just a substance you either have too much or too little of. Fat is a vital class of molecules that every cell uses; our bodies store fat in special cells, adipocytes, that have many functions. And to top it all, fat played a special role in human evolution. If we want to better understand obesity, we have to learn more about fat.

If you ask (as I have) hunter-gatherers what they prefer to eat, more often than not the answer is meat, especially fatty meat. Most wild animals are extremely lean, so hunter-gatherers crave the fat-rich parts of the body, including the marrow and liver. Like almost everyone, they think fat is delicious. They also appreciate that you need enough fat to be healthy, have children, and not be grumpy.

As a reminder, animals—humans included—store most of their energy as fat. Each fat molecule has three fatty acid chains that stick off like tines on a fork from a tiny glycerol molecule. Burning fat molecules in cells harnesses 9 calories per gram by retrieving energy from the bonds between the fatty acids’ many carbon atoms, which are strung together like beads. Because fat is denser than carbs, it stores more than twice as many calories per gram, making it our primary repository of energy.

Fat makes life possible but has one salient constraint: it doesn’t dissolve in water. As a result, we can’t stockpile fat inside tissues the way we store carbs. Instead, most of our body fat stores need to be sequestered in adipocytes. Keep in mind that these cells aren’t just little inflatable sacs of fat—they are dynamic cells with a nucleus, DNA, and organelles. Among fat cells’ many functions, they synthesize and secrete hormones and other molecular messengers that signal to the body how much fat they contain, whether they are shrinking or damaged, and much more.

At birth, we have about one billion of these fat cells, and by the time we reach adulthood, we have approximately ten to thirty billion. But there is variation. Women tend to have more fat cells than men, and people who are overweight or obese tend to have more fat cells than people who are classified as “normal” weight, especially if they developed obesity during childhood and adolescence. Although 10 percent of fat cells are replaced every year, the total number stays relatively constant as adults age. Like balloons, fat cells swell or shrink as we gain or lose weight.

The relative constancy of fat cell numbers combined with their ability to change size and function has implications for the evolutionarily unusual state of being obese. Although you can’t see or feel it, when fat cells store more fat than they should, they become damaged, like an overstuffed suitcase bursting at the seams. They also don’t get enough oxygen. Damaged, deprived fat cells then synthesize and secrete chemical messengers (adipokines) that activate an immune response, triggering inflammation. That inflammation is a low-grade version of the hot, red swelling you get from a hangnail or a sore throat, but because it’s in your fat cells, you don’t feel it.

What’s more, fat cells are everywhere, and that inflammation can spread to nearly every nook and cranny of the body. Despite its pervasiveness, scientists didn’t pay much attention to this kind of simmering, body-wide inflammation until modern technologies to measure it were developed. The ability to measure this sort of inflammation has been a game changer because if it persists for years, inflammation’s effects are pernicious. Bit by bit, low-grade systemic inflammation attacks cells all over the body, causing them to malfunction.

One common form of inflammatory damage is to obstruct fat and muscle cells’ ability to respond to the hormone insulin. As we will learn later, insulin is needed to transport sugar and fat molecules out of the bloodstream and into cells. Inflammation triggered by obesity gums up these actions, causing both blood sugar and cholesterol levels to rise silently, gradually, and imperceptibly, causing harm. One bad consequence is type 2 diabetes. Another is hardening of the arteries. Elsewhere, sustained low-grade inflammation impairs other kinds of receptors and tissues. Inflammation in the brain damages neurons, increasing the risk of Alzheimer’s disease and other forms of dementia. Inflammation also increases the risk of cancers, osteoarthritis, asthma, inflammatory bowel disease, and other illnesses. And, scarily, this kind of inflammation is mostly imperceptible. It slowly and surreptitiously damages your body.

hether we are normal weight or obese, shrinking fat cells induce powerful sensations of hunger and lethargy, countering efforts to lose weight.

To add insult to injury, the inflammation obesity causes also sets in motion a vicious cycle that can promote yet more weight gain. This happens because adipocytes produce hormones and other signaling molecules in response to how much fat they contain. The most important of these hormones is leptin, which signals to the brain information about the body’s energy supply. When your fat cells are adequately stocked with fat, leptin levels are high, and your brain suppresses appetite. When your fat cells shrink as you burn fat, leptin levels drop, stimulating hunger.

Normally this system helps us maintain a nearly constant weight, but an insidious effect of low-grade inflammation is to damage the brain’s leptin receptors. When this happens, the normal feedback system malfunctions like a broken thermostat. Even though an obese person’s fat cells may produce enough leptin to suppress their appetite, the brain doesn’t receive those signals, causing constant hunger. Hunger, in turn, fights against weight loss and possibly even spurs more weight gain.

