BMI Formula: How to Calculate Body Mass Index (Math + History)

Complete mathematical derivation of the BMI formula with Adolphe Quetelet’s 1832 origin story, metric and imperial calculations, three worked clinical examples, and the common calculation mistakes that distort results.

01The Mathematical Formula

Metric Formula (Standard)

The standard international BMI formula uses metric units. Body weight is expressed in kilograms (kg) and height in meters (m). The denominator squares the height value, producing units of square meters (m²). The complete formula reads:

A practical example: an adult weighing 70 kg (154 lbs) at 1.75 meters (5’9″) tall calculates BMI as 70 ÷ (1.75 × 1.75) = 70 ÷ 3.0625 = 22.86 kg/m². This value falls within the WHO Normal Weight classification range of 18.5 to 24.9.

Imperial Formula (US Standard)

Healthcare systems in the United States typically use imperial units — pounds for weight and inches for height. The imperial formula requires a conversion factor to maintain numerical equivalence with metric calculations:

The same individual measured in imperial units — 154 lb at 69 inches tall — calculates BMI as (154 ÷ 69²) × 703 = (154 ÷ 4761) × 703 = 0.03235 × 703 = 22.74. The minor discrepancy from the metric calculation (22.86 vs 22.74) reflects rounding in unit conversions.

Output Interpretation

The numerical result of either formula corresponds to standardized weight category thresholds established by the World Health Organization in 1995:

02Historical Origin — Adolphe Quetelet (1832)

Who Was Adolphe Quetelet?

Lambert Adolphe Jacques Quetelet (1796-1874) was a Belgian astronomer, mathematician, statistician, and sociologist. Trained at the University of Ghent and the Paris Observatory, Quetelet pioneered the application of statistical methods to human populations — a field he termed physique sociale (social physics). His work established the foundations of modern demographic research and biostatistics.

Quetelet was obsessed with finding mathematical regularities in human characteristics. He developed the concept of l’homme moyen (the average man), a statistical archetype representing population norms across physical, intellectual, and moral dimensions. The Quetelet Index emerged from this broader project to quantify the average human form.

The 1832 Publication

Quetelet’s foundational paper “Sur l’homme et le développement de ses facultés, ou Essai de physique sociale” (1832) established what would become BMI. He analyzed body measurements from European populations and tested various mathematical relationships between weight and height. After comparing weight scaling against multiple height exponents — h, h², h³ — Quetelet observed that weight correlated most consistently with the square of height in adult populations.

The original formulation in Quetelet’s notation differed slightly from modern BMI, but the mathematical structure remained identical: total body weight divided by stature squared. Quetelet derived this relationship empirically from population data, not theoretical biological principles.

Quetelet’s Own Warning

The most consequential aspect of Quetelet’s original work is the warning he embedded — and that the medical community subsequently ignored for nearly two centuries. Quetelet specifically cautioned that his index applied to population averages, not individual diagnosis. He noted that variations in muscle, bone, and overall composition would distort individual readings, making the formula unsuitable as a personal health metric.

Renaming to BMI (1972)

The Quetelet Index remained an obscure statistical tool for 140 years until American physiologist Ancel Keys revived it. In his landmark 1972 paper “Indices of relative weight and obesity” published in the Journal of Chronic Diseases, Keys evaluated various body weight indices against actual body composition data. He concluded that Quetelet’s formula performed best among simple ratios and renamed it “Body Mass Index” — the term used today.

Keys’s paper triggered widespread clinical adoption. The World Health Organization formally standardized BMI cutoffs in 1995, creating the now-familiar categories (Underweight, Normal, Overweight, Obese) that anchor global obesity surveillance. From a 19th-century demographic curiosity to a 21st-century clinical screening standard — the trajectory took 163 years.

03Why Height Is Squared (Not Cubed)

The Geometric Intuition

If humans were perfectly proportional three-dimensional objects, body weight would scale as height cubed (h³) — since volume scales as the cube of any linear dimension. A person twice as tall as another, geometrically scaled, would weigh eight times as much (2³ = 8). However, real human bodies do not maintain isometric proportions across heights.

Taller adults are not simply scaled-up versions of shorter adults. Their cross-sectional dimensions (limb circumference, torso width) do not increase proportionally with stature. Bone and muscle volume scales less aggressively than pure geometric prediction would suggest. The result: weight increases more slowly than h³ would predict but faster than h¹ (linear) would predict.

The Empirical Decision

Quetelet did not derive the squared exponent from biological theory. He tested the relationship statistically against population measurement data and selected the exponent producing the highest correlation coefficient. For adult European populations, h² produced approximately 0.7 to 0.8 correlation with body weight — substantially better than either linear or cubic scaling.

This empirical foundation creates one of BMI’s persistent vulnerabilities. The h² exponent fits average adult populations reasonably well but performs poorly for outliers — exceptionally tall individuals, children whose proportions differ developmentally, and athletes whose muscle mass deviates from population norms. The formula’s mathematical structure assumes a “standard” adult body that not everyone matches.

