What BMI Actually Measures, and When It's Wrong About You

That gap, between what BMI can do and what people ask it to do, causes most of the confusion around it. The number is not lying to you. It is answering a narrower question than you think it is.

Where the number came from

BMI is almost 200 years old, and no doctor invented it. Adolphe Quetelet was a Belgian astronomer and statistician, and in the 1830s he was trying to describe the average person in a population. He found that among adults, weight rises roughly with the square of height. The ratio he wrote down, weight divided by height squared, is the formula still used today.

Quetelet was studying groups. He never set out to diagnose anybody, and never suggested his ratio could.

The name arrived 140 years later. In 1972 the American physiologist Ancel Keys compared several weight-for-height formulas against measured body fat and found Quetelet's was the best of them. He called it the body mass index, and he recommended it for population studies rather than for diagnosing patients.

What does BMI actually measure?

BMI is your weight divided by the square of your height. It estimates whether your weight is in a healthy range for your height. It measures weight and height and nothing else — not fat, not muscle, not bone, and not where fat sits on your body. For adults, the standard healthy range is 18.5 to 25.

In metric, divide kilograms by metres squared. In imperial, divide pounds by inches squared and multiply by 703, which converts between the two scales.

That is the entire calculation. Two inputs, one output. Everything BMI does not know about you is left out at this step, and that single fact explains every complaint ever made about it.

The CalculatNow BMI calculator runs it in either pounds or kilos, and shows both the band your result falls in and the healthy weight range for your height.

What the categories mean

The CDC sorts adult BMI into four bands, and splits obesity into three classes. The split matters, because the health risk at 31 is not the health risk at 45.

The CDC is direct about what those bands are for. BMI is a screening measure, not a diagnosis. It flags who might benefit from a closer look, and that is the whole of its job.

Why the healthy range is 18.5 to 25

The bands are not arbitrary, and this is the part that critics of BMI tend to skip. They come from decades of data linking body weight to how long people live.

The largest study of it is the Global BMI Mortality Collaboration, published in The Lancet in 2016. It pooled 239 studies covering 10.6 million adults in 32 countries. Death rates from any cause were lowest between a BMI of 20 and 25, and rose on both sides of that band. The curve is J-shaped: risk climbs steeply above 30, and rises again below 20.

So the healthy band is real, and it is earned. Across millions of people BMI predicts risk well, and it does so using a set of scales and a mark on a wall. No blood test comes close on cost.

Why is my BMI high when I'm not fat?

Because muscle weighs more than fat for the same amount of space, and BMI sees only the weight. Put on 10 kg of muscle and your BMI rises by exactly as much as if you had put on 10 kg of fat.

The clearest demonstration comes from American football. Researchers at the Steadman Philippon Research Institute measured 1,958 athletes at the NFL Scouting Combine and compared the two methods on the same men. Their 2018 study is worth reading in full if you have ever been told your BMI is too high.

By BMI, more than half of them were obese. By body composition, fewer than one in ten. These are elite athletes, so the misclassification is not a rounding error at the edges. It is most of the group.

You do not have to be a professional athlete for this to reach you. Anyone who lifts regularly, does heavy physical work, or is simply built on a wide frame will read higher than their body fat justifies. If that describes you, body fat percentage is the more honest measure.

The reverse happens too, and it gets far less attention. Someone with little muscle and a lot of fat around the middle can sit comfortably inside the healthy band while carrying real metabolic risk. A normal BMI is not a clean bill of health.

Where BMI misreads whole groups

Individual variation is one problem. The second is that the thresholds were drawn from a narrow slice of humanity, and they do not transfer evenly to everyone else.

Ancestry

Risk does not begin at the same number everywhere. A World Health Organization expert consultation reviewed the evidence for Asian populations and published its findings in The Lancet in 2004. It concluded that heart disease and type 2 diabetes risk rise at a lower BMI than the standard cut-off of 25. Across different Asian populations, the point where that risk became visible ranged from 22 to 25.

WHO kept the international bands so countries could still compare figures, but added public health action points at 23.0 and 27.5. Several health systems across Asia now begin screening at a BMI of 23.

Age and sex

BMI does not distinguish men from women, and it does not adjust for age. Women carry more body fat than men at the same BMI. Older adults tend to lose muscle and gain fat while the scales barely move. A BMI that has not shifted since someone was 50 can therefore hide a real change in what their body is made of.

Children and teenagers

The adult bands do not apply to anyone under 20. A child's BMI is read against age- and sex-specific percentiles, because body composition changes throughout growth. Their number is compared with other children of the same age and sex, never against 18.5 and 25.

In June 2023 the American Medical Association adopted a policy on all of this. It called BMI an imperfect measure when used on its own. The cut-offs, it noted, rest largely on data from earlier generations of non-Hispanic white populations, and account for neither sex nor ethnicity. The AMA asked doctors to read BMI alongside other measures rather than alone.

What to measure alongside it

BMI is worth having. It is just not worth having on its own. The single most useful thing to add costs almost nothing: a tape measure.

Waist circumference captures what BMI structurally cannot, which is where the fat sits. Fat around the abdomen lies close to the organs and carries more metabolic risk than the same amount of fat on the hips and thighs. BMI cannot tell those two bodies apart. A tape measure can.

Measure at the top of the hip bone, just after breathing out normally, with the tape snug against the skin but not pressing into it.

After that, waist-to-hip ratio describes how fat is distributed, and a body fat estimate gets at composition directly. None of them replaces BMI. Together they answer the question BMI was never built to answer: what is this particular body made of, and where.

Common mistakes

Reading the category as a verdict. BMI knows two things about you. It does not know your blood pressure, your fitness, your bloodwork, or where your fat sits. Someone at 26 who trains regularly and tests well on everything else is in better health than someone at 23 who does none of it. The band is a prompt to look further, not a conclusion.

Comparing your BMI with someone else's. Two people at 24 can be built completely differently. The number compresses a whole body into one figure, and comparing compressed figures tells you nothing about what was compressed.

Tracking BMI while your body composition is changing. Gain muscle and lose fat at a similar rate and BMI will barely move, which reads as failure when it is the opposite. Track your waist, your strength and how your clothes fit alongside it.

Assuming a healthy BMI means there is nothing to check. Most criticism of BMI concerns people wrongly flagged as overweight. The quieter error runs the other way: low muscle and high abdominal fat can look fine on paper and is not.

Using the adult bands on a child. Under 20, BMI has to be read as a percentile for age and sex. Applying 18.5 and 25 to a 14-year-old produces a number that means nothing at all.

Key takeaways

• BMI is weight divided by height squared, and it was designed in the 1830s to describe populations rather than individuals.
• The CDC's healthy band for adults is 18.5 to under 25, and it is backed by mortality data covering 10.6 million people.
• It cannot separate muscle from fat. BMI classed 53.4% of NFL Combine athletes as obese; measured body fat classed 8.9%.
• The cut-offs do not transfer evenly. WHO added action points at 23.0 and 27.5 for Asian populations, where risk rises earlier.
• Add a waist measurement. Above 40 inches for men or 35 inches for women signals raised risk whatever BMI says.