And if that weren’t bad enough, fat cells cause problems not only when they get too big but also when they shrink. As people lose weight by burning fat and their fat cells deflate, leptin levels nosedive, warning the brain that the body’s energy stores are in decline. Whether we are normal weight or obese, shrinking fat cells induce powerful sensations of hunger and lethargy, countering efforts to lose weight. This response was presumably adaptive to prevent our ancestors from becoming underweight but is disastrous for today’s dieters.

It makes sense that we evolved to detect and respond to declining levels of fat because having too little fat is ruinous for reproduction and deleterious for other functions. However, because obesity was so rare in our evolutionary history, we apparently never evolved adaptations to prevent us from storing too much fat. But how much is too much?

The philosopher Ralph Waldo Emerson (1803–1882) wrote that a “foolish consistency is the hobgoblin of little minds.” Emerson was complaining about politicians and theologians with inflexible opinions closed to new ideas and facts, but he might as well have been writing about how we have medicalized obesity. Consider that a five-foot, ten-inch person weighing 208.5 pounds has a BMI of 29.9 and is technically overweight, but if they gain half a pound, thus increasing their BMI to 30.0, they will suddenly be obese. And according to the American Medical Association, they then have a disease. In addition to being arbitrary, reductionist, and biased, classifying individuals as obese or overweight based on a single number ignores many complexities about body fat.

We are evidently best adapted to store enough but not too much.

“A rose is a rose is a rose” according to Gertrude Stein, but the same cannot be said of body fat. There are several different types of body fat, and they behave differently. The most consequential distinction is between fat that lies under the skin, termed “subcutaneous fat,” and fat in and around organs like the liver, termed “abdominal fat” (also called “visceral fat,” “belly fat,” or “ectopic fat”).* We evolved to store mostly subcutaneous fat and have only a little abdominal fat. Abdominal fat, in addition, is more sensitive to hormones and has more blood vessels. In essence, abdominal fat is a readily accessed, short-term energy reserve for use in times of need. For this reason, during stressful situations like a famine or during conflicts when the body might require lots of extra calories, the brain directs cortisol levels to rise, which among other things causes the body to store more abdominal fat. That additional abdominal fat is an adaptation to help us cope with short-term food shortages or times when we need to be extra physically active.

But too much abdominal fat is a health concern. When abdominal fats cells over-swell, they ignite more chronic inflammation than do subcutaneous fat cells. Hundreds of studies have shown that individuals with higher levels of abdominal fat have elevated risks of developing inflammation-related illnesses, especially type 2 diabetes and heart disease.

The difference between abdominal and subcutaneous fat highlights a major concern about obesity along with a major limitation of BMI. While excess abdominal fat is more disease-causing than subcutaneous fat, BMI is calculated from just weight and height and thus fails to distinguish between these different fat types. That’s a problem because subcutaneous fat deposits in places like the hips and buttocks don’t cause as much inflammation as fat in muscles or in organs like the liver. For this reason, we should stop using just BMI to assess fatness and employ other measures to estimate fat distribution. Some hospitals have sophisticated scanners that can do this, but a simple way to estimate abdominal fat levels is to use a measuring tape. The ratio of a person’s waist to hip circumference, or their waist circumference to height are pretty good estimators of their abdominal fat deposits. One can also look in the mirror: someone storing more fat in the abdomen than elsewhere, including the hips, will have more of an apple than a pear shape.

All things considered, let’s be more thoughtful about how we measure and use body fat levels to assess someone’s health or anything else about them. At the same time, while it’s wrong to judge, stigmatize, or discriminate on the basis of weight, it’s true that high levels of body fat, especially abdominal fat, increase a person’s risk of many diseases. Even though BMI is a flawed measure for individuals, it is useful for assessing population-level associations between body fat levels and long-term health outcomes. Figure 14 shows the relationship between BMI and the risk of dying from any cause among more than 3.5 million women and men in the United Kingdom.

Keep in mind that this analysis includes people of all races; it measures only longevity, not health; it doesn’t distinguish between abdominal and subcutaneous fat; and it doesn’t account for exercise, access to healthcare, and other factors. Despite these limitations, you can see the J-shaped relationship between BMI and the risk of all-cause mortality. People with BMIs below 18 and above 30 are at greater risk of dying. The graph also shows that having a little more fat is beneficial as people age. We are evidently best adapted to store enough but not too much.

That, in turn, raises a difficult question: If excess fat isn’t good for us, why are humans so prone to storing too much?

__________________________________

From Fed Up by Daniel E Lieberman. Copyright © 2026 by Daniel Lieberman Published by arrangement with Alfred A Knopf, an imprint of The Knopf Doubleday Group, a division of Penguin Random House LLC