Guide
BMI: What It Measures and What It Misses
Body mass index is weight divided by height squared. It was designed to describe populations, not individuals, and it works for that purpose because it needs only two measurements and a multiplication. For a single person it is a screening signal, not a diagnosis: it cannot distinguish muscle from fat, cannot see where fat is stored, and says nothing about fitness, blood pressure or blood sugar.
Where the number comes from
The index is usually credited to the Belgian statistician Adolphe Quetelet, who in the 1830s was looking for a way to describe the "average man" across a population. It became a clinical shorthand in the twentieth century for a practical reason: an epidemiologist with a scale and a tape measure could classify thousands of people cheaply. Its virtue is that it is standardised and reproducible. Its vice is that standardised and reproducible is not the same as accurate for any one person.
Modern guidelines — from the World Health Organization and most national health services — treat BMI as a first-pass screen, to be followed by something more specific when it flags a question. They do not treat it as a diagnosis, and neither should you.
The arithmetic and the bands
BMI = weight (kg) / height (m)². Imperial input converts first: pounds to kilograms by dividing by 2.2046, and feet and inches to centimetres by multiplying total inches by 2.54. A 75 kg person at 178 cm scores 75 / 3.1684 = 23.7.
| BMI (adults) | WHO category | Common alternative thresholds |
|---|---|---|
| Below 18.5 | Underweight | Same |
| 18.5–24.9 | Normal weight | Some Asian guidelines use 18.5–22.9 |
| 25–29.9 | Overweight | 23–27.4 in several Asian countries |
| 30 and above | Obese | 27.5 and above in several Asian countries |
The alternative thresholds exist because cardiometabolic risk rises at a lower BMI in some populations, particularly South and East Asian. If your clinician uses 23 rather than 25, that is why — it is a calibration for population risk, not a stricter opinion.
What a ratio cannot see
BMI has three structural blind spots.
Composition. Muscle is denser than fat. A 90 kg, 180 cm weightlifter and a 90 kg, 180 cm sedentary person both score 27.8, and only one of them has an adiposity problem. Trained people routinely land in the "overweight" band at low body fat, which is why the index is close to useless for athletes.
Distribution. Fat stored around the internal organs (visceral fat) is associated with more metabolic risk than fat under the skin at the hips and thighs. Two people with the same BMI can carry their fat in different places, and the ratio has no way to tell them apart. Waist circumference is the cheap proxy for exactly this.
Scale. BMI divides weight by height squared, but body weight in a population scales closer to height cubed. The practical consequence is a mild bias: at the same body shape, taller people score slightly higher and shorter people slightly lower. It also means a 1 cm height-entry error moves the result, which matters when the measurement is self-reported.
Measurements that add context
- Waist circumference. Measured at the midpoint between the lowest rib and the top of the hip bone, after a normal exhale. Widely used thresholds are 94 cm for men and 80 cm for women of South Asian descent, with 102 cm and 88 cm as higher-risk cut-offs in other populations.
- Waist-to-height ratio. Waist divided by height, with 0.5 as a rough upper bound for adults. It is crude, but it accounts for frame size in a way BMI does not.
- Body-fat percentage. Measured properly — DEXA, hydrostatic weighing, or a clinician using calipers — separates the two components BMI merges. Consumer scales that "estimate" body fat from bioelectrical impedance are sensitive to hydration and best read as trends.
- Metabolic markers. Blood pressure, fasting glucose, HbA1c and a lipid panel say more about risk than any ratio of mass to height.
Where BMI is simply the wrong tool
Children and teenagers need age-and-sex percentile charts, because body composition changes with growth; an adult cut-off applied to a 12-year-old is meaningless. Pregnancy changes both weight and fluid distribution, and BMI is not used to assess it. People over about 70 lose muscle mass, so the same BMI can represent more fat than it did at 40. Amputees, people with significant oedema, and anyone with a very high muscle mass all break the index's assumptions.
Using it without misusing it
The reasonable use is as a trend line and a conversation starter. Track it quarterly at the same time of day, alongside waist measurement, and look at the direction rather than the decimal. The difference between 24.9 and 25.1 is one large glass of water, and treating that boundary as a verdict is a good way to make yourself anxious about nothing.
If the number concerns you, the next step is a measurement that separates fat from muscle, and a clinician who can look at the whole picture. The BMI calculator gives you the value, the band and the healthy-weight range for your height in both units; it deliberately does not tell you what to do about it.
FAQ
Is BMI accurate for athletes?
No. Muscle is denser than fat, so athletes commonly score in the overweight band at low body fat. A body-fat measurement, not a ratio of weight to height, is the right tool for that population.
What is a healthy BMI?
For adults, the WHO normal band is 18.5 to 24.9, which for a 178 cm person corresponds to roughly 58.6 to 78.9 kg. Some national guidelines use lower upper bounds, and the band is not a target to hit for its own sake.
Should children use this calculator?
No. Children and teenagers are assessed with age-and-sex BMI percentile charts, which is a different calculation with different thresholds. This page is for adults only.
Try it: metric or imperial, with the healthy weight range for your height. check your own BMI
Keep reading
Written by the CalcNest team. Last updated November 19, 2026.