The History of BMI

The formula you run in a BMI calculator is nearly two centuries old, and it was never designed to assess an individual's health. Its journey from statistical curiosity to the world's default weight metric explains most of its quirks — including why its own populariser warned against using it the way we now use it.

1830s: Quetelet's "average man"

Adolphe Quetelet was a Belgian astronomer and mathematician obsessed with applying statistics to human populations — he is also a reason the concept of the "average man" exists. Measuring Belgian conscripts, he observed that adult weight scales roughly with the square of height, and defined the ratio weight/height² — the Quetelet Index — to describe a population's build. He was explicit that this was social statistics, not medicine; the index was a tool for describing groups, never for judging one person.

Early 1900s: insurers discover weight

The first people to attach money to body weight were American life insurers. Companies like Metropolitan Life built "desirable weight" tables from policyholder mortality data, complete with unscientific "frame size" adjustments. These tables dominated clinical practice for half a century, but they were built on a narrow, self-selected population — people who bought life insurance — and the frame-size fudge was essentially unmeasurable.

1972: Ancel Keys renames the index

Physiologist Ancel Keys — already famous for work linking diet and heart disease — compared the available weight-for-height indices across large population datasets and concluded Quetelet's was the least-bad proxy for body fatness at population scale. His paper coined the modern name: body mass index. The nuance usually lost: Keys recommended BMI for population studies and explicitly cautioned that it was inappropriate for diagnosing individuals — the exact use it would soon acquire.

1980s–1990s: the WHO makes it official

BMI's virtues — free, fast, requiring only a scale and a tape — made it irresistible to public health. The WHO adopted it for obesity classification in the 1990s, fixing the now-familiar cut-offs: 18.5, 25, 30. Round numbers chosen for memorability as much as biology; the underlying risk curves are smooth, with no cliff at 24.9. In 2004 the WHO added lower action points (23/27.5) for Asian populations after evidence showed risk rising earlier — the story told in Asian BMI cut-offs.

Why imperial BMI has a 703 in it

The formula is metric: kg ÷ m². To use pounds and inches directly, you multiply by a conversion constant: BMI = 703 × lb ÷ in². The 703 is nothing more than (kilograms-per-pound × inches-per-metre²) rounded — a unit-conversion artifact, not a magic number.

2000s–today: workhorse and lightning rod

Modern critiques — that BMI misreads muscular athletes, ignores fat distribution, ages poorly (BMI by age) and was calibrated on European bodies — are all fair, and mostly rediscoveries of caveats Quetelet and Keys stated themselves. In 2023 the American Medical Association formally advised pairing BMI with other measures such as waist circumference and body composition. Yet BMI persists, because nothing equally free and universal has replaced it: alternatives like waist-to-height ratio complement it rather than displacing it.

Timeline at a glance

Bottom line

BMI is a 19th-century population statistic wearing a 20th-century medical coat. Used the way its own authors intended — a quick screen, read alongside waist, age and build — it still earns its keep. Used as a personal verdict, it was never fit for purpose, and the people who built it said so first.