BMI Is a Population Statistic, Not a Diagnosis: What It Actually Measures
BMI was built two centuries ago to describe average body proportions across populations, not to assess any one individual's health — which explains most of its well-known blind spots.
Body Mass Index gets treated as a personal health verdict, but it wasn't designed for that job at all — understanding where it actually came from explains most of the criticism it gets today.
Where the formula comes from
BMI is simply weight divided by height squared (kg/m²). It was developed in the 1830s by Adolphe Quetelet, a Belgian astronomer and statistician, as part of a broader project measuring average human physical traits across populations — he was looking for a simple formula whose result stayed roughly consistent across people of different heights, for population-level statistics, not for evaluating any single person. The name "Quetelet Index" was only later rebranded "Body Mass Index" when epidemiologists adopted it in the 1970s to study obesity trends across large groups.
What it's actually good at
At a population level — tracking obesity trends across a country over decades, comparing broad health statistics between large groups — BMI works reasonably well precisely because it's cheap to calculate from two easy measurements, and individual-level quirks average out across a big enough sample. This is the context it was built for, and where it remains genuinely useful.
Where it breaks down for an individual
BMI only uses height and weight — it has no way to distinguish what the weight is made of. A pound of muscle and a pound of fat weigh exactly the same but take up different volume and have very different implications, and BMI can't tell them apart. This is the most commonly cited limitation: a heavily muscled athlete can register as "overweight" or "obese" by BMI despite very low body fat, because the formula has no muscle/fat distinction to draw on in the first place — it's not an edge case bug, it's a direct consequence of using only two numbers to describe something more complicated.
BMI also doesn't account for where fat is distributed on the body, which several studies suggest can matter for measuring certain health risks independent of overall weight — nor does it adjust for age, sex, or ethnicity, despite average body composition varying across all three, which is part of why some health organizations increasingly present BMI alongside other measurements (waist circumference, waist-to-hip ratio) rather than as a number that stands alone.
Why it's still everywhere
Despite the well-documented limitations, BMI has stuck around for the same reason it was invented: it requires nothing more than a scale and a tape measure, produces a single comparable number, and correlates reasonably well with health risk on average across large groups — which is exactly the population-statistics use case it was designed for. The problem isn't the arithmetic; it's applying a population-level tool as if it were a precise individual diagnosis.
Reading your own number
A BMI number is a starting data point, not a verdict — it's most useful in the context it was built for (broad population categories) and least reliable exactly where people tend to use it (as a complete individual health assessment). Spellkit's BMI Calculator computes BMI in metric or imperial units and shows the standard WHO weight categories alongside it, entirely in your browser, as a quick reference — not a medical assessment.