Alternative Indices Proposed

Researchers have proposed alternative indices to address BMI’s limitations. The Ponderal Index (weight/height³) better captures very tall individuals. The Body Adiposity Index (BAI) incorporates hip circumference rather than height alone. The Surface-based Body Shape Index (SBSI) adjusts for both height and waist circumference. None has achieved BMI’s institutional adoption, despite measurable accuracy advantages in specific populations.

04Three Worked Examples

The athletic male example illustrates BMI’s primary failure mode — the Athlete Paradox. A 82 kg male at 1.78 m with substantial muscle mass calculates as Overweight despite carrying minimal adipose tissue. The formula cannot distinguish lean mass from fat mass, producing misleading classification for muscular individuals.

The average female example represents the formula’s intended use case. A 62 kg female at 1.65 m calculates a BMI of 22.8 — squarely within the Normal Weight range and aligned with the BMI 22 to 24 mortality optimum identified in the Global BMI Mortality Collaboration’s 2016 Lancet meta-analysis of 10.6 million participants.

The senior male example demonstrates BMI’s blindness to body composition changes with aging. A 72-year-old male calculating BMI 23 may carry 28 to 32 percent body fat due to age-related sarcopenia — qualifying as obese by composition while appearing Normal Weight by BMI. The formula’s mathematical structure cannot detect this clinically significant pattern.

05Metric vs Imperial — Why the Conversion Factor 703?

Deriving the 703 Factor

To convert imperial BMI to metric-equivalent values, two unit conversions must occur. Weight in pounds must convert to kilograms (1 kg = 2.20462 lb). Height in inches must convert to meters (1 m = 39.3701 inches). Because the formula squares height, the inches-to-meters conversion factor must be squared as well.

The factor 703 (sometimes 703.07 for higher precision) makes the imperial formula numerically equivalent to the metric formula. Using this conversion, a person with metric BMI 22.86 will produce imperial BMI 22.86 (within rounding tolerance). Without the conversion, imperial calculation would yield approximately 0.0325 — a number with no clinical meaning.

Practical Conversion Reference

The minor discrepancy between metric (22.86) and imperial (22.74) results reflects rounding in the unit conversions, not formula error. For clinical purposes, both values produce identical WHO classification — Normal Weight.

06Common Calculation Mistakes

Mistake 1: Using Centimeters Instead of Meters

The metric formula requires height in meters, not centimeters. A 175 cm individual must enter 1.75 m, not 175. Using centimeters produces vastly wrong results: 70 ÷ 175² = 70 ÷ 30,625 = 0.00229 — a meaningless value. Always convert centimeters to meters by dividing by 100.

Mistake 2: Forgetting to Square the Height

Some calculators omit the squaring operation, dividing weight by height directly. For 70 kg at 1.75 m, this would produce 70 ÷ 1.75 = 40 — falsely indicating severe obesity. The squaring of height is mathematically essential, not optional.

Mistake 3: Mixing Imperial and Metric Units

Combining pounds with meters (or kilograms with inches) without proper conversion produces nonsense values. The formula requires consistent units within either metric or imperial systems. Mixed-unit calculations rank among the most common online calculator errors.

Mistake 4: Wrong Conversion Factor

Imperial BMI requires the 703 conversion factor — not 700, 705, or any other approximation. Using incorrect factors produces results that shift WHO category boundaries by 1 to 2 points. For borderline cases (BMI 24.5 to 25.5), this can change clinical classification from Normal to Overweight.

Mistake 5: Premature Rounding

Rounding intermediate values during calculation introduces cumulative error. Maintain full decimal precision through all calculation steps, then round only the final BMI value. Premature rounding of height² (3.0625 → 3.06) or pre-final BMI (22.857 → 22.9) can shift category placement near threshold boundaries.

07Mathematical Limitations of the Formula

Composition Blindness

The Quetelet formula operates exclusively on two scalar inputs: mass and height. It cannot capture body composition variables that determine actual health risk — visceral fat percentage, lean muscle mass, bone density, or hydration status. Two adults with identical BMI may present dramatically different metabolic profiles. The formula’s mathematical structure precludes composition assessment.

The h² Approximation

The squared height denominator works reasonably for average adult populations but fails at extremes. Very tall individuals (over 6’4″) tend to be misclassified upward (BMI overestimated) because human bodies don’t perfectly maintain h² scaling at extreme stature. Very short adults face the opposite bias. Children require completely different formula parameters via age-specific percentile charts.

Population vs Individual Validity

BMI was designed for population-level analysis — the formula’s mathematical structure works adequately when averaged across large groups but fails at the individual level. Quetelet himself warned against individual application. The formula correlates 0.7 to 0.8 with body fat at population scale but only 0.4 to 0.6 for individual prediction. Clinical use of BMI for personal diagnosis violates the formula’s original epistemological scope.

Skip the Manual Calculation — Use the Clinical Calculator

VisualBody Lab’s BMI calculator handles metric and imperial units, applies Asian-adjusted thresholds where appropriate, and presents results alongside complementary body composition metrics like waist-to-hip ratio and body fat percentage.

Calculate BMI in Context

Frequently Asked Questions