Your BMI Was Invented by an Astronomer Who Never Meant It for Your Body
Body Mass Index runs modern medicine — but it began as a 19th-century population statistic by a Belgian stargazer, was renamed by a heart researcher in 1972, and reclassified millions overnight in 1998. Here's the strange, true story.
The number your doctor scribbles next to your weight, the figure that decides whether a health app paints your dot green or red, the metric governments use to declare an "obesity epidemic" — all of it traces back to a Belgian astronomer who was not studying health, was not studying fat, and never intended his little formula to be applied to a single human being.
That is the strange truth about Body Mass Index. It is one of the most consequential numbers in modern medicine, and it was essentially a side project in 19th-century statistics.
A formula born from stargazing, not medicine
Adolphe Quetelet was born in Ghent in 1796 and trained as a mathematician and astronomer. But his real obsession was bigger and stranger: he wanted to apply the rigorous, predictable mathematics of the heavens to people. He called the field "social physics," and his guiding idea was l'homme moyen — the "average man."
Quetelet had fallen in love with the bell curve. He noticed that if you measured enough soldiers' chest sizes or enough people's heights, the results clustered around an average in that familiar Gaussian shape, the same distribution astronomers used to handle measurement errors. To him, the average wasn't just a number; it was a kind of ideal that nature was aiming at, with individuals scattered around it like stars around a mean position.
Somewhere in this work — published in the 1830s — he made a simple observation. Across a population, people's weight tended to increase not in proportion to their height, but roughly in proportion to their height squared. Divide weight by height squared and you got a number that stayed fairly stable whether someone was short or tall. That ratio became the Quetelet Index.
Here is the part everyone forgets: Quetelet was describing a population. He was sketching the statistical shape of a whole society, not diagnosing the man in front of him. He never suggested the index measured fatness, fitness, or health. It was a tool for studying groups, full stop.
The 130-year nap
For more than a century, Quetelet's index mostly gathered dust in the world of statistics. Doctors who wanted to judge a patient's weight reached for something else entirely: insurance tables.
In the first half of the 20th century, life insurance companies — most famously Metropolitan Life — crunched their own data on which policyholders died early and built "ideal weight" charts. They listed a desirable weight for each height (and even "frame size": small, medium, large). For decades, "what should I weigh?" was answered not by a formula but by a grid printed on a card, derived from who was profitable to insure.
Those tables had real problems. The frame-size categories were vague, the data came from a self-selected slice of people who bought insurance, and the numbers shifted depending on which company you asked. Medicine wanted something cleaner.
How a heart researcher gave it the name
The rebrand came in 1972. Ancel Keys — the American physiologist behind the wartime K-ration and the sprawling Seven Countries Study on diet and heart disease — set out to find the best simple proxy for body fat. His team compared several weight-and-height formulas against actual body-fat measurements across thousands of men.
Quetelet's old ratio won. It wasn't perfect, but among the cheap, easy options it tracked body fat better than the rivals. In that paper Keys gave the 140-year-old index a snappier, more clinical name: the Body Mass Index.
And — this matters — Keys issued the same warning Quetelet implied. He stated plainly that the index was suitable for population studies but was not appropriate for evaluating individuals. The caution was baked in from the very paper that made BMI famous. The world adopted the formula and quietly ignored the asterisk.
If you want to see your own number on a proper color-coded scale — alongside your healthy weight range, body-fat estimate and BMR — you can run it through a BMI calculator in a few seconds. Just keep that asterisk in mind while you read the result.
The night millions of people "became" overweight
Because BMI is just arithmetic, the categories built on top of it are arbitrary lines drawn by committees — and those lines have moved.
In June 1998, the U.S. National Institutes of Health adopted the World Health Organization's cut-offs and lowered the threshold for "overweight" from a BMI of 27 down to 25. No one's body changed that night. No one ate a different dinner. Yet an estimated 29 million Americans went to sleep "normal weight" and woke up officially "overweight," purely because a definition shifted. It's one of the cleanest illustrations you'll ever find that a health category can be a policy decision as much as a biological fact.
Why tall people get a raw deal (and short people a free pass)
There's a quieter mathematical complaint baked into the formula itself. Dividing by height squared makes a hidden assumption: that people scale like flat shapes rather than three-dimensional bodies.
In reality, if you took a person and scaled them up uniformly, their weight would grow with height cubed. Real humans land somewhere in between — they don't get proportionally wider as they get taller. The upshot, as Oxford mathematician Nick Trefethen argued in a widely shared 2013 letter, is that the squared formula makes tall people look heavier than they really are and short people look lighter. He proposed tweaking the exponent closer to 2.5 to fix the bias. Quetelet himself had hinted at this nearly two centuries earlier, noting weight grew a touch faster than height squared. The "²" survives mostly because it's easy to compute in your head, not because it's the most accurate choice.
One number, very different bodies
The most famous failure of BMI is the muscle problem. Muscle is roughly 18% denser than fat, so a lean, heavily trained athlete can weigh a lot for their height and score as "obese" while carrying single-digit body fat. Squads of rugby players, sprinters and bodybuilders technically flunk the BMI test. The number simply can't tell a kilo of muscle from a kilo of fat — it only sees total mass. A pound of muscle and a pound of fat weigh exactly the same, of course, but the muscle takes up noticeably less room, which is why two people at the identical BMI can look — and be — completely different.
It also can't see where fat sits, which turns out to be the part that matters most. Fat packed around the organs in the abdomen (visceral fat) carries far more metabolic risk than the same amount on the hips and thighs. Two people can share an identical BMI while one carries a high-risk belly and the other doesn't. That's why many researchers now lean on waist circumference or the waist-to-height ratio — a rule of thumb being to keep your waist under half your height — as a sharper, almost-as-simple check.
It isn't even the same line everywhere
Health risk at a given BMI also varies by population. Research found that many people of South and East Asian descent develop diabetes and heart disease at lower BMIs than the standard cut-offs suggest. In response, the WHO flagged additional "action points" for Asian populations — with overweight risk rising closer to a BMI of 23 — and countries like Japan define obesity from a BMI of 25, not 30. A single global threshold, it turns out, fits no one perfectly.
So why won't BMI die?
Given all these caveats, you might expect medicine to have dropped BMI long ago. It hasn't, and the reason is brutally practical: it is almost free. It needs only a tape measure and a scale, no lab, no calipers, no scan. For studying millions of people or screening a whole clinic, nothing else comes close on cost and speed — exactly the population-scale job Quetelet designed it for in the first place.
The trouble was never the formula. It was the expectation we loaded onto it: that a single ratio could judge one specific person's health. Used as Quetelet and Keys intended — as a quick, rough flag that invites a closer look — BMI is genuinely useful. Treated as a verdict on your body, it overreaches every time.
The takeaway
Next time a number tells you which "category" you belong to, remember where it came from: a stargazer mapping the average citizen, a heart researcher hunting a cheap proxy, and a committee drawing a line in 1998. Your BMI is a useful starting point and a famously blunt instrument at the same time.
Treat it like a smoke alarm, not a diagnosis. If you're curious, calculate yours, note your healthy weight range, then look at the fuller picture — your waist, your strength, your energy, your bloodwork. A number invented to describe a crowd was never going to capture the one person it matters to most: you.
