Body mass index against all eight rows of the WHO adult table, each naming its source, with the pound and inch at their exact defined values and a note when the adult table does not apply.
Initializing in your browser…
Aspect ratios reduced exactly by the greatest common divisor, with the reduction multiplied back as a check, resizes that say whether they are exact, and named ratios that carry a note where the name is not the ratio.
Exact ages and date differences worked out on the calendar itself, with no arithmetic on milliseconds, so the answer is the same in every time zone. Shows which days are not 86,400 seconds long.
Exact arithmetic on rationals, with both evaluation conventions named and the precedence table the parser runs on printed on the page.
A BMI of 16.5. Which band is that, and how many bands does the published table have?
Input
weight 40 kg · height 1.56 m
Result
BMI 16.4 -> Moderate thinness (WHO TRS 894 table 2.1: 16.00 to 16.99)
The WHO adult table has eight rows: under 16, 16 to 17 and 17 to 18.5 are severe, moderate and mild thinness, and merging the middle two into a single "Underweight" loses a distinction the table draws deliberately. All eight are on the page and each one names the line it came from. The scale is drawn to the real widths of the bands rather than as equal segments, and the pound and inch are used at their exact defined values, 0.45359237 kg and 0.0254 m. The formula itself is one line; what a BMI calculator can get wrong is the units, the boundaries, and who the table applies to, which is why entering an age under 20 brings up a note that the reference there is a percentile for age and sex instead.
Body mass index against the WHO adult table, all eight bands, each naming the row it comes from. The arithmetic is trivial and exact; what a BMI calculator can get wrong is the units, the boundaries, and who the table applies to.
BMI is mass in kilograms divided by height in metres squared. That part is one line and hard to get wrong. Everything around it is not.
**The units are definitions, so they are exact here.** The international pound is exactly 0.45359237 kg and the international inch exactly 0.0254 m, both fixed by the 1959 international yard and pound agreement. A calculator holding a rounded 0.453592 is carrying a copy of a defined constant that is a part in a million out; there is no reason to.
**The WHO table has eight rows, not seven.** Under 16 is severe thinness, 16 to 17 is moderate thinness, 17 to 18.5 is mild thinness, 18.5 to 25 is the normal range, 25 to 30 is pre-obese, and 30 to 35, 35 to 40 and 40 upward are obese classes I, II and III. Merging the two middle thinness rows into a single "Underweight" loses a distinction the table draws deliberately. Every row here names the source line it came from, and the top band has no upper bound, so a BMI of 120 still lands somewhere.
**The scale is drawn to the real widths.** A bar of eight equal segments makes the normal range look like an eighth of the scale when it is 6.5 BMI points out of the 33 shown, and it puts the pointer in the wrong place inside its band. Here each band is as wide as it actually is.
**The table is an ADULT table, and the tool says when it does not apply.** For anyone under about 20 the reference is a percentile for age and sex, from the CDC growth charts or the WHO growth reference, and a fixed cut-off is simply the wrong instrument. The shipped version collected an age and ignored it entirely; the age here does not enter the formula, because it does not, but it does decide whether the table below means anything.
**And BMI is a population statistic.** Keys and colleagues named it in 1972 precisely to say it worked better across a group than for one person. It does not distinguish muscle from fat, it does not say where the mass is, and it scales with height squared while a body scales closer to height cubed, which is why it reads high for tall people and low for short ones. The ponderal index, the same measurement with the cube, is shown alongside for exactly that reason, along with BMI prime and the lower action points the 2004 WHO expert consultation identified for many Asian populations. Those are not a disclaimer bolted on: the arithmetic is exact and the answer is still not what most readers take it to be, which is the accuracy question for this tool.
BMI 22.9, in the normal range. The normal range at that height runs from 56.7 kg to 76.6 kg.
81.647 kg over 1.778 m squared is 25.83, which is pre-obese. Using a rounded pound of 0.453592 instead of the exact 0.45359237 changes it in the fifth decimal place; it is exact here because the constant is a definition.
Pre-obese. Every bound is inclusive at the bottom and exclusive at the top, which is what the published rows "18.50 to 24.99" and "25.00 to 29.99" mean once a value has more than two decimal places.
With the exact figure and the classification it falls under.
The whole table is on the page, with the source of each row.
Enter an age; under 20 it says the reference is a percentile instead.
It is a screening statistic for populations, not a diagnosis for a person. It does not distinguish muscle from fat, does not say where the mass is, and scales with height squared where a body scales closer to height cubed. Waist circumference and the waist to height ratio carry information it cannot, and both are easy to measure. A healthcare provider can measure what it misses.
18.5 up to but not including 25, in the WHO adult table. Below that the table splits into mild, moderate and severe thinness at 17 and 16; above it, pre-obese to 30 and then obese classes I, II and III at 30, 35 and 40.
The formula does not: it is mass over height squared for everyone. The interpretation does. For anyone under about 20 the reference is a percentile for age and sex rather than a fixed cut-off, so the adult table does not apply, and this tool says so when you enter an age under 20.
The 2004 WHO expert consultation found that the risk associated with a given BMI differs by population, and added public health action points at 23 and 27.5 for many Asian populations. It did not redefine the cut-off points, and both are shown here so the difference is visible rather than hidden.
Mass divided by height cubed, sometimes called Rohrer's index. It scales correctly with height where BMI does not, which is why BMI reads high for tall people and low for short ones. It is shown alongside so the two can be compared.
Exact. The international pound is 0.45359237 kg and the international inch 0.0254 m, both by definition since 1959, and those are the values used rather than rounded copies of them.
Every calculation runs locally in your browser. Your numbers and expressions are not transmitted or stored.