Body Mass Index (BMI)

Body mass index is weight in kilograms divided by the square of height in metres. The formula dates to Quetelet in the 1830s as a description of populations, and the square is an empirical fit rather than a geometric law: isometric scaling would give a cube, and the exponent that best removes residual height dependence varies by population. Nothing in the calculation carries information about tissue type, so one value can describe very different bodies.

The World Health Organization cut-offs: under 18.5 underweight, 18.5 to 24.9 normal, 25 to 29.9 overweight, 30 or above obesity, subdivided into class I to 34.9, class II to 39.9 and class III at 40 and over. Lower thresholds of 23 and 25 are applied in several Asian populations because cardiometabolic risk appears at a lower index. Those numbers are not merely descriptive; the obesity indications for semaglutide and tirzepatide, and the entry criteria for the STEP and SURMOUNT trials, are written as an index of 30 or above, or 27 or above with a weight-related comorbidity.

That makes the index an administrative object as much as a biological one. A patient at 26.8 with dyslipidaemia and one at 27.1 differ by rounding, not by physiology, yet only one meets a label. Knowing the threshold is a convenient line rather than a risk discontinuity is what lets eligibility and risk be read as separate questions.

Where it fails is individual classification. A muscular athlete is routinely placed in the obesity category, an older person with sarcopenia can sit in the normal range with substantial fat mass, and the index says nothing about where fat is deposited. The American Medical Association adopted policy in 2023 calling the measure imperfect and recommending it be used alongside waist circumference and visceral fat measures. That is the right reading: a cheap first filter, not a diagnosis.